Original Research

ABSTRACT

Objective: To teach internal medicine residents quality improvement (QI) principles in an effort to improve resident knowledge and comfort with QI, as well as address quality care gaps in resident clinic primary care patient panels.

Design: A QI curriculum was implemented for all residents rotating through a primary care block over a 6-month period. Residents completed Institute for Healthcare Improvement (IHI) modules, participated in a QI workshop, and received panel data reports, ultimately completing a plan-do-study-act (PDSA) cycle to improve colorectal cancer screening and hypertension control.

Setting and participants: This project was undertaken at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. All internal medicine residents were included, with 55 (73%) of the 75 residents completing the presurvey, and 39 (52%) completing the postsurvey.

Measurements: We administered a 10-question pre- and postsurvey looking at resident attitudes toward and comfort with QI and familiarity with their panel data as well as measured rates of colorectal cancer screening and hypertension control in resident panels.

Results: There was an increase in the numbers of residents who performed a PDSA cycle (P = .002), completed outreach based on their panel data (P = .02), and felt comfortable in both creating aim statements and designing and implementing PDSA cycles (P < .0001). The residents’ knowledge of their panel data significantly increased. There was no significant improvement in hypertension control, but there was an increase in colorectal cancer screening rates (P < .0001).

Conclusion: Providing panel data and performing targeted QI interventions can improve resident comfort with QI, translating to improvement in patient outcomes.

Keywords: quality improvement, resident education, medical education, care gaps, quality metrics.

As quality improvement (QI) has become an integral part of clinical practice, residency training programs have continued to evolve in how best to teach QI. The Accreditation Council for Graduate Medical Education (ACGME) Common Program requirements mandate that core competencies in residency programs include practice-based learning and improvement and systems-based practice.¹ Residents should receive education in QI, receive data on quality metrics and benchmarks related to their patient population, and participate in QI activities. The Clinical Learning Environment Review (CLER) program was established to provide feedback to institutions on 6 focused areas, including patient safety and health care quality. In visits to institutions across the United States, the CLER committees found that many residents had limited knowledge of QI concepts and limited access to data on quality metrics and benchmarks.²

There are many barriers to implementing a QI curriculum in residency programs, and creating and maintaining successful strategies has proven challenging.³ Many QI curricula for internal medicine residents have been described in the literature, but the results of many of these studies focus on resident self-assessment of QI knowledge and numbers of projects rather than on patient outcomes.^4-13 As there is some evidence suggesting that patients treated by residents have worse outcomes on ambulatory quality measures when compared with patients treated by staff physicians,^14,15 it is important to also look at patient outcomes when evaluating a QI curriculum. Experts in education recommend the following to optimize learning: exposure to both didactic and experiential opportunities, connection to health system improvement efforts, and assessment of patient outcomes in addition to learner feedback.^16,17 A study also found that providing panel data to residents could improve quality metrics.¹⁸

In this study, we sought to investigate the effects of a resident QI intervention during an ambulatory block on both residents’ self-assessments of QI knowledge and attitudes as well as on patient quality metrics.

Methods

Curriculum

We implemented this educational initiative at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. Co-located with the 415-bed academic medical center in downtown Boston, the practice serves more than 40,000 patients, approximately 7000 of whom are cared for by resident primary care physicians (PCPs). The internal medicine residents rotate through the primary care clinic as part of continuity clinic during ambulatory or elective blocks. In addition to continuity clinic, the residents have 2 dedicated 3-week primary care rotations during the course of an academic year. Primary care rotations consist of 5 clinic sessions a week as well as structured teaching sessions. Each resident inherits a panel of patients from an outgoing senior resident, with an average panel size of 96 patients per resident.

Prior to this study intervention, we did not do any formal QI teaching to our residents as part of their primary care curriculum, and previous panel management had focused more on chart reviews of patients whom residents perceived to be higher risk. Residents from all 3 years were included in the intervention. We taught a QI curriculum to our residents from January 2018 to June 2018 during the 3-week primary care rotation, which consisted of the following components:

• Institute for Healthcare Improvement (IHI) module QI 102 completed independently online.
• A 2-hour QI workshop led by 1 of 2 primary care faculty with backgrounds in QI, during which residents were taught basic principles of QI, including how to craft aim statements and design plan-do-study-act (PDSA) cycles, and participated in a hands-on QI activity designed to model rapid cycle improvement (the Paper Airplane Factory¹⁹).
• Distribution of individualized reports of residents’ patient panel data by email at the start of the primary care block that detailed patients’ overall rates of colorectal cancer screening and hypertension (HTN) control, along with the average resident panel rates and the average attending panel rates. The reports also included a list of all residents’ patients who were overdue for colorectal cancer screening or whose last blood pressure (BP) was uncontrolled (systolic BP ≥ 140 mm Hg or diastolic BP ≥  90 mm Hg). These reports were originally designed by our practice’s QI team and run and exported in Microsoft Excel format monthly by our information technology (IT) administrator.
• Instruction on aim statements as a group, followed by the expectation that each resident create an individualized aim statement tailored to each resident’s patient panel rates, with the PDSA cycle to be implemented during the remainder of the primary care rotation, focusing on improvement of colorectal cancer screening and HTN control (see supplementary eFigure 1 online for the worksheet used for the workshop).
• Residents were held accountable for their interventions by various check-ins. At the end of the primary care block, residents were required to submit their completed worksheets showing the intervention they had undertaken and when it was performed. The 2 primary care attendings primarily responsible for QI education would review the resident’s work approximately 1 to 2 months after they submitted their worksheets describing their intervention. These attendings sent the residents personalized feedback based on whether the intervention had been completed or successful as evidenced by documentation in the chart, including direct patient outreach by phone, letter, or portal; outreach to the resident coordinator; scheduled follow-up appointment; or booking or completion of colorectal cancer screening. Along with this feedback, residents were also sent suggestions for next steps. Resident preceptors were copied on the email to facilitate reinforcement of the goals and plans. Finally, the resident preceptors also helped with accountability by going through the residents’ worksheets and patient panel metrics with the residents during biannual evaluations.

Evaluation

Residents were surveyed with a 10-item questionnaire pre and post intervention regarding their attitudes toward QI, understanding of QI principles, and familiarity with their patient panel data. Surveys were anonymous and distributed via the SurveyMonkey platform (see supplementary eFigure 2 online). Residents were asked if they had ever performed a PDSA cycle, performed patient outreach, or performed an intervention and whether they knew the rates of diabetes, HTN, and colorectal cancer screening in their patient panels. Questions rated on a 5-point Likert scale were used to assess comfort with panel management, developing an aim statement, designing and implementing a PDSA cycle, as well as interest in pursuing QI as a career. For the purposes of analysis, these questions were dichotomized into “somewhat comfortable” and “very comfortable” vs “neutral,” “somewhat uncomfortable,” and “very uncomfortable.” Similarly, we dichotomized the question about interest in QI as a career into “somewhat interested” and “very interested” vs “neutral,” “somewhat disinterested,” and “very disinterested.” As the surveys were anonymous, we were unable to pair the pre- and postintervention surveys and used a chi-square test to evaluate whether there was an association between survey assessments pre intervention vs post intervention and a positive or negative response to the question.

We also examined rates of HTN control and colorectal cancer screening in all 75 resident panels pre and post intervention. The paired t-test was used to determine whether the mean change from pre to post intervention was significant. SAS 9.4 (SAS Institute Inc.) was used for all analyses. Institutional Review Board exemption was obtained from the Tufts Medical Center IRB. There was no funding received for this study.

Respondents

Of the 75 residents, 55 (73%) completed the survey prior to the intervention, and 39 (52%) completed the survey after the intervention.

Panel Knowledge and Intervention

Prior to the intervention, 45% of residents had performed a PDSA cycle, compared with 77% post intervention, which was a significant increase (P = .002) (Table 1). Sixty-two percent of residents had performed outreach or an intervention based on their patient panel reports prior to the intervention, compared with 85% of residents post intervention, which was also a significant increase (P = .02). The increase post intervention was not 100%, as there were residents who either missed the initial workshop or who did not follow through with their planned intervention. Common interventions included the residents giving their coordinators a list of patients to call to schedule appointments, utilizing fellow team members (eg, pharmacists, social workers) for targeted patient outreach, or calling patients themselves to reestablish a connection.

In terms of knowledge of their patient panels, prior to the intervention, 55%, 62%, and 62% of residents knew the rates of patients in their panel with diabetes, HTN, and colorectal cancer screening, respectively. After the intervention, the residents’ knowledge of these rates increased significantly, to 85% for diabetes (P = .002), 97% for HTN (P < .0001), and 97% for colorectal cancer screening (P < .0001).

Comfort With QI Approaches

Prior to the intervention, 82% of residents were comfortable managing their primary care panel, which did not change significantly post intervention (Table 2). The residents’ comfort with designing an aim statement did significantly increase, from 55% to 95% (P < .0001). The residents also had a significant increase in comfort with both designing and implementing a PDSA cycle. Prior to the intervention, 22% felt comfortable designing a PDSA cycle, which increased to 79% (P < .0001) post intervention, and 24% felt comfortable implementing a PDSA cycle, which increased to 77% (P < .0001) post intervention.

Patient Outcome Measures

The rate of HTN control in the residents' patient panels did not change significantly pre and post intervention (Table 3). The rate of resident patients who were up to date with colorectal cancer screening increased by 6.5% post intervention (P < .0001).

Interest in QI as a Career

As part of the survey, residents were asked how interested they were in making QI a part of their career. Fifty percent of residents indicated an interest in QI pre intervention, and 54% indicated an interest post intervention, which was not a significant difference (P = .72).

Discussion

In this study, we found that integration of a QI curriculum into a primary care rotation improved both residents’ knowledge of their patient panels and comfort with QI approaches, which translated to improvement in patient outcomes. Several previous studies have found improvements in resident self-assessment or knowledge after implementation of a QI curriculum.^4-13 Liao et al implemented a longitudinal curriculum including both didactic and experiential components and found an improvement in both QI confidence and knowledge.³ Similarly, Duello et al⁸ found that a curriculum including both didactic lectures and QI projects improved subjective QI knowledge and comfort. Interestingly, Fok and Wong⁹ found that resident knowledge could be sustained post curriculum after completion of a QI project, suggesting that experiential learning may be helpful in maintaining knowledge.

Studies also have looked at providing performance data to residents. Hwang et al¹⁸ found that providing audit and feedback in the form of individual panel performance data to residents compared with practice targets led to statistically significant improvement in cancer screening rates and composite quality score, indicating that there is tremendous potential in providing residents with their data. While the ACGME mandates that residents should receive data on their quality metrics, on CLER visits, many residents interviewed noted limited access to data on their metrics and benchmarks.^1,2

Though previous studies have individually looked at teaching QI concepts, providing panel data, or targeting select metrics, our study was unique in that it reviewed both self-reported resident outcomes data as well as actual patient outcomes. In addition to finding increased knowledge of patient panels and comfort with QI approaches, we found a significant increase in colorectal cancer screening rates post intervention. We thought this finding was particularly important given some data that residents' patients have been found to have worse outcomes on quality metrics compared with patients cared for by staff physicians.^14,15 Given that having a resident physician as a PCP has been associated with failing to meet quality measures, it is especially important to focus targeted quality improvement initiatives in this patient population to reduce disparities in care.

We found that residents had improved knowledge on their patient panels as a result of this initiative. The residents were noted to have a higher knowledge of their HTN and colorectal cancer screening rates in comparison to their diabetes metrics. We suspect this is because residents are provided with multiple metrics related to diabetes, including process measures such as A1c testing, as well as outcome measures such as A1c control, so it may be harder for them to elucidate exactly how they are doing with their diabetes patients, whereas in HTN control and colorectal cancer screening, there is only 1 associated metric. Interestingly, even though HTN and colorectal cancer screening were the 2 measures focused on in the study, the residents had a significant improvement in knowledge of the rates of diabetes in their panel as well. This suggests that even just receiving data alone is valuable, hopefully translating to better outcomes with better baseline understanding of panels. We believe that our intervention was successful because it included both a didactic and an experiential component, as well as the use of individual panel performance data.

There were several limitations to our study. It was performed at a single institution, translating to a small sample size. Our data analysis was limited because we were unable to pair our pre- and postintervention survey responses because we used an anonymous survey. We also did not have full participation in postintervention surveys from all residents, which may have biased the study in favor of high performers. Another limitation was that our survey relied on self-reported outcomes for the questions about the residents knowing their patient panels.

This study required a 2-hour workshop every 3 weeks led by a faculty member trained in QI. Given the amount of time needed for the curriculum, this study may be difficult to replicate at other institutions, especially if faculty with an interest or training in QI are not available. Given our finding that residents had increased knowledge of their patient panels after receiving panel metrics, simply providing data with the goal of smaller, focused interventions may be easier to implement. At our institution, we discontinued the longer 2-hour QI workshops designed to teach QI approaches more broadly. We continue to provide individualized panel data to all residents during their primary care rotations and conduct half-hour, small group workshops with the interns that focus on drafting aim statements and planning interventions. All residents are required to submit worksheets to us at the end of their primary care blocks listing their current rates of each predetermined metric and laying out their aim statements and planned interventions. Residents also continue to receive feedback from our faculty with expertise in QI afterward on their plans and evidence of follow-through in the chart, with their preceptors included on the feedback emails. Even without the larger QI workshop, this approach has continued to be successful and appreciated. In fact, it does appear as though improvement in colorectal cancer screening has been sustained over several years. At the end of our study period, the resident patient colorectal cancer screening rate rose from 34% to 43%, and for the 2021-2022 academic year, the rate rose further, from 46% to 50%.

Given that the resident clinic patient population is at higher risk overall, targeted outreach and approaches to improve quality must be continued. Future areas of research include looking at which interventions, whether QI curriculum, provision of panel data, or required panel management interventions, translate to the greatest improvements in patient outcomes in this vulnerable population.

Conclusion

Our study showed that a dedicated QI curriculum for the residents and access to quality metric data improved both resident knowledge and comfort with QI approaches. Beyond resident-centered outcomes, there was also translation to improved patient outcomes, with a significant increase in colon cancer screening rates post intervention.

Corresponding author: Kinjalika Sathi, MD, 800 Washington St., Boston, MA 02111; ksathi@tuftsmedicalcenter.org

Disclosures: None reported.

ABSTRACT

Objective: To teach internal medicine residents quality improvement (QI) principles in an effort to improve resident knowledge and comfort with QI, as well as address quality care gaps in resident clinic primary care patient panels.

Design: A QI curriculum was implemented for all residents rotating through a primary care block over a 6-month period. Residents completed Institute for Healthcare Improvement (IHI) modules, participated in a QI workshop, and received panel data reports, ultimately completing a plan-do-study-act (PDSA) cycle to improve colorectal cancer screening and hypertension control.

Setting and participants: This project was undertaken at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. All internal medicine residents were included, with 55 (73%) of the 75 residents completing the presurvey, and 39 (52%) completing the postsurvey.

Measurements: We administered a 10-question pre- and postsurvey looking at resident attitudes toward and comfort with QI and familiarity with their panel data as well as measured rates of colorectal cancer screening and hypertension control in resident panels.

Results: There was an increase in the numbers of residents who performed a PDSA cycle (P = .002), completed outreach based on their panel data (P = .02), and felt comfortable in both creating aim statements and designing and implementing PDSA cycles (P < .0001). The residents’ knowledge of their panel data significantly increased. There was no significant improvement in hypertension control, but there was an increase in colorectal cancer screening rates (P < .0001).

Conclusion: Providing panel data and performing targeted QI interventions can improve resident comfort with QI, translating to improvement in patient outcomes.

Keywords: quality improvement, resident education, medical education, care gaps, quality metrics.

As quality improvement (QI) has become an integral part of clinical practice, residency training programs have continued to evolve in how best to teach QI. The Accreditation Council for Graduate Medical Education (ACGME) Common Program requirements mandate that core competencies in residency programs include practice-based learning and improvement and systems-based practice.¹ Residents should receive education in QI, receive data on quality metrics and benchmarks related to their patient population, and participate in QI activities. The Clinical Learning Environment Review (CLER) program was established to provide feedback to institutions on 6 focused areas, including patient safety and health care quality. In visits to institutions across the United States, the CLER committees found that many residents had limited knowledge of QI concepts and limited access to data on quality metrics and benchmarks.²

There are many barriers to implementing a QI curriculum in residency programs, and creating and maintaining successful strategies has proven challenging.³ Many QI curricula for internal medicine residents have been described in the literature, but the results of many of these studies focus on resident self-assessment of QI knowledge and numbers of projects rather than on patient outcomes.^4-13 As there is some evidence suggesting that patients treated by residents have worse outcomes on ambulatory quality measures when compared with patients treated by staff physicians,^14,15 it is important to also look at patient outcomes when evaluating a QI curriculum. Experts in education recommend the following to optimize learning: exposure to both didactic and experiential opportunities, connection to health system improvement efforts, and assessment of patient outcomes in addition to learner feedback.^16,17 A study also found that providing panel data to residents could improve quality metrics.¹⁸

In this study, we sought to investigate the effects of a resident QI intervention during an ambulatory block on both residents’ self-assessments of QI knowledge and attitudes as well as on patient quality metrics.

Methods

Curriculum

We implemented this educational initiative at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. Co-located with the 415-bed academic medical center in downtown Boston, the practice serves more than 40,000 patients, approximately 7000 of whom are cared for by resident primary care physicians (PCPs). The internal medicine residents rotate through the primary care clinic as part of continuity clinic during ambulatory or elective blocks. In addition to continuity clinic, the residents have 2 dedicated 3-week primary care rotations during the course of an academic year. Primary care rotations consist of 5 clinic sessions a week as well as structured teaching sessions. Each resident inherits a panel of patients from an outgoing senior resident, with an average panel size of 96 patients per resident.

Prior to this study intervention, we did not do any formal QI teaching to our residents as part of their primary care curriculum, and previous panel management had focused more on chart reviews of patients whom residents perceived to be higher risk. Residents from all 3 years were included in the intervention. We taught a QI curriculum to our residents from January 2018 to June 2018 during the 3-week primary care rotation, which consisted of the following components:

• Institute for Healthcare Improvement (IHI) module QI 102 completed independently online.
• A 2-hour QI workshop led by 1 of 2 primary care faculty with backgrounds in QI, during which residents were taught basic principles of QI, including how to craft aim statements and design plan-do-study-act (PDSA) cycles, and participated in a hands-on QI activity designed to model rapid cycle improvement (the Paper Airplane Factory¹⁹).
• Distribution of individualized reports of residents’ patient panel data by email at the start of the primary care block that detailed patients’ overall rates of colorectal cancer screening and hypertension (HTN) control, along with the average resident panel rates and the average attending panel rates. The reports also included a list of all residents’ patients who were overdue for colorectal cancer screening or whose last blood pressure (BP) was uncontrolled (systolic BP ≥ 140 mm Hg or diastolic BP ≥  90 mm Hg). These reports were originally designed by our practice’s QI team and run and exported in Microsoft Excel format monthly by our information technology (IT) administrator.
• Instruction on aim statements as a group, followed by the expectation that each resident create an individualized aim statement tailored to each resident’s patient panel rates, with the PDSA cycle to be implemented during the remainder of the primary care rotation, focusing on improvement of colorectal cancer screening and HTN control (see supplementary eFigure 1 online for the worksheet used for the workshop).
• Residents were held accountable for their interventions by various check-ins. At the end of the primary care block, residents were required to submit their completed worksheets showing the intervention they had undertaken and when it was performed. The 2 primary care attendings primarily responsible for QI education would review the resident’s work approximately 1 to 2 months after they submitted their worksheets describing their intervention. These attendings sent the residents personalized feedback based on whether the intervention had been completed or successful as evidenced by documentation in the chart, including direct patient outreach by phone, letter, or portal; outreach to the resident coordinator; scheduled follow-up appointment; or booking or completion of colorectal cancer screening. Along with this feedback, residents were also sent suggestions for next steps. Resident preceptors were copied on the email to facilitate reinforcement of the goals and plans. Finally, the resident preceptors also helped with accountability by going through the residents’ worksheets and patient panel metrics with the residents during biannual evaluations.

Evaluation

Residents were surveyed with a 10-item questionnaire pre and post intervention regarding their attitudes toward QI, understanding of QI principles, and familiarity with their patient panel data. Surveys were anonymous and distributed via the SurveyMonkey platform (see supplementary eFigure 2 online). Residents were asked if they had ever performed a PDSA cycle, performed patient outreach, or performed an intervention and whether they knew the rates of diabetes, HTN, and colorectal cancer screening in their patient panels. Questions rated on a 5-point Likert scale were used to assess comfort with panel management, developing an aim statement, designing and implementing a PDSA cycle, as well as interest in pursuing QI as a career. For the purposes of analysis, these questions were dichotomized into “somewhat comfortable” and “very comfortable” vs “neutral,” “somewhat uncomfortable,” and “very uncomfortable.” Similarly, we dichotomized the question about interest in QI as a career into “somewhat interested” and “very interested” vs “neutral,” “somewhat disinterested,” and “very disinterested.” As the surveys were anonymous, we were unable to pair the pre- and postintervention surveys and used a chi-square test to evaluate whether there was an association between survey assessments pre intervention vs post intervention and a positive or negative response to the question.

We also examined rates of HTN control and colorectal cancer screening in all 75 resident panels pre and post intervention. The paired t-test was used to determine whether the mean change from pre to post intervention was significant. SAS 9.4 (SAS Institute Inc.) was used for all analyses. Institutional Review Board exemption was obtained from the Tufts Medical Center IRB. There was no funding received for this study.

Respondents

Of the 75 residents, 55 (73%) completed the survey prior to the intervention, and 39 (52%) completed the survey after the intervention.

Panel Knowledge and Intervention

Prior to the intervention, 45% of residents had performed a PDSA cycle, compared with 77% post intervention, which was a significant increase (P = .002) (Table 1). Sixty-two percent of residents had performed outreach or an intervention based on their patient panel reports prior to the intervention, compared with 85% of residents post intervention, which was also a significant increase (P = .02). The increase post intervention was not 100%, as there were residents who either missed the initial workshop or who did not follow through with their planned intervention. Common interventions included the residents giving their coordinators a list of patients to call to schedule appointments, utilizing fellow team members (eg, pharmacists, social workers) for targeted patient outreach, or calling patients themselves to reestablish a connection.

In terms of knowledge of their patient panels, prior to the intervention, 55%, 62%, and 62% of residents knew the rates of patients in their panel with diabetes, HTN, and colorectal cancer screening, respectively. After the intervention, the residents’ knowledge of these rates increased significantly, to 85% for diabetes (P = .002), 97% for HTN (P < .0001), and 97% for colorectal cancer screening (P < .0001).

Comfort With QI Approaches

Prior to the intervention, 82% of residents were comfortable managing their primary care panel, which did not change significantly post intervention (Table 2). The residents’ comfort with designing an aim statement did significantly increase, from 55% to 95% (P < .0001). The residents also had a significant increase in comfort with both designing and implementing a PDSA cycle. Prior to the intervention, 22% felt comfortable designing a PDSA cycle, which increased to 79% (P < .0001) post intervention, and 24% felt comfortable implementing a PDSA cycle, which increased to 77% (P < .0001) post intervention.

Patient Outcome Measures

The rate of HTN control in the residents' patient panels did not change significantly pre and post intervention (Table 3). The rate of resident patients who were up to date with colorectal cancer screening increased by 6.5% post intervention (P < .0001).

Interest in QI as a Career

As part of the survey, residents were asked how interested they were in making QI a part of their career. Fifty percent of residents indicated an interest in QI pre intervention, and 54% indicated an interest post intervention, which was not a significant difference (P = .72).

Discussion

In this study, we found that integration of a QI curriculum into a primary care rotation improved both residents’ knowledge of their patient panels and comfort with QI approaches, which translated to improvement in patient outcomes. Several previous studies have found improvements in resident self-assessment or knowledge after implementation of a QI curriculum.^4-13 Liao et al implemented a longitudinal curriculum including both didactic and experiential components and found an improvement in both QI confidence and knowledge.³ Similarly, Duello et al⁸ found that a curriculum including both didactic lectures and QI projects improved subjective QI knowledge and comfort. Interestingly, Fok and Wong⁹ found that resident knowledge could be sustained post curriculum after completion of a QI project, suggesting that experiential learning may be helpful in maintaining knowledge.

Studies also have looked at providing performance data to residents. Hwang et al¹⁸ found that providing audit and feedback in the form of individual panel performance data to residents compared with practice targets led to statistically significant improvement in cancer screening rates and composite quality score, indicating that there is tremendous potential in providing residents with their data. While the ACGME mandates that residents should receive data on their quality metrics, on CLER visits, many residents interviewed noted limited access to data on their metrics and benchmarks.^1,2

Though previous studies have individually looked at teaching QI concepts, providing panel data, or targeting select metrics, our study was unique in that it reviewed both self-reported resident outcomes data as well as actual patient outcomes. In addition to finding increased knowledge of patient panels and comfort with QI approaches, we found a significant increase in colorectal cancer screening rates post intervention. We thought this finding was particularly important given some data that residents' patients have been found to have worse outcomes on quality metrics compared with patients cared for by staff physicians.^14,15 Given that having a resident physician as a PCP has been associated with failing to meet quality measures, it is especially important to focus targeted quality improvement initiatives in this patient population to reduce disparities in care.

We found that residents had improved knowledge on their patient panels as a result of this initiative. The residents were noted to have a higher knowledge of their HTN and colorectal cancer screening rates in comparison to their diabetes metrics. We suspect this is because residents are provided with multiple metrics related to diabetes, including process measures such as A1c testing, as well as outcome measures such as A1c control, so it may be harder for them to elucidate exactly how they are doing with their diabetes patients, whereas in HTN control and colorectal cancer screening, there is only 1 associated metric. Interestingly, even though HTN and colorectal cancer screening were the 2 measures focused on in the study, the residents had a significant improvement in knowledge of the rates of diabetes in their panel as well. This suggests that even just receiving data alone is valuable, hopefully translating to better outcomes with better baseline understanding of panels. We believe that our intervention was successful because it included both a didactic and an experiential component, as well as the use of individual panel performance data.

There were several limitations to our study. It was performed at a single institution, translating to a small sample size. Our data analysis was limited because we were unable to pair our pre- and postintervention survey responses because we used an anonymous survey. We also did not have full participation in postintervention surveys from all residents, which may have biased the study in favor of high performers. Another limitation was that our survey relied on self-reported outcomes for the questions about the residents knowing their patient panels.

This study required a 2-hour workshop every 3 weeks led by a faculty member trained in QI. Given the amount of time needed for the curriculum, this study may be difficult to replicate at other institutions, especially if faculty with an interest or training in QI are not available. Given our finding that residents had increased knowledge of their patient panels after receiving panel metrics, simply providing data with the goal of smaller, focused interventions may be easier to implement. At our institution, we discontinued the longer 2-hour QI workshops designed to teach QI approaches more broadly. We continue to provide individualized panel data to all residents during their primary care rotations and conduct half-hour, small group workshops with the interns that focus on drafting aim statements and planning interventions. All residents are required to submit worksheets to us at the end of their primary care blocks listing their current rates of each predetermined metric and laying out their aim statements and planned interventions. Residents also continue to receive feedback from our faculty with expertise in QI afterward on their plans and evidence of follow-through in the chart, with their preceptors included on the feedback emails. Even without the larger QI workshop, this approach has continued to be successful and appreciated. In fact, it does appear as though improvement in colorectal cancer screening has been sustained over several years. At the end of our study period, the resident patient colorectal cancer screening rate rose from 34% to 43%, and for the 2021-2022 academic year, the rate rose further, from 46% to 50%.

Given that the resident clinic patient population is at higher risk overall, targeted outreach and approaches to improve quality must be continued. Future areas of research include looking at which interventions, whether QI curriculum, provision of panel data, or required panel management interventions, translate to the greatest improvements in patient outcomes in this vulnerable population.

Conclusion

Our study showed that a dedicated QI curriculum for the residents and access to quality metric data improved both resident knowledge and comfort with QI approaches. Beyond resident-centered outcomes, there was also translation to improved patient outcomes, with a significant increase in colon cancer screening rates post intervention.

Corresponding author: Kinjalika Sathi, MD, 800 Washington St., Boston, MA 02111; ksathi@tuftsmedicalcenter.org

Disclosures: None reported.

ABSTRACT

Objective: To teach internal medicine residents quality improvement (QI) principles in an effort to improve resident knowledge and comfort with QI, as well as address quality care gaps in resident clinic primary care patient panels.

Design: A QI curriculum was implemented for all residents rotating through a primary care block over a 6-month period. Residents completed Institute for Healthcare Improvement (IHI) modules, participated in a QI workshop, and received panel data reports, ultimately completing a plan-do-study-act (PDSA) cycle to improve colorectal cancer screening and hypertension control.

Setting and participants: This project was undertaken at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. All internal medicine residents were included, with 55 (73%) of the 75 residents completing the presurvey, and 39 (52%) completing the postsurvey.

Measurements: We administered a 10-question pre- and postsurvey looking at resident attitudes toward and comfort with QI and familiarity with their panel data as well as measured rates of colorectal cancer screening and hypertension control in resident panels.

Results: There was an increase in the numbers of residents who performed a PDSA cycle (P = .002), completed outreach based on their panel data (P = .02), and felt comfortable in both creating aim statements and designing and implementing PDSA cycles (P < .0001). The residents’ knowledge of their panel data significantly increased. There was no significant improvement in hypertension control, but there was an increase in colorectal cancer screening rates (P < .0001).

Conclusion: Providing panel data and performing targeted QI interventions can improve resident comfort with QI, translating to improvement in patient outcomes.

Keywords: quality improvement, resident education, medical education, care gaps, quality metrics.

As quality improvement (QI) has become an integral part of clinical practice, residency training programs have continued to evolve in how best to teach QI. The Accreditation Council for Graduate Medical Education (ACGME) Common Program requirements mandate that core competencies in residency programs include practice-based learning and improvement and systems-based practice.¹ Residents should receive education in QI, receive data on quality metrics and benchmarks related to their patient population, and participate in QI activities. The Clinical Learning Environment Review (CLER) program was established to provide feedback to institutions on 6 focused areas, including patient safety and health care quality. In visits to institutions across the United States, the CLER committees found that many residents had limited knowledge of QI concepts and limited access to data on quality metrics and benchmarks.²

There are many barriers to implementing a QI curriculum in residency programs, and creating and maintaining successful strategies has proven challenging.³ Many QI curricula for internal medicine residents have been described in the literature, but the results of many of these studies focus on resident self-assessment of QI knowledge and numbers of projects rather than on patient outcomes.^4-13 As there is some evidence suggesting that patients treated by residents have worse outcomes on ambulatory quality measures when compared with patients treated by staff physicians,^14,15 it is important to also look at patient outcomes when evaluating a QI curriculum. Experts in education recommend the following to optimize learning: exposure to both didactic and experiential opportunities, connection to health system improvement efforts, and assessment of patient outcomes in addition to learner feedback.^16,17 A study also found that providing panel data to residents could improve quality metrics.¹⁸

In this study, we sought to investigate the effects of a resident QI intervention during an ambulatory block on both residents’ self-assessments of QI knowledge and attitudes as well as on patient quality metrics.

Methods

Curriculum

We implemented this educational initiative at Tufts Medical Center Primary Care, Boston, Massachusetts, the primary care teaching practice for all 75 internal medicine residents at Tufts Medical Center. Co-located with the 415-bed academic medical center in downtown Boston, the practice serves more than 40,000 patients, approximately 7000 of whom are cared for by resident primary care physicians (PCPs). The internal medicine residents rotate through the primary care clinic as part of continuity clinic during ambulatory or elective blocks. In addition to continuity clinic, the residents have 2 dedicated 3-week primary care rotations during the course of an academic year. Primary care rotations consist of 5 clinic sessions a week as well as structured teaching sessions. Each resident inherits a panel of patients from an outgoing senior resident, with an average panel size of 96 patients per resident.

Prior to this study intervention, we did not do any formal QI teaching to our residents as part of their primary care curriculum, and previous panel management had focused more on chart reviews of patients whom residents perceived to be higher risk. Residents from all 3 years were included in the intervention. We taught a QI curriculum to our residents from January 2018 to June 2018 during the 3-week primary care rotation, which consisted of the following components:

• Institute for Healthcare Improvement (IHI) module QI 102 completed independently online.
• A 2-hour QI workshop led by 1 of 2 primary care faculty with backgrounds in QI, during which residents were taught basic principles of QI, including how to craft aim statements and design plan-do-study-act (PDSA) cycles, and participated in a hands-on QI activity designed to model rapid cycle improvement (the Paper Airplane Factory¹⁹).
• Distribution of individualized reports of residents’ patient panel data by email at the start of the primary care block that detailed patients’ overall rates of colorectal cancer screening and hypertension (HTN) control, along with the average resident panel rates and the average attending panel rates. The reports also included a list of all residents’ patients who were overdue for colorectal cancer screening or whose last blood pressure (BP) was uncontrolled (systolic BP ≥ 140 mm Hg or diastolic BP ≥  90 mm Hg). These reports were originally designed by our practice’s QI team and run and exported in Microsoft Excel format monthly by our information technology (IT) administrator.
• Instruction on aim statements as a group, followed by the expectation that each resident create an individualized aim statement tailored to each resident’s patient panel rates, with the PDSA cycle to be implemented during the remainder of the primary care rotation, focusing on improvement of colorectal cancer screening and HTN control (see supplementary eFigure 1 online for the worksheet used for the workshop).
• Residents were held accountable for their interventions by various check-ins. At the end of the primary care block, residents were required to submit their completed worksheets showing the intervention they had undertaken and when it was performed. The 2 primary care attendings primarily responsible for QI education would review the resident’s work approximately 1 to 2 months after they submitted their worksheets describing their intervention. These attendings sent the residents personalized feedback based on whether the intervention had been completed or successful as evidenced by documentation in the chart, including direct patient outreach by phone, letter, or portal; outreach to the resident coordinator; scheduled follow-up appointment; or booking or completion of colorectal cancer screening. Along with this feedback, residents were also sent suggestions for next steps. Resident preceptors were copied on the email to facilitate reinforcement of the goals and plans. Finally, the resident preceptors also helped with accountability by going through the residents’ worksheets and patient panel metrics with the residents during biannual evaluations.

Evaluation

Residents were surveyed with a 10-item questionnaire pre and post intervention regarding their attitudes toward QI, understanding of QI principles, and familiarity with their patient panel data. Surveys were anonymous and distributed via the SurveyMonkey platform (see supplementary eFigure 2 online). Residents were asked if they had ever performed a PDSA cycle, performed patient outreach, or performed an intervention and whether they knew the rates of diabetes, HTN, and colorectal cancer screening in their patient panels. Questions rated on a 5-point Likert scale were used to assess comfort with panel management, developing an aim statement, designing and implementing a PDSA cycle, as well as interest in pursuing QI as a career. For the purposes of analysis, these questions were dichotomized into “somewhat comfortable” and “very comfortable” vs “neutral,” “somewhat uncomfortable,” and “very uncomfortable.” Similarly, we dichotomized the question about interest in QI as a career into “somewhat interested” and “very interested” vs “neutral,” “somewhat disinterested,” and “very disinterested.” As the surveys were anonymous, we were unable to pair the pre- and postintervention surveys and used a chi-square test to evaluate whether there was an association between survey assessments pre intervention vs post intervention and a positive or negative response to the question.

We also examined rates of HTN control and colorectal cancer screening in all 75 resident panels pre and post intervention. The paired t-test was used to determine whether the mean change from pre to post intervention was significant. SAS 9.4 (SAS Institute Inc.) was used for all analyses. Institutional Review Board exemption was obtained from the Tufts Medical Center IRB. There was no funding received for this study.

Respondents

Of the 75 residents, 55 (73%) completed the survey prior to the intervention, and 39 (52%) completed the survey after the intervention.

Panel Knowledge and Intervention

Prior to the intervention, 45% of residents had performed a PDSA cycle, compared with 77% post intervention, which was a significant increase (P = .002) (Table 1). Sixty-two percent of residents had performed outreach or an intervention based on their patient panel reports prior to the intervention, compared with 85% of residents post intervention, which was also a significant increase (P = .02). The increase post intervention was not 100%, as there were residents who either missed the initial workshop or who did not follow through with their planned intervention. Common interventions included the residents giving their coordinators a list of patients to call to schedule appointments, utilizing fellow team members (eg, pharmacists, social workers) for targeted patient outreach, or calling patients themselves to reestablish a connection.

In terms of knowledge of their patient panels, prior to the intervention, 55%, 62%, and 62% of residents knew the rates of patients in their panel with diabetes, HTN, and colorectal cancer screening, respectively. After the intervention, the residents’ knowledge of these rates increased significantly, to 85% for diabetes (P = .002), 97% for HTN (P < .0001), and 97% for colorectal cancer screening (P < .0001).

Comfort With QI Approaches

Prior to the intervention, 82% of residents were comfortable managing their primary care panel, which did not change significantly post intervention (Table 2). The residents’ comfort with designing an aim statement did significantly increase, from 55% to 95% (P < .0001). The residents also had a significant increase in comfort with both designing and implementing a PDSA cycle. Prior to the intervention, 22% felt comfortable designing a PDSA cycle, which increased to 79% (P < .0001) post intervention, and 24% felt comfortable implementing a PDSA cycle, which increased to 77% (P < .0001) post intervention.

Patient Outcome Measures

The rate of HTN control in the residents' patient panels did not change significantly pre and post intervention (Table 3). The rate of resident patients who were up to date with colorectal cancer screening increased by 6.5% post intervention (P < .0001).

Interest in QI as a Career

As part of the survey, residents were asked how interested they were in making QI a part of their career. Fifty percent of residents indicated an interest in QI pre intervention, and 54% indicated an interest post intervention, which was not a significant difference (P = .72).

Discussion

In this study, we found that integration of a QI curriculum into a primary care rotation improved both residents’ knowledge of their patient panels and comfort with QI approaches, which translated to improvement in patient outcomes. Several previous studies have found improvements in resident self-assessment or knowledge after implementation of a QI curriculum.^4-13 Liao et al implemented a longitudinal curriculum including both didactic and experiential components and found an improvement in both QI confidence and knowledge.³ Similarly, Duello et al⁸ found that a curriculum including both didactic lectures and QI projects improved subjective QI knowledge and comfort. Interestingly, Fok and Wong⁹ found that resident knowledge could be sustained post curriculum after completion of a QI project, suggesting that experiential learning may be helpful in maintaining knowledge.

Studies also have looked at providing performance data to residents. Hwang et al¹⁸ found that providing audit and feedback in the form of individual panel performance data to residents compared with practice targets led to statistically significant improvement in cancer screening rates and composite quality score, indicating that there is tremendous potential in providing residents with their data. While the ACGME mandates that residents should receive data on their quality metrics, on CLER visits, many residents interviewed noted limited access to data on their metrics and benchmarks.^1,2

Though previous studies have individually looked at teaching QI concepts, providing panel data, or targeting select metrics, our study was unique in that it reviewed both self-reported resident outcomes data as well as actual patient outcomes. In addition to finding increased knowledge of patient panels and comfort with QI approaches, we found a significant increase in colorectal cancer screening rates post intervention. We thought this finding was particularly important given some data that residents' patients have been found to have worse outcomes on quality metrics compared with patients cared for by staff physicians.^14,15 Given that having a resident physician as a PCP has been associated with failing to meet quality measures, it is especially important to focus targeted quality improvement initiatives in this patient population to reduce disparities in care.

We found that residents had improved knowledge on their patient panels as a result of this initiative. The residents were noted to have a higher knowledge of their HTN and colorectal cancer screening rates in comparison to their diabetes metrics. We suspect this is because residents are provided with multiple metrics related to diabetes, including process measures such as A1c testing, as well as outcome measures such as A1c control, so it may be harder for them to elucidate exactly how they are doing with their diabetes patients, whereas in HTN control and colorectal cancer screening, there is only 1 associated metric. Interestingly, even though HTN and colorectal cancer screening were the 2 measures focused on in the study, the residents had a significant improvement in knowledge of the rates of diabetes in their panel as well. This suggests that even just receiving data alone is valuable, hopefully translating to better outcomes with better baseline understanding of panels. We believe that our intervention was successful because it included both a didactic and an experiential component, as well as the use of individual panel performance data.

There were several limitations to our study. It was performed at a single institution, translating to a small sample size. Our data analysis was limited because we were unable to pair our pre- and postintervention survey responses because we used an anonymous survey. We also did not have full participation in postintervention surveys from all residents, which may have biased the study in favor of high performers. Another limitation was that our survey relied on self-reported outcomes for the questions about the residents knowing their patient panels.

This study required a 2-hour workshop every 3 weeks led by a faculty member trained in QI. Given the amount of time needed for the curriculum, this study may be difficult to replicate at other institutions, especially if faculty with an interest or training in QI are not available. Given our finding that residents had increased knowledge of their patient panels after receiving panel metrics, simply providing data with the goal of smaller, focused interventions may be easier to implement. At our institution, we discontinued the longer 2-hour QI workshops designed to teach QI approaches more broadly. We continue to provide individualized panel data to all residents during their primary care rotations and conduct half-hour, small group workshops with the interns that focus on drafting aim statements and planning interventions. All residents are required to submit worksheets to us at the end of their primary care blocks listing their current rates of each predetermined metric and laying out their aim statements and planned interventions. Residents also continue to receive feedback from our faculty with expertise in QI afterward on their plans and evidence of follow-through in the chart, with their preceptors included on the feedback emails. Even without the larger QI workshop, this approach has continued to be successful and appreciated. In fact, it does appear as though improvement in colorectal cancer screening has been sustained over several years. At the end of our study period, the resident patient colorectal cancer screening rate rose from 34% to 43%, and for the 2021-2022 academic year, the rate rose further, from 46% to 50%.

Given that the resident clinic patient population is at higher risk overall, targeted outreach and approaches to improve quality must be continued. Future areas of research include looking at which interventions, whether QI curriculum, provision of panel data, or required panel management interventions, translate to the greatest improvements in patient outcomes in this vulnerable population.

Conclusion

Our study showed that a dedicated QI curriculum for the residents and access to quality metric data improved both resident knowledge and comfort with QI approaches. Beyond resident-centered outcomes, there was also translation to improved patient outcomes, with a significant increase in colon cancer screening rates post intervention.

Corresponding author: Kinjalika Sathi, MD, 800 Washington St., Boston, MA 02111; ksathi@tuftsmedicalcenter.org

Disclosures: None reported.

Dermatology residency continues to be one of the most competitive specialties, with a match rate of 84.7% for US allopathic seniors in the 2019-2020 academic year.¹ In the 2019-2020 cycle, dermatology applicants were tied with plastic surgery for the highest median US Medical Licensing Examination (USMLE) Step 1 score compared with other specialties, which suggests that the top medical students are applying, yet only approximately 5 of 6 students are matching.

Factors that have been cited with successful dermatology matching include USMLE Step 1 and Step 2 Clinical Knowledge (CK) scores,² research accomplishments,³ letters of recommendation,⁴ medical school performance, personal statement, grades in required clerkships, and volunteer/extracurricular experiences, among others.⁵

The National Resident Matching Program (NRMP) publishes data each year regarding different academic factors—USMLE scores; number of abstracts, presentations, and papers; work, volunteer, and research experiences—and compares the mean between matched and nonmatched applicants.¹ However, the USMLE does not report any demographic information of the applicants and the implication it has for matching. Additionally, the number of couples participating in the couples match continues to increase each year. In the 2019-2020 cycle, 1224 couples participated in the couples match.¹ However, NRMP reports only limited data regarding the couples match, and it is not specialty specific.

We aimed to determine the characteristics of matched vs nonmatched dermatology applicants. Secondarily, we aimed to determine any differences among demographics regarding matching rates, academic performance, and research publications. We also aimed to characterize the strategy and outcomes of applicants that couples matched.

Materials and Methods

The Mayo Clinic institutional review board deemed this study exempt. All applicants who applied to Mayo Clinic dermatology residency in Scottsdale, Arizona, during the 2018-2019 cycle were emailed an initial survey (N=475) before Match Day that obtained demographic information, geographic information, gap-year information, USMLE Step 1 score, publications, medical school grades, number of away rotations, and number of interviews. A follow-up survey gathering match data and couples matching data was sent to the applicants who completed the first survey on Match Day. The survey was repeated for the 2019-2020 cycle. In the second survey, Step 2 CK data were obtained. The survey was sent to 629 applicants who applied to Mayo Clinic dermatology residencies in Arizona, Minnesota, and Florida to include a broader group of applicants. For publications, applicants were asked to count only published or accepted manuscripts, not abstracts, posters, conference presentations, or submitted manuscripts. Applicants who did not respond to the second survey (match data) were not included in that part of the analysis. One survey was excluded because of implausible answers (eg, scores outside of range for USMLE Step scores).

Statistical Analysis—For statistical analyses, the applicants from both applications cycles were combined. Descriptive statistics were reported in the form of mean, median, or counts (percentages), as applicable. Means were compared using 2-sided t tests. Group comparisons were examined using χ² tests for categorical variables. Statistical analyses were performed using the BlueSky Statistics version 6.30. P<.05 was considered significant.

Results

In 2019, a total of 149 applicants completed the initial survey (31.4% response rate), and 112 completed the follow-up survey (75.2% response rate). In 2020, a total of 142 applicants completed the initial survey (22.6% response rate), and 124 completed the follow-up survey (87.3% response rate). Combining the 2 years, after removing 1 survey with implausible answers, there were 290 respondents from the initial survey and 235 from the follow-up survey. The median (SD) age for the total applicants over both years was 27 (3.0) years, and 180 applicants were female (61.9%).

USMLE Scores—The median USMLE Step 1 score was 250, and scores ranged from 196 to 271. The median USMLE Step 2 CK score was 257, and scores ranged from 213 to 281. Higher USMLE Step 1 and Step 2 CK scores and more interviews were associated with higher match rates (Table 1). In addition, students with a dermatology program at their medical school were more likely to match than those without a home dermatology program.

Gender Differences—There were 180 females and 110 males who completed the surveys. Males and females had similar match rates (85.2% vs 89.0%; P=.39)(Table 2).

Family Life—In comparing marital status, applicants who were divorced had a higher median age (38.5 years) compared with applicants who were single, married, or in a domestic partnership (all 27 years; P<.01). Differences are outlined in Table 3.

On average, applicants with children (n=27 [15 male, 12 female]; P=.13) were 3 years older than those without (30.5 vs 27; P<.01) and were more likely to be married (88.9% vs 21.5%; P<.01). Applicants with children had a mean USMLE Step 1 score of 241 compared to 251 for those without children (P=.02) and a mean USMLE Step 2 CK score of 246 compared to 258 for those without children (P<.01). Applicants with children had similar debt, number of publications, number of honored rotations, and match rates compared to applicants without children (Figure).

Couples Match—Seventeen individuals in our survey participated in the couples match (7.8%), and all 17 (100%) matched into dermatology. The mean age was 26.7 years, 12 applicants were female, 2 applicants were married, and 1 applicant had children. The mean number of interviews offered was 13.6, and the mean number of interviews attended was 11.3. This was higher than participants who were not couples matching (13.6 vs 9.8 [P=.02] and 11.3 vs 8.9 [P=.04], respectively). Applicants and their partners applied to programs and received interviews in a mean of 10 cities. Sixteen applicants reported that they contacted programs where their partner had interview offers. All participants’ rank lists included programs located in different cities than their partners’ ranked programs, and all but 1 participant ranked programs located in a different state than their partners’ ranked programs. Fifteen participants had options in their rank list for the applicant not to match, even if the partner would match. Similarly, 12 had the option for the applicant to match, even if the partner would not match. Fourteen (82.4%) matched at the same institution as their significant other. Three (17.6%) applicants matched to a program in a different state than the partner’s matched program. Two (11.8%) participants felt their relationship with their partner suffered because of the match, and 1 (5.9%) applicant was undetermined. One applicant described their relationship suffering from “unnecessary tension and anxiety” and noted “difficult conversations” about potentially matching into dermatology in a different location from their partner that could have been “devastating and not something [he or she] should have to choose.”

Comment

Factors for Matching in Dermatology—In our survey, we found the statistically significant factors of matching into dermatology included high USMLE Step 1 and Step 2 CK scores (P<.01), having a home dermatology program (P=.04), and attending a higher number of dermatology interviews (P<.01). These data are similar to NRMP results¹; however, the higher likelihood of matching if the medical school has a home dermatology program has not been reported. This finding could be due to multiple factors such as students have less access to academic dermatologists for research projects, letters of recommendations, mentorship, and clinical rotations.

Gender and having children were factors that had no correlation with the match rate. There was a statistical difference of matching based on marital status (P<.01), but this is likely due to the low number of applicants in the divorced category. There were differences among demographics with USMLE Step 1 and Step 2 CK scores, which is a known factor in matching.^1,2 Applicants with children had lower USMLE Step 1 and Step 2 CK scores compared to applicants without children. Females also had lower median USMLE Step 1 scores compared to males. This finding may serve as a reminder to programs when comparing USMLE Step examination scores that demographic factors may play a role. The race and ethnicity of applicants likely play a role. It has been reported that underrepresented minorities had lower match rates than White and Asian applicants in dermatology.⁶ There have been several published articles discussing the lack of diversity in dermatology, with a call to action.^7-9

Factors for Couples Matching—The number of applicants participating in the couples match continues to increase yearly. The NMRP does publish data regarding “successful” couples matching but does not specify how many couples match together. There also is little published regarding advice for participation in the couples match. Although we had a limited number of couples that participated in the match, it is interesting to note they had similar strategies, including contacting programs at institutions that had offered interviews to their partners. This strategy may be effective, as dermatology programs offer interviews relatively late compared with other specialties.⁵ Additionally, this strategy may increase the number of interviews offered and received, as evidenced by the higher number of interviews offered compared with those who were not couples matching. Additionally, this survey highlights the sacrifice often needed by couples in the couples match as revealed by the inclusion of rank-list options in which the couples reside long distance or in which 1 partner does not match. This information may be helpful to applicants who are planning a strategy for the couples match in dermatology. Although this study does not encompass all dermatology applicants in the 2019-2020 cycle, we do believe it may be representative. The USMLE Step 1 scores in this study were similar to the published NRMP data.^1,10 According to NRMP data from the 2019-2020 cycle, the mean USMLE Step 1 score was 248 for matched applicants and 239 for unmatched.¹ The NRMP reported the mean USMLE Step 2 CK score for matched was 256 and 248 for unmatched, which also is similar to our data. The NRMP reported the mean number of programs ranked was 9.9 for matched and 4.5 for unmatched applicants.¹ Again, our data were similar for number of dermatology interviews attended.

Limitations—There are limitations to this study. The main limitation is that the survey is from a single institution and had a limited number of respondents. Given the nature of the study, the accuracy of the data is dependent on the applicants’ honesty in self-reporting academic performance and other variables. There also may be a selection bias given the low response rate. The subanalyses—children and couples matching—were underpowered with the limited number of participants. Further studies that include multiple residency programs and multiple years could be helpful to provide more power and less risk of bias. We did not gather information such as the Medical Student Performance Evaluation letter, letters of recommendation, or personal statements, which do play an important role in the assessment of an applicant. However, because the applicants completed these surveys, and given these are largely blinded to applicants, we did not feel the applicants could accurately respond to those aspects of the application.

Conclusion

Our survey finds that factors associated with matching included a higher USMLE Step 1 score, having a home dermatology program, and a higher number of interviews offered and attended. Some demographics had varying USMLE Step 1 scores but similar match rates.

Dermatology residency continues to be one of the most competitive specialties, with a match rate of 84.7% for US allopathic seniors in the 2019-2020 academic year.¹ In the 2019-2020 cycle, dermatology applicants were tied with plastic surgery for the highest median US Medical Licensing Examination (USMLE) Step 1 score compared with other specialties, which suggests that the top medical students are applying, yet only approximately 5 of 6 students are matching.

Factors that have been cited with successful dermatology matching include USMLE Step 1 and Step 2 Clinical Knowledge (CK) scores,² research accomplishments,³ letters of recommendation,⁴ medical school performance, personal statement, grades in required clerkships, and volunteer/extracurricular experiences, among others.⁵

The National Resident Matching Program (NRMP) publishes data each year regarding different academic factors—USMLE scores; number of abstracts, presentations, and papers; work, volunteer, and research experiences—and compares the mean between matched and nonmatched applicants.¹ However, the USMLE does not report any demographic information of the applicants and the implication it has for matching. Additionally, the number of couples participating in the couples match continues to increase each year. In the 2019-2020 cycle, 1224 couples participated in the couples match.¹ However, NRMP reports only limited data regarding the couples match, and it is not specialty specific.

We aimed to determine the characteristics of matched vs nonmatched dermatology applicants. Secondarily, we aimed to determine any differences among demographics regarding matching rates, academic performance, and research publications. We also aimed to characterize the strategy and outcomes of applicants that couples matched.

Materials and Methods

The Mayo Clinic institutional review board deemed this study exempt. All applicants who applied to Mayo Clinic dermatology residency in Scottsdale, Arizona, during the 2018-2019 cycle were emailed an initial survey (N=475) before Match Day that obtained demographic information, geographic information, gap-year information, USMLE Step 1 score, publications, medical school grades, number of away rotations, and number of interviews. A follow-up survey gathering match data and couples matching data was sent to the applicants who completed the first survey on Match Day. The survey was repeated for the 2019-2020 cycle. In the second survey, Step 2 CK data were obtained. The survey was sent to 629 applicants who applied to Mayo Clinic dermatology residencies in Arizona, Minnesota, and Florida to include a broader group of applicants. For publications, applicants were asked to count only published or accepted manuscripts, not abstracts, posters, conference presentations, or submitted manuscripts. Applicants who did not respond to the second survey (match data) were not included in that part of the analysis. One survey was excluded because of implausible answers (eg, scores outside of range for USMLE Step scores).

Statistical Analysis—For statistical analyses, the applicants from both applications cycles were combined. Descriptive statistics were reported in the form of mean, median, or counts (percentages), as applicable. Means were compared using 2-sided t tests. Group comparisons were examined using χ² tests for categorical variables. Statistical analyses were performed using the BlueSky Statistics version 6.30. P<.05 was considered significant.

Results

In 2019, a total of 149 applicants completed the initial survey (31.4% response rate), and 112 completed the follow-up survey (75.2% response rate). In 2020, a total of 142 applicants completed the initial survey (22.6% response rate), and 124 completed the follow-up survey (87.3% response rate). Combining the 2 years, after removing 1 survey with implausible answers, there were 290 respondents from the initial survey and 235 from the follow-up survey. The median (SD) age for the total applicants over both years was 27 (3.0) years, and 180 applicants were female (61.9%).

USMLE Scores—The median USMLE Step 1 score was 250, and scores ranged from 196 to 271. The median USMLE Step 2 CK score was 257, and scores ranged from 213 to 281. Higher USMLE Step 1 and Step 2 CK scores and more interviews were associated with higher match rates (Table 1). In addition, students with a dermatology program at their medical school were more likely to match than those without a home dermatology program.

Gender Differences—There were 180 females and 110 males who completed the surveys. Males and females had similar match rates (85.2% vs 89.0%; P=.39)(Table 2).

Family Life—In comparing marital status, applicants who were divorced had a higher median age (38.5 years) compared with applicants who were single, married, or in a domestic partnership (all 27 years; P<.01). Differences are outlined in Table 3.

On average, applicants with children (n=27 [15 male, 12 female]; P=.13) were 3 years older than those without (30.5 vs 27; P<.01) and were more likely to be married (88.9% vs 21.5%; P<.01). Applicants with children had a mean USMLE Step 1 score of 241 compared to 251 for those without children (P=.02) and a mean USMLE Step 2 CK score of 246 compared to 258 for those without children (P<.01). Applicants with children had similar debt, number of publications, number of honored rotations, and match rates compared to applicants without children (Figure).

Couples Match—Seventeen individuals in our survey participated in the couples match (7.8%), and all 17 (100%) matched into dermatology. The mean age was 26.7 years, 12 applicants were female, 2 applicants were married, and 1 applicant had children. The mean number of interviews offered was 13.6, and the mean number of interviews attended was 11.3. This was higher than participants who were not couples matching (13.6 vs 9.8 [P=.02] and 11.3 vs 8.9 [P=.04], respectively). Applicants and their partners applied to programs and received interviews in a mean of 10 cities. Sixteen applicants reported that they contacted programs where their partner had interview offers. All participants’ rank lists included programs located in different cities than their partners’ ranked programs, and all but 1 participant ranked programs located in a different state than their partners’ ranked programs. Fifteen participants had options in their rank list for the applicant not to match, even if the partner would match. Similarly, 12 had the option for the applicant to match, even if the partner would not match. Fourteen (82.4%) matched at the same institution as their significant other. Three (17.6%) applicants matched to a program in a different state than the partner’s matched program. Two (11.8%) participants felt their relationship with their partner suffered because of the match, and 1 (5.9%) applicant was undetermined. One applicant described their relationship suffering from “unnecessary tension and anxiety” and noted “difficult conversations” about potentially matching into dermatology in a different location from their partner that could have been “devastating and not something [he or she] should have to choose.”

Comment

Factors for Matching in Dermatology—In our survey, we found the statistically significant factors of matching into dermatology included high USMLE Step 1 and Step 2 CK scores (P<.01), having a home dermatology program (P=.04), and attending a higher number of dermatology interviews (P<.01). These data are similar to NRMP results¹; however, the higher likelihood of matching if the medical school has a home dermatology program has not been reported. This finding could be due to multiple factors such as students have less access to academic dermatologists for research projects, letters of recommendations, mentorship, and clinical rotations.

Gender and having children were factors that had no correlation with the match rate. There was a statistical difference of matching based on marital status (P<.01), but this is likely due to the low number of applicants in the divorced category. There were differences among demographics with USMLE Step 1 and Step 2 CK scores, which is a known factor in matching.^1,2 Applicants with children had lower USMLE Step 1 and Step 2 CK scores compared to applicants without children. Females also had lower median USMLE Step 1 scores compared to males. This finding may serve as a reminder to programs when comparing USMLE Step examination scores that demographic factors may play a role. The race and ethnicity of applicants likely play a role. It has been reported that underrepresented minorities had lower match rates than White and Asian applicants in dermatology.⁶ There have been several published articles discussing the lack of diversity in dermatology, with a call to action.^7-9

Factors for Couples Matching—The number of applicants participating in the couples match continues to increase yearly. The NMRP does publish data regarding “successful” couples matching but does not specify how many couples match together. There also is little published regarding advice for participation in the couples match. Although we had a limited number of couples that participated in the match, it is interesting to note they had similar strategies, including contacting programs at institutions that had offered interviews to their partners. This strategy may be effective, as dermatology programs offer interviews relatively late compared with other specialties.⁵ Additionally, this strategy may increase the number of interviews offered and received, as evidenced by the higher number of interviews offered compared with those who were not couples matching. Additionally, this survey highlights the sacrifice often needed by couples in the couples match as revealed by the inclusion of rank-list options in which the couples reside long distance or in which 1 partner does not match. This information may be helpful to applicants who are planning a strategy for the couples match in dermatology. Although this study does not encompass all dermatology applicants in the 2019-2020 cycle, we do believe it may be representative. The USMLE Step 1 scores in this study were similar to the published NRMP data.^1,10 According to NRMP data from the 2019-2020 cycle, the mean USMLE Step 1 score was 248 for matched applicants and 239 for unmatched.¹ The NRMP reported the mean USMLE Step 2 CK score for matched was 256 and 248 for unmatched, which also is similar to our data. The NRMP reported the mean number of programs ranked was 9.9 for matched and 4.5 for unmatched applicants.¹ Again, our data were similar for number of dermatology interviews attended.

Limitations—There are limitations to this study. The main limitation is that the survey is from a single institution and had a limited number of respondents. Given the nature of the study, the accuracy of the data is dependent on the applicants’ honesty in self-reporting academic performance and other variables. There also may be a selection bias given the low response rate. The subanalyses—children and couples matching—were underpowered with the limited number of participants. Further studies that include multiple residency programs and multiple years could be helpful to provide more power and less risk of bias. We did not gather information such as the Medical Student Performance Evaluation letter, letters of recommendation, or personal statements, which do play an important role in the assessment of an applicant. However, because the applicants completed these surveys, and given these are largely blinded to applicants, we did not feel the applicants could accurately respond to those aspects of the application.

Conclusion

Our survey finds that factors associated with matching included a higher USMLE Step 1 score, having a home dermatology program, and a higher number of interviews offered and attended. Some demographics had varying USMLE Step 1 scores but similar match rates.

Dermatology residency continues to be one of the most competitive specialties, with a match rate of 84.7% for US allopathic seniors in the 2019-2020 academic year.¹ In the 2019-2020 cycle, dermatology applicants were tied with plastic surgery for the highest median US Medical Licensing Examination (USMLE) Step 1 score compared with other specialties, which suggests that the top medical students are applying, yet only approximately 5 of 6 students are matching.

Factors that have been cited with successful dermatology matching include USMLE Step 1 and Step 2 Clinical Knowledge (CK) scores,² research accomplishments,³ letters of recommendation,⁴ medical school performance, personal statement, grades in required clerkships, and volunteer/extracurricular experiences, among others.⁵

The National Resident Matching Program (NRMP) publishes data each year regarding different academic factors—USMLE scores; number of abstracts, presentations, and papers; work, volunteer, and research experiences—and compares the mean between matched and nonmatched applicants.¹ However, the USMLE does not report any demographic information of the applicants and the implication it has for matching. Additionally, the number of couples participating in the couples match continues to increase each year. In the 2019-2020 cycle, 1224 couples participated in the couples match.¹ However, NRMP reports only limited data regarding the couples match, and it is not specialty specific.

We aimed to determine the characteristics of matched vs nonmatched dermatology applicants. Secondarily, we aimed to determine any differences among demographics regarding matching rates, academic performance, and research publications. We also aimed to characterize the strategy and outcomes of applicants that couples matched.

Materials and Methods

The Mayo Clinic institutional review board deemed this study exempt. All applicants who applied to Mayo Clinic dermatology residency in Scottsdale, Arizona, during the 2018-2019 cycle were emailed an initial survey (N=475) before Match Day that obtained demographic information, geographic information, gap-year information, USMLE Step 1 score, publications, medical school grades, number of away rotations, and number of interviews. A follow-up survey gathering match data and couples matching data was sent to the applicants who completed the first survey on Match Day. The survey was repeated for the 2019-2020 cycle. In the second survey, Step 2 CK data were obtained. The survey was sent to 629 applicants who applied to Mayo Clinic dermatology residencies in Arizona, Minnesota, and Florida to include a broader group of applicants. For publications, applicants were asked to count only published or accepted manuscripts, not abstracts, posters, conference presentations, or submitted manuscripts. Applicants who did not respond to the second survey (match data) were not included in that part of the analysis. One survey was excluded because of implausible answers (eg, scores outside of range for USMLE Step scores).

Statistical Analysis—For statistical analyses, the applicants from both applications cycles were combined. Descriptive statistics were reported in the form of mean, median, or counts (percentages), as applicable. Means were compared using 2-sided t tests. Group comparisons were examined using χ² tests for categorical variables. Statistical analyses were performed using the BlueSky Statistics version 6.30. P<.05 was considered significant.

Results

In 2019, a total of 149 applicants completed the initial survey (31.4% response rate), and 112 completed the follow-up survey (75.2% response rate). In 2020, a total of 142 applicants completed the initial survey (22.6% response rate), and 124 completed the follow-up survey (87.3% response rate). Combining the 2 years, after removing 1 survey with implausible answers, there were 290 respondents from the initial survey and 235 from the follow-up survey. The median (SD) age for the total applicants over both years was 27 (3.0) years, and 180 applicants were female (61.9%).

USMLE Scores—The median USMLE Step 1 score was 250, and scores ranged from 196 to 271. The median USMLE Step 2 CK score was 257, and scores ranged from 213 to 281. Higher USMLE Step 1 and Step 2 CK scores and more interviews were associated with higher match rates (Table 1). In addition, students with a dermatology program at their medical school were more likely to match than those without a home dermatology program.

Gender Differences—There were 180 females and 110 males who completed the surveys. Males and females had similar match rates (85.2% vs 89.0%; P=.39)(Table 2).

Family Life—In comparing marital status, applicants who were divorced had a higher median age (38.5 years) compared with applicants who were single, married, or in a domestic partnership (all 27 years; P<.01). Differences are outlined in Table 3.

On average, applicants with children (n=27 [15 male, 12 female]; P=.13) were 3 years older than those without (30.5 vs 27; P<.01) and were more likely to be married (88.9% vs 21.5%; P<.01). Applicants with children had a mean USMLE Step 1 score of 241 compared to 251 for those without children (P=.02) and a mean USMLE Step 2 CK score of 246 compared to 258 for those without children (P<.01). Applicants with children had similar debt, number of publications, number of honored rotations, and match rates compared to applicants without children (Figure).

Couples Match—Seventeen individuals in our survey participated in the couples match (7.8%), and all 17 (100%) matched into dermatology. The mean age was 26.7 years, 12 applicants were female, 2 applicants were married, and 1 applicant had children. The mean number of interviews offered was 13.6, and the mean number of interviews attended was 11.3. This was higher than participants who were not couples matching (13.6 vs 9.8 [P=.02] and 11.3 vs 8.9 [P=.04], respectively). Applicants and their partners applied to programs and received interviews in a mean of 10 cities. Sixteen applicants reported that they contacted programs where their partner had interview offers. All participants’ rank lists included programs located in different cities than their partners’ ranked programs, and all but 1 participant ranked programs located in a different state than their partners’ ranked programs. Fifteen participants had options in their rank list for the applicant not to match, even if the partner would match. Similarly, 12 had the option for the applicant to match, even if the partner would not match. Fourteen (82.4%) matched at the same institution as their significant other. Three (17.6%) applicants matched to a program in a different state than the partner’s matched program. Two (11.8%) participants felt their relationship with their partner suffered because of the match, and 1 (5.9%) applicant was undetermined. One applicant described their relationship suffering from “unnecessary tension and anxiety” and noted “difficult conversations” about potentially matching into dermatology in a different location from their partner that could have been “devastating and not something [he or she] should have to choose.”

Comment

Factors for Matching in Dermatology—In our survey, we found the statistically significant factors of matching into dermatology included high USMLE Step 1 and Step 2 CK scores (P<.01), having a home dermatology program (P=.04), and attending a higher number of dermatology interviews (P<.01). These data are similar to NRMP results¹; however, the higher likelihood of matching if the medical school has a home dermatology program has not been reported. This finding could be due to multiple factors such as students have less access to academic dermatologists for research projects, letters of recommendations, mentorship, and clinical rotations.

Gender and having children were factors that had no correlation with the match rate. There was a statistical difference of matching based on marital status (P<.01), but this is likely due to the low number of applicants in the divorced category. There were differences among demographics with USMLE Step 1 and Step 2 CK scores, which is a known factor in matching.^1,2 Applicants with children had lower USMLE Step 1 and Step 2 CK scores compared to applicants without children. Females also had lower median USMLE Step 1 scores compared to males. This finding may serve as a reminder to programs when comparing USMLE Step examination scores that demographic factors may play a role. The race and ethnicity of applicants likely play a role. It has been reported that underrepresented minorities had lower match rates than White and Asian applicants in dermatology.⁶ There have been several published articles discussing the lack of diversity in dermatology, with a call to action.^7-9

Factors for Couples Matching—The number of applicants participating in the couples match continues to increase yearly. The NMRP does publish data regarding “successful” couples matching but does not specify how many couples match together. There also is little published regarding advice for participation in the couples match. Although we had a limited number of couples that participated in the match, it is interesting to note they had similar strategies, including contacting programs at institutions that had offered interviews to their partners. This strategy may be effective, as dermatology programs offer interviews relatively late compared with other specialties.⁵ Additionally, this strategy may increase the number of interviews offered and received, as evidenced by the higher number of interviews offered compared with those who were not couples matching. Additionally, this survey highlights the sacrifice often needed by couples in the couples match as revealed by the inclusion of rank-list options in which the couples reside long distance or in which 1 partner does not match. This information may be helpful to applicants who are planning a strategy for the couples match in dermatology. Although this study does not encompass all dermatology applicants in the 2019-2020 cycle, we do believe it may be representative. The USMLE Step 1 scores in this study were similar to the published NRMP data.^1,10 According to NRMP data from the 2019-2020 cycle, the mean USMLE Step 1 score was 248 for matched applicants and 239 for unmatched.¹ The NRMP reported the mean USMLE Step 2 CK score for matched was 256 and 248 for unmatched, which also is similar to our data. The NRMP reported the mean number of programs ranked was 9.9 for matched and 4.5 for unmatched applicants.¹ Again, our data were similar for number of dermatology interviews attended.

Limitations—There are limitations to this study. The main limitation is that the survey is from a single institution and had a limited number of respondents. Given the nature of the study, the accuracy of the data is dependent on the applicants’ honesty in self-reporting academic performance and other variables. There also may be a selection bias given the low response rate. The subanalyses—children and couples matching—were underpowered with the limited number of participants. Further studies that include multiple residency programs and multiple years could be helpful to provide more power and less risk of bias. We did not gather information such as the Medical Student Performance Evaluation letter, letters of recommendation, or personal statements, which do play an important role in the assessment of an applicant. However, because the applicants completed these surveys, and given these are largely blinded to applicants, we did not feel the applicants could accurately respond to those aspects of the application.

Conclusion

Our survey finds that factors associated with matching included a higher USMLE Step 1 score, having a home dermatology program, and a higher number of interviews offered and attended. Some demographics had varying USMLE Step 1 scores but similar match rates.

Vaccinations have substantially reduced morbidity and mortality from many infectious diseases. Despite the clear value of vaccinations in public health, efforts to better understand adverse events (AEs) following immunization are important to sustain public trust and vaccine confidence. Noninfectious inflammation of the heart may manifest as myocarditis or pericarditis, or occasionally, with shared signs and symptoms of each, as myopericarditis. This is a rare AE following some immunizations. Vaccine-associated myocarditis, pericarditis, or myopericarditis (VAMP) has been most clearly associated with smallpox vaccines and mRNA COVID-19 vaccines.^1-6 Although extremely rare, VAMP also has been associated with other vaccines.^7,8 Limited information exists to guide shared clinical decision making on COVID-19 vaccination in persons with a history of VAMP. It is unknown whether individuals with a history of VAMP are at higher risk for developing a recurrence or experiencing a more severe outcome following COVID-19 vaccination.

Methods

As part of the collaborative public health mission with the Centers for Disease Control and Prevention (CDC) for enhanced vaccine AE surveillance, the Defense Health Agency Immunization Healthcare Division (IHD) maintains a clinical database of service members and beneficiaries referred for suspected AEs following immunizations. A review of all AEs following immunization cases in this database from January 1, 2003, through February 28, 2022, identified individuals meeting the following criteria: (a) VAMP prior to receipt of COVID-19 vaccine; (b) receipt of COVID-19 vaccine in 2021; and (c) medical documentation in available electronic health records sufficient to describe health status at least 30 days following COVID-19 vaccination.⁹ If medical entries suggested cardiac symptoms following a COVID-19 vaccine, additional information was sought to verify VAMP based on current published criteria.^10,11 Both the initial VAMP cases and the suspected COVID-19 VAMP cases were adjudicated by a team of vaccine experts and specialists in immunology, cardiology, and preventive medicine.

This retrospective review was approved and conducted in accordance with the Walter Reed National Military Medical Center Institutional Review Board protocol #20664. All individuals with recurrent VAMP consented to share their health records and clinical details.

Results

Among 9260 cases in the IHD database, 431 met the case definition for VAMP.

Among the 179 patients included in this analysis, 171 (96%) were male. Their median age was 39 years at the time of COVID-19 vaccination.

Within 1 month of receipt of any COVID-19 vaccine, 11 individuals had documented symptoms suggesting cardiac involvement, specifically, chest pain, palpitations, or dyspnea. After cardiac evaluation, 4 patients met the criteria for VAMP after COVID-19 vaccination.^10,11 Seven patients either did not meet the criteria for VAMP or had alternative causes for their symptoms.

Two men aged 49 and 50 years with a history of vaccine-associated myocarditis following smallpox vaccination (Dryvax and ACAM2000) developed myocarditis 3 days after their second dose of the Moderna vaccine. One of these patients received a Pfizer-BioNTech booster 10 months later with no recurrence of symptoms. A 55-year-old man with a history of vaccine-associated myocarditis following Dryvax vaccination developed myocarditis 2 days after his Pfizer-BioNTech booster. None of the patients who developed post-COVID-19 VAMP reported residual symptoms from their initial VAMP episode, which occurred 12 to 18 years earlier. All were hospitalized briefly for observation and had complete symptom resolution within 6 weeks.

A 25-year-old man developed pericarditis 4 days after his second Pfizer-BioNTech vaccination. His previous ACAM2000 vaccine-associated myocarditis occurred 3 years earlier, with no residual symptoms. Of note, he had a mild COVID-19 infection 78 days before the onset of his pericarditis. After the onset of his COVID-19 vaccine-associated pericarditis, he continued to experience transient bouts of chest pressure and exertional dyspnea that resolved within 7 months of onset.

The median interval between COVID-19 vaccine doses in those who developed post-COVID-19 VAMP was within the recommended mRNA vaccine dosing intervals of 3 to 4 weeks and was consistent with the median mRNA vaccine dosing intervals among the entire cohort.

Due to the small cohort size and other limitations of this study, the suggested rate of cardiac injury in this review (4 cases in 179 persons, or 2.2%) is an imprecise estimate of risk in a small population (95% CI, 0.1%-4.4%). While this rate may seem higher than expected within the general population after COVID-19 vaccination, it is lower than the estimated lifetime risk of recurrent myocarditis from any cause.^6,12

Discussion

To our knowledge, this is the first report describing cardiac outcomes after COVID-19 vaccination among a cohort of individuals with prior history of VAMP. Four cases of COVID-19 VAMP were identified among 179 patients with previous VAMP. All cases had experienced VAMP after the smallpox vaccine several years earlier, with complete resolution of symptoms. Three cases presented with recurrent VAMP after their second dose of an mRNA COVID-19 vaccine, and one after an mRNA booster dose. All fully recovered over the course of several months.

Myocarditis is a heterogeneous inflammatory injury with diverse, sometimes idiopathic, etiologies.¹³ In contrast to infection-related cardiac injury, prior reports of vaccine-associated myocarditis have suggested a hypersensitivity reaction characterized by patchy eosinophilic infiltrates, a benign clinical course, and good prognosis.^2,3

There are several common features between VAMP after smallpox and COVID-19 vaccination. Cases occur predominantly in young men. The onset of symptoms after smallpox vaccine (mean, 10 days) and after mRNA COVID-19 vaccine (mean, 3 days) appears to correspond to the timing of peak postvaccination pro-inflammatory cytokine elevation.¹⁴ While all VAMP cases are serious events, the majority of patients appear to have a relatively benign clinical course with rapid and full recovery.¹³

Patients who have experienced an inflammatory cardiac injury may be at higher risk for recurrence, but quantifying risk of this rare phenomenon is challenging. Cases of VAMP after the COVID-19 vaccine have occasionally been reported in patients with previous cardiac injury unrelated to vaccination.^15-17 The cases presented here represent the first report of recurrent VAMP following prior non-COVID-19 vaccinations.

Most patients with prior VAMP in this cohort did not experience cardiac-suggestive symptoms following COVID-19 vaccination. Among 11 patients who developed symptoms, 3 had confirmed myocarditis and 1 had confirmed pericarditis. The clinical course for these patients with recurrent VAMP was observed to be no different in severity or duration from those who experience new-onset VAMP.⁴ All other patients not meeting criteria for VAMP or having alternative explanations for their symptoms also had a benign clinical course. Nonetheless, of the study cohort of 179, recurrent VAMP was diagnosed in 4 of the 11 who developed cardiac-suggestive symptoms following COVID-19 vaccination. The importance of cardiac evaluation should be emphasized for any patient presenting with chest pain, dyspnea, or other cardiac-suggestive symptoms following vaccination.

Strengths and Limitations

The strength of this review of VAMP recurrence associated with COVID-19 vaccination derives from our large and unique longitudinal database of VAMP among current and prior service members. Additionally, the IHD’s ongoing enhanced vaccine AEs surveillance provides the opportunity to contact patients and review their electronic health records over an extended interval of time.

When interpreting this report’s implications, limitations inherent to any retrospective case review should be considered. The cohort of cases of prior VAMP included primarily healthy, fit, young service members; this population is not representative of the general population. The cohort included prior VAMP cases that generally occurred after smallpox vaccination. Experiences after smallpox vaccine may not apply to cardiac injury from other vaccines or etiologies. By the nature of this review, the population studied at the time of COVID-19 vaccination was somewhat older than those most likely to develop an initial bout of VAMP.² This review was limited by information available in the electronic health records of a small number of patients. Subclinical cases of VAMP and cases without adequate clinical evaluation also could not be included.

Conclusions

Noninfectious inflammation of the heart (myocarditis, pericarditis, or myopericarditis) is a rare AE following certain vaccines, especially live replicating smallpox vaccine and mRNA COVID-19 vaccines. In this observational analysis, the majority of patients with previous VAMP successfully received a COVID-19 vaccine without recurrence. The 4 patients who were identified with recurrent VAMP following COVID-19 vaccination all recovered with supportive care. While the CDC endorses that individuals with a history of infectious myocarditis may receive COVID-19 vaccine after symptoms have resolved, there is currently insufficient safety data regarding COVID-19 vaccination of those with prior non-COVID-19 VAMP or following subsequent COVID-19 vaccination in those with prior VAMP related to COVID-19.¹⁰ For these individuals, COVID-19 vaccination is a precaution.¹⁰ Although insufficient to determine a precise level of risk, this report does provide data on which to base the CDC-recommended shared decision-making counseling of these patients. More research is needed to better define factors that increase risk for, or protection from, immune-mediated AEs following immunization, including VAMP. While benefits of vaccination have clearly outweighed risks during the COVID-19 pandemic, such research may optimize future vaccine recommendations.¹⁸

Vaccinations have substantially reduced morbidity and mortality from many infectious diseases. Despite the clear value of vaccinations in public health, efforts to better understand adverse events (AEs) following immunization are important to sustain public trust and vaccine confidence. Noninfectious inflammation of the heart may manifest as myocarditis or pericarditis, or occasionally, with shared signs and symptoms of each, as myopericarditis. This is a rare AE following some immunizations. Vaccine-associated myocarditis, pericarditis, or myopericarditis (VAMP) has been most clearly associated with smallpox vaccines and mRNA COVID-19 vaccines.^1-6 Although extremely rare, VAMP also has been associated with other vaccines.^7,8 Limited information exists to guide shared clinical decision making on COVID-19 vaccination in persons with a history of VAMP. It is unknown whether individuals with a history of VAMP are at higher risk for developing a recurrence or experiencing a more severe outcome following COVID-19 vaccination.

Methods

As part of the collaborative public health mission with the Centers for Disease Control and Prevention (CDC) for enhanced vaccine AE surveillance, the Defense Health Agency Immunization Healthcare Division (IHD) maintains a clinical database of service members and beneficiaries referred for suspected AEs following immunizations. A review of all AEs following immunization cases in this database from January 1, 2003, through February 28, 2022, identified individuals meeting the following criteria: (a) VAMP prior to receipt of COVID-19 vaccine; (b) receipt of COVID-19 vaccine in 2021; and (c) medical documentation in available electronic health records sufficient to describe health status at least 30 days following COVID-19 vaccination.⁹ If medical entries suggested cardiac symptoms following a COVID-19 vaccine, additional information was sought to verify VAMP based on current published criteria.^10,11 Both the initial VAMP cases and the suspected COVID-19 VAMP cases were adjudicated by a team of vaccine experts and specialists in immunology, cardiology, and preventive medicine.

This retrospective review was approved and conducted in accordance with the Walter Reed National Military Medical Center Institutional Review Board protocol #20664. All individuals with recurrent VAMP consented to share their health records and clinical details.

Results

Among 9260 cases in the IHD database, 431 met the case definition for VAMP.

Among the 179 patients included in this analysis, 171 (96%) were male. Their median age was 39 years at the time of COVID-19 vaccination.

Within 1 month of receipt of any COVID-19 vaccine, 11 individuals had documented symptoms suggesting cardiac involvement, specifically, chest pain, palpitations, or dyspnea. After cardiac evaluation, 4 patients met the criteria for VAMP after COVID-19 vaccination.^10,11 Seven patients either did not meet the criteria for VAMP or had alternative causes for their symptoms.

Two men aged 49 and 50 years with a history of vaccine-associated myocarditis following smallpox vaccination (Dryvax and ACAM2000) developed myocarditis 3 days after their second dose of the Moderna vaccine. One of these patients received a Pfizer-BioNTech booster 10 months later with no recurrence of symptoms. A 55-year-old man with a history of vaccine-associated myocarditis following Dryvax vaccination developed myocarditis 2 days after his Pfizer-BioNTech booster. None of the patients who developed post-COVID-19 VAMP reported residual symptoms from their initial VAMP episode, which occurred 12 to 18 years earlier. All were hospitalized briefly for observation and had complete symptom resolution within 6 weeks.

A 25-year-old man developed pericarditis 4 days after his second Pfizer-BioNTech vaccination. His previous ACAM2000 vaccine-associated myocarditis occurred 3 years earlier, with no residual symptoms. Of note, he had a mild COVID-19 infection 78 days before the onset of his pericarditis. After the onset of his COVID-19 vaccine-associated pericarditis, he continued to experience transient bouts of chest pressure and exertional dyspnea that resolved within 7 months of onset.

The median interval between COVID-19 vaccine doses in those who developed post-COVID-19 VAMP was within the recommended mRNA vaccine dosing intervals of 3 to 4 weeks and was consistent with the median mRNA vaccine dosing intervals among the entire cohort.

Due to the small cohort size and other limitations of this study, the suggested rate of cardiac injury in this review (4 cases in 179 persons, or 2.2%) is an imprecise estimate of risk in a small population (95% CI, 0.1%-4.4%). While this rate may seem higher than expected within the general population after COVID-19 vaccination, it is lower than the estimated lifetime risk of recurrent myocarditis from any cause.^6,12

Discussion

To our knowledge, this is the first report describing cardiac outcomes after COVID-19 vaccination among a cohort of individuals with prior history of VAMP. Four cases of COVID-19 VAMP were identified among 179 patients with previous VAMP. All cases had experienced VAMP after the smallpox vaccine several years earlier, with complete resolution of symptoms. Three cases presented with recurrent VAMP after their second dose of an mRNA COVID-19 vaccine, and one after an mRNA booster dose. All fully recovered over the course of several months.

Myocarditis is a heterogeneous inflammatory injury with diverse, sometimes idiopathic, etiologies.¹³ In contrast to infection-related cardiac injury, prior reports of vaccine-associated myocarditis have suggested a hypersensitivity reaction characterized by patchy eosinophilic infiltrates, a benign clinical course, and good prognosis.^2,3

There are several common features between VAMP after smallpox and COVID-19 vaccination. Cases occur predominantly in young men. The onset of symptoms after smallpox vaccine (mean, 10 days) and after mRNA COVID-19 vaccine (mean, 3 days) appears to correspond to the timing of peak postvaccination pro-inflammatory cytokine elevation.¹⁴ While all VAMP cases are serious events, the majority of patients appear to have a relatively benign clinical course with rapid and full recovery.¹³

Patients who have experienced an inflammatory cardiac injury may be at higher risk for recurrence, but quantifying risk of this rare phenomenon is challenging. Cases of VAMP after the COVID-19 vaccine have occasionally been reported in patients with previous cardiac injury unrelated to vaccination.^15-17 The cases presented here represent the first report of recurrent VAMP following prior non-COVID-19 vaccinations.

Most patients with prior VAMP in this cohort did not experience cardiac-suggestive symptoms following COVID-19 vaccination. Among 11 patients who developed symptoms, 3 had confirmed myocarditis and 1 had confirmed pericarditis. The clinical course for these patients with recurrent VAMP was observed to be no different in severity or duration from those who experience new-onset VAMP.⁴ All other patients not meeting criteria for VAMP or having alternative explanations for their symptoms also had a benign clinical course. Nonetheless, of the study cohort of 179, recurrent VAMP was diagnosed in 4 of the 11 who developed cardiac-suggestive symptoms following COVID-19 vaccination. The importance of cardiac evaluation should be emphasized for any patient presenting with chest pain, dyspnea, or other cardiac-suggestive symptoms following vaccination.

Strengths and Limitations

The strength of this review of VAMP recurrence associated with COVID-19 vaccination derives from our large and unique longitudinal database of VAMP among current and prior service members. Additionally, the IHD’s ongoing enhanced vaccine AEs surveillance provides the opportunity to contact patients and review their electronic health records over an extended interval of time.

When interpreting this report’s implications, limitations inherent to any retrospective case review should be considered. The cohort of cases of prior VAMP included primarily healthy, fit, young service members; this population is not representative of the general population. The cohort included prior VAMP cases that generally occurred after smallpox vaccination. Experiences after smallpox vaccine may not apply to cardiac injury from other vaccines or etiologies. By the nature of this review, the population studied at the time of COVID-19 vaccination was somewhat older than those most likely to develop an initial bout of VAMP.² This review was limited by information available in the electronic health records of a small number of patients. Subclinical cases of VAMP and cases without adequate clinical evaluation also could not be included.

Conclusions

Noninfectious inflammation of the heart (myocarditis, pericarditis, or myopericarditis) is a rare AE following certain vaccines, especially live replicating smallpox vaccine and mRNA COVID-19 vaccines. In this observational analysis, the majority of patients with previous VAMP successfully received a COVID-19 vaccine without recurrence. The 4 patients who were identified with recurrent VAMP following COVID-19 vaccination all recovered with supportive care. While the CDC endorses that individuals with a history of infectious myocarditis may receive COVID-19 vaccine after symptoms have resolved, there is currently insufficient safety data regarding COVID-19 vaccination of those with prior non-COVID-19 VAMP or following subsequent COVID-19 vaccination in those with prior VAMP related to COVID-19.¹⁰ For these individuals, COVID-19 vaccination is a precaution.¹⁰ Although insufficient to determine a precise level of risk, this report does provide data on which to base the CDC-recommended shared decision-making counseling of these patients. More research is needed to better define factors that increase risk for, or protection from, immune-mediated AEs following immunization, including VAMP. While benefits of vaccination have clearly outweighed risks during the COVID-19 pandemic, such research may optimize future vaccine recommendations.¹⁸

Vaccinations have substantially reduced morbidity and mortality from many infectious diseases. Despite the clear value of vaccinations in public health, efforts to better understand adverse events (AEs) following immunization are important to sustain public trust and vaccine confidence. Noninfectious inflammation of the heart may manifest as myocarditis or pericarditis, or occasionally, with shared signs and symptoms of each, as myopericarditis. This is a rare AE following some immunizations. Vaccine-associated myocarditis, pericarditis, or myopericarditis (VAMP) has been most clearly associated with smallpox vaccines and mRNA COVID-19 vaccines.^1-6 Although extremely rare, VAMP also has been associated with other vaccines.^7,8 Limited information exists to guide shared clinical decision making on COVID-19 vaccination in persons with a history of VAMP. It is unknown whether individuals with a history of VAMP are at higher risk for developing a recurrence or experiencing a more severe outcome following COVID-19 vaccination.

Methods

As part of the collaborative public health mission with the Centers for Disease Control and Prevention (CDC) for enhanced vaccine AE surveillance, the Defense Health Agency Immunization Healthcare Division (IHD) maintains a clinical database of service members and beneficiaries referred for suspected AEs following immunizations. A review of all AEs following immunization cases in this database from January 1, 2003, through February 28, 2022, identified individuals meeting the following criteria: (a) VAMP prior to receipt of COVID-19 vaccine; (b) receipt of COVID-19 vaccine in 2021; and (c) medical documentation in available electronic health records sufficient to describe health status at least 30 days following COVID-19 vaccination.⁹ If medical entries suggested cardiac symptoms following a COVID-19 vaccine, additional information was sought to verify VAMP based on current published criteria.^10,11 Both the initial VAMP cases and the suspected COVID-19 VAMP cases were adjudicated by a team of vaccine experts and specialists in immunology, cardiology, and preventive medicine.

This retrospective review was approved and conducted in accordance with the Walter Reed National Military Medical Center Institutional Review Board protocol #20664. All individuals with recurrent VAMP consented to share their health records and clinical details.

Results

Among 9260 cases in the IHD database, 431 met the case definition for VAMP.

Among the 179 patients included in this analysis, 171 (96%) were male. Their median age was 39 years at the time of COVID-19 vaccination.

Within 1 month of receipt of any COVID-19 vaccine, 11 individuals had documented symptoms suggesting cardiac involvement, specifically, chest pain, palpitations, or dyspnea. After cardiac evaluation, 4 patients met the criteria for VAMP after COVID-19 vaccination.^10,11 Seven patients either did not meet the criteria for VAMP or had alternative causes for their symptoms.

Two men aged 49 and 50 years with a history of vaccine-associated myocarditis following smallpox vaccination (Dryvax and ACAM2000) developed myocarditis 3 days after their second dose of the Moderna vaccine. One of these patients received a Pfizer-BioNTech booster 10 months later with no recurrence of symptoms. A 55-year-old man with a history of vaccine-associated myocarditis following Dryvax vaccination developed myocarditis 2 days after his Pfizer-BioNTech booster. None of the patients who developed post-COVID-19 VAMP reported residual symptoms from their initial VAMP episode, which occurred 12 to 18 years earlier. All were hospitalized briefly for observation and had complete symptom resolution within 6 weeks.

A 25-year-old man developed pericarditis 4 days after his second Pfizer-BioNTech vaccination. His previous ACAM2000 vaccine-associated myocarditis occurred 3 years earlier, with no residual symptoms. Of note, he had a mild COVID-19 infection 78 days before the onset of his pericarditis. After the onset of his COVID-19 vaccine-associated pericarditis, he continued to experience transient bouts of chest pressure and exertional dyspnea that resolved within 7 months of onset.

The median interval between COVID-19 vaccine doses in those who developed post-COVID-19 VAMP was within the recommended mRNA vaccine dosing intervals of 3 to 4 weeks and was consistent with the median mRNA vaccine dosing intervals among the entire cohort.

Due to the small cohort size and other limitations of this study, the suggested rate of cardiac injury in this review (4 cases in 179 persons, or 2.2%) is an imprecise estimate of risk in a small population (95% CI, 0.1%-4.4%). While this rate may seem higher than expected within the general population after COVID-19 vaccination, it is lower than the estimated lifetime risk of recurrent myocarditis from any cause.^6,12

Discussion

To our knowledge, this is the first report describing cardiac outcomes after COVID-19 vaccination among a cohort of individuals with prior history of VAMP. Four cases of COVID-19 VAMP were identified among 179 patients with previous VAMP. All cases had experienced VAMP after the smallpox vaccine several years earlier, with complete resolution of symptoms. Three cases presented with recurrent VAMP after their second dose of an mRNA COVID-19 vaccine, and one after an mRNA booster dose. All fully recovered over the course of several months.

Myocarditis is a heterogeneous inflammatory injury with diverse, sometimes idiopathic, etiologies.¹³ In contrast to infection-related cardiac injury, prior reports of vaccine-associated myocarditis have suggested a hypersensitivity reaction characterized by patchy eosinophilic infiltrates, a benign clinical course, and good prognosis.^2,3

There are several common features between VAMP after smallpox and COVID-19 vaccination. Cases occur predominantly in young men. The onset of symptoms after smallpox vaccine (mean, 10 days) and after mRNA COVID-19 vaccine (mean, 3 days) appears to correspond to the timing of peak postvaccination pro-inflammatory cytokine elevation.¹⁴ While all VAMP cases are serious events, the majority of patients appear to have a relatively benign clinical course with rapid and full recovery.¹³

Patients who have experienced an inflammatory cardiac injury may be at higher risk for recurrence, but quantifying risk of this rare phenomenon is challenging. Cases of VAMP after the COVID-19 vaccine have occasionally been reported in patients with previous cardiac injury unrelated to vaccination.^15-17 The cases presented here represent the first report of recurrent VAMP following prior non-COVID-19 vaccinations.

Most patients with prior VAMP in this cohort did not experience cardiac-suggestive symptoms following COVID-19 vaccination. Among 11 patients who developed symptoms, 3 had confirmed myocarditis and 1 had confirmed pericarditis. The clinical course for these patients with recurrent VAMP was observed to be no different in severity or duration from those who experience new-onset VAMP.⁴ All other patients not meeting criteria for VAMP or having alternative explanations for their symptoms also had a benign clinical course. Nonetheless, of the study cohort of 179, recurrent VAMP was diagnosed in 4 of the 11 who developed cardiac-suggestive symptoms following COVID-19 vaccination. The importance of cardiac evaluation should be emphasized for any patient presenting with chest pain, dyspnea, or other cardiac-suggestive symptoms following vaccination.

Strengths and Limitations

The strength of this review of VAMP recurrence associated with COVID-19 vaccination derives from our large and unique longitudinal database of VAMP among current and prior service members. Additionally, the IHD’s ongoing enhanced vaccine AEs surveillance provides the opportunity to contact patients and review their electronic health records over an extended interval of time.

When interpreting this report’s implications, limitations inherent to any retrospective case review should be considered. The cohort of cases of prior VAMP included primarily healthy, fit, young service members; this population is not representative of the general population. The cohort included prior VAMP cases that generally occurred after smallpox vaccination. Experiences after smallpox vaccine may not apply to cardiac injury from other vaccines or etiologies. By the nature of this review, the population studied at the time of COVID-19 vaccination was somewhat older than those most likely to develop an initial bout of VAMP.² This review was limited by information available in the electronic health records of a small number of patients. Subclinical cases of VAMP and cases without adequate clinical evaluation also could not be included.

Conclusions

Noninfectious inflammation of the heart (myocarditis, pericarditis, or myopericarditis) is a rare AE following certain vaccines, especially live replicating smallpox vaccine and mRNA COVID-19 vaccines. In this observational analysis, the majority of patients with previous VAMP successfully received a COVID-19 vaccine without recurrence. The 4 patients who were identified with recurrent VAMP following COVID-19 vaccination all recovered with supportive care. While the CDC endorses that individuals with a history of infectious myocarditis may receive COVID-19 vaccine after symptoms have resolved, there is currently insufficient safety data regarding COVID-19 vaccination of those with prior non-COVID-19 VAMP or following subsequent COVID-19 vaccination in those with prior VAMP related to COVID-19.¹⁰ For these individuals, COVID-19 vaccination is a precaution.¹⁰ Although insufficient to determine a precise level of risk, this report does provide data on which to base the CDC-recommended shared decision-making counseling of these patients. More research is needed to better define factors that increase risk for, or protection from, immune-mediated AEs following immunization, including VAMP. While benefits of vaccination have clearly outweighed risks during the COVID-19 pandemic, such research may optimize future vaccine recommendations.¹⁸

Vancomycin is a commonly used glycopeptide antibiotic used to treat infections caused by gram-positive organisms. Vancomycin is most often used as a parenteral agent for empiric or definitive treatment of methicillin-resistant Staphylococcus aureus (MRSA). It can also be used for the treatment of other susceptible Staphylococcus or Enterococcus species. Adverse effects of parenteral vancomycin include infusion-related reactions, ototoxicity, and nephrotoxicity.¹ Higher vancomycin trough levels have been associated with an increased risk of nephrotoxicity.^1-4 The major safety concern with vancomycin is acute kidney injury (AKI). Even mild AKI can prolong hospitalizations, increase the cost of health care, and increase morbidity.²

In March 2020, the American Society of Health-System Pharmacists, the Infectious Diseases Society of America (IDSA), the Pediatric Infectious Disease Society, and the Society of Infectious Diseases Pharmacists released a consensus statement and guidelines regarding the optimization of vancomycin dosing and monitoring for patients with suspected or definitive serious MRSA infections. Based on these guidelines, it is recommended to target an individualized area under the curve/minimum inhibitory concentration (AUC/MIC) ratio of 400 to 600 mg × h/L to maximize clinical efficacy and minimize the risk of AKI.²

Before March 2020, the vancomycin monitoring recommendation was to target trough levels of 10 to 20 mg/L. A goal trough of 15 to 20 mg/L was recommended for severe infections, including sepsis, endocarditis, hospital-acquired pneumonia, meningitis, and osteomyelitis, caused by MRSA. A goal trough of 10 to 15 mg/L was recommended for noninvasive infections, such as skin and soft tissue infections and urinary tract infections, caused by MRSA. Targeting these trough levels was thought to achieve an AUC/MIC ≥ 400 mg × h/L.⁵ Evidence has since shown that trough values may not be an optimal marker for AUC/MIC values.²

The updated vancomycin therapeutic drug monitoring (TDM) guidelines recommend that health systems transition to AUC/MIC-guided monitoring for suspected or confirmed infections caused by MRSA. There is not enough evidence to recommend AUC/MIC-guided monitoring in patients with noninvasive infections or infections caused by other microbes.²

AUC/MIC-guided monitoring can be achieved in 2 ways. The first method is collecting C_max (peak level) and C_min (trough level) serum concentrations, preferably during the same dosing interval. Ideally, C_max should be drawn 1 to 2 hours after the vancomycin infusion and C_min should be drawn at the end of the dosing interval. First-order pharmacokinetic equations are used to estimate the AUC/MIC with this method. Bayesian software pharmacokinetic modeling based on 1 or 2 vancomycin concentrations with 1 trough level also can be used for monitoring. Preferably, 2 levels would be obtained to estimate the AUC/MIC when using Bayesian modeling.²

The bactericidal activity of vancomycin was achieved with AUC/MIC ratios of ≥ 400 mg × h/L. AUC/MIC ratios of < 400 mg × h/L increase the incidence of resistant and intermediate strains of S aureus. AUC/MIC-guided monitoring assumes an MIC of 1 mg/L. When the MIC is > 1 mg/L, it is less likely that an AUC/MIC ≥ 400 mg × h/L is achievable. Regardless of the TDM method used, AUC/MIC ratios ≥ 400 mg × h/L are not achievable with conventional dosing methods if the vancomycin MIC is > 2 mg/L in patients with normal renal function. Alternative therapy is recommended to be used for these patients.²

There are multiple studies investigating the therapeutic dosing of vancomycin and the associated incidence of AKI. Previous studies have correlated vancomycin AUC/MICs of 400 mg to 600 mg × h/L with clinical effectiveness.^2,6 In 2017, Neely and colleagues looked at the therapeutic dosing of vancomycin in 252 adults with ≥ 1 vancomycin level.⁷ During this prospective trial, they evaluated patients for 1 year and targeted trough concentrations of 10 to 20 mg/L with infection-specific goal ranges of 10 to 15 mg/L and 15 to 20 mg/L for noninvasive and invasive infections, respectively. They also targeted AUC/MIC ratios ≥ 400 mg × h/L regardless of trough concentration using Bayesian estimated AUC/MICs for 2 years. They found only 19% of trough concentrations to be therapeutic compared with 70% of AUC/MICs. A secondary outcome assessed by Neely and colleagues was nephrotoxicity, which was identified in 8% of patients with trough targets and 2% of patients with AUC/MIC targets.⁸

Previous studies evaluating the use of vancomycin in the veteran population have focused on AKI incidence, general nephrotoxicity, and 30-day readmission rates.^4,7,9,10 Poston-Blahnik and colleagues investigated the rates of AKI in 200 veterans using AUC/MIC-guided vancomycin TDM.⁵ They found an AKI incidence of 42% of patients with AUC/MICs ≥ 550 mg × h/L and 2% of patients with AUC/MICs < 550 mg × h/L.⁵ Gyamlani and colleagues investigated the rates of AKI in 33,527 veterans and found that serum vancomycin trough levels ≥ 20 mg/L were associated with a higher risk of AKI.⁸ Prabaker and colleagues investigated the association between vancomycin trough levels and nephrotoxicity, defined as 0.5 mg/L or a 50% increase in serum creatinine (sCr) in 348 veterans. They found nephrotoxicity in 8.9% of patients.¹⁰ Patel and colleagues investigated the effect of AKI on 30-day readmission rates in 216 veterans.¹⁰ AKI occurred in 8.8% of patients and of those 19.4% were readmitted within 30 days.¹⁰ Current literature lacks evidence regarding the comparison of the safety and efficacy of vancomycin trough-guided vs AUC/MIC-guided TDM in the veteran population. Therefore, the objective of this study was to investigate the differences in the safety and efficacy of vancomycin TDM in the veteran population based on the different monitoring methods used.

METHODS

This study was a retrospective, single-center, quasi-experimental chart review conducted at the Sioux Falls Veterans Affairs Health Care System (SFVAHCS) in South Dakota. Data were collected from the Computerized Patient Record System (CPRS). The SFVAHCS transitioned from trough-guided to AUC/MIC-guided TDM in November 2020.

Patients included in this study were veterans aged ≥ 18 years with orders for parenteral vancomycin between February 1, 2020, and October 31, 2020, for the trough-guided TDM group and between December 1, 2020, and August 31, 2021, for the AUC/MIC-guided TDM group. Patients with vancomycin courses initiated during November 2020 were excluded as both TDM methods were being used at that time. Patients were excluded if their vancomycin course began before February 1, 2020, for the trough-guided TDM group or began during November 2020 for the AUC/MIC-guided TDM group. Patients were excluded if their vancomycin course extended past October 31, 2020, for the trough group or past August 31, 2021, for the AUC/MIC group. Patients on dialysis or missing C_max, C_min, or sCr levels were excluded.

This study evaluated both safety (AKI incidence) and effectiveness (time spent in therapeutic range and time to therapeutic range). The primary endpoint was presence of vancomycin-induced AKI, which was based on the most recent Kidney Disease: Improving Global Outcomes (KDIGO) AKI definition: increased sCr of ≥ 0.3 mg/dL or by 50% from baseline sustained over 48 hours without any other explanation for the change.¹¹ A secondary endpoint was the absence or presence of AKI.

Additional secondary endpoints included the presence of the initial trough or AUC/MIC of each vancomycin course within the therapeutic range and the percentage of all trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges. The therapeutic range for AUC/MIC-guided TDM was 400 to 600 mg × h/L and 10 to 20 mg/L depending on indication for trough-guided TDM (15-20 mg/L for severe infections and 10-15 mg/L for less invasive infections). The percentage of trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges were calculated as a ratio of levels within each range to total levels taken for each patient.

For AUC/MIC-guided TDM the C_max levels were ideally drawn 1 to 2 hours after vancomycin infusion and C_min levels were ideally drawn 30 minutes before the next dose. First-order pharmacokinetic equations were used to estimate the AUC/MIC.¹² If the timing of a vancomycin level was inappropriate, actual levels were extrapolated based on the timing of the blood draw compared with the ideal C_min or C_max time. Extrapolated levels were used for both trough-guided and AUC/MIC-guided TDM groups when appropriate. Vancomycin levels were excluded if they were drawn during the vancomycin infusion.

Study participant age, sex, race, weight, baseline estimated glomerular filtration (eGFR) rate, baseline sCr, concomitant nephrotoxic medications, duration of vancomycin course, indication of vancomycin, and acuity of illness based on indication were collected. sCr levels were collected from the initial day vancomycin was ordered through 72 hours following completion of a vancomycin course to evaluate for AKI. Patients’ charts were reviewed for the use of the following nephrotoxic medications: nonsteroidal anti-inflammatories, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, aminoglycosides, piperacillin/tazobactam, loop diuretics, amphotericin B, acyclovir, intravenous contrast, and nephrotoxic chemotherapy (cisplatin). The category of concomitant nephrotoxic medications was also collected including the continuation of a home nephrotoxic medication vs the initiation of a new nephrotoxic medication.

Statistical Analysis

The primary endpoint of the incidence of vancomycin-induced AKI was compared using a Fisher exact test. The secondary endpoint of the percentage of trough levels or AUC/MICs in the therapeutic, subtherapeutic, and supratherapeutic range were compared using a student t test. The secondary endpoint of first level or AUC/MIC within goal range was compared using a χ² test. Continuous baseline characteristics were reported as a mean and compared using a student t test. Nominal baseline characteristics were reported as a percentage and compared using the χ² test. P values < .05 were considered statistically significant.

RESULTS

This study included 97 patients, 43 in the AUC/MIC group and 54 in the trough group.

One (2%) patient in the AUC/MIC group and 2 (4%) patients in the trough group experienced vancomycin-induced AKI (P = .10) (Table 2).

DISCUSSION

There was no statistically significant difference between the 2 groups for the vancomycin-induced AKI (P = .10), the primary endpoint, or overall AKI (P = .29), the secondary endpoint. It should be noted that there was more overall AKI in the AUC/MIC group. Veterans in the AUC/MIC group were found to have their first AUC/MIC within the therapeutic range statistically significantly more often than the first trough level in the trough group (P = .04). The percentage of time spent within therapeutic range was statistically significantly higher in the AUC/MIC-guided TDM group (P = .02). The percentage of time spent subtherapeutic of goal range was statistically significantly higher in the trough-guided TDM group (P < .001). There was no statistically significant difference found in the percent of time spent supratherapeutic of goal range (P = .25). However, the observed percentage of time spent supratherapeutic of goal range was higher in the AUC/MIC group. These results indicate that AUC/MIC-guided TDM may be more efficacious with regard to time in therapeutic range and time to therapeutic range.

The finding of increased AKI with AUC/MIC-guided TDM does not align with previous studies.⁸ The prospective study by Neely and colleagues found that AUC/MIC-guided TDM resulted in more time in the therapeutic range as well as less nephrotoxicity compared with trough-guided TDM, although it was limited by its lack of randomization and did not account for other causes of nephrotoxicity.⁸ They found that only 19% of trough concentrations were therapeutic compared with 70% of AUC/MICs and found nephrotoxicity in 8% of trough-guided TDM patients compared with 2% of AUC/MIC-guided TDM patients.⁸

Unlike Nealy and colleagues, our study did not find lower nephrotoxicity associated with AUC/MIC-guided TDM. Multiple factors may have influenced our results. Our AUC/MIC group had significantly more newly started concomitant nephrotoxins and other nephrotoxic medications used during the vancomycin courses compared with the trough-guided group, which may have influenced AKI outcomes. It also should be noted that there was significantly more time spent subtherapeutic of the goal range and significantly less time in the goal range in the trough group compared with the AUC/MIC group. In our study, the trough-guided group had significantly more patients with acute illness compared with the AUC/MIC group (skin, soft tissue, and joint infections were similar between the groups). The group with more acutely ill patients would have been expected to have more nephrotoxicity. However, despite the acute illnesses, patients in the trough-guided group spent more time in the subtherapeutic range. This may explain the increased nephrotoxicity in the AUC/MIC group since those patients spent more time in the therapeutic range.

This study used the most recent KDIGO AKI definition: either an increase in sCr of ≥ 0.3 mg/dL or a 50% increase in sCr from baseline sustained over 48 hours without any other explanation for the change in renal function.¹¹ This AKI definition is stricter than the previous definition, which was used by earlier studies, including Neely and colleagues, to evaluate rates of vancomycin-induced AKI.^2,3 Therefore, the rates of overall AKI found in this study may be higher than in previous studies due to the definition of AKI used.

Limitations

This study was limited by its retrospective nature, lack of randomization, and small sample size. To decrease the potential for error in this study, analysis of power and a larger study sample would have been beneficial. During the COVID-19 pandemic, increased pneumonia cases may have hidden bacterial causes and caused an undercount. Nephrotoxicity may also be related to volume depletion, severe systemic illness, dehydration, or hypotension. Screening was completed via chart review for these alternative causes of nephrotoxicity in this study but may not be completely accounted for due to lack of documentation and the retrospective nature of this study.

CONCLUSIONS

This study did not find a significant difference in the rates of vancomycin-induced or overall AKI between AUC/MIC-guided and trough-guided TDM. However, this study may not have been powered to detect a significant difference in the primary endpoint. This study indicated that AUC/MIC-guided TDM of vancomycin resulted in a quicker time to the therapeutic range and a higher percentage of overall time in the therapeutic range as compared with trough-guided TDM. The results of this study indicated that trough-guided monitoring resulted in a higher percentage of time in a subtherapeutic range. This study also found that the first AUC/MIC calculated was within therapeutic range more often than the first trough level collected.

These results indicate that AUC/MIC-guided TDM may be more effective than trough-guided TDM in the veteran population. However, while AUC/MIC-guided TDM may be more effective with regards to time in therapeutic range and time to therapeutic range, this study did not indicate any safety benefit of AUC/MIC-guided over trough-guided TDM with regards to AKI incidence. Our data indicate that AUC/MIC-guided TDM increases the amount of time in the therapeutic range compared with trough-guided TDM and is not more nephrotoxic. The findings of this study support the recommendation to transition to the use of AUC/MIC-guided TDM of vancomycin in the veteran population.

Acknowledgments

This material is the result of work supported with the use of facilities and resources from the Sioux Falls Veterans Affairs Health Care System.

Vancomycin is a commonly used glycopeptide antibiotic used to treat infections caused by gram-positive organisms. Vancomycin is most often used as a parenteral agent for empiric or definitive treatment of methicillin-resistant Staphylococcus aureus (MRSA). It can also be used for the treatment of other susceptible Staphylococcus or Enterococcus species. Adverse effects of parenteral vancomycin include infusion-related reactions, ototoxicity, and nephrotoxicity.¹ Higher vancomycin trough levels have been associated with an increased risk of nephrotoxicity.^1-4 The major safety concern with vancomycin is acute kidney injury (AKI). Even mild AKI can prolong hospitalizations, increase the cost of health care, and increase morbidity.²

In March 2020, the American Society of Health-System Pharmacists, the Infectious Diseases Society of America (IDSA), the Pediatric Infectious Disease Society, and the Society of Infectious Diseases Pharmacists released a consensus statement and guidelines regarding the optimization of vancomycin dosing and monitoring for patients with suspected or definitive serious MRSA infections. Based on these guidelines, it is recommended to target an individualized area under the curve/minimum inhibitory concentration (AUC/MIC) ratio of 400 to 600 mg × h/L to maximize clinical efficacy and minimize the risk of AKI.²

Before March 2020, the vancomycin monitoring recommendation was to target trough levels of 10 to 20 mg/L. A goal trough of 15 to 20 mg/L was recommended for severe infections, including sepsis, endocarditis, hospital-acquired pneumonia, meningitis, and osteomyelitis, caused by MRSA. A goal trough of 10 to 15 mg/L was recommended for noninvasive infections, such as skin and soft tissue infections and urinary tract infections, caused by MRSA. Targeting these trough levels was thought to achieve an AUC/MIC ≥ 400 mg × h/L.⁵ Evidence has since shown that trough values may not be an optimal marker for AUC/MIC values.²

The updated vancomycin therapeutic drug monitoring (TDM) guidelines recommend that health systems transition to AUC/MIC-guided monitoring for suspected or confirmed infections caused by MRSA. There is not enough evidence to recommend AUC/MIC-guided monitoring in patients with noninvasive infections or infections caused by other microbes.²

AUC/MIC-guided monitoring can be achieved in 2 ways. The first method is collecting C_max (peak level) and C_min (trough level) serum concentrations, preferably during the same dosing interval. Ideally, C_max should be drawn 1 to 2 hours after the vancomycin infusion and C_min should be drawn at the end of the dosing interval. First-order pharmacokinetic equations are used to estimate the AUC/MIC with this method. Bayesian software pharmacokinetic modeling based on 1 or 2 vancomycin concentrations with 1 trough level also can be used for monitoring. Preferably, 2 levels would be obtained to estimate the AUC/MIC when using Bayesian modeling.²

The bactericidal activity of vancomycin was achieved with AUC/MIC ratios of ≥ 400 mg × h/L. AUC/MIC ratios of < 400 mg × h/L increase the incidence of resistant and intermediate strains of S aureus. AUC/MIC-guided monitoring assumes an MIC of 1 mg/L. When the MIC is > 1 mg/L, it is less likely that an AUC/MIC ≥ 400 mg × h/L is achievable. Regardless of the TDM method used, AUC/MIC ratios ≥ 400 mg × h/L are not achievable with conventional dosing methods if the vancomycin MIC is > 2 mg/L in patients with normal renal function. Alternative therapy is recommended to be used for these patients.²

There are multiple studies investigating the therapeutic dosing of vancomycin and the associated incidence of AKI. Previous studies have correlated vancomycin AUC/MICs of 400 mg to 600 mg × h/L with clinical effectiveness.^2,6 In 2017, Neely and colleagues looked at the therapeutic dosing of vancomycin in 252 adults with ≥ 1 vancomycin level.⁷ During this prospective trial, they evaluated patients for 1 year and targeted trough concentrations of 10 to 20 mg/L with infection-specific goal ranges of 10 to 15 mg/L and 15 to 20 mg/L for noninvasive and invasive infections, respectively. They also targeted AUC/MIC ratios ≥ 400 mg × h/L regardless of trough concentration using Bayesian estimated AUC/MICs for 2 years. They found only 19% of trough concentrations to be therapeutic compared with 70% of AUC/MICs. A secondary outcome assessed by Neely and colleagues was nephrotoxicity, which was identified in 8% of patients with trough targets and 2% of patients with AUC/MIC targets.⁸

Previous studies evaluating the use of vancomycin in the veteran population have focused on AKI incidence, general nephrotoxicity, and 30-day readmission rates.^4,7,9,10 Poston-Blahnik and colleagues investigated the rates of AKI in 200 veterans using AUC/MIC-guided vancomycin TDM.⁵ They found an AKI incidence of 42% of patients with AUC/MICs ≥ 550 mg × h/L and 2% of patients with AUC/MICs < 550 mg × h/L.⁵ Gyamlani and colleagues investigated the rates of AKI in 33,527 veterans and found that serum vancomycin trough levels ≥ 20 mg/L were associated with a higher risk of AKI.⁸ Prabaker and colleagues investigated the association between vancomycin trough levels and nephrotoxicity, defined as 0.5 mg/L or a 50% increase in serum creatinine (sCr) in 348 veterans. They found nephrotoxicity in 8.9% of patients.¹⁰ Patel and colleagues investigated the effect of AKI on 30-day readmission rates in 216 veterans.¹⁰ AKI occurred in 8.8% of patients and of those 19.4% were readmitted within 30 days.¹⁰ Current literature lacks evidence regarding the comparison of the safety and efficacy of vancomycin trough-guided vs AUC/MIC-guided TDM in the veteran population. Therefore, the objective of this study was to investigate the differences in the safety and efficacy of vancomycin TDM in the veteran population based on the different monitoring methods used.

METHODS

This study was a retrospective, single-center, quasi-experimental chart review conducted at the Sioux Falls Veterans Affairs Health Care System (SFVAHCS) in South Dakota. Data were collected from the Computerized Patient Record System (CPRS). The SFVAHCS transitioned from trough-guided to AUC/MIC-guided TDM in November 2020.

Patients included in this study were veterans aged ≥ 18 years with orders for parenteral vancomycin between February 1, 2020, and October 31, 2020, for the trough-guided TDM group and between December 1, 2020, and August 31, 2021, for the AUC/MIC-guided TDM group. Patients with vancomycin courses initiated during November 2020 were excluded as both TDM methods were being used at that time. Patients were excluded if their vancomycin course began before February 1, 2020, for the trough-guided TDM group or began during November 2020 for the AUC/MIC-guided TDM group. Patients were excluded if their vancomycin course extended past October 31, 2020, for the trough group or past August 31, 2021, for the AUC/MIC group. Patients on dialysis or missing C_max, C_min, or sCr levels were excluded.

This study evaluated both safety (AKI incidence) and effectiveness (time spent in therapeutic range and time to therapeutic range). The primary endpoint was presence of vancomycin-induced AKI, which was based on the most recent Kidney Disease: Improving Global Outcomes (KDIGO) AKI definition: increased sCr of ≥ 0.3 mg/dL or by 50% from baseline sustained over 48 hours without any other explanation for the change.¹¹ A secondary endpoint was the absence or presence of AKI.

Additional secondary endpoints included the presence of the initial trough or AUC/MIC of each vancomycin course within the therapeutic range and the percentage of all trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges. The therapeutic range for AUC/MIC-guided TDM was 400 to 600 mg × h/L and 10 to 20 mg/L depending on indication for trough-guided TDM (15-20 mg/L for severe infections and 10-15 mg/L for less invasive infections). The percentage of trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges were calculated as a ratio of levels within each range to total levels taken for each patient.

For AUC/MIC-guided TDM the C_max levels were ideally drawn 1 to 2 hours after vancomycin infusion and C_min levels were ideally drawn 30 minutes before the next dose. First-order pharmacokinetic equations were used to estimate the AUC/MIC.¹² If the timing of a vancomycin level was inappropriate, actual levels were extrapolated based on the timing of the blood draw compared with the ideal C_min or C_max time. Extrapolated levels were used for both trough-guided and AUC/MIC-guided TDM groups when appropriate. Vancomycin levels were excluded if they were drawn during the vancomycin infusion.

Study participant age, sex, race, weight, baseline estimated glomerular filtration (eGFR) rate, baseline sCr, concomitant nephrotoxic medications, duration of vancomycin course, indication of vancomycin, and acuity of illness based on indication were collected. sCr levels were collected from the initial day vancomycin was ordered through 72 hours following completion of a vancomycin course to evaluate for AKI. Patients’ charts were reviewed for the use of the following nephrotoxic medications: nonsteroidal anti-inflammatories, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, aminoglycosides, piperacillin/tazobactam, loop diuretics, amphotericin B, acyclovir, intravenous contrast, and nephrotoxic chemotherapy (cisplatin). The category of concomitant nephrotoxic medications was also collected including the continuation of a home nephrotoxic medication vs the initiation of a new nephrotoxic medication.

Statistical Analysis

The primary endpoint of the incidence of vancomycin-induced AKI was compared using a Fisher exact test. The secondary endpoint of the percentage of trough levels or AUC/MICs in the therapeutic, subtherapeutic, and supratherapeutic range were compared using a student t test. The secondary endpoint of first level or AUC/MIC within goal range was compared using a χ² test. Continuous baseline characteristics were reported as a mean and compared using a student t test. Nominal baseline characteristics were reported as a percentage and compared using the χ² test. P values < .05 were considered statistically significant.

RESULTS

This study included 97 patients, 43 in the AUC/MIC group and 54 in the trough group.

One (2%) patient in the AUC/MIC group and 2 (4%) patients in the trough group experienced vancomycin-induced AKI (P = .10) (Table 2).

DISCUSSION

There was no statistically significant difference between the 2 groups for the vancomycin-induced AKI (P = .10), the primary endpoint, or overall AKI (P = .29), the secondary endpoint. It should be noted that there was more overall AKI in the AUC/MIC group. Veterans in the AUC/MIC group were found to have their first AUC/MIC within the therapeutic range statistically significantly more often than the first trough level in the trough group (P = .04). The percentage of time spent within therapeutic range was statistically significantly higher in the AUC/MIC-guided TDM group (P = .02). The percentage of time spent subtherapeutic of goal range was statistically significantly higher in the trough-guided TDM group (P < .001). There was no statistically significant difference found in the percent of time spent supratherapeutic of goal range (P = .25). However, the observed percentage of time spent supratherapeutic of goal range was higher in the AUC/MIC group. These results indicate that AUC/MIC-guided TDM may be more efficacious with regard to time in therapeutic range and time to therapeutic range.

The finding of increased AKI with AUC/MIC-guided TDM does not align with previous studies.⁸ The prospective study by Neely and colleagues found that AUC/MIC-guided TDM resulted in more time in the therapeutic range as well as less nephrotoxicity compared with trough-guided TDM, although it was limited by its lack of randomization and did not account for other causes of nephrotoxicity.⁸ They found that only 19% of trough concentrations were therapeutic compared with 70% of AUC/MICs and found nephrotoxicity in 8% of trough-guided TDM patients compared with 2% of AUC/MIC-guided TDM patients.⁸

Unlike Nealy and colleagues, our study did not find lower nephrotoxicity associated with AUC/MIC-guided TDM. Multiple factors may have influenced our results. Our AUC/MIC group had significantly more newly started concomitant nephrotoxins and other nephrotoxic medications used during the vancomycin courses compared with the trough-guided group, which may have influenced AKI outcomes. It also should be noted that there was significantly more time spent subtherapeutic of the goal range and significantly less time in the goal range in the trough group compared with the AUC/MIC group. In our study, the trough-guided group had significantly more patients with acute illness compared with the AUC/MIC group (skin, soft tissue, and joint infections were similar between the groups). The group with more acutely ill patients would have been expected to have more nephrotoxicity. However, despite the acute illnesses, patients in the trough-guided group spent more time in the subtherapeutic range. This may explain the increased nephrotoxicity in the AUC/MIC group since those patients spent more time in the therapeutic range.

This study used the most recent KDIGO AKI definition: either an increase in sCr of ≥ 0.3 mg/dL or a 50% increase in sCr from baseline sustained over 48 hours without any other explanation for the change in renal function.¹¹ This AKI definition is stricter than the previous definition, which was used by earlier studies, including Neely and colleagues, to evaluate rates of vancomycin-induced AKI.^2,3 Therefore, the rates of overall AKI found in this study may be higher than in previous studies due to the definition of AKI used.

Limitations

This study was limited by its retrospective nature, lack of randomization, and small sample size. To decrease the potential for error in this study, analysis of power and a larger study sample would have been beneficial. During the COVID-19 pandemic, increased pneumonia cases may have hidden bacterial causes and caused an undercount. Nephrotoxicity may also be related to volume depletion, severe systemic illness, dehydration, or hypotension. Screening was completed via chart review for these alternative causes of nephrotoxicity in this study but may not be completely accounted for due to lack of documentation and the retrospective nature of this study.

CONCLUSIONS

This study did not find a significant difference in the rates of vancomycin-induced or overall AKI between AUC/MIC-guided and trough-guided TDM. However, this study may not have been powered to detect a significant difference in the primary endpoint. This study indicated that AUC/MIC-guided TDM of vancomycin resulted in a quicker time to the therapeutic range and a higher percentage of overall time in the therapeutic range as compared with trough-guided TDM. The results of this study indicated that trough-guided monitoring resulted in a higher percentage of time in a subtherapeutic range. This study also found that the first AUC/MIC calculated was within therapeutic range more often than the first trough level collected.

These results indicate that AUC/MIC-guided TDM may be more effective than trough-guided TDM in the veteran population. However, while AUC/MIC-guided TDM may be more effective with regards to time in therapeutic range and time to therapeutic range, this study did not indicate any safety benefit of AUC/MIC-guided over trough-guided TDM with regards to AKI incidence. Our data indicate that AUC/MIC-guided TDM increases the amount of time in the therapeutic range compared with trough-guided TDM and is not more nephrotoxic. The findings of this study support the recommendation to transition to the use of AUC/MIC-guided TDM of vancomycin in the veteran population.

Acknowledgments

This material is the result of work supported with the use of facilities and resources from the Sioux Falls Veterans Affairs Health Care System.

Vancomycin is a commonly used glycopeptide antibiotic used to treat infections caused by gram-positive organisms. Vancomycin is most often used as a parenteral agent for empiric or definitive treatment of methicillin-resistant Staphylococcus aureus (MRSA). It can also be used for the treatment of other susceptible Staphylococcus or Enterococcus species. Adverse effects of parenteral vancomycin include infusion-related reactions, ototoxicity, and nephrotoxicity.¹ Higher vancomycin trough levels have been associated with an increased risk of nephrotoxicity.^1-4 The major safety concern with vancomycin is acute kidney injury (AKI). Even mild AKI can prolong hospitalizations, increase the cost of health care, and increase morbidity.²

In March 2020, the American Society of Health-System Pharmacists, the Infectious Diseases Society of America (IDSA), the Pediatric Infectious Disease Society, and the Society of Infectious Diseases Pharmacists released a consensus statement and guidelines regarding the optimization of vancomycin dosing and monitoring for patients with suspected or definitive serious MRSA infections. Based on these guidelines, it is recommended to target an individualized area under the curve/minimum inhibitory concentration (AUC/MIC) ratio of 400 to 600 mg × h/L to maximize clinical efficacy and minimize the risk of AKI.²

Before March 2020, the vancomycin monitoring recommendation was to target trough levels of 10 to 20 mg/L. A goal trough of 15 to 20 mg/L was recommended for severe infections, including sepsis, endocarditis, hospital-acquired pneumonia, meningitis, and osteomyelitis, caused by MRSA. A goal trough of 10 to 15 mg/L was recommended for noninvasive infections, such as skin and soft tissue infections and urinary tract infections, caused by MRSA. Targeting these trough levels was thought to achieve an AUC/MIC ≥ 400 mg × h/L.⁵ Evidence has since shown that trough values may not be an optimal marker for AUC/MIC values.²

The updated vancomycin therapeutic drug monitoring (TDM) guidelines recommend that health systems transition to AUC/MIC-guided monitoring for suspected or confirmed infections caused by MRSA. There is not enough evidence to recommend AUC/MIC-guided monitoring in patients with noninvasive infections or infections caused by other microbes.²

AUC/MIC-guided monitoring can be achieved in 2 ways. The first method is collecting C_max (peak level) and C_min (trough level) serum concentrations, preferably during the same dosing interval. Ideally, C_max should be drawn 1 to 2 hours after the vancomycin infusion and C_min should be drawn at the end of the dosing interval. First-order pharmacokinetic equations are used to estimate the AUC/MIC with this method. Bayesian software pharmacokinetic modeling based on 1 or 2 vancomycin concentrations with 1 trough level also can be used for monitoring. Preferably, 2 levels would be obtained to estimate the AUC/MIC when using Bayesian modeling.²

The bactericidal activity of vancomycin was achieved with AUC/MIC ratios of ≥ 400 mg × h/L. AUC/MIC ratios of < 400 mg × h/L increase the incidence of resistant and intermediate strains of S aureus. AUC/MIC-guided monitoring assumes an MIC of 1 mg/L. When the MIC is > 1 mg/L, it is less likely that an AUC/MIC ≥ 400 mg × h/L is achievable. Regardless of the TDM method used, AUC/MIC ratios ≥ 400 mg × h/L are not achievable with conventional dosing methods if the vancomycin MIC is > 2 mg/L in patients with normal renal function. Alternative therapy is recommended to be used for these patients.²

There are multiple studies investigating the therapeutic dosing of vancomycin and the associated incidence of AKI. Previous studies have correlated vancomycin AUC/MICs of 400 mg to 600 mg × h/L with clinical effectiveness.^2,6 In 2017, Neely and colleagues looked at the therapeutic dosing of vancomycin in 252 adults with ≥ 1 vancomycin level.⁷ During this prospective trial, they evaluated patients for 1 year and targeted trough concentrations of 10 to 20 mg/L with infection-specific goal ranges of 10 to 15 mg/L and 15 to 20 mg/L for noninvasive and invasive infections, respectively. They also targeted AUC/MIC ratios ≥ 400 mg × h/L regardless of trough concentration using Bayesian estimated AUC/MICs for 2 years. They found only 19% of trough concentrations to be therapeutic compared with 70% of AUC/MICs. A secondary outcome assessed by Neely and colleagues was nephrotoxicity, which was identified in 8% of patients with trough targets and 2% of patients with AUC/MIC targets.⁸

Previous studies evaluating the use of vancomycin in the veteran population have focused on AKI incidence, general nephrotoxicity, and 30-day readmission rates.^4,7,9,10 Poston-Blahnik and colleagues investigated the rates of AKI in 200 veterans using AUC/MIC-guided vancomycin TDM.⁵ They found an AKI incidence of 42% of patients with AUC/MICs ≥ 550 mg × h/L and 2% of patients with AUC/MICs < 550 mg × h/L.⁵ Gyamlani and colleagues investigated the rates of AKI in 33,527 veterans and found that serum vancomycin trough levels ≥ 20 mg/L were associated with a higher risk of AKI.⁸ Prabaker and colleagues investigated the association between vancomycin trough levels and nephrotoxicity, defined as 0.5 mg/L or a 50% increase in serum creatinine (sCr) in 348 veterans. They found nephrotoxicity in 8.9% of patients.¹⁰ Patel and colleagues investigated the effect of AKI on 30-day readmission rates in 216 veterans.¹⁰ AKI occurred in 8.8% of patients and of those 19.4% were readmitted within 30 days.¹⁰ Current literature lacks evidence regarding the comparison of the safety and efficacy of vancomycin trough-guided vs AUC/MIC-guided TDM in the veteran population. Therefore, the objective of this study was to investigate the differences in the safety and efficacy of vancomycin TDM in the veteran population based on the different monitoring methods used.

METHODS

This study was a retrospective, single-center, quasi-experimental chart review conducted at the Sioux Falls Veterans Affairs Health Care System (SFVAHCS) in South Dakota. Data were collected from the Computerized Patient Record System (CPRS). The SFVAHCS transitioned from trough-guided to AUC/MIC-guided TDM in November 2020.

Patients included in this study were veterans aged ≥ 18 years with orders for parenteral vancomycin between February 1, 2020, and October 31, 2020, for the trough-guided TDM group and between December 1, 2020, and August 31, 2021, for the AUC/MIC-guided TDM group. Patients with vancomycin courses initiated during November 2020 were excluded as both TDM methods were being used at that time. Patients were excluded if their vancomycin course began before February 1, 2020, for the trough-guided TDM group or began during November 2020 for the AUC/MIC-guided TDM group. Patients were excluded if their vancomycin course extended past October 31, 2020, for the trough group or past August 31, 2021, for the AUC/MIC group. Patients on dialysis or missing C_max, C_min, or sCr levels were excluded.

This study evaluated both safety (AKI incidence) and effectiveness (time spent in therapeutic range and time to therapeutic range). The primary endpoint was presence of vancomycin-induced AKI, which was based on the most recent Kidney Disease: Improving Global Outcomes (KDIGO) AKI definition: increased sCr of ≥ 0.3 mg/dL or by 50% from baseline sustained over 48 hours without any other explanation for the change.¹¹ A secondary endpoint was the absence or presence of AKI.

Additional secondary endpoints included the presence of the initial trough or AUC/MIC of each vancomycin course within the therapeutic range and the percentage of all trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges. The therapeutic range for AUC/MIC-guided TDM was 400 to 600 mg × h/L and 10 to 20 mg/L depending on indication for trough-guided TDM (15-20 mg/L for severe infections and 10-15 mg/L for less invasive infections). The percentage of trough levels or AUC/MICs within therapeutic, subtherapeutic, and supratherapeutic ranges were calculated as a ratio of levels within each range to total levels taken for each patient.

For AUC/MIC-guided TDM the C_max levels were ideally drawn 1 to 2 hours after vancomycin infusion and C_min levels were ideally drawn 30 minutes before the next dose. First-order pharmacokinetic equations were used to estimate the AUC/MIC.¹² If the timing of a vancomycin level was inappropriate, actual levels were extrapolated based on the timing of the blood draw compared with the ideal C_min or C_max time. Extrapolated levels were used for both trough-guided and AUC/MIC-guided TDM groups when appropriate. Vancomycin levels were excluded if they were drawn during the vancomycin infusion.

Study participant age, sex, race, weight, baseline estimated glomerular filtration (eGFR) rate, baseline sCr, concomitant nephrotoxic medications, duration of vancomycin course, indication of vancomycin, and acuity of illness based on indication were collected. sCr levels were collected from the initial day vancomycin was ordered through 72 hours following completion of a vancomycin course to evaluate for AKI. Patients’ charts were reviewed for the use of the following nephrotoxic medications: nonsteroidal anti-inflammatories, angiotensin-converting enzyme inhibitors, angiotensin II receptor blockers, aminoglycosides, piperacillin/tazobactam, loop diuretics, amphotericin B, acyclovir, intravenous contrast, and nephrotoxic chemotherapy (cisplatin). The category of concomitant nephrotoxic medications was also collected including the continuation of a home nephrotoxic medication vs the initiation of a new nephrotoxic medication.

Statistical Analysis

The primary endpoint of the incidence of vancomycin-induced AKI was compared using a Fisher exact test. The secondary endpoint of the percentage of trough levels or AUC/MICs in the therapeutic, subtherapeutic, and supratherapeutic range were compared using a student t test. The secondary endpoint of first level or AUC/MIC within goal range was compared using a χ² test. Continuous baseline characteristics were reported as a mean and compared using a student t test. Nominal baseline characteristics were reported as a percentage and compared using the χ² test. P values < .05 were considered statistically significant.

RESULTS

This study included 97 patients, 43 in the AUC/MIC group and 54 in the trough group.

One (2%) patient in the AUC/MIC group and 2 (4%) patients in the trough group experienced vancomycin-induced AKI (P = .10) (Table 2).

DISCUSSION

There was no statistically significant difference between the 2 groups for the vancomycin-induced AKI (P = .10), the primary endpoint, or overall AKI (P = .29), the secondary endpoint. It should be noted that there was more overall AKI in the AUC/MIC group. Veterans in the AUC/MIC group were found to have their first AUC/MIC within the therapeutic range statistically significantly more often than the first trough level in the trough group (P = .04). The percentage of time spent within therapeutic range was statistically significantly higher in the AUC/MIC-guided TDM group (P = .02). The percentage of time spent subtherapeutic of goal range was statistically significantly higher in the trough-guided TDM group (P < .001). There was no statistically significant difference found in the percent of time spent supratherapeutic of goal range (P = .25). However, the observed percentage of time spent supratherapeutic of goal range was higher in the AUC/MIC group. These results indicate that AUC/MIC-guided TDM may be more efficacious with regard to time in therapeutic range and time to therapeutic range.

The finding of increased AKI with AUC/MIC-guided TDM does not align with previous studies.⁸ The prospective study by Neely and colleagues found that AUC/MIC-guided TDM resulted in more time in the therapeutic range as well as less nephrotoxicity compared with trough-guided TDM, although it was limited by its lack of randomization and did not account for other causes of nephrotoxicity.⁸ They found that only 19% of trough concentrations were therapeutic compared with 70% of AUC/MICs and found nephrotoxicity in 8% of trough-guided TDM patients compared with 2% of AUC/MIC-guided TDM patients.⁸

Unlike Nealy and colleagues, our study did not find lower nephrotoxicity associated with AUC/MIC-guided TDM. Multiple factors may have influenced our results. Our AUC/MIC group had significantly more newly started concomitant nephrotoxins and other nephrotoxic medications used during the vancomycin courses compared with the trough-guided group, which may have influenced AKI outcomes. It also should be noted that there was significantly more time spent subtherapeutic of the goal range and significantly less time in the goal range in the trough group compared with the AUC/MIC group. In our study, the trough-guided group had significantly more patients with acute illness compared with the AUC/MIC group (skin, soft tissue, and joint infections were similar between the groups). The group with more acutely ill patients would have been expected to have more nephrotoxicity. However, despite the acute illnesses, patients in the trough-guided group spent more time in the subtherapeutic range. This may explain the increased nephrotoxicity in the AUC/MIC group since those patients spent more time in the therapeutic range.

This study used the most recent KDIGO AKI definition: either an increase in sCr of ≥ 0.3 mg/dL or a 50% increase in sCr from baseline sustained over 48 hours without any other explanation for the change in renal function.¹¹ This AKI definition is stricter than the previous definition, which was used by earlier studies, including Neely and colleagues, to evaluate rates of vancomycin-induced AKI.^2,3 Therefore, the rates of overall AKI found in this study may be higher than in previous studies due to the definition of AKI used.

Limitations

This study was limited by its retrospective nature, lack of randomization, and small sample size. To decrease the potential for error in this study, analysis of power and a larger study sample would have been beneficial. During the COVID-19 pandemic, increased pneumonia cases may have hidden bacterial causes and caused an undercount. Nephrotoxicity may also be related to volume depletion, severe systemic illness, dehydration, or hypotension. Screening was completed via chart review for these alternative causes of nephrotoxicity in this study but may not be completely accounted for due to lack of documentation and the retrospective nature of this study.

CONCLUSIONS

This study did not find a significant difference in the rates of vancomycin-induced or overall AKI between AUC/MIC-guided and trough-guided TDM. However, this study may not have been powered to detect a significant difference in the primary endpoint. This study indicated that AUC/MIC-guided TDM of vancomycin resulted in a quicker time to the therapeutic range and a higher percentage of overall time in the therapeutic range as compared with trough-guided TDM. The results of this study indicated that trough-guided monitoring resulted in a higher percentage of time in a subtherapeutic range. This study also found that the first AUC/MIC calculated was within therapeutic range more often than the first trough level collected.

These results indicate that AUC/MIC-guided TDM may be more effective than trough-guided TDM in the veteran population. However, while AUC/MIC-guided TDM may be more effective with regards to time in therapeutic range and time to therapeutic range, this study did not indicate any safety benefit of AUC/MIC-guided over trough-guided TDM with regards to AKI incidence. Our data indicate that AUC/MIC-guided TDM increases the amount of time in the therapeutic range compared with trough-guided TDM and is not more nephrotoxic. The findings of this study support the recommendation to transition to the use of AUC/MIC-guided TDM of vancomycin in the veteran population.

Acknowledgments

This material is the result of work supported with the use of facilities and resources from the Sioux Falls Veterans Affairs Health Care System.

Pemphigus is a group of autoimmune blistering diseases characterized by the development of painful and flaccid blisters on the skin and/or mucous membranes. Pemphigus vulgaris (PV) and pemphigus foliaceus (PF) are 2 major subtypes and can be distinguished by the location of blister formation or the specificity of autoantibodies directed against different desmogleins.^1,2 Although rare, pemphigus is considered a serious and life-threatening condition with a great impact on quality of life (QOL) due to disease symptoms (eg, painful lesions, physical appearance of skin lesions) as well as treatment complications (eg, adverse drug effects, cost of treatment).^3-6 Moreover, the physical and psychological effects can lead to marked functional morbidity and work-related disability during patients’ productive years.⁷ Therefore, affected individuals usually have a remarkably compromised health-related quality of life (HRQOL).⁸ Effective treatments may considerably improve the QOL of patients with pemphigus.⁶

Despite the available treatment options, finding the best regimen for pemphigus remains a challenge. Corticosteroids are assumed to be the main treatment, though they have considerable side effects.^9,10 Adjuvant therapies are used to suppress or modulate immune responses, leading to remission with the least possible need for corticosteroids. Finding an optimal steroid-sparing agent has been the aim of research, and biologic agents seem to be the best option.⁸ Rituximab (RTX), an anti-CD20 monoclonal antibody, has shown great promise in several studies of its clinical efficacy and has become a first-line treatment in new guidelines.^11-14 Rituximab treatment has been associated with notable improvement in physician-assessed outcome measures with a favorable safety profile in patients with pemphigus.^11-15 However, it is important to assess response to treatment from a patient’s perspective through the use of outcome-assessment measures that encompass patient-reported outcomes to reflect the complete patient experience and establish the overall impact of RTX as well as its likelihood of acceptance by patients with pemphigus.

In our study, we compared clinical outcomes and HRQOL through the use of disease-specific measures as well as comprehensive generic health status measures among patients with PV and PF who received RTX treatment 3 months earlier and those who received RTX in the last 2 weeks. The clinical relevance of the patient-reported outcomes is discussed.

MATERIALS AND METHODS

Study Design

We conducted a single-center cross-sectional study of 96 patients with pemphigus aged 18 to 65 years of either sex who were willing to participate in this study. Patients with a confirmed diagnosis of PV or PF who received RTX 3 months earlier or in the last 2 weeks were enrolled in the study. Patients were identified using Dermatry.ir, an archiving software that contains patients’ medical data. Exclusion criteria included lack of sufficient knowledge of the concepts of the questionnaires as well as age younger than 16 years. The study was conducted from October 2019 to April 2020 by the Autoimmune Bullous Disease Research Center at Razi Hospital in Tehran, Iran, which is the main dermatology-specific center and teaching hospital of Iran. The study protocol was approved by the relevant ethics committee.

Patients were categorized into 2 groups: (1) those who received RTX 3 months earlier (3M group); and (2) those who received RTX in the last 2 weeks (R group).

After an explanation of the study to participants, informed written consent was signed by each patient, and their personal data (eg, age, sex, education, marital status, smoking status), as well as clinical data (eg, type of pemphigus, duration of disease, site of onset, prednisolone dosage, presence of Nikolsky sign, anti-DSG1 and anti-DSG3 values, Pemphigus Disease Area Index [PDAI] score, RTX treatment protocol); any known comorbidities such as hypertension, diabetes mellitus, or morbid obesity; and any chronic pulmonary, cardiac, endocrinologic, renal, or hepatic condition, were collected and recorded in a predefined Case Record.

Patient-Reported Outcome Measures

The effect of RTX on QOL in patients with pemphigus was assessed using 2 HRQOL instruments: (1) a general health status indicator, the 36-Item Short Form Survey (SF-36), and (2) a validated, Persian version of a dermatology-specific questionnaire, Dermatology Life Quality Index (DLQI). The questionnaires were completed by each patient or by an assistant if needed.

The SF-36 is a widely used 36-item questionnaire measuring functional health and well-being across 8 domains—mental health, pain, physical function, role emotional, role physical, social functioning, vitality, and general health perception—with scores for each ranging from 0 to 100. The physical component scores (PCSs) and mental component scores (MCSs) were derived from these 8 subscales, each ranging from 0 to 400, with higher scores indicating better health status.⁶

The DLQI, one of the most frequently used QOL measures in dermatology, contains 10 questions, each referring to the prior week and classified in the following 6 subscales: symptoms and feelings, daily activities, leisure, personal relationships, work and school, and treatment.¹⁶ The total score ranges from 0 (no impact) to 30 (very high impact), with a higher score indicating a lower QOL (eTable 1). The minimal clinically important difference (MCD) for the DLQI was considered to be 2- to 5-point changes in prior studies.^17,18 In this study, we used an MCD of a 5-point change or more between study groups.

Moreover, the patient general assessment (PGA) of disease severity was identified using a 3-point scale (1=mild, 2=moderate, 3=severe).

Statistical Analysis

Data were analyzed using SPSS Statistics version 23. P≤.05 was considered significant. Mean and SD were calculated for descriptive data. The t test, Fisher exact test, analysis of variance, multiple regression analysis, and logistic regression analysis were used to identify the relationship between variables.

RESULTS

Patient Characteristics

A total of 96 patients were enrolled in this study. The mean (SD) age of participants was 41.42 (15.1) years (range, 18–58 years). Of 96 patients whose data were included, 55 (57.29%) patients had received RTX 3 months earlier (3M group) and 41 (42.71%) received RTX in the last 2 weeks (R group). A summary of study patient characteristics in each group is provided in eTable 2. There was no significant difference between the 2 groups in terms of age, sex, type of pemphigus, marital status, education, positive Nikolsky sign, smoking status, existence of comorbidities, site of lesions, and RTX treatment protocol. However, a significant difference was found for duration of disease (P=.0124) and mean prednisolone dosage (P=.001) as well as severity of disease measured by PDAI score (P=.003) and anti-DSG1 (P=.003) and anti-DSG3 (P=.021) values.

Patient-Reported Outcomes

Physical and mental component scores are summarized in eTable 3. Generally, SF-36 scores were improved with RTX treatment in all dimensions except for mental health, though these differences were not statistically significant, with the greatest mean improvement in the role physical index (75.45 in the 3M group vs 53.04 in the R group; P=.009). Mean SF-36 PCS and MCS scores were higher in the 3M group vs the R group, though the difference in MCS score did not reach the level of significance (eTable 3).

Mean DLQI scores in the R and 3M groups were 12.31 and 6.96, respectively, indicating a considerable burden on HRQOL in both groups. However, a statistically significant difference between these values was seen that also was clinically meaningful, indicating a significant improvement of QOL in patients receiving RTX 3 months earlier (P=.005)(eTable 3).

The PGA scores indicated that patients in the 3M group were significantly more likely to report less severe disease vs the R group (P=.008)(eTable 3).

Multivariate Analysis—Effect of the patient characteristics and some disease features on indices of QOL was evaluated using the multiple linear regression model. eTable 4 shows the P values of those analyses.

COMMENT

Pemphigus is a chronic disabling disease with notable QOL impairment due to disease burden as well as the need for long-term use of immunosuppressive agents during the disease course. To study the effect of RTX on QOL of patients with pemphigus, we compared 2 sets of patients. Prior studies have shown that clinically significant effects of RTX take 4 to 12 weeks to appear.^19,20 Therefore, we selected patients who received RTX 3 months earlier to measure their HRQOL indices and compare them with patients who had received RTX in the last 2 weeks as a control group to investigate the effect of RTX intrinsically, as this was the focus of this study.

In our study, one of the research tools was the DLQI. Healthy patients typically have an average score of 0.5.²¹ The mean DLQI score of the patients in R group was 12.31, which was similar to prior analysis⁸ and reflects a substantial burden of disease comparable to atopic dermatitis and psoriasis.^21,22 In patients in the 3M group, the mean DLQI score was lower than the R group (6.96 vs 12.31), indicating a significant (P=.005) and clinically meaningful improvement in QOL of patients due to the dramatic therapeutic effect of RTX. However, this score indicated a moderate effect on HRQOL, even in the context of clinical improvement due to RTX treatment, which may reflect that the short duration of treatment in the 3M group was a limitation of this study. Although the 12-week treatment duration was comparable with other studies^19,20 and major differences in objective measures of treatment efficacy were found in PDAI as well as anti-DSG1 and anti-DSG3 values, longer treatment duration may be needed for a more comprehensive assessment of the benefit of RTX on HRQOL indices in patients with pemphigus.

Based on results of the SF-36 questionnaire, PCS and MCS scores were not substantially impaired in the R group considering the fact that a mean score of 50 has been articulated as a normative value for all scales.²³ These data demonstrated the importance of using a dermatologic-specific instrument such as the DLQI instead of a general questionnaire to assess QOL in patients with pemphigus. However, better indices were reported with RTX treatment in the 3 SF-36 domains—role physical (P=.009), role emotional (P=.03), and general health perception (P=.03)—with the role physical showing the greatest magnitude of mean change (75.45 in the 3M group vs 53.04 in the R group). Notably, PCS was impaired to a greater extent than MCS in patients in the R group and showed a greater magnitude of improvement after 3 months of treatment. These results could be explained by the fact that MCS can be largely changed in diseases with a direct effect on the central nervous system.²³

Our results also revealed that the dose of corticosteroid correlated to HRQOL of patients with pemphigus who recently received RTX therapy. Indeed, it is more likely that patients on lower-dose prednisolone have a higher QOL, especially on physical function and social function dimensions of SF-36. This finding is highly expectable by less severe disease due to RTX treatment and also lower potential dose-dependent adverse effects of long-term steroid therapy.

One of the most striking findings of this study was the correlation of location of lesions to QOL indices. We found that the mucocutaneous phenotype was significantly correlated to greater improvement in role emotional, role physical, and social functioning scores due to RTX treatment compared with cutaneous or mucosal types (P=.02, P=.025, and P=.017, respectively). Although mucosal involvement of the disease can be the most burdensome feature because of its large impact on essential activities such as eating and speaking, cutaneous lesions with unpleasant appearance and undesirable symptoms may have a similar impact on QOL. Therefore, having both mucosal and cutaneous lesions causes a worsened QOL and decreased treatment efficacy vs having only one area involved. This may explain the greater improvement in some QOL indices with RTX treatment.

Limitations—Given the cross-sectional design of this study in which patients were observed at a single time point during their treatment course, it is not possible to establish a clear cause-effect relationship between variables. Moreover, we did not evaluate the impact of RTX or prednisolone adverse effects on QOL. Therefore, further prospective studies with longer treatment durations may help to validate our findings. In addition, MCDs for DLQI and SF-36 in pemphigus need to be determined and validated in future studies.

CONCLUSION

The results of our study demonstrated that patients with pemphigus may benefit from taking RTX, not only in terms of clinical improvement of their disease measured by objective indices such as PDAI and anti-DSG1 and anti-DSG3 values but also in several domains that are important to patients, including physical and mental health status (SF-36), HRQOL (DLQI), and overall disease severity (PGA). Rituximab administration in patients with pemphigus can lead to rapid and significant improvement in HRQOL as well as patient- and physician-assessed measures. Its favorable safety profile along with its impact on patients’ daily lives and mental health makes RTX a suitable treatment option for patients with pemphigus. Moreover, we recommend taking QOL indices into account while evaluating the efficacy of new medications to improve our insight into the patient experience and provide better patient adherence to treatment, which is an important issue for optimal control of chronic disorders.

Pemphigus is a group of autoimmune blistering diseases characterized by the development of painful and flaccid blisters on the skin and/or mucous membranes. Pemphigus vulgaris (PV) and pemphigus foliaceus (PF) are 2 major subtypes and can be distinguished by the location of blister formation or the specificity of autoantibodies directed against different desmogleins.^1,2 Although rare, pemphigus is considered a serious and life-threatening condition with a great impact on quality of life (QOL) due to disease symptoms (eg, painful lesions, physical appearance of skin lesions) as well as treatment complications (eg, adverse drug effects, cost of treatment).^3-6 Moreover, the physical and psychological effects can lead to marked functional morbidity and work-related disability during patients’ productive years.⁷ Therefore, affected individuals usually have a remarkably compromised health-related quality of life (HRQOL).⁸ Effective treatments may considerably improve the QOL of patients with pemphigus.⁶

Despite the available treatment options, finding the best regimen for pemphigus remains a challenge. Corticosteroids are assumed to be the main treatment, though they have considerable side effects.^9,10 Adjuvant therapies are used to suppress or modulate immune responses, leading to remission with the least possible need for corticosteroids. Finding an optimal steroid-sparing agent has been the aim of research, and biologic agents seem to be the best option.⁸ Rituximab (RTX), an anti-CD20 monoclonal antibody, has shown great promise in several studies of its clinical efficacy and has become a first-line treatment in new guidelines.^11-14 Rituximab treatment has been associated with notable improvement in physician-assessed outcome measures with a favorable safety profile in patients with pemphigus.^11-15 However, it is important to assess response to treatment from a patient’s perspective through the use of outcome-assessment measures that encompass patient-reported outcomes to reflect the complete patient experience and establish the overall impact of RTX as well as its likelihood of acceptance by patients with pemphigus.

In our study, we compared clinical outcomes and HRQOL through the use of disease-specific measures as well as comprehensive generic health status measures among patients with PV and PF who received RTX treatment 3 months earlier and those who received RTX in the last 2 weeks. The clinical relevance of the patient-reported outcomes is discussed.

MATERIALS AND METHODS

Study Design

We conducted a single-center cross-sectional study of 96 patients with pemphigus aged 18 to 65 years of either sex who were willing to participate in this study. Patients with a confirmed diagnosis of PV or PF who received RTX 3 months earlier or in the last 2 weeks were enrolled in the study. Patients were identified using Dermatry.ir, an archiving software that contains patients’ medical data. Exclusion criteria included lack of sufficient knowledge of the concepts of the questionnaires as well as age younger than 16 years. The study was conducted from October 2019 to April 2020 by the Autoimmune Bullous Disease Research Center at Razi Hospital in Tehran, Iran, which is the main dermatology-specific center and teaching hospital of Iran. The study protocol was approved by the relevant ethics committee.

Patients were categorized into 2 groups: (1) those who received RTX 3 months earlier (3M group); and (2) those who received RTX in the last 2 weeks (R group).

After an explanation of the study to participants, informed written consent was signed by each patient, and their personal data (eg, age, sex, education, marital status, smoking status), as well as clinical data (eg, type of pemphigus, duration of disease, site of onset, prednisolone dosage, presence of Nikolsky sign, anti-DSG1 and anti-DSG3 values, Pemphigus Disease Area Index [PDAI] score, RTX treatment protocol); any known comorbidities such as hypertension, diabetes mellitus, or morbid obesity; and any chronic pulmonary, cardiac, endocrinologic, renal, or hepatic condition, were collected and recorded in a predefined Case Record.

Patient-Reported Outcome Measures

The effect of RTX on QOL in patients with pemphigus was assessed using 2 HRQOL instruments: (1) a general health status indicator, the 36-Item Short Form Survey (SF-36), and (2) a validated, Persian version of a dermatology-specific questionnaire, Dermatology Life Quality Index (DLQI). The questionnaires were completed by each patient or by an assistant if needed.

The SF-36 is a widely used 36-item questionnaire measuring functional health and well-being across 8 domains—mental health, pain, physical function, role emotional, role physical, social functioning, vitality, and general health perception—with scores for each ranging from 0 to 100. The physical component scores (PCSs) and mental component scores (MCSs) were derived from these 8 subscales, each ranging from 0 to 400, with higher scores indicating better health status.⁶

The DLQI, one of the most frequently used QOL measures in dermatology, contains 10 questions, each referring to the prior week and classified in the following 6 subscales: symptoms and feelings, daily activities, leisure, personal relationships, work and school, and treatment.¹⁶ The total score ranges from 0 (no impact) to 30 (very high impact), with a higher score indicating a lower QOL (eTable 1). The minimal clinically important difference (MCD) for the DLQI was considered to be 2- to 5-point changes in prior studies.^17,18 In this study, we used an MCD of a 5-point change or more between study groups.

Moreover, the patient general assessment (PGA) of disease severity was identified using a 3-point scale (1=mild, 2=moderate, 3=severe).

Statistical Analysis

Data were analyzed using SPSS Statistics version 23. P≤.05 was considered significant. Mean and SD were calculated for descriptive data. The t test, Fisher exact test, analysis of variance, multiple regression analysis, and logistic regression analysis were used to identify the relationship between variables.

RESULTS

Patient Characteristics

A total of 96 patients were enrolled in this study. The mean (SD) age of participants was 41.42 (15.1) years (range, 18–58 years). Of 96 patients whose data were included, 55 (57.29%) patients had received RTX 3 months earlier (3M group) and 41 (42.71%) received RTX in the last 2 weeks (R group). A summary of study patient characteristics in each group is provided in eTable 2. There was no significant difference between the 2 groups in terms of age, sex, type of pemphigus, marital status, education, positive Nikolsky sign, smoking status, existence of comorbidities, site of lesions, and RTX treatment protocol. However, a significant difference was found for duration of disease (P=.0124) and mean prednisolone dosage (P=.001) as well as severity of disease measured by PDAI score (P=.003) and anti-DSG1 (P=.003) and anti-DSG3 (P=.021) values.

Patient-Reported Outcomes

Physical and mental component scores are summarized in eTable 3. Generally, SF-36 scores were improved with RTX treatment in all dimensions except for mental health, though these differences were not statistically significant, with the greatest mean improvement in the role physical index (75.45 in the 3M group vs 53.04 in the R group; P=.009). Mean SF-36 PCS and MCS scores were higher in the 3M group vs the R group, though the difference in MCS score did not reach the level of significance (eTable 3).

Mean DLQI scores in the R and 3M groups were 12.31 and 6.96, respectively, indicating a considerable burden on HRQOL in both groups. However, a statistically significant difference between these values was seen that also was clinically meaningful, indicating a significant improvement of QOL in patients receiving RTX 3 months earlier (P=.005)(eTable 3).

The PGA scores indicated that patients in the 3M group were significantly more likely to report less severe disease vs the R group (P=.008)(eTable 3).

Multivariate Analysis—Effect of the patient characteristics and some disease features on indices of QOL was evaluated using the multiple linear regression model. eTable 4 shows the P values of those analyses.

COMMENT

Pemphigus is a chronic disabling disease with notable QOL impairment due to disease burden as well as the need for long-term use of immunosuppressive agents during the disease course. To study the effect of RTX on QOL of patients with pemphigus, we compared 2 sets of patients. Prior studies have shown that clinically significant effects of RTX take 4 to 12 weeks to appear.^19,20 Therefore, we selected patients who received RTX 3 months earlier to measure their HRQOL indices and compare them with patients who had received RTX in the last 2 weeks as a control group to investigate the effect of RTX intrinsically, as this was the focus of this study.

In our study, one of the research tools was the DLQI. Healthy patients typically have an average score of 0.5.²¹ The mean DLQI score of the patients in R group was 12.31, which was similar to prior analysis⁸ and reflects a substantial burden of disease comparable to atopic dermatitis and psoriasis.^21,22 In patients in the 3M group, the mean DLQI score was lower than the R group (6.96 vs 12.31), indicating a significant (P=.005) and clinically meaningful improvement in QOL of patients due to the dramatic therapeutic effect of RTX. However, this score indicated a moderate effect on HRQOL, even in the context of clinical improvement due to RTX treatment, which may reflect that the short duration of treatment in the 3M group was a limitation of this study. Although the 12-week treatment duration was comparable with other studies^19,20 and major differences in objective measures of treatment efficacy were found in PDAI as well as anti-DSG1 and anti-DSG3 values, longer treatment duration may be needed for a more comprehensive assessment of the benefit of RTX on HRQOL indices in patients with pemphigus.

Based on results of the SF-36 questionnaire, PCS and MCS scores were not substantially impaired in the R group considering the fact that a mean score of 50 has been articulated as a normative value for all scales.²³ These data demonstrated the importance of using a dermatologic-specific instrument such as the DLQI instead of a general questionnaire to assess QOL in patients with pemphigus. However, better indices were reported with RTX treatment in the 3 SF-36 domains—role physical (P=.009), role emotional (P=.03), and general health perception (P=.03)—with the role physical showing the greatest magnitude of mean change (75.45 in the 3M group vs 53.04 in the R group). Notably, PCS was impaired to a greater extent than MCS in patients in the R group and showed a greater magnitude of improvement after 3 months of treatment. These results could be explained by the fact that MCS can be largely changed in diseases with a direct effect on the central nervous system.²³

Our results also revealed that the dose of corticosteroid correlated to HRQOL of patients with pemphigus who recently received RTX therapy. Indeed, it is more likely that patients on lower-dose prednisolone have a higher QOL, especially on physical function and social function dimensions of SF-36. This finding is highly expectable by less severe disease due to RTX treatment and also lower potential dose-dependent adverse effects of long-term steroid therapy.

One of the most striking findings of this study was the correlation of location of lesions to QOL indices. We found that the mucocutaneous phenotype was significantly correlated to greater improvement in role emotional, role physical, and social functioning scores due to RTX treatment compared with cutaneous or mucosal types (P=.02, P=.025, and P=.017, respectively). Although mucosal involvement of the disease can be the most burdensome feature because of its large impact on essential activities such as eating and speaking, cutaneous lesions with unpleasant appearance and undesirable symptoms may have a similar impact on QOL. Therefore, having both mucosal and cutaneous lesions causes a worsened QOL and decreased treatment efficacy vs having only one area involved. This may explain the greater improvement in some QOL indices with RTX treatment.

Limitations—Given the cross-sectional design of this study in which patients were observed at a single time point during their treatment course, it is not possible to establish a clear cause-effect relationship between variables. Moreover, we did not evaluate the impact of RTX or prednisolone adverse effects on QOL. Therefore, further prospective studies with longer treatment durations may help to validate our findings. In addition, MCDs for DLQI and SF-36 in pemphigus need to be determined and validated in future studies.

CONCLUSION

The results of our study demonstrated that patients with pemphigus may benefit from taking RTX, not only in terms of clinical improvement of their disease measured by objective indices such as PDAI and anti-DSG1 and anti-DSG3 values but also in several domains that are important to patients, including physical and mental health status (SF-36), HRQOL (DLQI), and overall disease severity (PGA). Rituximab administration in patients with pemphigus can lead to rapid and significant improvement in HRQOL as well as patient- and physician-assessed measures. Its favorable safety profile along with its impact on patients’ daily lives and mental health makes RTX a suitable treatment option for patients with pemphigus. Moreover, we recommend taking QOL indices into account while evaluating the efficacy of new medications to improve our insight into the patient experience and provide better patient adherence to treatment, which is an important issue for optimal control of chronic disorders.

Pemphigus is a group of autoimmune blistering diseases characterized by the development of painful and flaccid blisters on the skin and/or mucous membranes. Pemphigus vulgaris (PV) and pemphigus foliaceus (PF) are 2 major subtypes and can be distinguished by the location of blister formation or the specificity of autoantibodies directed against different desmogleins.^1,2 Although rare, pemphigus is considered a serious and life-threatening condition with a great impact on quality of life (QOL) due to disease symptoms (eg, painful lesions, physical appearance of skin lesions) as well as treatment complications (eg, adverse drug effects, cost of treatment).^3-6 Moreover, the physical and psychological effects can lead to marked functional morbidity and work-related disability during patients’ productive years.⁷ Therefore, affected individuals usually have a remarkably compromised health-related quality of life (HRQOL).⁸ Effective treatments may considerably improve the QOL of patients with pemphigus.⁶

Despite the available treatment options, finding the best regimen for pemphigus remains a challenge. Corticosteroids are assumed to be the main treatment, though they have considerable side effects.^9,10 Adjuvant therapies are used to suppress or modulate immune responses, leading to remission with the least possible need for corticosteroids. Finding an optimal steroid-sparing agent has been the aim of research, and biologic agents seem to be the best option.⁸ Rituximab (RTX), an anti-CD20 monoclonal antibody, has shown great promise in several studies of its clinical efficacy and has become a first-line treatment in new guidelines.^11-14 Rituximab treatment has been associated with notable improvement in physician-assessed outcome measures with a favorable safety profile in patients with pemphigus.^11-15 However, it is important to assess response to treatment from a patient’s perspective through the use of outcome-assessment measures that encompass patient-reported outcomes to reflect the complete patient experience and establish the overall impact of RTX as well as its likelihood of acceptance by patients with pemphigus.

In our study, we compared clinical outcomes and HRQOL through the use of disease-specific measures as well as comprehensive generic health status measures among patients with PV and PF who received RTX treatment 3 months earlier and those who received RTX in the last 2 weeks. The clinical relevance of the patient-reported outcomes is discussed.

MATERIALS AND METHODS

Study Design

We conducted a single-center cross-sectional study of 96 patients with pemphigus aged 18 to 65 years of either sex who were willing to participate in this study. Patients with a confirmed diagnosis of PV or PF who received RTX 3 months earlier or in the last 2 weeks were enrolled in the study. Patients were identified using Dermatry.ir, an archiving software that contains patients’ medical data. Exclusion criteria included lack of sufficient knowledge of the concepts of the questionnaires as well as age younger than 16 years. The study was conducted from October 2019 to April 2020 by the Autoimmune Bullous Disease Research Center at Razi Hospital in Tehran, Iran, which is the main dermatology-specific center and teaching hospital of Iran. The study protocol was approved by the relevant ethics committee.

Patients were categorized into 2 groups: (1) those who received RTX 3 months earlier (3M group); and (2) those who received RTX in the last 2 weeks (R group).

After an explanation of the study to participants, informed written consent was signed by each patient, and their personal data (eg, age, sex, education, marital status, smoking status), as well as clinical data (eg, type of pemphigus, duration of disease, site of onset, prednisolone dosage, presence of Nikolsky sign, anti-DSG1 and anti-DSG3 values, Pemphigus Disease Area Index [PDAI] score, RTX treatment protocol); any known comorbidities such as hypertension, diabetes mellitus, or morbid obesity; and any chronic pulmonary, cardiac, endocrinologic, renal, or hepatic condition, were collected and recorded in a predefined Case Record.

Patient-Reported Outcome Measures

The effect of RTX on QOL in patients with pemphigus was assessed using 2 HRQOL instruments: (1) a general health status indicator, the 36-Item Short Form Survey (SF-36), and (2) a validated, Persian version of a dermatology-specific questionnaire, Dermatology Life Quality Index (DLQI). The questionnaires were completed by each patient or by an assistant if needed.

The SF-36 is a widely used 36-item questionnaire measuring functional health and well-being across 8 domains—mental health, pain, physical function, role emotional, role physical, social functioning, vitality, and general health perception—with scores for each ranging from 0 to 100. The physical component scores (PCSs) and mental component scores (MCSs) were derived from these 8 subscales, each ranging from 0 to 400, with higher scores indicating better health status.⁶

The DLQI, one of the most frequently used QOL measures in dermatology, contains 10 questions, each referring to the prior week and classified in the following 6 subscales: symptoms and feelings, daily activities, leisure, personal relationships, work and school, and treatment.¹⁶ The total score ranges from 0 (no impact) to 30 (very high impact), with a higher score indicating a lower QOL (eTable 1). The minimal clinically important difference (MCD) for the DLQI was considered to be 2- to 5-point changes in prior studies.^17,18 In this study, we used an MCD of a 5-point change or more between study groups.

Moreover, the patient general assessment (PGA) of disease severity was identified using a 3-point scale (1=mild, 2=moderate, 3=severe).

Statistical Analysis

Data were analyzed using SPSS Statistics version 23. P≤.05 was considered significant. Mean and SD were calculated for descriptive data. The t test, Fisher exact test, analysis of variance, multiple regression analysis, and logistic regression analysis were used to identify the relationship between variables.

RESULTS

Patient Characteristics

A total of 96 patients were enrolled in this study. The mean (SD) age of participants was 41.42 (15.1) years (range, 18–58 years). Of 96 patients whose data were included, 55 (57.29%) patients had received RTX 3 months earlier (3M group) and 41 (42.71%) received RTX in the last 2 weeks (R group). A summary of study patient characteristics in each group is provided in eTable 2. There was no significant difference between the 2 groups in terms of age, sex, type of pemphigus, marital status, education, positive Nikolsky sign, smoking status, existence of comorbidities, site of lesions, and RTX treatment protocol. However, a significant difference was found for duration of disease (P=.0124) and mean prednisolone dosage (P=.001) as well as severity of disease measured by PDAI score (P=.003) and anti-DSG1 (P=.003) and anti-DSG3 (P=.021) values.

Patient-Reported Outcomes

Physical and mental component scores are summarized in eTable 3. Generally, SF-36 scores were improved with RTX treatment in all dimensions except for mental health, though these differences were not statistically significant, with the greatest mean improvement in the role physical index (75.45 in the 3M group vs 53.04 in the R group; P=.009). Mean SF-36 PCS and MCS scores were higher in the 3M group vs the R group, though the difference in MCS score did not reach the level of significance (eTable 3).

Mean DLQI scores in the R and 3M groups were 12.31 and 6.96, respectively, indicating a considerable burden on HRQOL in both groups. However, a statistically significant difference between these values was seen that also was clinically meaningful, indicating a significant improvement of QOL in patients receiving RTX 3 months earlier (P=.005)(eTable 3).

The PGA scores indicated that patients in the 3M group were significantly more likely to report less severe disease vs the R group (P=.008)(eTable 3).

Multivariate Analysis—Effect of the patient characteristics and some disease features on indices of QOL was evaluated using the multiple linear regression model. eTable 4 shows the P values of those analyses.

COMMENT

Pemphigus is a chronic disabling disease with notable QOL impairment due to disease burden as well as the need for long-term use of immunosuppressive agents during the disease course. To study the effect of RTX on QOL of patients with pemphigus, we compared 2 sets of patients. Prior studies have shown that clinically significant effects of RTX take 4 to 12 weeks to appear.^19,20 Therefore, we selected patients who received RTX 3 months earlier to measure their HRQOL indices and compare them with patients who had received RTX in the last 2 weeks as a control group to investigate the effect of RTX intrinsically, as this was the focus of this study.

In our study, one of the research tools was the DLQI. Healthy patients typically have an average score of 0.5.²¹ The mean DLQI score of the patients in R group was 12.31, which was similar to prior analysis⁸ and reflects a substantial burden of disease comparable to atopic dermatitis and psoriasis.^21,22 In patients in the 3M group, the mean DLQI score was lower than the R group (6.96 vs 12.31), indicating a significant (P=.005) and clinically meaningful improvement in QOL of patients due to the dramatic therapeutic effect of RTX. However, this score indicated a moderate effect on HRQOL, even in the context of clinical improvement due to RTX treatment, which may reflect that the short duration of treatment in the 3M group was a limitation of this study. Although the 12-week treatment duration was comparable with other studies^19,20 and major differences in objective measures of treatment efficacy were found in PDAI as well as anti-DSG1 and anti-DSG3 values, longer treatment duration may be needed for a more comprehensive assessment of the benefit of RTX on HRQOL indices in patients with pemphigus.

Based on results of the SF-36 questionnaire, PCS and MCS scores were not substantially impaired in the R group considering the fact that a mean score of 50 has been articulated as a normative value for all scales.²³ These data demonstrated the importance of using a dermatologic-specific instrument such as the DLQI instead of a general questionnaire to assess QOL in patients with pemphigus. However, better indices were reported with RTX treatment in the 3 SF-36 domains—role physical (P=.009), role emotional (P=.03), and general health perception (P=.03)—with the role physical showing the greatest magnitude of mean change (75.45 in the 3M group vs 53.04 in the R group). Notably, PCS was impaired to a greater extent than MCS in patients in the R group and showed a greater magnitude of improvement after 3 months of treatment. These results could be explained by the fact that MCS can be largely changed in diseases with a direct effect on the central nervous system.²³

Our results also revealed that the dose of corticosteroid correlated to HRQOL of patients with pemphigus who recently received RTX therapy. Indeed, it is more likely that patients on lower-dose prednisolone have a higher QOL, especially on physical function and social function dimensions of SF-36. This finding is highly expectable by less severe disease due to RTX treatment and also lower potential dose-dependent adverse effects of long-term steroid therapy.

One of the most striking findings of this study was the correlation of location of lesions to QOL indices. We found that the mucocutaneous phenotype was significantly correlated to greater improvement in role emotional, role physical, and social functioning scores due to RTX treatment compared with cutaneous or mucosal types (P=.02, P=.025, and P=.017, respectively). Although mucosal involvement of the disease can be the most burdensome feature because of its large impact on essential activities such as eating and speaking, cutaneous lesions with unpleasant appearance and undesirable symptoms may have a similar impact on QOL. Therefore, having both mucosal and cutaneous lesions causes a worsened QOL and decreased treatment efficacy vs having only one area involved. This may explain the greater improvement in some QOL indices with RTX treatment.

Limitations—Given the cross-sectional design of this study in which patients were observed at a single time point during their treatment course, it is not possible to establish a clear cause-effect relationship between variables. Moreover, we did not evaluate the impact of RTX or prednisolone adverse effects on QOL. Therefore, further prospective studies with longer treatment durations may help to validate our findings. In addition, MCDs for DLQI and SF-36 in pemphigus need to be determined and validated in future studies.

CONCLUSION

The results of our study demonstrated that patients with pemphigus may benefit from taking RTX, not only in terms of clinical improvement of their disease measured by objective indices such as PDAI and anti-DSG1 and anti-DSG3 values but also in several domains that are important to patients, including physical and mental health status (SF-36), HRQOL (DLQI), and overall disease severity (PGA). Rituximab administration in patients with pemphigus can lead to rapid and significant improvement in HRQOL as well as patient- and physician-assessed measures. Its favorable safety profile along with its impact on patients’ daily lives and mental health makes RTX a suitable treatment option for patients with pemphigus. Moreover, we recommend taking QOL indices into account while evaluating the efficacy of new medications to improve our insight into the patient experience and provide better patient adherence to treatment, which is an important issue for optimal control of chronic disorders.

Phototherapy—particularly UVB phototherapy, which utilizes UVB rays of specific wavelengths within the UV spectrum—is indicated for a wide variety of dermatoses. In-office and at-home UVB treatments commonly are used, as are salon tanning and sunbathing. When selecting a form of phototherapy, patients are likely to consider safety, cost, effectiveness, insurance issues, and convenience. Research on patient preferences; the reasons for these preferences; and which options patients perceive to be the safest, most cost-effective, efficacious, and convenient is lacking. We aimed to assess the forms of phototherapy that patients would most consider using; the factors influencing patient preferences; and the forms patients perceived as the safest and most cost-effective, efficacious, and convenient.

Methods

Study Participants—We recruited 500 Amazon Mechanical Turk users who were 18 years or older to complete our REDCap-generated survey. The study was approved by the Wake Forest University institutional review board (Winston-Salem, North Carolina).

Evaluation—Participants were asked, “If you were diagnosed with a skin disease that benefited from UV therapy, which of the following forms of UV therapy would you consider choosing?” Participants were instructed to choose all of the forms they would consider using. Available options included in-office UV, at-home UV, home tanning, salon tanning, sunbathing, and other. Participants were asked to select which factors—from safety, cost, effectiveness, issues with insurance, convenience, and other—influenced their decision-making; which form of phototherapy they would most consider along with the factors that influenced their preference for this specific form of phototherapy; and which options they considered to be safest and most cost-effective, efficacious, and convenient. Participants were asked to provide basic sociodemographic information, level of education, income, insurance status (private, Medicare, Medicaid, Veterans Affairs, and uninsured), and distance from the nearest dermatologist.

Statistical Analysis—Descriptive and inferential statistics (χ² test) were used to analyze the data with a significance set at P<.05.

Results

Five hundred participants completed the survey (Table 1).

Factors Influencing Patient Preferences—When asked to select all forms of phototherapy they would consider, 186 (37.2%) participants selected in-office UVB, 263 (52.6%) selected at-home UV, 141 (28.2%) selected home tanning, 117 (23.4%) selected salon tanning, 191 (38.2%) selected sunbathing, and 3 (0.6%) selected other. Participants who selected in-office UVB as an option were more likely to also select salon tanning (P<.012). No other relationship was found between the UVB options and the tanning options. When asked which factors influenced their phototherapy preferences, 295 (59%) selected convenience, 266 (53.2%) selected effectiveness, 220 (44%) selected safety, 218 (43.6%) selected cost, 72 (14.4%) selected issues with insurance, and 4 (0.8%) selected other. Forms of Phototherapy Patients Consider Using—When asked which form of phototherapy they would most consider using, 179 (35.8%) participants selected at-home UVB, 108 (21.6%) selected sunbathing, 92 (18.4%) selected in-office UVB, 62 (12.4%) selected home-tanning, 57 (11.4%) selected salon tanning, 1 (0.2%) selected other, and 1 participant provided no response (P<.001).

Reasons for Using Phototherapy—Of the 179 who selected at-home UVB, 125 (70%) cited convenience as a reason. Of the 108 who selected salon tanning as their top choice, 62 (57%) cited cost as a reason. Convenience (P<.001), cost (P<.001), and safety (P=.023) were related to top preference. Issues with insurance did not have a statistically significant relationship with the top preference. However, participant insurance type was related to top phototherapy preference (P=.021), with privately insured patients more likely to select in-office UVB, whereas those with Medicaid and Medicare were more likely to select home or salon tanning. Efficacy was not related to top preference. Furthermore, age, gender, education, income, and distance from nearest dermatologist were not related to top preference.

In-office UVB was perceived to be safest (P<.001) and most efficacious (P<.001). Meanwhile, at-home UVB was selected as most convenient (P<.001). Lastly, sunbathing was determined to be most cost-effective (P<.001)(Table 2). Cost-effectiveness had a relationship (P<.001) with the participant’s insurance, as those with private insurance were more likely to select at-home UVB, whereas those with Medicare or Medicaid were more likely to select the tanning options. Additionally, of the54 uninsured participants in the survey, 29 selected sunbathing as the most cost-effective option.

Comment

Phototherapy Treatment—UVB phototherapy at a wavelength of 290 to 320 nm (311–313 nm for narrowband UVB) is used to treat various dermatoses, including psoriasis and atopic dermatitis. UVB alters skin cytokines, induces apoptosis, promotes immunosuppression, causes DNA damage, and decreases the proliferation of dendritic cells and other cells of the innate immune system.¹ In-office and at-home UV therapies make use of UVB wavelengths for treatment, while tanning and sunbathing contain not only UVB but also potentially harmful UVA rays. The wavelengths for indoor tanning devices include UVB at 280 to 315 nm and UVA at 315 to 400 nm, which are similar to those of the sun but with a different ratio of UVB to UVA and more intense total UV.² When in-office and at-home UVB options are not available, various forms of tanning such as salon tanning and sunbathing may be alternatives that are widely used.³ One of the main reasons patients consider alternative phototherapy options is cost, as 1 in-office UVB treatment may cost $140, but a month of unlimited tanning may cost $30 or perhaps nothing if a patient has a gym membership with access to a tanning bed. Lack of insurance benefits covering phototherapy can exacerbate cost burden.⁴ However, tanning beds are associated with an increased risk for melanoma and nonmelanoma cancers.^5,6 Additionally, all forms of phototherapy are associated with photoaging, but it is more intense with tanning and heliotherapy because of the presence of UVA, which penetrates deeper into the dermis.⁷ Meanwhile, for those who choose UVB therapy, deciding between an in-office and at-home UVB treatment could be a matter of convenience, as patients must consider long trips to the physician’s office; insurance status, as some insurances may not cover at-home UVB; or efficacy, which might be influenced by the presence of a physician or other medical staff. In many cases, patients may not be informed that at-home UVB is an option.

Patient Preferences—At-home UVB therapy was the most popular option in our study population, with most participants (52.6%) considering using it, and 35.9% choosing it as their top choice over all other phototherapy options. Safety, cost, and convenience were all found to be related to the option participants would most consider using. Prior analysis between at-home UVB and in-office UVB for the treatment of psoriasis determined that at-home UVB is as safe and cost-effective as in-office UVB without the inconvenience of the patient having to take time out of the week to visit the physician’s office,^8,9 making at-home UVB an option dermatologists may strongly consider for patients who value safety, cost, and convenience. Oddly, efficacy was not related to the top preference, despite being the second highest–cited factor (53.2%) for which forms of phototherapy participants would consider using. For insurance coverage, those with Medicaid and Medicare selected the cheaper tanning options with higher-than-expected frequencies. Although problems with insurance were not related to the top preference, insurance status was related, suggesting that preferences are tied to cost. Of note, while the number of dermatologists that accept Medicare has increased in the last few years, there still remains an uneven distribution of phototherapy clinics. As of 2015, there were 19 million individuals who qualified for Medicare without a clinic within driving distance.¹⁰ This problem likely also exists for many Medicaid patients who may not qualify for at-home UVB. In this scenario, tanning or heliotherapy may be effective alternatives.

In-Office vs At-Home Options—Although in-office UVB was the option considered safest (26.2%) and most efficacious (26.8%), it was followed closely by at-home UVB in both categories (safest, 23.8%; most efficacious, 24.2%). Meanwhile, at-home UVB (40.2%) was chosen as the most convenient. Some patients consider tanning options over in-office UVB because of the inconvenience of traveling to an appointment.¹¹ Therefore, at-home tanning may be a convenient alternative for these patients.

Considerations—Although our study was limited to an adult population, issues with convenience exist for the pediatric population as well, as children may need to miss multiple days of school each week to be treated in the office. For these pediatric patients, an at-home unit is preferable; however; issues with insurance coverage remain a challenge.¹² Increasing insurance coverage of at-home units for the pediatric population therefore would be most prudent. However, when other options have been exhausted, including in-office UVB, tanning and sunbathing may be viable alternatives because of cost and convenience. In our study, sunbathing (33.2%) was considered the most cost-effective, likely because it does not require expensive equipment or a visit to a salon or physician’s office. Sunbathing has been effective in treating some dermatologic conditions, such as atopic dermatitis.¹³ However, it may only be effective during certain months and at different latitudes—conditions that make UVB sun rays more accessible—particularly when treating psoriasis.¹⁴ Furthermore, sunbathing may not be as cost-effective in patients with average-severity psoriasis compared with conventional psoriasis therapy because of the costs of travel to areas with sufficient UVB rays for treatment.¹⁵ Additionally, insurance status was related to which option was selected as the most cost-effective, as 29 (53.7%) of 54 uninsured participants chose sunbathing as the most cost-effective option, while only 92 (34.2%) of 269 privately insured patients selected sunbathing. Therefore, insurance status may be a factor for dermatologists to consider if a patient prefers a treatment that is cost-effective. Overall, dermatologists could perhaps consider guiding patients and optimizing their treatment plans based on the factors most important to the patients while understanding that costs and insurance status may ultimately determine the treatment option.

Limitations—Survey participants were recruited on Amazon Mechanical Turk, which could create sampling bias. Furthermore, these participants were representative of the general public and not exclusively patients on phototherapy, therefore representing the opinions of the general public and not those who may require phototherapy. Furthermore, given the nature of the survey, the study was limited to the adult population.

Phototherapy—particularly UVB phototherapy, which utilizes UVB rays of specific wavelengths within the UV spectrum—is indicated for a wide variety of dermatoses. In-office and at-home UVB treatments commonly are used, as are salon tanning and sunbathing. When selecting a form of phototherapy, patients are likely to consider safety, cost, effectiveness, insurance issues, and convenience. Research on patient preferences; the reasons for these preferences; and which options patients perceive to be the safest, most cost-effective, efficacious, and convenient is lacking. We aimed to assess the forms of phototherapy that patients would most consider using; the factors influencing patient preferences; and the forms patients perceived as the safest and most cost-effective, efficacious, and convenient.

Methods

Study Participants—We recruited 500 Amazon Mechanical Turk users who were 18 years or older to complete our REDCap-generated survey. The study was approved by the Wake Forest University institutional review board (Winston-Salem, North Carolina).

Evaluation—Participants were asked, “If you were diagnosed with a skin disease that benefited from UV therapy, which of the following forms of UV therapy would you consider choosing?” Participants were instructed to choose all of the forms they would consider using. Available options included in-office UV, at-home UV, home tanning, salon tanning, sunbathing, and other. Participants were asked to select which factors—from safety, cost, effectiveness, issues with insurance, convenience, and other—influenced their decision-making; which form of phototherapy they would most consider along with the factors that influenced their preference for this specific form of phototherapy; and which options they considered to be safest and most cost-effective, efficacious, and convenient. Participants were asked to provide basic sociodemographic information, level of education, income, insurance status (private, Medicare, Medicaid, Veterans Affairs, and uninsured), and distance from the nearest dermatologist.

Statistical Analysis—Descriptive and inferential statistics (χ² test) were used to analyze the data with a significance set at P<.05.

Results

Five hundred participants completed the survey (Table 1).

Factors Influencing Patient Preferences—When asked to select all forms of phototherapy they would consider, 186 (37.2%) participants selected in-office UVB, 263 (52.6%) selected at-home UV, 141 (28.2%) selected home tanning, 117 (23.4%) selected salon tanning, 191 (38.2%) selected sunbathing, and 3 (0.6%) selected other. Participants who selected in-office UVB as an option were more likely to also select salon tanning (P<.012). No other relationship was found between the UVB options and the tanning options. When asked which factors influenced their phototherapy preferences, 295 (59%) selected convenience, 266 (53.2%) selected effectiveness, 220 (44%) selected safety, 218 (43.6%) selected cost, 72 (14.4%) selected issues with insurance, and 4 (0.8%) selected other. Forms of Phototherapy Patients Consider Using—When asked which form of phototherapy they would most consider using, 179 (35.8%) participants selected at-home UVB, 108 (21.6%) selected sunbathing, 92 (18.4%) selected in-office UVB, 62 (12.4%) selected home-tanning, 57 (11.4%) selected salon tanning, 1 (0.2%) selected other, and 1 participant provided no response (P<.001).

Reasons for Using Phototherapy—Of the 179 who selected at-home UVB, 125 (70%) cited convenience as a reason. Of the 108 who selected salon tanning as their top choice, 62 (57%) cited cost as a reason. Convenience (P<.001), cost (P<.001), and safety (P=.023) were related to top preference. Issues with insurance did not have a statistically significant relationship with the top preference. However, participant insurance type was related to top phototherapy preference (P=.021), with privately insured patients more likely to select in-office UVB, whereas those with Medicaid and Medicare were more likely to select home or salon tanning. Efficacy was not related to top preference. Furthermore, age, gender, education, income, and distance from nearest dermatologist were not related to top preference.

In-office UVB was perceived to be safest (P<.001) and most efficacious (P<.001). Meanwhile, at-home UVB was selected as most convenient (P<.001). Lastly, sunbathing was determined to be most cost-effective (P<.001)(Table 2). Cost-effectiveness had a relationship (P<.001) with the participant’s insurance, as those with private insurance were more likely to select at-home UVB, whereas those with Medicare or Medicaid were more likely to select the tanning options. Additionally, of the54 uninsured participants in the survey, 29 selected sunbathing as the most cost-effective option.

Comment

Phototherapy Treatment—UVB phototherapy at a wavelength of 290 to 320 nm (311–313 nm for narrowband UVB) is used to treat various dermatoses, including psoriasis and atopic dermatitis. UVB alters skin cytokines, induces apoptosis, promotes immunosuppression, causes DNA damage, and decreases the proliferation of dendritic cells and other cells of the innate immune system.¹ In-office and at-home UV therapies make use of UVB wavelengths for treatment, while tanning and sunbathing contain not only UVB but also potentially harmful UVA rays. The wavelengths for indoor tanning devices include UVB at 280 to 315 nm and UVA at 315 to 400 nm, which are similar to those of the sun but with a different ratio of UVB to UVA and more intense total UV.² When in-office and at-home UVB options are not available, various forms of tanning such as salon tanning and sunbathing may be alternatives that are widely used.³ One of the main reasons patients consider alternative phototherapy options is cost, as 1 in-office UVB treatment may cost $140, but a month of unlimited tanning may cost $30 or perhaps nothing if a patient has a gym membership with access to a tanning bed. Lack of insurance benefits covering phototherapy can exacerbate cost burden.⁴ However, tanning beds are associated with an increased risk for melanoma and nonmelanoma cancers.^5,6 Additionally, all forms of phototherapy are associated with photoaging, but it is more intense with tanning and heliotherapy because of the presence of UVA, which penetrates deeper into the dermis.⁷ Meanwhile, for those who choose UVB therapy, deciding between an in-office and at-home UVB treatment could be a matter of convenience, as patients must consider long trips to the physician’s office; insurance status, as some insurances may not cover at-home UVB; or efficacy, which might be influenced by the presence of a physician or other medical staff. In many cases, patients may not be informed that at-home UVB is an option.

Patient Preferences—At-home UVB therapy was the most popular option in our study population, with most participants (52.6%) considering using it, and 35.9% choosing it as their top choice over all other phototherapy options. Safety, cost, and convenience were all found to be related to the option participants would most consider using. Prior analysis between at-home UVB and in-office UVB for the treatment of psoriasis determined that at-home UVB is as safe and cost-effective as in-office UVB without the inconvenience of the patient having to take time out of the week to visit the physician’s office,^8,9 making at-home UVB an option dermatologists may strongly consider for patients who value safety, cost, and convenience. Oddly, efficacy was not related to the top preference, despite being the second highest–cited factor (53.2%) for which forms of phototherapy participants would consider using. For insurance coverage, those with Medicaid and Medicare selected the cheaper tanning options with higher-than-expected frequencies. Although problems with insurance were not related to the top preference, insurance status was related, suggesting that preferences are tied to cost. Of note, while the number of dermatologists that accept Medicare has increased in the last few years, there still remains an uneven distribution of phototherapy clinics. As of 2015, there were 19 million individuals who qualified for Medicare without a clinic within driving distance.¹⁰ This problem likely also exists for many Medicaid patients who may not qualify for at-home UVB. In this scenario, tanning or heliotherapy may be effective alternatives.

In-Office vs At-Home Options—Although in-office UVB was the option considered safest (26.2%) and most efficacious (26.8%), it was followed closely by at-home UVB in both categories (safest, 23.8%; most efficacious, 24.2%). Meanwhile, at-home UVB (40.2%) was chosen as the most convenient. Some patients consider tanning options over in-office UVB because of the inconvenience of traveling to an appointment.¹¹ Therefore, at-home tanning may be a convenient alternative for these patients.

Considerations—Although our study was limited to an adult population, issues with convenience exist for the pediatric population as well, as children may need to miss multiple days of school each week to be treated in the office. For these pediatric patients, an at-home unit is preferable; however; issues with insurance coverage remain a challenge.¹² Increasing insurance coverage of at-home units for the pediatric population therefore would be most prudent. However, when other options have been exhausted, including in-office UVB, tanning and sunbathing may be viable alternatives because of cost and convenience. In our study, sunbathing (33.2%) was considered the most cost-effective, likely because it does not require expensive equipment or a visit to a salon or physician’s office. Sunbathing has been effective in treating some dermatologic conditions, such as atopic dermatitis.¹³ However, it may only be effective during certain months and at different latitudes—conditions that make UVB sun rays more accessible—particularly when treating psoriasis.¹⁴ Furthermore, sunbathing may not be as cost-effective in patients with average-severity psoriasis compared with conventional psoriasis therapy because of the costs of travel to areas with sufficient UVB rays for treatment.¹⁵ Additionally, insurance status was related to which option was selected as the most cost-effective, as 29 (53.7%) of 54 uninsured participants chose sunbathing as the most cost-effective option, while only 92 (34.2%) of 269 privately insured patients selected sunbathing. Therefore, insurance status may be a factor for dermatologists to consider if a patient prefers a treatment that is cost-effective. Overall, dermatologists could perhaps consider guiding patients and optimizing their treatment plans based on the factors most important to the patients while understanding that costs and insurance status may ultimately determine the treatment option.

Limitations—Survey participants were recruited on Amazon Mechanical Turk, which could create sampling bias. Furthermore, these participants were representative of the general public and not exclusively patients on phototherapy, therefore representing the opinions of the general public and not those who may require phototherapy. Furthermore, given the nature of the survey, the study was limited to the adult population.

Phototherapy—particularly UVB phototherapy, which utilizes UVB rays of specific wavelengths within the UV spectrum—is indicated for a wide variety of dermatoses. In-office and at-home UVB treatments commonly are used, as are salon tanning and sunbathing. When selecting a form of phototherapy, patients are likely to consider safety, cost, effectiveness, insurance issues, and convenience. Research on patient preferences; the reasons for these preferences; and which options patients perceive to be the safest, most cost-effective, efficacious, and convenient is lacking. We aimed to assess the forms of phototherapy that patients would most consider using; the factors influencing patient preferences; and the forms patients perceived as the safest and most cost-effective, efficacious, and convenient.

Methods

Study Participants—We recruited 500 Amazon Mechanical Turk users who were 18 years or older to complete our REDCap-generated survey. The study was approved by the Wake Forest University institutional review board (Winston-Salem, North Carolina).

Evaluation—Participants were asked, “If you were diagnosed with a skin disease that benefited from UV therapy, which of the following forms of UV therapy would you consider choosing?” Participants were instructed to choose all of the forms they would consider using. Available options included in-office UV, at-home UV, home tanning, salon tanning, sunbathing, and other. Participants were asked to select which factors—from safety, cost, effectiveness, issues with insurance, convenience, and other—influenced their decision-making; which form of phototherapy they would most consider along with the factors that influenced their preference for this specific form of phototherapy; and which options they considered to be safest and most cost-effective, efficacious, and convenient. Participants were asked to provide basic sociodemographic information, level of education, income, insurance status (private, Medicare, Medicaid, Veterans Affairs, and uninsured), and distance from the nearest dermatologist.

Statistical Analysis—Descriptive and inferential statistics (χ² test) were used to analyze the data with a significance set at P<.05.

Results

Five hundred participants completed the survey (Table 1).

Factors Influencing Patient Preferences—When asked to select all forms of phototherapy they would consider, 186 (37.2%) participants selected in-office UVB, 263 (52.6%) selected at-home UV, 141 (28.2%) selected home tanning, 117 (23.4%) selected salon tanning, 191 (38.2%) selected sunbathing, and 3 (0.6%) selected other. Participants who selected in-office UVB as an option were more likely to also select salon tanning (P<.012). No other relationship was found between the UVB options and the tanning options. When asked which factors influenced their phototherapy preferences, 295 (59%) selected convenience, 266 (53.2%) selected effectiveness, 220 (44%) selected safety, 218 (43.6%) selected cost, 72 (14.4%) selected issues with insurance, and 4 (0.8%) selected other. Forms of Phototherapy Patients Consider Using—When asked which form of phototherapy they would most consider using, 179 (35.8%) participants selected at-home UVB, 108 (21.6%) selected sunbathing, 92 (18.4%) selected in-office UVB, 62 (12.4%) selected home-tanning, 57 (11.4%) selected salon tanning, 1 (0.2%) selected other, and 1 participant provided no response (P<.001).

Reasons for Using Phototherapy—Of the 179 who selected at-home UVB, 125 (70%) cited convenience as a reason. Of the 108 who selected salon tanning as their top choice, 62 (57%) cited cost as a reason. Convenience (P<.001), cost (P<.001), and safety (P=.023) were related to top preference. Issues with insurance did not have a statistically significant relationship with the top preference. However, participant insurance type was related to top phototherapy preference (P=.021), with privately insured patients more likely to select in-office UVB, whereas those with Medicaid and Medicare were more likely to select home or salon tanning. Efficacy was not related to top preference. Furthermore, age, gender, education, income, and distance from nearest dermatologist were not related to top preference.

In-office UVB was perceived to be safest (P<.001) and most efficacious (P<.001). Meanwhile, at-home UVB was selected as most convenient (P<.001). Lastly, sunbathing was determined to be most cost-effective (P<.001)(Table 2). Cost-effectiveness had a relationship (P<.001) with the participant’s insurance, as those with private insurance were more likely to select at-home UVB, whereas those with Medicare or Medicaid were more likely to select the tanning options. Additionally, of the54 uninsured participants in the survey, 29 selected sunbathing as the most cost-effective option.

Comment

Phototherapy Treatment—UVB phototherapy at a wavelength of 290 to 320 nm (311–313 nm for narrowband UVB) is used to treat various dermatoses, including psoriasis and atopic dermatitis. UVB alters skin cytokines, induces apoptosis, promotes immunosuppression, causes DNA damage, and decreases the proliferation of dendritic cells and other cells of the innate immune system.¹ In-office and at-home UV therapies make use of UVB wavelengths for treatment, while tanning and sunbathing contain not only UVB but also potentially harmful UVA rays. The wavelengths for indoor tanning devices include UVB at 280 to 315 nm and UVA at 315 to 400 nm, which are similar to those of the sun but with a different ratio of UVB to UVA and more intense total UV.² When in-office and at-home UVB options are not available, various forms of tanning such as salon tanning and sunbathing may be alternatives that are widely used.³ One of the main reasons patients consider alternative phototherapy options is cost, as 1 in-office UVB treatment may cost $140, but a month of unlimited tanning may cost $30 or perhaps nothing if a patient has a gym membership with access to a tanning bed. Lack of insurance benefits covering phototherapy can exacerbate cost burden.⁴ However, tanning beds are associated with an increased risk for melanoma and nonmelanoma cancers.^5,6 Additionally, all forms of phototherapy are associated with photoaging, but it is more intense with tanning and heliotherapy because of the presence of UVA, which penetrates deeper into the dermis.⁷ Meanwhile, for those who choose UVB therapy, deciding between an in-office and at-home UVB treatment could be a matter of convenience, as patients must consider long trips to the physician’s office; insurance status, as some insurances may not cover at-home UVB; or efficacy, which might be influenced by the presence of a physician or other medical staff. In many cases, patients may not be informed that at-home UVB is an option.

Patient Preferences—At-home UVB therapy was the most popular option in our study population, with most participants (52.6%) considering using it, and 35.9% choosing it as their top choice over all other phototherapy options. Safety, cost, and convenience were all found to be related to the option participants would most consider using. Prior analysis between at-home UVB and in-office UVB for the treatment of psoriasis determined that at-home UVB is as safe and cost-effective as in-office UVB without the inconvenience of the patient having to take time out of the week to visit the physician’s office,^8,9 making at-home UVB an option dermatologists may strongly consider for patients who value safety, cost, and convenience. Oddly, efficacy was not related to the top preference, despite being the second highest–cited factor (53.2%) for which forms of phototherapy participants would consider using. For insurance coverage, those with Medicaid and Medicare selected the cheaper tanning options with higher-than-expected frequencies. Although problems with insurance were not related to the top preference, insurance status was related, suggesting that preferences are tied to cost. Of note, while the number of dermatologists that accept Medicare has increased in the last few years, there still remains an uneven distribution of phototherapy clinics. As of 2015, there were 19 million individuals who qualified for Medicare without a clinic within driving distance.¹⁰ This problem likely also exists for many Medicaid patients who may not qualify for at-home UVB. In this scenario, tanning or heliotherapy may be effective alternatives.

In-Office vs At-Home Options—Although in-office UVB was the option considered safest (26.2%) and most efficacious (26.8%), it was followed closely by at-home UVB in both categories (safest, 23.8%; most efficacious, 24.2%). Meanwhile, at-home UVB (40.2%) was chosen as the most convenient. Some patients consider tanning options over in-office UVB because of the inconvenience of traveling to an appointment.¹¹ Therefore, at-home tanning may be a convenient alternative for these patients.

Considerations—Although our study was limited to an adult population, issues with convenience exist for the pediatric population as well, as children may need to miss multiple days of school each week to be treated in the office. For these pediatric patients, an at-home unit is preferable; however; issues with insurance coverage remain a challenge.¹² Increasing insurance coverage of at-home units for the pediatric population therefore would be most prudent. However, when other options have been exhausted, including in-office UVB, tanning and sunbathing may be viable alternatives because of cost and convenience. In our study, sunbathing (33.2%) was considered the most cost-effective, likely because it does not require expensive equipment or a visit to a salon or physician’s office. Sunbathing has been effective in treating some dermatologic conditions, such as atopic dermatitis.¹³ However, it may only be effective during certain months and at different latitudes—conditions that make UVB sun rays more accessible—particularly when treating psoriasis.¹⁴ Furthermore, sunbathing may not be as cost-effective in patients with average-severity psoriasis compared with conventional psoriasis therapy because of the costs of travel to areas with sufficient UVB rays for treatment.¹⁵ Additionally, insurance status was related to which option was selected as the most cost-effective, as 29 (53.7%) of 54 uninsured participants chose sunbathing as the most cost-effective option, while only 92 (34.2%) of 269 privately insured patients selected sunbathing. Therefore, insurance status may be a factor for dermatologists to consider if a patient prefers a treatment that is cost-effective. Overall, dermatologists could perhaps consider guiding patients and optimizing their treatment plans based on the factors most important to the patients while understanding that costs and insurance status may ultimately determine the treatment option.

Limitations—Survey participants were recruited on Amazon Mechanical Turk, which could create sampling bias. Furthermore, these participants were representative of the general public and not exclusively patients on phototherapy, therefore representing the opinions of the general public and not those who may require phototherapy. Furthermore, given the nature of the survey, the study was limited to the adult population.

Colorectal cancer is the third leading cause of cancer-related death in both men and women.¹ Colonoscopy is the current gold standard for screening due to the ability to remove precancerous lesions but remains highly dependent on the quality of bowel preparation.² Poor bowel preparation has been associated with impaired adenoma detection as well as increased health care utilization due to the need for a repeat colonoscopy.³

Multiple patient factors are associated with increased risk of poor bowel preparation, including age > 60 years, male sex, diabetes mellitus, and presence of a mental health diagnosis, factors that are prevalent among the veteran population.^3-5 Text messages have been shown to improve the quality of bowel preparation by increasing patients' understanding and adherence with the preparation process. Improved adherence with bowel preparation directions is associated with a cleaner colon prior to colonoscopy, leading to a thorough examination. Studies using text messaging instructions prior to colonoscopies have also shown measurable improvement in adenoma detection rate, patient preparation-associated discomfort, and completion of colonoscopy.^6-10

In 2016, the Veterans Health Administration (VHA) introduced Annie, one of the first automated text messaging services, named after Army Lieutenant Annie Fox, the first woman to receive the Purple Heart for combat. The Annie platform allows for notifications, instructions, and simple data collection. The development of this platform allows VHA practitioners to engage and educate veterans in a similar way to other health care systems using text messaging protocols. Annie text messages have been piloted for the use of hepatitis C treatment, demonstrating promise of improved medication adherence and patient satisfaction.¹¹ We aimed to develop and pilot the Annie bowel preparation protocol to improve the quality of colonoscopy bowel preparation for outpatients at the Minneapolis Veterans Affairs Medical Center (MVAMC) in Minnesota. A secondary goal included measuring patient satisfaction with the text messaging instructions for outpatient colonoscopy preparation.

Methods

We conducted a single center, prospective, endoscopist-blinded, study with two 3-month long Plan-Do-Study-Act (PDSA) cycles to improve the text messaging bowel preparation protocol at MVAMC between January 2019 and April 2020. The MVAMC Institutional Review Board determined the quality improvement project was exempt. Veterans who had outpatient colonoscopies scheduled were included. Veterans undergoing inpatient colonoscopies or outpatients who could not be reached to obtain informed consent, lacked text message capability, declined participation, or required extended colonoscopy preparation were excluded. Per MVAMC procedures, extended colonoscopy preparation was provided to patients receiving general or monitored anesthesia care, with a history of poor bowel preparation, or with risk factors for poor preparation as determined by the ordering health care professional (HCP). Standard bowel preparation involves ingestion of 4 L of polyethylene glycol 3350 with electrolytes; extended bowel preparation requires ingestion of an additional 2 L to total 6 L and uses a different set of instructions. Additionally, the patient population requiring extended bowel preparation also includes patients with spinal cord injuries, who often are admitted for assistance with extended preparation. Patients who consented to receiving text messages were placed in the Annie intervention group, and all others were placed in the control group.

The control group received standardized patient education, including a mailed copy of bowel preparation instructions and a phone call from a gastroenterology service nurse about 1 to 2 weeks before the procedure. Current MVAMC standard of care involves a phone call from a nurse to confirm that patients have received the polyethylene glycol preparation solution, the mailed instructions, have an escort and transportation, and to answer any questions. Both the usual care and intervention group received the phone call. During this call, the Annie text messaging bowel preparation protocol was introduced; if the veteran chose to participate, consent and enrollment were completed.

On the day of the colonoscopy, veterans in the intervention group were surveyed in the waiting room about their experience receiving the text messages and soliciting feedback for improvement or surveyed via telephone call within 3 days of their procedure. Patient satisfaction was quantified with a scale from 1 (low) to 10 (high), including questions about how helpful the texts were in relation to total number, timing, and content of messages as well as whether veterans would like to receive the text messages again for future procedures.

We reviewed individual charts and collected Boston Bowel Preparation Scale (BBPS) scores to determine adequate preparation. BBPS assigns a score of 0 to 3 for the right, transverse, and left colon applied upon withdrawal after flushing and suctioning have been completed.¹² Adequate preparation is considered a total score of ≥ 6 with no segment scoring < 2. This method of preparation assessment is preferred due to its ability to account for difference in preparation quality among colonic segments, well-defined scoring characteristics, and several studies validating its use showing inter- and intraobserver reliability.¹² Follow-up studies have shown validity of the BBPS when compared with relevant outcomes such as polyp detection rate and recommended timing for repeat procedure.¹³ Variables associated with poor bowel preparation (ie, gender, prior abdominal surgery, impaired mobility, high body mass index, diabetes mellitus, stroke, dementia, any neurologic diagnosis, cirrhosis, smoking, polypharmacy [> 8 active medications], and narcotic or tricyclic antidepressant medication use) were also collected through chart review.^3-5 We note that immobility was defined by International Classification of Diseases (ICD)-9 and ICD-10 codes and prescriptions for assistive devices (ie, canes, wheelchairs, 4-wheeled walkers).

Veterans assent to be enrolled in Annie. After enrollment, veterans must text back a specific word response to an initial text message to receive the protocolized messages from the Annie program. A contact phone number to the gastrointestinal nurse line was provided for questions during business hours. The start date for the text message protocol is 6 days prior to the procedure date. If a patient rescheduled their colonoscopy, the Annie database was updated manually.

Statistical Analysis

We used both Pearson χ² test and 2-sample t test analyses to compare demographic information and patient satisfaction scores between the control and intervention groups. We compared continuous BBPS scores between Annie intervention vs control group using parametric and nonparametric independent t tests using the Mann-Whitney U test. We repeated this analysis controlling for both mental health diagnoses and age using linear regression. We were unable to survey 61 of the 187 veterans who received Annie text messages. 

RESULTS

During PDSA cycles 1 and 2, 640 veterans were scheduled for outpatient colonoscopy: 453 veterans were in the control group; 187 veterans were in the intervention group, of which 126 were surveyed. A significant percentage of veterans declined participation because they felt like they did not need reinforced education; others were not eligible for Annie due to requirement for extended bowel preparation, cancelled colonoscopy, inability to physically read text messages, or lack of cell phone.

The mean (SD) age was 65 (8) years; 184 (28.8%) had a diabetes mellitus diagnosis, and the mean (SD) body mass index was 31.6 (6.4). The Annie group was slightly more likely to have mental health diagnoses and lower age compared with the control group (Table 1).

Patient Feedback

We collected feedback from veterans after each PDSA cycle to identify areas for improvement by both in-person and telephone surveys. Based on feedback from PDSA cycle 1, we decreased the total number of text messages to create a more succinct set of instructions. The most frequently requested change involved timing the text messages to align with the exact morning a specific instruction should take place.

Patient satisfaction with the Annie text messaging service was high.

DISCUSSION

To our knowledge, this is the first report of using Annie at a VAMC for colonoscopy bowel preparation improvement. We found a statistically significant improvement in the average BBPS in those receiving Annie text messages compared with the routine care control group. We also found high levels of patient satisfaction with most patients requesting to receive them again for future procedures.

The clinical significance of a BBPS of 7.8 vs 8.2 is unclear, although any score > 6 is considered to be adequate. However, subjectively speaking, the higher the BBPS the cleaner the colon, and theoretically the easier it is to see small or flat polyps. Future steps could include calculating adenoma detection rates for those enrolled in the Annie program vs the control group.

We have received inquiries regarding potential program implementation at other facilities. Success and sustainability of the program will require long-term commitment and ideally protected time for staff. It is helpful to remember that for each person who chooses to enroll in the intervention, the program currently requires that a brief consent note is placed in the patient’s chart. Thus, depending on the facilities’ resources, it is ideal for one staff member to be the designated lead to help oversee, troubleshoot, and train additional personnel. Surveys can be intermittently used to obtain feedback for improvement but are not required for sustainability. Automated text messaging is a promising addition to medicine for clinical education and communication. Future studies should examine the clinical significance (ie, adenoma detection rates) of text messaging bowel preparation protocols.

Limitations

Our study has several limitations. First, this was a single center study, thus generalizability is limited. MVAMC represents a predominantly White, male, and rural population. Second, data are likely an underestimation of the true impact of intervention, because results do not account for patients who were turned away on day of procedure (typically still reporting brown stools at time of check-in for procedure) due to poor preparation or aborted procedures secondary to poor preparation. Only about one-third of the 640 veterans opted to receive Annie text messages.

Studies have shown veterans are willing to use technology for health care; however, access to technology and lack of training remain barriers to use.¹⁴ This has been most robustly studied at the VA in veterans experiencing mental illness and homelessness. Targeted strategies to improve veteran adoption of technology within their health care include supplying veterans with cell phones and paid data plans and providing training on specific technology-based resources.^15-17 Future improvement for the Annie platform should include improved integration with CPRS. Integration will facilitate automatic import of key information such as mobile phone number or colonoscopy procedure date. Unfortunately, this is not currently an automated process, and the manual workload of staff limits sustainability. Since our study ended, the Annie database now allows an “event date” to be programmed in to center the text message series around. This will be entered at the time of Annie enrollment and eliminate manual activation of the protocol. The issue of updating information for rescheduled procedures remains.

Conclusions

There is increasing evidence that automated text messaging is a promising addition to medicine for clinical education and communication. It continues to gain traction as a readily available and acceptable option, and many patients are willing to incorporate the technology platform into their care plan. We found high patient satisfaction with our protocol, and Annie patients had cleaner bowel preparations compared with control patients. Our study supports the use of text message reminders as an effective intervention for improving patient adherence with bowel preparation instructions. We suspect that creation of a text messaging protocol designed for patients requiring outpatient extended bowel preparation will yield great benefit. As technology continues to improve, future implementation of Annie text messaging will become increasingly seamless within the field of gastroenterology and beyond.

Colorectal cancer is the third leading cause of cancer-related death in both men and women.¹ Colonoscopy is the current gold standard for screening due to the ability to remove precancerous lesions but remains highly dependent on the quality of bowel preparation.² Poor bowel preparation has been associated with impaired adenoma detection as well as increased health care utilization due to the need for a repeat colonoscopy.³

Multiple patient factors are associated with increased risk of poor bowel preparation, including age > 60 years, male sex, diabetes mellitus, and presence of a mental health diagnosis, factors that are prevalent among the veteran population.^3-5 Text messages have been shown to improve the quality of bowel preparation by increasing patients' understanding and adherence with the preparation process. Improved adherence with bowel preparation directions is associated with a cleaner colon prior to colonoscopy, leading to a thorough examination. Studies using text messaging instructions prior to colonoscopies have also shown measurable improvement in adenoma detection rate, patient preparation-associated discomfort, and completion of colonoscopy.^6-10

In 2016, the Veterans Health Administration (VHA) introduced Annie, one of the first automated text messaging services, named after Army Lieutenant Annie Fox, the first woman to receive the Purple Heart for combat. The Annie platform allows for notifications, instructions, and simple data collection. The development of this platform allows VHA practitioners to engage and educate veterans in a similar way to other health care systems using text messaging protocols. Annie text messages have been piloted for the use of hepatitis C treatment, demonstrating promise of improved medication adherence and patient satisfaction.¹¹ We aimed to develop and pilot the Annie bowel preparation protocol to improve the quality of colonoscopy bowel preparation for outpatients at the Minneapolis Veterans Affairs Medical Center (MVAMC) in Minnesota. A secondary goal included measuring patient satisfaction with the text messaging instructions for outpatient colonoscopy preparation.

Methods

We conducted a single center, prospective, endoscopist-blinded, study with two 3-month long Plan-Do-Study-Act (PDSA) cycles to improve the text messaging bowel preparation protocol at MVAMC between January 2019 and April 2020. The MVAMC Institutional Review Board determined the quality improvement project was exempt. Veterans who had outpatient colonoscopies scheduled were included. Veterans undergoing inpatient colonoscopies or outpatients who could not be reached to obtain informed consent, lacked text message capability, declined participation, or required extended colonoscopy preparation were excluded. Per MVAMC procedures, extended colonoscopy preparation was provided to patients receiving general or monitored anesthesia care, with a history of poor bowel preparation, or with risk factors for poor preparation as determined by the ordering health care professional (HCP). Standard bowel preparation involves ingestion of 4 L of polyethylene glycol 3350 with electrolytes; extended bowel preparation requires ingestion of an additional 2 L to total 6 L and uses a different set of instructions. Additionally, the patient population requiring extended bowel preparation also includes patients with spinal cord injuries, who often are admitted for assistance with extended preparation. Patients who consented to receiving text messages were placed in the Annie intervention group, and all others were placed in the control group.

The control group received standardized patient education, including a mailed copy of bowel preparation instructions and a phone call from a gastroenterology service nurse about 1 to 2 weeks before the procedure. Current MVAMC standard of care involves a phone call from a nurse to confirm that patients have received the polyethylene glycol preparation solution, the mailed instructions, have an escort and transportation, and to answer any questions. Both the usual care and intervention group received the phone call. During this call, the Annie text messaging bowel preparation protocol was introduced; if the veteran chose to participate, consent and enrollment were completed.

On the day of the colonoscopy, veterans in the intervention group were surveyed in the waiting room about their experience receiving the text messages and soliciting feedback for improvement or surveyed via telephone call within 3 days of their procedure. Patient satisfaction was quantified with a scale from 1 (low) to 10 (high), including questions about how helpful the texts were in relation to total number, timing, and content of messages as well as whether veterans would like to receive the text messages again for future procedures.

We reviewed individual charts and collected Boston Bowel Preparation Scale (BBPS) scores to determine adequate preparation. BBPS assigns a score of 0 to 3 for the right, transverse, and left colon applied upon withdrawal after flushing and suctioning have been completed.¹² Adequate preparation is considered a total score of ≥ 6 with no segment scoring < 2. This method of preparation assessment is preferred due to its ability to account for difference in preparation quality among colonic segments, well-defined scoring characteristics, and several studies validating its use showing inter- and intraobserver reliability.¹² Follow-up studies have shown validity of the BBPS when compared with relevant outcomes such as polyp detection rate and recommended timing for repeat procedure.¹³ Variables associated with poor bowel preparation (ie, gender, prior abdominal surgery, impaired mobility, high body mass index, diabetes mellitus, stroke, dementia, any neurologic diagnosis, cirrhosis, smoking, polypharmacy [> 8 active medications], and narcotic or tricyclic antidepressant medication use) were also collected through chart review.^3-5 We note that immobility was defined by International Classification of Diseases (ICD)-9 and ICD-10 codes and prescriptions for assistive devices (ie, canes, wheelchairs, 4-wheeled walkers).

Veterans assent to be enrolled in Annie. After enrollment, veterans must text back a specific word response to an initial text message to receive the protocolized messages from the Annie program. A contact phone number to the gastrointestinal nurse line was provided for questions during business hours. The start date for the text message protocol is 6 days prior to the procedure date. If a patient rescheduled their colonoscopy, the Annie database was updated manually.

Statistical Analysis

We used both Pearson χ² test and 2-sample t test analyses to compare demographic information and patient satisfaction scores between the control and intervention groups. We compared continuous BBPS scores between Annie intervention vs control group using parametric and nonparametric independent t tests using the Mann-Whitney U test. We repeated this analysis controlling for both mental health diagnoses and age using linear regression. We were unable to survey 61 of the 187 veterans who received Annie text messages. 

RESULTS

During PDSA cycles 1 and 2, 640 veterans were scheduled for outpatient colonoscopy: 453 veterans were in the control group; 187 veterans were in the intervention group, of which 126 were surveyed. A significant percentage of veterans declined participation because they felt like they did not need reinforced education; others were not eligible for Annie due to requirement for extended bowel preparation, cancelled colonoscopy, inability to physically read text messages, or lack of cell phone.

The mean (SD) age was 65 (8) years; 184 (28.8%) had a diabetes mellitus diagnosis, and the mean (SD) body mass index was 31.6 (6.4). The Annie group was slightly more likely to have mental health diagnoses and lower age compared with the control group (Table 1).

Patient Feedback

We collected feedback from veterans after each PDSA cycle to identify areas for improvement by both in-person and telephone surveys. Based on feedback from PDSA cycle 1, we decreased the total number of text messages to create a more succinct set of instructions. The most frequently requested change involved timing the text messages to align with the exact morning a specific instruction should take place.

Patient satisfaction with the Annie text messaging service was high.

DISCUSSION

To our knowledge, this is the first report of using Annie at a VAMC for colonoscopy bowel preparation improvement. We found a statistically significant improvement in the average BBPS in those receiving Annie text messages compared with the routine care control group. We also found high levels of patient satisfaction with most patients requesting to receive them again for future procedures.

The clinical significance of a BBPS of 7.8 vs 8.2 is unclear, although any score > 6 is considered to be adequate. However, subjectively speaking, the higher the BBPS the cleaner the colon, and theoretically the easier it is to see small or flat polyps. Future steps could include calculating adenoma detection rates for those enrolled in the Annie program vs the control group.

We have received inquiries regarding potential program implementation at other facilities. Success and sustainability of the program will require long-term commitment and ideally protected time for staff. It is helpful to remember that for each person who chooses to enroll in the intervention, the program currently requires that a brief consent note is placed in the patient’s chart. Thus, depending on the facilities’ resources, it is ideal for one staff member to be the designated lead to help oversee, troubleshoot, and train additional personnel. Surveys can be intermittently used to obtain feedback for improvement but are not required for sustainability. Automated text messaging is a promising addition to medicine for clinical education and communication. Future studies should examine the clinical significance (ie, adenoma detection rates) of text messaging bowel preparation protocols.

Limitations

Our study has several limitations. First, this was a single center study, thus generalizability is limited. MVAMC represents a predominantly White, male, and rural population. Second, data are likely an underestimation of the true impact of intervention, because results do not account for patients who were turned away on day of procedure (typically still reporting brown stools at time of check-in for procedure) due to poor preparation or aborted procedures secondary to poor preparation. Only about one-third of the 640 veterans opted to receive Annie text messages.

Studies have shown veterans are willing to use technology for health care; however, access to technology and lack of training remain barriers to use.¹⁴ This has been most robustly studied at the VA in veterans experiencing mental illness and homelessness. Targeted strategies to improve veteran adoption of technology within their health care include supplying veterans with cell phones and paid data plans and providing training on specific technology-based resources.^15-17 Future improvement for the Annie platform should include improved integration with CPRS. Integration will facilitate automatic import of key information such as mobile phone number or colonoscopy procedure date. Unfortunately, this is not currently an automated process, and the manual workload of staff limits sustainability. Since our study ended, the Annie database now allows an “event date” to be programmed in to center the text message series around. This will be entered at the time of Annie enrollment and eliminate manual activation of the protocol. The issue of updating information for rescheduled procedures remains.

Conclusions

There is increasing evidence that automated text messaging is a promising addition to medicine for clinical education and communication. It continues to gain traction as a readily available and acceptable option, and many patients are willing to incorporate the technology platform into their care plan. We found high patient satisfaction with our protocol, and Annie patients had cleaner bowel preparations compared with control patients. Our study supports the use of text message reminders as an effective intervention for improving patient adherence with bowel preparation instructions. We suspect that creation of a text messaging protocol designed for patients requiring outpatient extended bowel preparation will yield great benefit. As technology continues to improve, future implementation of Annie text messaging will become increasingly seamless within the field of gastroenterology and beyond.

Colorectal cancer is the third leading cause of cancer-related death in both men and women.¹ Colonoscopy is the current gold standard for screening due to the ability to remove precancerous lesions but remains highly dependent on the quality of bowel preparation.² Poor bowel preparation has been associated with impaired adenoma detection as well as increased health care utilization due to the need for a repeat colonoscopy.³

Multiple patient factors are associated with increased risk of poor bowel preparation, including age > 60 years, male sex, diabetes mellitus, and presence of a mental health diagnosis, factors that are prevalent among the veteran population.^3-5 Text messages have been shown to improve the quality of bowel preparation by increasing patients' understanding and adherence with the preparation process. Improved adherence with bowel preparation directions is associated with a cleaner colon prior to colonoscopy, leading to a thorough examination. Studies using text messaging instructions prior to colonoscopies have also shown measurable improvement in adenoma detection rate, patient preparation-associated discomfort, and completion of colonoscopy.^6-10

In 2016, the Veterans Health Administration (VHA) introduced Annie, one of the first automated text messaging services, named after Army Lieutenant Annie Fox, the first woman to receive the Purple Heart for combat. The Annie platform allows for notifications, instructions, and simple data collection. The development of this platform allows VHA practitioners to engage and educate veterans in a similar way to other health care systems using text messaging protocols. Annie text messages have been piloted for the use of hepatitis C treatment, demonstrating promise of improved medication adherence and patient satisfaction.¹¹ We aimed to develop and pilot the Annie bowel preparation protocol to improve the quality of colonoscopy bowel preparation for outpatients at the Minneapolis Veterans Affairs Medical Center (MVAMC) in Minnesota. A secondary goal included measuring patient satisfaction with the text messaging instructions for outpatient colonoscopy preparation.

Methods

We conducted a single center, prospective, endoscopist-blinded, study with two 3-month long Plan-Do-Study-Act (PDSA) cycles to improve the text messaging bowel preparation protocol at MVAMC between January 2019 and April 2020. The MVAMC Institutional Review Board determined the quality improvement project was exempt. Veterans who had outpatient colonoscopies scheduled were included. Veterans undergoing inpatient colonoscopies or outpatients who could not be reached to obtain informed consent, lacked text message capability, declined participation, or required extended colonoscopy preparation were excluded. Per MVAMC procedures, extended colonoscopy preparation was provided to patients receiving general or monitored anesthesia care, with a history of poor bowel preparation, or with risk factors for poor preparation as determined by the ordering health care professional (HCP). Standard bowel preparation involves ingestion of 4 L of polyethylene glycol 3350 with electrolytes; extended bowel preparation requires ingestion of an additional 2 L to total 6 L and uses a different set of instructions. Additionally, the patient population requiring extended bowel preparation also includes patients with spinal cord injuries, who often are admitted for assistance with extended preparation. Patients who consented to receiving text messages were placed in the Annie intervention group, and all others were placed in the control group.

The control group received standardized patient education, including a mailed copy of bowel preparation instructions and a phone call from a gastroenterology service nurse about 1 to 2 weeks before the procedure. Current MVAMC standard of care involves a phone call from a nurse to confirm that patients have received the polyethylene glycol preparation solution, the mailed instructions, have an escort and transportation, and to answer any questions. Both the usual care and intervention group received the phone call. During this call, the Annie text messaging bowel preparation protocol was introduced; if the veteran chose to participate, consent and enrollment were completed.

On the day of the colonoscopy, veterans in the intervention group were surveyed in the waiting room about their experience receiving the text messages and soliciting feedback for improvement or surveyed via telephone call within 3 days of their procedure. Patient satisfaction was quantified with a scale from 1 (low) to 10 (high), including questions about how helpful the texts were in relation to total number, timing, and content of messages as well as whether veterans would like to receive the text messages again for future procedures.

We reviewed individual charts and collected Boston Bowel Preparation Scale (BBPS) scores to determine adequate preparation. BBPS assigns a score of 0 to 3 for the right, transverse, and left colon applied upon withdrawal after flushing and suctioning have been completed.¹² Adequate preparation is considered a total score of ≥ 6 with no segment scoring < 2. This method of preparation assessment is preferred due to its ability to account for difference in preparation quality among colonic segments, well-defined scoring characteristics, and several studies validating its use showing inter- and intraobserver reliability.¹² Follow-up studies have shown validity of the BBPS when compared with relevant outcomes such as polyp detection rate and recommended timing for repeat procedure.¹³ Variables associated with poor bowel preparation (ie, gender, prior abdominal surgery, impaired mobility, high body mass index, diabetes mellitus, stroke, dementia, any neurologic diagnosis, cirrhosis, smoking, polypharmacy [> 8 active medications], and narcotic or tricyclic antidepressant medication use) were also collected through chart review.^3-5 We note that immobility was defined by International Classification of Diseases (ICD)-9 and ICD-10 codes and prescriptions for assistive devices (ie, canes, wheelchairs, 4-wheeled walkers).

Veterans assent to be enrolled in Annie. After enrollment, veterans must text back a specific word response to an initial text message to receive the protocolized messages from the Annie program. A contact phone number to the gastrointestinal nurse line was provided for questions during business hours. The start date for the text message protocol is 6 days prior to the procedure date. If a patient rescheduled their colonoscopy, the Annie database was updated manually.

Statistical Analysis

We used both Pearson χ² test and 2-sample t test analyses to compare demographic information and patient satisfaction scores between the control and intervention groups. We compared continuous BBPS scores between Annie intervention vs control group using parametric and nonparametric independent t tests using the Mann-Whitney U test. We repeated this analysis controlling for both mental health diagnoses and age using linear regression. We were unable to survey 61 of the 187 veterans who received Annie text messages. 

RESULTS

During PDSA cycles 1 and 2, 640 veterans were scheduled for outpatient colonoscopy: 453 veterans were in the control group; 187 veterans were in the intervention group, of which 126 were surveyed. A significant percentage of veterans declined participation because they felt like they did not need reinforced education; others were not eligible for Annie due to requirement for extended bowel preparation, cancelled colonoscopy, inability to physically read text messages, or lack of cell phone.

The mean (SD) age was 65 (8) years; 184 (28.8%) had a diabetes mellitus diagnosis, and the mean (SD) body mass index was 31.6 (6.4). The Annie group was slightly more likely to have mental health diagnoses and lower age compared with the control group (Table 1).

Patient Feedback

We collected feedback from veterans after each PDSA cycle to identify areas for improvement by both in-person and telephone surveys. Based on feedback from PDSA cycle 1, we decreased the total number of text messages to create a more succinct set of instructions. The most frequently requested change involved timing the text messages to align with the exact morning a specific instruction should take place.

Patient satisfaction with the Annie text messaging service was high.

DISCUSSION

To our knowledge, this is the first report of using Annie at a VAMC for colonoscopy bowel preparation improvement. We found a statistically significant improvement in the average BBPS in those receiving Annie text messages compared with the routine care control group. We also found high levels of patient satisfaction with most patients requesting to receive them again for future procedures.

The clinical significance of a BBPS of 7.8 vs 8.2 is unclear, although any score > 6 is considered to be adequate. However, subjectively speaking, the higher the BBPS the cleaner the colon, and theoretically the easier it is to see small or flat polyps. Future steps could include calculating adenoma detection rates for those enrolled in the Annie program vs the control group.

We have received inquiries regarding potential program implementation at other facilities. Success and sustainability of the program will require long-term commitment and ideally protected time for staff. It is helpful to remember that for each person who chooses to enroll in the intervention, the program currently requires that a brief consent note is placed in the patient’s chart. Thus, depending on the facilities’ resources, it is ideal for one staff member to be the designated lead to help oversee, troubleshoot, and train additional personnel. Surveys can be intermittently used to obtain feedback for improvement but are not required for sustainability. Automated text messaging is a promising addition to medicine for clinical education and communication. Future studies should examine the clinical significance (ie, adenoma detection rates) of text messaging bowel preparation protocols.

Limitations

Our study has several limitations. First, this was a single center study, thus generalizability is limited. MVAMC represents a predominantly White, male, and rural population. Second, data are likely an underestimation of the true impact of intervention, because results do not account for patients who were turned away on day of procedure (typically still reporting brown stools at time of check-in for procedure) due to poor preparation or aborted procedures secondary to poor preparation. Only about one-third of the 640 veterans opted to receive Annie text messages.

Studies have shown veterans are willing to use technology for health care; however, access to technology and lack of training remain barriers to use.¹⁴ This has been most robustly studied at the VA in veterans experiencing mental illness and homelessness. Targeted strategies to improve veteran adoption of technology within their health care include supplying veterans with cell phones and paid data plans and providing training on specific technology-based resources.^15-17 Future improvement for the Annie platform should include improved integration with CPRS. Integration will facilitate automatic import of key information such as mobile phone number or colonoscopy procedure date. Unfortunately, this is not currently an automated process, and the manual workload of staff limits sustainability. Since our study ended, the Annie database now allows an “event date” to be programmed in to center the text message series around. This will be entered at the time of Annie enrollment and eliminate manual activation of the protocol. The issue of updating information for rescheduled procedures remains.

Conclusions

There is increasing evidence that automated text messaging is a promising addition to medicine for clinical education and communication. It continues to gain traction as a readily available and acceptable option, and many patients are willing to incorporate the technology platform into their care plan. We found high patient satisfaction with our protocol, and Annie patients had cleaner bowel preparations compared with control patients. Our study supports the use of text message reminders as an effective intervention for improving patient adherence with bowel preparation instructions. We suspect that creation of a text messaging protocol designed for patients requiring outpatient extended bowel preparation will yield great benefit. As technology continues to improve, future implementation of Annie text messaging will become increasingly seamless within the field of gastroenterology and beyond.

Radiation therapy can cause long-term dysphagia that seriously affects quality of life for survivors of head and neck (H&N) cancer.^1-3 Numerous studies have linked pharyngeal constrictor dose to long-term dysphagia, but conclusions about the dose distribution that can be safely tolerated have been inconsistent. For example, a group from the Netherlands found that the mean dose to the superior pharyngeal constrictor muscle and the supraglottic larynx were each predictive of dysphagia.⁴ A subsequent Vanderbilt study refuted these findings, reporting that these structures were not predictive but that dose to the inferior pharyngeal constrictor muscle was.⁵ Other studies have connected late dysphagia with dose to the middle pharyngeal constrictor muscle, total larynx, oral cavity, contralateral submandibular gland, contralateral parotid gland, or a combination of these structures.^6-14 NRG Oncology trials commonly evaluate dose to the “uninvolved pharynx,” which is the total pharyngeal constrictor muscle volume minus the planning target volume for the lowest dose target volume. NRG H&N trials 3, 4, 5, 6, 8, and 9 all use uninvolved pharynx mean dose  ≤ 45 Gy as a constraint to judge radiation plan quality.

Differences in methodology or patient population may explain the inconsistency of prior studies on dosimetric predictors of dysphagia, but it is possible that these studies did not evaluate the optimal metric for dysphagia. This study evaluates a novel organ at risk, the contralateral pharyngeal constrictor muscle, to determine whether dose to this structure is predictive of late swallowing function. The study also compares a constraint based on this structure to the NRG uninvolved pharynx constraint mentioned earlier.

Methods

This study is a retrospective review of patients treated at the Richard L. Roudebush Veterans Affairs (VA) Medical Center in Indianapolis, Indiana. Patients were identified by searching the VA Cancer Registry for patients treated for H&N squamous cell carcinoma between September 1, 2016, and August 30, 2019. Eligible sites included cancers of the nasopharynx, oropharynx, hypopharynx, larynx and oral cavity, as well as H&N cancer of an unknown primary site. Only patients treated with primary radiation with concurrent systemic therapy were included. Patients were excluded if they had prior surgery or radiation to the H&N.

The pharyngeal constrictor muscles were contoured per the techniques described by Bhide and colleagues.¹¹ The contralateral constrictor was defined as the half of the constrictor volume contralateral to the primary site. For midline tumors, the side of the neck with a lower volume of lymph node metastases was judged to be the contralateral side.

Results

The VA Cancer Registry identified 113 patients treated for H&N cancer during the study period. Of these, 55 patients met the inclusion criteria. No patients were lost to follow-up. The median follow-up was 29 months. The median age was 67 years (range, 41-83) (Table 1).

All patients were treated with intensity-modulated radiotherapy (IMRT). Patients treated with a sequential boost had an initial dose of 54 Gy and/or 50 Gy, followed by a boost to a total of 70 Gy at 2 Gy per fraction. Patients treated with a simultaneous integrated boost (SIB) technique received 69.96 Gy in 33 fractions, with elective volumes treated to 54.45 Gy in 33 fractions. Both patients with nasopharyngeal cancer were treated with SIB plans and had an intermediate dose volume of 59.4 Gy.

Systemic therapy was weekly cisplatin in 41 patients (75%) and cetuximab in 14 (25%). Twenty percent of patients receiving cisplatin switched to an alternative agent during treatment, most commonly carboplatin.

Forty-nine patients (89%) had a G-tube placed before starting radiation. G-tubes were in place for an interval of 0 to 47 months (mean, 8.6); 12 (22%) had a G-tube > 12 months. After completion of radiation, 18 patients (33%) had an abnormal MBS. These were done 1 to 50 months (mean, 14.8) after completion of radiation. Abnormal MBS occurred ≥ 12 months after radiation in 9 patients, 5 of whom had their G-tube in place for less than a year.

Forty-six patients (84%) survived more than 1 year and could be evaluated for late swallowing function. One-year dysphagia was seen in 17 (37%) of these patients. Recurrence was seen in 20 patients (36%), with locoregional recurrence in 12 (60%) of these cases. Recurrence occurred at a range of 0 to 15 months (mean, 5.6). Neither recurrence (P = .69) nor locoregional recurrence (P = .11) was associated with increased 1-year dysphagia.

In patients who could be evaluated for long-term swallowing function, contralateral constrictor V60 ranged from 0% to 100% (median, 51%). V60 was < 40% in 18 patients (39%). With V60 < 40%, there was a 6% rate of 1-year dysphagia compared with 57% for V60 ≥ 40% (P < .001).

Patients with contralateral constrictor V60 < 40 and V60 ≥ 40 both had a mean age of 65 years. χ² analysis did not show a difference in T stage or systemic treatment but did show that patients with V60 < 40% were more likely to have N1 disease (P = .01), and less likely to have N2 disease (P = .01) compared with patients with V60 ≥ 40%. The difference in 1-year dysphagia between N0 to N1 patients (27%) and N2 to N3 patients (46%) was not statistically significant (P = .19).

In patients who could be evaluated for long-term swallowing function, the uninvolved pharynx volume median of the total constrictor volume was 32% (range, < 1%-62%). The uninvolved pharynx mean dose ranged from 28 to 68 Gy (median, 45). When the uninvolved pharynx mean dose was < 45 Gy, 1-year dysphagia was 22% compared with 52% with a dose ≥ 45 Gy (P = .03).

Discussion

This is the first study to link contralateral constrictor dose to long-term dysphagia in patients treated with radiation for H&N cancer. Editing the boost volume off air cavities was associated with lower contralateral constrictor V60 and with less long-term dysphagia. This may indicate that optimizing plans to meet a contralateral constrictor constraint can reduce rates of long-term dysphagia.

The most useful clinical predictors are those that identify a patient at low risk for toxicity. These constraints are useful because they reassure physicians that treatments will have a favorable risk/benefit ratio while identifying plans that may need modification before starting treatment.

The contralateral constrictor outperformed the uninvolved pharynx in identifying patients at low risk for long-term dysphagia. This difference could not be overcome by decreasing the threshold of the pharynx constraint, as 17% of patients with dysphagia had a mean dose of < 40 Gy to the uninvolved pharynx, which was not statistically significant.

An advantage of contralateral constrictor is that it is independent of planning target volume (PTV) size. The uninvolved pharynx structure depends on the PTV contour, so it may obscure a connection between PTV size and dysphagia.

In the context of a clinical trial, only measuring dose to the uninvolved pharynx may allow more plans to meet constraints, but even in NRG trials, physicians have some control over target volumes. For example, NRG HN009, a national trial for patients with H&N cancer, recommends editing the CTV_7000 (clinical target volume treated to 70 Gy) off air cavities but does not define how much the volume should be cropped or specify protocol violations if the volume is not cropped.¹⁵ Furthermore, constraints used in clinical trials are often adopted for use outside the trial, where physicians have extensive control over target volumes.

The broad range of uninvolved pharynx volume relative to total constrictor volume confounds predictions using this variable. For example, according to the NRG constraint, a patient with an uninvolved pharynx mean dose of 44 Gy will have a low risk of dysphagia even if this structure is only 1% of the total constrictor. The contralateral constrictor is always about 50% of the total constrictor volume, which means that predictions using this structure will not be confounded by the same variation in volume size.

Figure 2 shows a representative patient who met the NRG uninvolved pharynx constraint but developed long-term dysphagia.

Limitations

This study is limited by its single institution, retrospective design, small sample size, and by all patients being male. The high correlation between air cavity editing and the use of SIB makes it impossible to assess the impact of each technique individually. Patients with contralateral constrictor V60 < 40% were less likely to have N2 disease, but N2 to N3 disease did not predict higher 1-year dysphagia, so the difference in N-category cannot fully explain the difference in 1-year dysphagia. It is possible that unreported factors, such as CTV, may contribute significantly to swallowing function. Nevertheless, within the study population, contralateral constrictor dose was able to identify a group with a low rate of long-term dysphagia.

Conclusions

Contralateral constrictor dose is a promising predictor of late dysphagia for patients with H&N cancer treated with radiation with concurrent systemic therapy. Contralateral constrictor V60 < 40% was able to identify a group of patients with a low rate of 1-year dysphagia in this single-center retrospective study. The correlation between air cavity editing and contralateral constrictor V60 suggests that contralateral constrictor dose may depend partly on technique. Further studies are needed to see if the contralateral constrictor dose can be used to predict long-term dysphagia prospectively and in other patient populations.

Radiation therapy can cause long-term dysphagia that seriously affects quality of life for survivors of head and neck (H&N) cancer.^1-3 Numerous studies have linked pharyngeal constrictor dose to long-term dysphagia, but conclusions about the dose distribution that can be safely tolerated have been inconsistent. For example, a group from the Netherlands found that the mean dose to the superior pharyngeal constrictor muscle and the supraglottic larynx were each predictive of dysphagia.⁴ A subsequent Vanderbilt study refuted these findings, reporting that these structures were not predictive but that dose to the inferior pharyngeal constrictor muscle was.⁵ Other studies have connected late dysphagia with dose to the middle pharyngeal constrictor muscle, total larynx, oral cavity, contralateral submandibular gland, contralateral parotid gland, or a combination of these structures.^6-14 NRG Oncology trials commonly evaluate dose to the “uninvolved pharynx,” which is the total pharyngeal constrictor muscle volume minus the planning target volume for the lowest dose target volume. NRG H&N trials 3, 4, 5, 6, 8, and 9 all use uninvolved pharynx mean dose  ≤ 45 Gy as a constraint to judge radiation plan quality.

Differences in methodology or patient population may explain the inconsistency of prior studies on dosimetric predictors of dysphagia, but it is possible that these studies did not evaluate the optimal metric for dysphagia. This study evaluates a novel organ at risk, the contralateral pharyngeal constrictor muscle, to determine whether dose to this structure is predictive of late swallowing function. The study also compares a constraint based on this structure to the NRG uninvolved pharynx constraint mentioned earlier.

Methods

This study is a retrospective review of patients treated at the Richard L. Roudebush Veterans Affairs (VA) Medical Center in Indianapolis, Indiana. Patients were identified by searching the VA Cancer Registry for patients treated for H&N squamous cell carcinoma between September 1, 2016, and August 30, 2019. Eligible sites included cancers of the nasopharynx, oropharynx, hypopharynx, larynx and oral cavity, as well as H&N cancer of an unknown primary site. Only patients treated with primary radiation with concurrent systemic therapy were included. Patients were excluded if they had prior surgery or radiation to the H&N.

The pharyngeal constrictor muscles were contoured per the techniques described by Bhide and colleagues.¹¹ The contralateral constrictor was defined as the half of the constrictor volume contralateral to the primary site. For midline tumors, the side of the neck with a lower volume of lymph node metastases was judged to be the contralateral side.

Results

The VA Cancer Registry identified 113 patients treated for H&N cancer during the study period. Of these, 55 patients met the inclusion criteria. No patients were lost to follow-up. The median follow-up was 29 months. The median age was 67 years (range, 41-83) (Table 1).

All patients were treated with intensity-modulated radiotherapy (IMRT). Patients treated with a sequential boost had an initial dose of 54 Gy and/or 50 Gy, followed by a boost to a total of 70 Gy at 2 Gy per fraction. Patients treated with a simultaneous integrated boost (SIB) technique received 69.96 Gy in 33 fractions, with elective volumes treated to 54.45 Gy in 33 fractions. Both patients with nasopharyngeal cancer were treated with SIB plans and had an intermediate dose volume of 59.4 Gy.

Systemic therapy was weekly cisplatin in 41 patients (75%) and cetuximab in 14 (25%). Twenty percent of patients receiving cisplatin switched to an alternative agent during treatment, most commonly carboplatin.

Forty-nine patients (89%) had a G-tube placed before starting radiation. G-tubes were in place for an interval of 0 to 47 months (mean, 8.6); 12 (22%) had a G-tube > 12 months. After completion of radiation, 18 patients (33%) had an abnormal MBS. These were done 1 to 50 months (mean, 14.8) after completion of radiation. Abnormal MBS occurred ≥ 12 months after radiation in 9 patients, 5 of whom had their G-tube in place for less than a year.

Forty-six patients (84%) survived more than 1 year and could be evaluated for late swallowing function. One-year dysphagia was seen in 17 (37%) of these patients. Recurrence was seen in 20 patients (36%), with locoregional recurrence in 12 (60%) of these cases. Recurrence occurred at a range of 0 to 15 months (mean, 5.6). Neither recurrence (P = .69) nor locoregional recurrence (P = .11) was associated with increased 1-year dysphagia.

In patients who could be evaluated for long-term swallowing function, contralateral constrictor V60 ranged from 0% to 100% (median, 51%). V60 was < 40% in 18 patients (39%). With V60 < 40%, there was a 6% rate of 1-year dysphagia compared with 57% for V60 ≥ 40% (P < .001).

Patients with contralateral constrictor V60 < 40 and V60 ≥ 40 both had a mean age of 65 years. χ² analysis did not show a difference in T stage or systemic treatment but did show that patients with V60 < 40% were more likely to have N1 disease (P = .01), and less likely to have N2 disease (P = .01) compared with patients with V60 ≥ 40%. The difference in 1-year dysphagia between N0 to N1 patients (27%) and N2 to N3 patients (46%) was not statistically significant (P = .19).

In patients who could be evaluated for long-term swallowing function, the uninvolved pharynx volume median of the total constrictor volume was 32% (range, < 1%-62%). The uninvolved pharynx mean dose ranged from 28 to 68 Gy (median, 45). When the uninvolved pharynx mean dose was < 45 Gy, 1-year dysphagia was 22% compared with 52% with a dose ≥ 45 Gy (P = .03).

Discussion

This is the first study to link contralateral constrictor dose to long-term dysphagia in patients treated with radiation for H&N cancer. Editing the boost volume off air cavities was associated with lower contralateral constrictor V60 and with less long-term dysphagia. This may indicate that optimizing plans to meet a contralateral constrictor constraint can reduce rates of long-term dysphagia.

The most useful clinical predictors are those that identify a patient at low risk for toxicity. These constraints are useful because they reassure physicians that treatments will have a favorable risk/benefit ratio while identifying plans that may need modification before starting treatment.

The contralateral constrictor outperformed the uninvolved pharynx in identifying patients at low risk for long-term dysphagia. This difference could not be overcome by decreasing the threshold of the pharynx constraint, as 17% of patients with dysphagia had a mean dose of < 40 Gy to the uninvolved pharynx, which was not statistically significant.

An advantage of contralateral constrictor is that it is independent of planning target volume (PTV) size. The uninvolved pharynx structure depends on the PTV contour, so it may obscure a connection between PTV size and dysphagia.

In the context of a clinical trial, only measuring dose to the uninvolved pharynx may allow more plans to meet constraints, but even in NRG trials, physicians have some control over target volumes. For example, NRG HN009, a national trial for patients with H&N cancer, recommends editing the CTV_7000 (clinical target volume treated to 70 Gy) off air cavities but does not define how much the volume should be cropped or specify protocol violations if the volume is not cropped.¹⁵ Furthermore, constraints used in clinical trials are often adopted for use outside the trial, where physicians have extensive control over target volumes.

The broad range of uninvolved pharynx volume relative to total constrictor volume confounds predictions using this variable. For example, according to the NRG constraint, a patient with an uninvolved pharynx mean dose of 44 Gy will have a low risk of dysphagia even if this structure is only 1% of the total constrictor. The contralateral constrictor is always about 50% of the total constrictor volume, which means that predictions using this structure will not be confounded by the same variation in volume size.

Figure 2 shows a representative patient who met the NRG uninvolved pharynx constraint but developed long-term dysphagia.

Limitations

This study is limited by its single institution, retrospective design, small sample size, and by all patients being male. The high correlation between air cavity editing and the use of SIB makes it impossible to assess the impact of each technique individually. Patients with contralateral constrictor V60 < 40% were less likely to have N2 disease, but N2 to N3 disease did not predict higher 1-year dysphagia, so the difference in N-category cannot fully explain the difference in 1-year dysphagia. It is possible that unreported factors, such as CTV, may contribute significantly to swallowing function. Nevertheless, within the study population, contralateral constrictor dose was able to identify a group with a low rate of long-term dysphagia.

Conclusions

Contralateral constrictor dose is a promising predictor of late dysphagia for patients with H&N cancer treated with radiation with concurrent systemic therapy. Contralateral constrictor V60 < 40% was able to identify a group of patients with a low rate of 1-year dysphagia in this single-center retrospective study. The correlation between air cavity editing and contralateral constrictor V60 suggests that contralateral constrictor dose may depend partly on technique. Further studies are needed to see if the contralateral constrictor dose can be used to predict long-term dysphagia prospectively and in other patient populations.

Radiation therapy can cause long-term dysphagia that seriously affects quality of life for survivors of head and neck (H&N) cancer.^1-3 Numerous studies have linked pharyngeal constrictor dose to long-term dysphagia, but conclusions about the dose distribution that can be safely tolerated have been inconsistent. For example, a group from the Netherlands found that the mean dose to the superior pharyngeal constrictor muscle and the supraglottic larynx were each predictive of dysphagia.⁴ A subsequent Vanderbilt study refuted these findings, reporting that these structures were not predictive but that dose to the inferior pharyngeal constrictor muscle was.⁵ Other studies have connected late dysphagia with dose to the middle pharyngeal constrictor muscle, total larynx, oral cavity, contralateral submandibular gland, contralateral parotid gland, or a combination of these structures.^6-14 NRG Oncology trials commonly evaluate dose to the “uninvolved pharynx,” which is the total pharyngeal constrictor muscle volume minus the planning target volume for the lowest dose target volume. NRG H&N trials 3, 4, 5, 6, 8, and 9 all use uninvolved pharynx mean dose  ≤ 45 Gy as a constraint to judge radiation plan quality.

Differences in methodology or patient population may explain the inconsistency of prior studies on dosimetric predictors of dysphagia, but it is possible that these studies did not evaluate the optimal metric for dysphagia. This study evaluates a novel organ at risk, the contralateral pharyngeal constrictor muscle, to determine whether dose to this structure is predictive of late swallowing function. The study also compares a constraint based on this structure to the NRG uninvolved pharynx constraint mentioned earlier.

Methods

This study is a retrospective review of patients treated at the Richard L. Roudebush Veterans Affairs (VA) Medical Center in Indianapolis, Indiana. Patients were identified by searching the VA Cancer Registry for patients treated for H&N squamous cell carcinoma between September 1, 2016, and August 30, 2019. Eligible sites included cancers of the nasopharynx, oropharynx, hypopharynx, larynx and oral cavity, as well as H&N cancer of an unknown primary site. Only patients treated with primary radiation with concurrent systemic therapy were included. Patients were excluded if they had prior surgery or radiation to the H&N.

The pharyngeal constrictor muscles were contoured per the techniques described by Bhide and colleagues.¹¹ The contralateral constrictor was defined as the half of the constrictor volume contralateral to the primary site. For midline tumors, the side of the neck with a lower volume of lymph node metastases was judged to be the contralateral side.

Results

The VA Cancer Registry identified 113 patients treated for H&N cancer during the study period. Of these, 55 patients met the inclusion criteria. No patients were lost to follow-up. The median follow-up was 29 months. The median age was 67 years (range, 41-83) (Table 1).

All patients were treated with intensity-modulated radiotherapy (IMRT). Patients treated with a sequential boost had an initial dose of 54 Gy and/or 50 Gy, followed by a boost to a total of 70 Gy at 2 Gy per fraction. Patients treated with a simultaneous integrated boost (SIB) technique received 69.96 Gy in 33 fractions, with elective volumes treated to 54.45 Gy in 33 fractions. Both patients with nasopharyngeal cancer were treated with SIB plans and had an intermediate dose volume of 59.4 Gy.

Systemic therapy was weekly cisplatin in 41 patients (75%) and cetuximab in 14 (25%). Twenty percent of patients receiving cisplatin switched to an alternative agent during treatment, most commonly carboplatin.

Forty-nine patients (89%) had a G-tube placed before starting radiation. G-tubes were in place for an interval of 0 to 47 months (mean, 8.6); 12 (22%) had a G-tube > 12 months. After completion of radiation, 18 patients (33%) had an abnormal MBS. These were done 1 to 50 months (mean, 14.8) after completion of radiation. Abnormal MBS occurred ≥ 12 months after radiation in 9 patients, 5 of whom had their G-tube in place for less than a year.

Forty-six patients (84%) survived more than 1 year and could be evaluated for late swallowing function. One-year dysphagia was seen in 17 (37%) of these patients. Recurrence was seen in 20 patients (36%), with locoregional recurrence in 12 (60%) of these cases. Recurrence occurred at a range of 0 to 15 months (mean, 5.6). Neither recurrence (P = .69) nor locoregional recurrence (P = .11) was associated with increased 1-year dysphagia.

In patients who could be evaluated for long-term swallowing function, contralateral constrictor V60 ranged from 0% to 100% (median, 51%). V60 was < 40% in 18 patients (39%). With V60 < 40%, there was a 6% rate of 1-year dysphagia compared with 57% for V60 ≥ 40% (P < .001).

Patients with contralateral constrictor V60 < 40 and V60 ≥ 40 both had a mean age of 65 years. χ² analysis did not show a difference in T stage or systemic treatment but did show that patients with V60 < 40% were more likely to have N1 disease (P = .01), and less likely to have N2 disease (P = .01) compared with patients with V60 ≥ 40%. The difference in 1-year dysphagia between N0 to N1 patients (27%) and N2 to N3 patients (46%) was not statistically significant (P = .19).

In patients who could be evaluated for long-term swallowing function, the uninvolved pharynx volume median of the total constrictor volume was 32% (range, < 1%-62%). The uninvolved pharynx mean dose ranged from 28 to 68 Gy (median, 45). When the uninvolved pharynx mean dose was < 45 Gy, 1-year dysphagia was 22% compared with 52% with a dose ≥ 45 Gy (P = .03).

Discussion

This is the first study to link contralateral constrictor dose to long-term dysphagia in patients treated with radiation for H&N cancer. Editing the boost volume off air cavities was associated with lower contralateral constrictor V60 and with less long-term dysphagia. This may indicate that optimizing plans to meet a contralateral constrictor constraint can reduce rates of long-term dysphagia.

The most useful clinical predictors are those that identify a patient at low risk for toxicity. These constraints are useful because they reassure physicians that treatments will have a favorable risk/benefit ratio while identifying plans that may need modification before starting treatment.

The contralateral constrictor outperformed the uninvolved pharynx in identifying patients at low risk for long-term dysphagia. This difference could not be overcome by decreasing the threshold of the pharynx constraint, as 17% of patients with dysphagia had a mean dose of < 40 Gy to the uninvolved pharynx, which was not statistically significant.

An advantage of contralateral constrictor is that it is independent of planning target volume (PTV) size. The uninvolved pharynx structure depends on the PTV contour, so it may obscure a connection between PTV size and dysphagia.

In the context of a clinical trial, only measuring dose to the uninvolved pharynx may allow more plans to meet constraints, but even in NRG trials, physicians have some control over target volumes. For example, NRG HN009, a national trial for patients with H&N cancer, recommends editing the CTV_7000 (clinical target volume treated to 70 Gy) off air cavities but does not define how much the volume should be cropped or specify protocol violations if the volume is not cropped.¹⁵ Furthermore, constraints used in clinical trials are often adopted for use outside the trial, where physicians have extensive control over target volumes.

The broad range of uninvolved pharynx volume relative to total constrictor volume confounds predictions using this variable. For example, according to the NRG constraint, a patient with an uninvolved pharynx mean dose of 44 Gy will have a low risk of dysphagia even if this structure is only 1% of the total constrictor. The contralateral constrictor is always about 50% of the total constrictor volume, which means that predictions using this structure will not be confounded by the same variation in volume size.

Figure 2 shows a representative patient who met the NRG uninvolved pharynx constraint but developed long-term dysphagia.

Limitations

This study is limited by its single institution, retrospective design, small sample size, and by all patients being male. The high correlation between air cavity editing and the use of SIB makes it impossible to assess the impact of each technique individually. Patients with contralateral constrictor V60 < 40% were less likely to have N2 disease, but N2 to N3 disease did not predict higher 1-year dysphagia, so the difference in N-category cannot fully explain the difference in 1-year dysphagia. It is possible that unreported factors, such as CTV, may contribute significantly to swallowing function. Nevertheless, within the study population, contralateral constrictor dose was able to identify a group with a low rate of long-term dysphagia.

Conclusions

Contralateral constrictor dose is a promising predictor of late dysphagia for patients with H&N cancer treated with radiation with concurrent systemic therapy. Contralateral constrictor V60 < 40% was able to identify a group of patients with a low rate of 1-year dysphagia in this single-center retrospective study. The correlation between air cavity editing and contralateral constrictor V60 suggests that contralateral constrictor dose may depend partly on technique. Further studies are needed to see if the contralateral constrictor dose can be used to predict long-term dysphagia prospectively and in other patient populations.

Pseudofolliculitis barbae (PFB)(also known as razor bumps or shaving bumps)¹ is a skin condition that consists of papules resulting from ingrown hairs.² In more severe cases, papules become pustules, then abscesses, which can cause scarring.^1,2 The condition can be distressing for patients, with considerable negative impact on their daily lives.³ The condition also is associated with shaving-related stinging, burning, pruritus, and cuts on the skin.⁴

Pseudofolliculitis barbae is most common in men of African descent due to the curved nature of the hair follicle,^2,5,6 with an estimated prevalence in this population of 45% to 83%,^1,6 but it can affect men of other ethnicities.⁷ A genetic polymorphism in a gene encoding a keratin specific to the hair follicle also has been found to predispose some individuals to PFB.⁵ When hair from a curved or destabilized hair follicle is cut to form a sharp tip, it is susceptible to extrafollicular and/or transfollicular penetration,^5,6,8 as illustrated in Figure 1.

With extrafollicular or transfollicular penetration, the hair shaft re-enters or retracts into the dermis, triggering an inflammatory response that may be exacerbated by subsequent shaving.² Few studies have been published that aim to identify potential shaving solutions for individuals with PFB who elect to or need to continue shaving.

A new razor technology comprising 2 blades separated by a bridge feature has been designed specifically for men with razor bumps (SkinGuard [Procter & Gamble]). The SkinGuard razor redistributes shaving pressure so that there is less force from the blades on the skin and inflamed lesions than without the bridge, as seen in Figure 2. The razor has been designed to protect the skin from the blades, thereby minimizing the occurrence of new lesions and allowing existing lesions to heal.

The primary purpose of this study was to assess the appearance of males with razor bumps and shaving irritation when using the new razor technology in a regular shaving routine. The secondary objective was to measure satisfaction of the shaving experience when using the new razor by means of assessing itching, burning, and stinging using the participant global severity assessment (PGSA) and the impact on quality of life (QOL) measures.

Methods

Participants—Eligible participants were male, aged 20 to 60 years, and had clinically diagnosed PFB as well as symptoms of skin irritation from shaving. Participants were recruited from a dermatology clinic and via institutional review board–approved advertising.

Those eligible for inclusion in the study had a shaving routine that comprised shaving at least 3 times a week using a wet-shave, blade-razor technique accompanied by only a shave gel or foam. In addition, eligible participants had mild to moderate symptoms of skin irritation (a minimum of 10 razor bumps) from shaving based on investigator global severity assessment (IGSA) rating scales and were willing to shave at least 5 times a week during the study period. Participants could continue certain topical and systemic interventions for their skin.

Participants were excluded from the study if they had an underlying inflammatory disease that could manifest with a skin rash or were using any of these medications: topical benzoyl peroxide, topical clindamycin, topical retinoids, or oral antibiotics.

Study Design—A prospective, open-label study was conducted over a period of 12 weeks at a single site in the United States. Investigators instructed participants to shave 5 or more times per week with the test razor and to keep a daily shaving journal to track the number of shaves and compliance.

Participants were evaluated at the baseline screening visit, then at 4, 8, and 12 weeks. Evaluations included an investigator lesion count, the IGSA, and the PGSA. The PGSA was used to evaluate subjective clinical measurements (ie, indicate how much postshave burning/itching/stinging the participant was experiencing). The impact of shaving on daily life was evaluated at the baseline screening visit and at 12 weeks with the Participant Quality of Life Questionnaire comprised of 22 QOL statements. eTable 1 summarizes the investigator assessments used in the study, and eTable 2 summarizes the participant self-assessments. Both tables include the scale details and results interpretation for each assessment.

The study was approved by the local institutional review board, and all participants provided written informed consent in accordance with Title 21 of the Code of Federal Regulations, Part 50.

Study Visits—At the baseline screening visit, participants provided written informed consent and completed a prestudy shave questionnaire concerning shaving preparations, techniques, and opinions. Participants also provided a medical history, including prior and concomitant medications, and were evaluated using the inclusion/exclusion criteria. Investigators explained adverse event reporting to the participants. Participants were provided with an adequate supply of test razors for the 12-week period.

Data Analysis—Means and SDs were calculated for the study measures assessed at each visit. Analyses were performed evaluating change from baseline in repeated-measures analysis of variance models. These models were adjusted for baseline levels of the outcome measure and visit number. The magnitude of change from baseline was evaluated against a null hypothesis of 0% change. This longitudinal model adjusted for any potential differing baseline levels among participants. Statistical significance was defined as P<.05. SAS version 9.4 (SAS Institute Inc) was used for all analyses.

Results

In total, 21 individuals were enrolled, and 20 completed the study. Participants who completed the study were non-Hispanic Black (n=10); non-Hispanic White (n=8); Asian (n=1); or White, American Indian (n=1). All participants adhered to the protocol and reported shaving at least 5 times a week for 12 weeks using the test razor. One participant was removed after he was found to have a history of sarcoidosis, making him ineligible for the study. No study-related adverse events were reported.

Papules and Pustules—Over the course of the 12-week study, the papule count decreased significantly from baseline. Results from the investigator lesion count (see eTable 1 for key) indicated that by week 12—adjusted for number of papules at baseline—the mean percentage reduction was estimated to be 59.6% (P<.0001). A significant decrease in papule count also was observed between the baseline visit and week 8 (57.2%; P<.0001). A nonsignificant decrease was observed at week 4 (18.9%; P=.17). Only 3 participants presented with pustules at baseline, and the pustule count remained low over the course of the study. No significant change was noted at week 12 vs baseline (P=.98). Notably, there was no increase in pustule count at the end of the study compared with baseline (Table 1).

Skin Appearance—An improvement in the skin’s appearance over the course of the study from baseline was consistent with an improvement in the IGSA. The IGSA score significantly improved from a mean (SD) measurement of 2.5 (0.6) (indicating mild to moderate inflammation) at baseline to 1.4 (0.8) at week 8 (P<.0001) and 1.2 (1.1) (indicating mild inflammation to almost clear) at week 12 (P<.0001). The observed decrease in severity of skin condition and skin inflammation is shown in Figure 3.

Significant improvements were observed in every category of the PGSA at week 12 vs baseline (P≤.0007)(Table 2). At week 12, there was a significant (P≤.05) increase from baseline in participant agreement for all 22 QOL metrics describing positive shave experience, achieving results, skin feel, self-confidence, and social interactions (Figure 4), which supports the positive impact of adopting a shaving regimen with the test razor. Notably, after using the test razor for 12 weeks, men reported that they were more likely to agree with the statements “my skin felt smooth,” “my skin felt good to touch,” and “I was able to achieve a consistently good shave.” Other meaningful increases occurred in “shaving was something I looked forward to doing,” “others thought I looked clean cut,” “I looked my best for my family/others/work,” and “I felt comfortable/confident getting closer to others.” All QOL statements are shown in Figure 4.

Improvement With Novel Razor Technology—For the first time, frequent use of a novel razor technology designed specifically for men with PFB was found to significantly improve skin appearance, shave satisfaction, and QOL after 12 weeks vs baseline in participants clinically diagnosed with PFB. In men with shave-related skin irritation and razor bumps who typically wet-shaved with a razor at least 3 times a week, use of the test razor with their regular shaving preparation product 5 or more times per week for 12 weeks was associated with significant improvements from baseline in investigator lesion count, IGSA, PGSA, and Participant Quality of Life Questionnaire measurements.

Study strengths included the quantification of the change in the number of lesions and the degree of severity by a trained investigator in a prospective clinical study along with an assessment of the impact on participant QOL. A lack of a control arm could be considered a limitation of the study; however, study end points were evaluated compared with baseline, with each participant serving as their own control. Spontaneous resolution of the condition with their standard routine was considered highly unlikely in these participants; therefore, in the absence of any other changes, improvements were attributed to regular use of the test product over the course of the study. The results presented here provide strong support for the effectiveness of the new razor technology in improving the appearance of men with razor bumps and shaving irritation.

Hair Removal Tools for the Management of PFB—Although various tools and techniques have been proposed in the past for men with PFB, the current test razor technology provided unique benefits, including improvements in appearance and severity of the condition as well as a positive impact on QOL. In 1979, Conte and Lawrence⁹ evaluated the effect of using an electric hair clipper and twice-daily use of a skin-cleansing pad on the occurrence of PFB. Participants (n=96) allowed their beards to grow out for 1 month, after which they started shaving with an electric clipper with a triple O head. The authors reported a favorable response in 95% (91/96) of cases. However, the electric clippers left 1 mm of beard at the skin level,⁹ which may not be acceptable for those who prefer a clean-shaven appearance.⁶

A prospective survey of 22 men of African descent with PFB found use of a safety razor was preferred over an electric razor.¹⁰ The single-arm study evaluated use of a foil-guarded shaver (single-razor blade) in the management of PFB based on investigator lesion counts and a participant questionnaire. Participants were asked to shave at least every other day and use a specially designed preshave brush. A mean reduction in lesion counts was observed at 2 weeks (29.6%), 4 weeks (38.1%), and 6 weeks (47.1%); statistical significance was not reported. At 6 weeks, 77.3% (17/22) of participants judged the foil-guarded shaver to be superior to other shaving devices in controlling their razor bumps, and 90.9% (20/22) indicated they would recommend the shaver to others with PFB. The authors hypothesized that the guard buffered the skin from the blade, which might otherwise facilitate the penetration of ingrowing hairs and cause trauma to existing lesions.

The mean reduction in lesion count from baseline observed at week 4 was greater in the study with the foil-guarded shaver and preshave brush (38% reduction)¹⁰ than in our study (19% reduction in papule count). Different methodologies, use of a preshave brush in the earlier study, and a difference in lesion severity at baseline may have contributed to this difference. The study with the foil-guarded shaver concluded after 6 weeks, and there was a 47.1% reduction in lesion counts vs baseline.¹⁰ In contrast, the current study continued for 12 weeks, and a 59.6% reduction in lesion counts was reported. Participants from both studies reported an improved shaving experience compared with their usual practice,¹⁰ though only the current study explored the positive impact of the new razor technology on participant QOL.

Preventing Hairs From Being Cut Too Close—The closeness of the shave is believed to be a contributory factor in the development and persistence of PFB^6,8,11 based on a tendency for the distal portion of tightly curled hair shafts to re-enter the skin after shaving via transfollicular penetration.¹² Inclusion of a buffer in the razor between the sharp blades and the skin has been proposed to prevent hairs from being cut too close and causing transfollicular penetration.¹²

In the test razor used in the current study, the bridge technology acted as the buffer to prevent hairs from being cut too close to the skin and to reduce blade contact with the skin (Figure 2). Having only 2 blades also reduced the closeness of the shave compared with 5-bladed technologies,¹³ as each hair can only be pulled and cut up to a maximum of 2 times per shaving stroke. Notably, this did not impact the participants’ QOL scores related to achieving a close shave or skin feeling smooth; both attributes were significantly improved at 12 weeks vs baseline (Figure 4).

By reducing blade contact with the skin, the bridge technology in the test razor was designed to prevent excessive force from being applied to the skin through the blades. Reduced blade loading minimizes contact with and impact on sensitive skin.¹⁴ Additional design features of the test razor to minimize the impact of shaving on the skin include treatment of the 2 blades with low-friction coatings, which allows the blades to cut through the beard hair with minimal force, helping to reduce the tug-and-pull effect that may otherwise result in irritation and inflammation.^13,15 Lubrication strips before and after the blades in the test razor reduce friction between the blades and the skin to further protect the skin from the blades.¹⁵

Shaving With Multiblade Razors Does Not Exacerbate PFB—In a 1-week, split-faced, randomized study of 45 Black men, shaving with a manual 3-bladed razor was compared with use of 3 different chemical depilatory formulations.¹⁶ Shaving every other day for 1 week with the manual razor resulted in more papule formation but less irritation than use of the depilatories. The authors concluded that a study with longer duration was needed to explore the impact of shaving on papule formation in participants with a history of PFB.¹⁶

In 2013, an investigator-blinded study of 90 African American men with PFB compared the impact of different shaving regimens on the signs and symptoms of PFB over a 12-week period.⁴ Participants were randomized to 1 of 3 arms: (1) shaving 2 to 3 times per week with a triple-blade razor and standard products (control group); (2) shaving daily with a 5-bladed razor and standard products; and (3) shaving daily with a 5-bladed razor and “advanced” specific pre- and postshave products. The researchers found that the mean papule measurement significantly decreased from baseline in the advanced (P=.01) and control (P=.016) groups. Between-group comparison revealed no significant differences for papule or pustule count among each arm. For the investigator-graded severity, the change from baseline was significant for all 3 groups (P≤.04); however, the differences among groups were not significant. Importantly, these data demonstrated that PFB was not exacerbated by multiblade razors used as part of a daily shaving regimen.⁴

The findings of the current study were consistent with those of Daniel et al⁴ in that there was no exacerbation of the signs and symptoms of PFB associated with daily shaving. However, rather than requiring participants to change their entire shaving regimen, the present study only required a change of razor type. Moreover, the use of the new razor technology significantly decreased papule counts at week 12 vs the baseline measurement (P<.0001) and was associated with an improvement in subjective skin severity measurements. The participants in the present study reported significantly less burning, stinging, and itching after using the test product for 12 weeks (P<.0001).

Impact of Treatment on QOL—The current study further expanded on prior findings by combining these clinical end points with the QOL results to assess the test razor’s impact on participants’ lives. Results showed that over the course of 12 weeks, the new razor technology significantly improved the participants’ QOL in all questions related to shaving experience, achieving results, skin feel, self-confidence, and social interactions. The significant improvement in QOL included statements such as “shaving was a pleasant experience,” “I was able to achieve a consistently good shave,” and “my skin felt smooth.” Participants also reported improvements in meaningful categories such as “my shave made me feel attractive” and “I felt comfortable/confident getting closer to others.” As the current study showed, a shave regimen has the potential to change participants’ overall assessment of their QOL, a variable that must not be overlooked.

Conclusion

In men with clinically diagnosed PFB, regular shaving with a razor designed to protect the skin was found to significantly decrease lesion counts, increase shave satisfaction, and improve QOL after 12 weeks compared with their usual shaving practice (baseline measures). This razor technology provides another option to help manage PFB for men who wish to or need to continue shaving.

Acknowledgments—The clinical study was funded by the Procter & Gamble Company. Editorial writing assistance, supported financially by the Procter & Gamble Company, was provided by Gill McFeat, PhD, of McFeat Science Ltd (Devon, United Kingdom).

Pseudofolliculitis barbae (PFB)(also known as razor bumps or shaving bumps)¹ is a skin condition that consists of papules resulting from ingrown hairs.² In more severe cases, papules become pustules, then abscesses, which can cause scarring.^1,2 The condition can be distressing for patients, with considerable negative impact on their daily lives.³ The condition also is associated with shaving-related stinging, burning, pruritus, and cuts on the skin.⁴

Pseudofolliculitis barbae is most common in men of African descent due to the curved nature of the hair follicle,^2,5,6 with an estimated prevalence in this population of 45% to 83%,^1,6 but it can affect men of other ethnicities.⁷ A genetic polymorphism in a gene encoding a keratin specific to the hair follicle also has been found to predispose some individuals to PFB.⁵ When hair from a curved or destabilized hair follicle is cut to form a sharp tip, it is susceptible to extrafollicular and/or transfollicular penetration,^5,6,8 as illustrated in Figure 1.

With extrafollicular or transfollicular penetration, the hair shaft re-enters or retracts into the dermis, triggering an inflammatory response that may be exacerbated by subsequent shaving.² Few studies have been published that aim to identify potential shaving solutions for individuals with PFB who elect to or need to continue shaving.

A new razor technology comprising 2 blades separated by a bridge feature has been designed specifically for men with razor bumps (SkinGuard [Procter & Gamble]). The SkinGuard razor redistributes shaving pressure so that there is less force from the blades on the skin and inflamed lesions than without the bridge, as seen in Figure 2. The razor has been designed to protect the skin from the blades, thereby minimizing the occurrence of new lesions and allowing existing lesions to heal.

The primary purpose of this study was to assess the appearance of males with razor bumps and shaving irritation when using the new razor technology in a regular shaving routine. The secondary objective was to measure satisfaction of the shaving experience when using the new razor by means of assessing itching, burning, and stinging using the participant global severity assessment (PGSA) and the impact on quality of life (QOL) measures.

Methods

Participants—Eligible participants were male, aged 20 to 60 years, and had clinically diagnosed PFB as well as symptoms of skin irritation from shaving. Participants were recruited from a dermatology clinic and via institutional review board–approved advertising.

Those eligible for inclusion in the study had a shaving routine that comprised shaving at least 3 times a week using a wet-shave, blade-razor technique accompanied by only a shave gel or foam. In addition, eligible participants had mild to moderate symptoms of skin irritation (a minimum of 10 razor bumps) from shaving based on investigator global severity assessment (IGSA) rating scales and were willing to shave at least 5 times a week during the study period. Participants could continue certain topical and systemic interventions for their skin.

Participants were excluded from the study if they had an underlying inflammatory disease that could manifest with a skin rash or were using any of these medications: topical benzoyl peroxide, topical clindamycin, topical retinoids, or oral antibiotics.

Study Design—A prospective, open-label study was conducted over a period of 12 weeks at a single site in the United States. Investigators instructed participants to shave 5 or more times per week with the test razor and to keep a daily shaving journal to track the number of shaves and compliance.

Participants were evaluated at the baseline screening visit, then at 4, 8, and 12 weeks. Evaluations included an investigator lesion count, the IGSA, and the PGSA. The PGSA was used to evaluate subjective clinical measurements (ie, indicate how much postshave burning/itching/stinging the participant was experiencing). The impact of shaving on daily life was evaluated at the baseline screening visit and at 12 weeks with the Participant Quality of Life Questionnaire comprised of 22 QOL statements. eTable 1 summarizes the investigator assessments used in the study, and eTable 2 summarizes the participant self-assessments. Both tables include the scale details and results interpretation for each assessment.

The study was approved by the local institutional review board, and all participants provided written informed consent in accordance with Title 21 of the Code of Federal Regulations, Part 50.

Study Visits—At the baseline screening visit, participants provided written informed consent and completed a prestudy shave questionnaire concerning shaving preparations, techniques, and opinions. Participants also provided a medical history, including prior and concomitant medications, and were evaluated using the inclusion/exclusion criteria. Investigators explained adverse event reporting to the participants. Participants were provided with an adequate supply of test razors for the 12-week period.

Data Analysis—Means and SDs were calculated for the study measures assessed at each visit. Analyses were performed evaluating change from baseline in repeated-measures analysis of variance models. These models were adjusted for baseline levels of the outcome measure and visit number. The magnitude of change from baseline was evaluated against a null hypothesis of 0% change. This longitudinal model adjusted for any potential differing baseline levels among participants. Statistical significance was defined as P<.05. SAS version 9.4 (SAS Institute Inc) was used for all analyses.

Results

In total, 21 individuals were enrolled, and 20 completed the study. Participants who completed the study were non-Hispanic Black (n=10); non-Hispanic White (n=8); Asian (n=1); or White, American Indian (n=1). All participants adhered to the protocol and reported shaving at least 5 times a week for 12 weeks using the test razor. One participant was removed after he was found to have a history of sarcoidosis, making him ineligible for the study. No study-related adverse events were reported.

Papules and Pustules—Over the course of the 12-week study, the papule count decreased significantly from baseline. Results from the investigator lesion count (see eTable 1 for key) indicated that by week 12—adjusted for number of papules at baseline—the mean percentage reduction was estimated to be 59.6% (P<.0001). A significant decrease in papule count also was observed between the baseline visit and week 8 (57.2%; P<.0001). A nonsignificant decrease was observed at week 4 (18.9%; P=.17). Only 3 participants presented with pustules at baseline, and the pustule count remained low over the course of the study. No significant change was noted at week 12 vs baseline (P=.98). Notably, there was no increase in pustule count at the end of the study compared with baseline (Table 1).

Skin Appearance—An improvement in the skin’s appearance over the course of the study from baseline was consistent with an improvement in the IGSA. The IGSA score significantly improved from a mean (SD) measurement of 2.5 (0.6) (indicating mild to moderate inflammation) at baseline to 1.4 (0.8) at week 8 (P<.0001) and 1.2 (1.1) (indicating mild inflammation to almost clear) at week 12 (P<.0001). The observed decrease in severity of skin condition and skin inflammation is shown in Figure 3.

Significant improvements were observed in every category of the PGSA at week 12 vs baseline (P≤.0007)(Table 2). At week 12, there was a significant (P≤.05) increase from baseline in participant agreement for all 22 QOL metrics describing positive shave experience, achieving results, skin feel, self-confidence, and social interactions (Figure 4), which supports the positive impact of adopting a shaving regimen with the test razor. Notably, after using the test razor for 12 weeks, men reported that they were more likely to agree with the statements “my skin felt smooth,” “my skin felt good to touch,” and “I was able to achieve a consistently good shave.” Other meaningful increases occurred in “shaving was something I looked forward to doing,” “others thought I looked clean cut,” “I looked my best for my family/others/work,” and “I felt comfortable/confident getting closer to others.” All QOL statements are shown in Figure 4.

Improvement With Novel Razor Technology—For the first time, frequent use of a novel razor technology designed specifically for men with PFB was found to significantly improve skin appearance, shave satisfaction, and QOL after 12 weeks vs baseline in participants clinically diagnosed with PFB. In men with shave-related skin irritation and razor bumps who typically wet-shaved with a razor at least 3 times a week, use of the test razor with their regular shaving preparation product 5 or more times per week for 12 weeks was associated with significant improvements from baseline in investigator lesion count, IGSA, PGSA, and Participant Quality of Life Questionnaire measurements.

Study strengths included the quantification of the change in the number of lesions and the degree of severity by a trained investigator in a prospective clinical study along with an assessment of the impact on participant QOL. A lack of a control arm could be considered a limitation of the study; however, study end points were evaluated compared with baseline, with each participant serving as their own control. Spontaneous resolution of the condition with their standard routine was considered highly unlikely in these participants; therefore, in the absence of any other changes, improvements were attributed to regular use of the test product over the course of the study. The results presented here provide strong support for the effectiveness of the new razor technology in improving the appearance of men with razor bumps and shaving irritation.

Hair Removal Tools for the Management of PFB—Although various tools and techniques have been proposed in the past for men with PFB, the current test razor technology provided unique benefits, including improvements in appearance and severity of the condition as well as a positive impact on QOL. In 1979, Conte and Lawrence⁹ evaluated the effect of using an electric hair clipper and twice-daily use of a skin-cleansing pad on the occurrence of PFB. Participants (n=96) allowed their beards to grow out for 1 month, after which they started shaving with an electric clipper with a triple O head. The authors reported a favorable response in 95% (91/96) of cases. However, the electric clippers left 1 mm of beard at the skin level,⁹ which may not be acceptable for those who prefer a clean-shaven appearance.⁶

A prospective survey of 22 men of African descent with PFB found use of a safety razor was preferred over an electric razor.¹⁰ The single-arm study evaluated use of a foil-guarded shaver (single-razor blade) in the management of PFB based on investigator lesion counts and a participant questionnaire. Participants were asked to shave at least every other day and use a specially designed preshave brush. A mean reduction in lesion counts was observed at 2 weeks (29.6%), 4 weeks (38.1%), and 6 weeks (47.1%); statistical significance was not reported. At 6 weeks, 77.3% (17/22) of participants judged the foil-guarded shaver to be superior to other shaving devices in controlling their razor bumps, and 90.9% (20/22) indicated they would recommend the shaver to others with PFB. The authors hypothesized that the guard buffered the skin from the blade, which might otherwise facilitate the penetration of ingrowing hairs and cause trauma to existing lesions.

The mean reduction in lesion count from baseline observed at week 4 was greater in the study with the foil-guarded shaver and preshave brush (38% reduction)¹⁰ than in our study (19% reduction in papule count). Different methodologies, use of a preshave brush in the earlier study, and a difference in lesion severity at baseline may have contributed to this difference. The study with the foil-guarded shaver concluded after 6 weeks, and there was a 47.1% reduction in lesion counts vs baseline.¹⁰ In contrast, the current study continued for 12 weeks, and a 59.6% reduction in lesion counts was reported. Participants from both studies reported an improved shaving experience compared with their usual practice,¹⁰ though only the current study explored the positive impact of the new razor technology on participant QOL.

Preventing Hairs From Being Cut Too Close—The closeness of the shave is believed to be a contributory factor in the development and persistence of PFB^6,8,11 based on a tendency for the distal portion of tightly curled hair shafts to re-enter the skin after shaving via transfollicular penetration.¹² Inclusion of a buffer in the razor between the sharp blades and the skin has been proposed to prevent hairs from being cut too close and causing transfollicular penetration.¹²

In the test razor used in the current study, the bridge technology acted as the buffer to prevent hairs from being cut too close to the skin and to reduce blade contact with the skin (Figure 2). Having only 2 blades also reduced the closeness of the shave compared with 5-bladed technologies,¹³ as each hair can only be pulled and cut up to a maximum of 2 times per shaving stroke. Notably, this did not impact the participants’ QOL scores related to achieving a close shave or skin feeling smooth; both attributes were significantly improved at 12 weeks vs baseline (Figure 4).

By reducing blade contact with the skin, the bridge technology in the test razor was designed to prevent excessive force from being applied to the skin through the blades. Reduced blade loading minimizes contact with and impact on sensitive skin.¹⁴ Additional design features of the test razor to minimize the impact of shaving on the skin include treatment of the 2 blades with low-friction coatings, which allows the blades to cut through the beard hair with minimal force, helping to reduce the tug-and-pull effect that may otherwise result in irritation and inflammation.^13,15 Lubrication strips before and after the blades in the test razor reduce friction between the blades and the skin to further protect the skin from the blades.¹⁵

Shaving With Multiblade Razors Does Not Exacerbate PFB—In a 1-week, split-faced, randomized study of 45 Black men, shaving with a manual 3-bladed razor was compared with use of 3 different chemical depilatory formulations.¹⁶ Shaving every other day for 1 week with the manual razor resulted in more papule formation but less irritation than use of the depilatories. The authors concluded that a study with longer duration was needed to explore the impact of shaving on papule formation in participants with a history of PFB.¹⁶

In 2013, an investigator-blinded study of 90 African American men with PFB compared the impact of different shaving regimens on the signs and symptoms of PFB over a 12-week period.⁴ Participants were randomized to 1 of 3 arms: (1) shaving 2 to 3 times per week with a triple-blade razor and standard products (control group); (2) shaving daily with a 5-bladed razor and standard products; and (3) shaving daily with a 5-bladed razor and “advanced” specific pre- and postshave products. The researchers found that the mean papule measurement significantly decreased from baseline in the advanced (P=.01) and control (P=.016) groups. Between-group comparison revealed no significant differences for papule or pustule count among each arm. For the investigator-graded severity, the change from baseline was significant for all 3 groups (P≤.04); however, the differences among groups were not significant. Importantly, these data demonstrated that PFB was not exacerbated by multiblade razors used as part of a daily shaving regimen.⁴

The findings of the current study were consistent with those of Daniel et al⁴ in that there was no exacerbation of the signs and symptoms of PFB associated with daily shaving. However, rather than requiring participants to change their entire shaving regimen, the present study only required a change of razor type. Moreover, the use of the new razor technology significantly decreased papule counts at week 12 vs the baseline measurement (P<.0001) and was associated with an improvement in subjective skin severity measurements. The participants in the present study reported significantly less burning, stinging, and itching after using the test product for 12 weeks (P<.0001).

Impact of Treatment on QOL—The current study further expanded on prior findings by combining these clinical end points with the QOL results to assess the test razor’s impact on participants’ lives. Results showed that over the course of 12 weeks, the new razor technology significantly improved the participants’ QOL in all questions related to shaving experience, achieving results, skin feel, self-confidence, and social interactions. The significant improvement in QOL included statements such as “shaving was a pleasant experience,” “I was able to achieve a consistently good shave,” and “my skin felt smooth.” Participants also reported improvements in meaningful categories such as “my shave made me feel attractive” and “I felt comfortable/confident getting closer to others.” As the current study showed, a shave regimen has the potential to change participants’ overall assessment of their QOL, a variable that must not be overlooked.

Conclusion

In men with clinically diagnosed PFB, regular shaving with a razor designed to protect the skin was found to significantly decrease lesion counts, increase shave satisfaction, and improve QOL after 12 weeks compared with their usual shaving practice (baseline measures). This razor technology provides another option to help manage PFB for men who wish to or need to continue shaving.

Acknowledgments—The clinical study was funded by the Procter & Gamble Company. Editorial writing assistance, supported financially by the Procter & Gamble Company, was provided by Gill McFeat, PhD, of McFeat Science Ltd (Devon, United Kingdom).

Pseudofolliculitis barbae (PFB)(also known as razor bumps or shaving bumps)¹ is a skin condition that consists of papules resulting from ingrown hairs.² In more severe cases, papules become pustules, then abscesses, which can cause scarring.^1,2 The condition can be distressing for patients, with considerable negative impact on their daily lives.³ The condition also is associated with shaving-related stinging, burning, pruritus, and cuts on the skin.⁴

Pseudofolliculitis barbae is most common in men of African descent due to the curved nature of the hair follicle,^2,5,6 with an estimated prevalence in this population of 45% to 83%,^1,6 but it can affect men of other ethnicities.⁷ A genetic polymorphism in a gene encoding a keratin specific to the hair follicle also has been found to predispose some individuals to PFB.⁵ When hair from a curved or destabilized hair follicle is cut to form a sharp tip, it is susceptible to extrafollicular and/or transfollicular penetration,^5,6,8 as illustrated in Figure 1.

With extrafollicular or transfollicular penetration, the hair shaft re-enters or retracts into the dermis, triggering an inflammatory response that may be exacerbated by subsequent shaving.² Few studies have been published that aim to identify potential shaving solutions for individuals with PFB who elect to or need to continue shaving.

A new razor technology comprising 2 blades separated by a bridge feature has been designed specifically for men with razor bumps (SkinGuard [Procter & Gamble]). The SkinGuard razor redistributes shaving pressure so that there is less force from the blades on the skin and inflamed lesions than without the bridge, as seen in Figure 2. The razor has been designed to protect the skin from the blades, thereby minimizing the occurrence of new lesions and allowing existing lesions to heal.

The primary purpose of this study was to assess the appearance of males with razor bumps and shaving irritation when using the new razor technology in a regular shaving routine. The secondary objective was to measure satisfaction of the shaving experience when using the new razor by means of assessing itching, burning, and stinging using the participant global severity assessment (PGSA) and the impact on quality of life (QOL) measures.

Methods

Participants—Eligible participants were male, aged 20 to 60 years, and had clinically diagnosed PFB as well as symptoms of skin irritation from shaving. Participants were recruited from a dermatology clinic and via institutional review board–approved advertising.

Those eligible for inclusion in the study had a shaving routine that comprised shaving at least 3 times a week using a wet-shave, blade-razor technique accompanied by only a shave gel or foam. In addition, eligible participants had mild to moderate symptoms of skin irritation (a minimum of 10 razor bumps) from shaving based on investigator global severity assessment (IGSA) rating scales and were willing to shave at least 5 times a week during the study period. Participants could continue certain topical and systemic interventions for their skin.

Participants were excluded from the study if they had an underlying inflammatory disease that could manifest with a skin rash or were using any of these medications: topical benzoyl peroxide, topical clindamycin, topical retinoids, or oral antibiotics.

Study Design—A prospective, open-label study was conducted over a period of 12 weeks at a single site in the United States. Investigators instructed participants to shave 5 or more times per week with the test razor and to keep a daily shaving journal to track the number of shaves and compliance.

Participants were evaluated at the baseline screening visit, then at 4, 8, and 12 weeks. Evaluations included an investigator lesion count, the IGSA, and the PGSA. The PGSA was used to evaluate subjective clinical measurements (ie, indicate how much postshave burning/itching/stinging the participant was experiencing). The impact of shaving on daily life was evaluated at the baseline screening visit and at 12 weeks with the Participant Quality of Life Questionnaire comprised of 22 QOL statements. eTable 1 summarizes the investigator assessments used in the study, and eTable 2 summarizes the participant self-assessments. Both tables include the scale details and results interpretation for each assessment.

The study was approved by the local institutional review board, and all participants provided written informed consent in accordance with Title 21 of the Code of Federal Regulations, Part 50.

Study Visits—At the baseline screening visit, participants provided written informed consent and completed a prestudy shave questionnaire concerning shaving preparations, techniques, and opinions. Participants also provided a medical history, including prior and concomitant medications, and were evaluated using the inclusion/exclusion criteria. Investigators explained adverse event reporting to the participants. Participants were provided with an adequate supply of test razors for the 12-week period.

Data Analysis—Means and SDs were calculated for the study measures assessed at each visit. Analyses were performed evaluating change from baseline in repeated-measures analysis of variance models. These models were adjusted for baseline levels of the outcome measure and visit number. The magnitude of change from baseline was evaluated against a null hypothesis of 0% change. This longitudinal model adjusted for any potential differing baseline levels among participants. Statistical significance was defined as P<.05. SAS version 9.4 (SAS Institute Inc) was used for all analyses.

Results

In total, 21 individuals were enrolled, and 20 completed the study. Participants who completed the study were non-Hispanic Black (n=10); non-Hispanic White (n=8); Asian (n=1); or White, American Indian (n=1). All participants adhered to the protocol and reported shaving at least 5 times a week for 12 weeks using the test razor. One participant was removed after he was found to have a history of sarcoidosis, making him ineligible for the study. No study-related adverse events were reported.

Papules and Pustules—Over the course of the 12-week study, the papule count decreased significantly from baseline. Results from the investigator lesion count (see eTable 1 for key) indicated that by week 12—adjusted for number of papules at baseline—the mean percentage reduction was estimated to be 59.6% (P<.0001). A significant decrease in papule count also was observed between the baseline visit and week 8 (57.2%; P<.0001). A nonsignificant decrease was observed at week 4 (18.9%; P=.17). Only 3 participants presented with pustules at baseline, and the pustule count remained low over the course of the study. No significant change was noted at week 12 vs baseline (P=.98). Notably, there was no increase in pustule count at the end of the study compared with baseline (Table 1).

Skin Appearance—An improvement in the skin’s appearance over the course of the study from baseline was consistent with an improvement in the IGSA. The IGSA score significantly improved from a mean (SD) measurement of 2.5 (0.6) (indicating mild to moderate inflammation) at baseline to 1.4 (0.8) at week 8 (P<.0001) and 1.2 (1.1) (indicating mild inflammation to almost clear) at week 12 (P<.0001). The observed decrease in severity of skin condition and skin inflammation is shown in Figure 3.

Significant improvements were observed in every category of the PGSA at week 12 vs baseline (P≤.0007)(Table 2). At week 12, there was a significant (P≤.05) increase from baseline in participant agreement for all 22 QOL metrics describing positive shave experience, achieving results, skin feel, self-confidence, and social interactions (Figure 4), which supports the positive impact of adopting a shaving regimen with the test razor. Notably, after using the test razor for 12 weeks, men reported that they were more likely to agree with the statements “my skin felt smooth,” “my skin felt good to touch,” and “I was able to achieve a consistently good shave.” Other meaningful increases occurred in “shaving was something I looked forward to doing,” “others thought I looked clean cut,” “I looked my best for my family/others/work,” and “I felt comfortable/confident getting closer to others.” All QOL statements are shown in Figure 4.

Improvement With Novel Razor Technology—For the first time, frequent use of a novel razor technology designed specifically for men with PFB was found to significantly improve skin appearance, shave satisfaction, and QOL after 12 weeks vs baseline in participants clinically diagnosed with PFB. In men with shave-related skin irritation and razor bumps who typically wet-shaved with a razor at least 3 times a week, use of the test razor with their regular shaving preparation product 5 or more times per week for 12 weeks was associated with significant improvements from baseline in investigator lesion count, IGSA, PGSA, and Participant Quality of Life Questionnaire measurements.

Study strengths included the quantification of the change in the number of lesions and the degree of severity by a trained investigator in a prospective clinical study along with an assessment of the impact on participant QOL. A lack of a control arm could be considered a limitation of the study; however, study end points were evaluated compared with baseline, with each participant serving as their own control. Spontaneous resolution of the condition with their standard routine was considered highly unlikely in these participants; therefore, in the absence of any other changes, improvements were attributed to regular use of the test product over the course of the study. The results presented here provide strong support for the effectiveness of the new razor technology in improving the appearance of men with razor bumps and shaving irritation.

Hair Removal Tools for the Management of PFB—Although various tools and techniques have been proposed in the past for men with PFB, the current test razor technology provided unique benefits, including improvements in appearance and severity of the condition as well as a positive impact on QOL. In 1979, Conte and Lawrence⁹ evaluated the effect of using an electric hair clipper and twice-daily use of a skin-cleansing pad on the occurrence of PFB. Participants (n=96) allowed their beards to grow out for 1 month, after which they started shaving with an electric clipper with a triple O head. The authors reported a favorable response in 95% (91/96) of cases. However, the electric clippers left 1 mm of beard at the skin level,⁹ which may not be acceptable for those who prefer a clean-shaven appearance.⁶

A prospective survey of 22 men of African descent with PFB found use of a safety razor was preferred over an electric razor.¹⁰ The single-arm study evaluated use of a foil-guarded shaver (single-razor blade) in the management of PFB based on investigator lesion counts and a participant questionnaire. Participants were asked to shave at least every other day and use a specially designed preshave brush. A mean reduction in lesion counts was observed at 2 weeks (29.6%), 4 weeks (38.1%), and 6 weeks (47.1%); statistical significance was not reported. At 6 weeks, 77.3% (17/22) of participants judged the foil-guarded shaver to be superior to other shaving devices in controlling their razor bumps, and 90.9% (20/22) indicated they would recommend the shaver to others with PFB. The authors hypothesized that the guard buffered the skin from the blade, which might otherwise facilitate the penetration of ingrowing hairs and cause trauma to existing lesions.

The mean reduction in lesion count from baseline observed at week 4 was greater in the study with the foil-guarded shaver and preshave brush (38% reduction)¹⁰ than in our study (19% reduction in papule count). Different methodologies, use of a preshave brush in the earlier study, and a difference in lesion severity at baseline may have contributed to this difference. The study with the foil-guarded shaver concluded after 6 weeks, and there was a 47.1% reduction in lesion counts vs baseline.¹⁰ In contrast, the current study continued for 12 weeks, and a 59.6% reduction in lesion counts was reported. Participants from both studies reported an improved shaving experience compared with their usual practice,¹⁰ though only the current study explored the positive impact of the new razor technology on participant QOL.

Preventing Hairs From Being Cut Too Close—The closeness of the shave is believed to be a contributory factor in the development and persistence of PFB^6,8,11 based on a tendency for the distal portion of tightly curled hair shafts to re-enter the skin after shaving via transfollicular penetration.¹² Inclusion of a buffer in the razor between the sharp blades and the skin has been proposed to prevent hairs from being cut too close and causing transfollicular penetration.¹²

In the test razor used in the current study, the bridge technology acted as the buffer to prevent hairs from being cut too close to the skin and to reduce blade contact with the skin (Figure 2). Having only 2 blades also reduced the closeness of the shave compared with 5-bladed technologies,¹³ as each hair can only be pulled and cut up to a maximum of 2 times per shaving stroke. Notably, this did not impact the participants’ QOL scores related to achieving a close shave or skin feeling smooth; both attributes were significantly improved at 12 weeks vs baseline (Figure 4).

By reducing blade contact with the skin, the bridge technology in the test razor was designed to prevent excessive force from being applied to the skin through the blades. Reduced blade loading minimizes contact with and impact on sensitive skin.¹⁴ Additional design features of the test razor to minimize the impact of shaving on the skin include treatment of the 2 blades with low-friction coatings, which allows the blades to cut through the beard hair with minimal force, helping to reduce the tug-and-pull effect that may otherwise result in irritation and inflammation.^13,15 Lubrication strips before and after the blades in the test razor reduce friction between the blades and the skin to further protect the skin from the blades.¹⁵

Shaving With Multiblade Razors Does Not Exacerbate PFB—In a 1-week, split-faced, randomized study of 45 Black men, shaving with a manual 3-bladed razor was compared with use of 3 different chemical depilatory formulations.¹⁶ Shaving every other day for 1 week with the manual razor resulted in more papule formation but less irritation than use of the depilatories. The authors concluded that a study with longer duration was needed to explore the impact of shaving on papule formation in participants with a history of PFB.¹⁶

In 2013, an investigator-blinded study of 90 African American men with PFB compared the impact of different shaving regimens on the signs and symptoms of PFB over a 12-week period.⁴ Participants were randomized to 1 of 3 arms: (1) shaving 2 to 3 times per week with a triple-blade razor and standard products (control group); (2) shaving daily with a 5-bladed razor and standard products; and (3) shaving daily with a 5-bladed razor and “advanced” specific pre- and postshave products. The researchers found that the mean papule measurement significantly decreased from baseline in the advanced (P=.01) and control (P=.016) groups. Between-group comparison revealed no significant differences for papule or pustule count among each arm. For the investigator-graded severity, the change from baseline was significant for all 3 groups (P≤.04); however, the differences among groups were not significant. Importantly, these data demonstrated that PFB was not exacerbated by multiblade razors used as part of a daily shaving regimen.⁴

The findings of the current study were consistent with those of Daniel et al⁴ in that there was no exacerbation of the signs and symptoms of PFB associated with daily shaving. However, rather than requiring participants to change their entire shaving regimen, the present study only required a change of razor type. Moreover, the use of the new razor technology significantly decreased papule counts at week 12 vs the baseline measurement (P<.0001) and was associated with an improvement in subjective skin severity measurements. The participants in the present study reported significantly less burning, stinging, and itching after using the test product for 12 weeks (P<.0001).

Impact of Treatment on QOL—The current study further expanded on prior findings by combining these clinical end points with the QOL results to assess the test razor’s impact on participants’ lives. Results showed that over the course of 12 weeks, the new razor technology significantly improved the participants’ QOL in all questions related to shaving experience, achieving results, skin feel, self-confidence, and social interactions. The significant improvement in QOL included statements such as “shaving was a pleasant experience,” “I was able to achieve a consistently good shave,” and “my skin felt smooth.” Participants also reported improvements in meaningful categories such as “my shave made me feel attractive” and “I felt comfortable/confident getting closer to others.” As the current study showed, a shave regimen has the potential to change participants’ overall assessment of their QOL, a variable that must not be overlooked.

Conclusion

In men with clinically diagnosed PFB, regular shaving with a razor designed to protect the skin was found to significantly decrease lesion counts, increase shave satisfaction, and improve QOL after 12 weeks compared with their usual shaving practice (baseline measures). This razor technology provides another option to help manage PFB for men who wish to or need to continue shaving.

Acknowledgments—The clinical study was funded by the Procter & Gamble Company. Editorial writing assistance, supported financially by the Procter & Gamble Company, was provided by Gill McFeat, PhD, of McFeat Science Ltd (Devon, United Kingdom).

From the Department of Neurological Surgery, Vanderbilt University Medical Center, Nashville, TN (Stefan W. Koester, Puja Jagasia, and Drs. Liles, Dambrino IV, Feldman, and Chambless), and the Department of Anesthesiology, Vanderbilt University Medical Center, Nashville, TN (Drs. Mathews and Tiwari).

ABSTRACT

Background: The COVID-19 pandemic has had broad effects on surgical care, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and newly implemented anti-infective measures. Our aim was to assess neurosurgery OR efficiency before the COVID-19 pandemic, during peak COVID-19, and during current times.

Methods: Institutional perioperative databases at a single, high-volume neurosurgical center were queried for operations performed from December 2019 until October 2021. March 12, 2020, the day that the state of Tennessee declared a state of emergency, was chosen as the onset of the COVID-19 pandemic. The 90-day periods before and after this day were used to define the pre-COVID-19, peak-COVID-19, and post-peak restrictions time periods for comparative analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover). Univariate analysis used Wilcoxon rank-sum test for continuous outcomes, while chi-square test and Fisher’s exact test were used for categorical comparisons. Significance was defined as P < .05.

Results: First-start time was analyzed in 426 pre-COVID-19, 357 peak-restrictions, and 2304 post-peak-restrictions cases. The unadjusted mean delay length was found to be significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different. The proportion of cases that started early, as well as significantly early past a 15-minute threshold, have not been impacted. There was no significant change in turnover time during peak restrictions relative to the pre-COVID-19 period (88 [100] minutes vs 85 [95] minutes), and turnover time has since remained unchanged (83 [87] minutes).

Conclusion: Our center was able to maintain OR efficiency before, during, and after peak restrictions even while instituting advanced infection-control strategies. While there were significant changes, delays were relatively small in magnitude.

Keywords: operating room timing, hospital efficiency, socioeconomics, pandemic.

The COVID-19 pandemic has led to major changes in patient care both from a surgical perspective and in regard to inpatient hospital course. Safety protocols nationwide have been implemented to protect both patients and providers. Some elements of surgical care have drastically changed, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and increased sterilization measures. Furloughs, layoffs, and reassignments due to the focus on nonelective and COVID-19–related cases challenged OR staffing and efficiency. Operating room staff with COVID-19 exposures or COVID-19 infections also caused last-minute changes in staffing. All of these scenarios can cause issues due to actual understaffing or due to staff members being pushed into highly specialized areas, such as neurosurgery, in which they have very little experience. A further obstacle to OR efficiency included policy changes involving PPE utilization, sterilization measures, and supply chain shortages of necessary resources such as PPE.

Neurosurgery in particular has been susceptible to COVID-19–related system-wide changes given operator proximity to the patient’s respiratory passages, frequency of emergent cases, and varying anesthetic needs, as well as the high level of specialization needed to perform neurosurgical care. Previous studies have shown a change in the makeup of neurosurgical patients seeking care, as well as in the acuity of neurological consult of these patients.¹ A study in orthopedic surgery by Andreata et al demonstrated worsened OR efficiency, with significantly increased first-start and turnover times.² In the COVID-19 era, OR quality and safety are crucially important to both patients and providers. Providing this safe and effective care in an efficient manner is important for optimal neurosurgical management in the long term.³ Moreover, the financial burden of implementing new protocols and standards can be compounded by additional financial losses due to reduced OR efficiency.

Methods

To analyze the effect of COVID-19 on neurosurgical OR efficiency, institutional perioperative databases at a single high-volume center were queried for operations performed from December 2019 until October 2021. March 12, 2020, was chosen as the onset of COVID-19 for analytic purposes, as this was the date when the state of Tennessee declared a state of emergency. The 90-day periods before and after this date were used for comparative analysis for pre-COVID-19, peak COVID-19, and post-peak-restrictions time periods. The peak COVID-19 period was defined as the 90-day period following the initial onset of COVID-19 and the surge of cases. For comparison purposes, post-peak COVID-19 was defined as the months following the first peak until October 2021 (approximately 17 months). COVID-19 burden was determined using a COVID-19 single-institution census of confirmed cases by polymerase chain reaction (PCR) for which the average number of cases of COVID-19 during a given month was determined. This number is a scaled trend, and a true number of COVID-19 cases in our hospital was not reported.

Neurosurgical and neuroendovascular cases were included in the analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases, defined as the time from the patient leaving the room until the next patient entered the room. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover, which is a standard for our single-institution perioperative center). Statistical analyses, including data aggregation, were performed using R, version 4.0.1 (R Foundation for Statistical Computing). Patients’ demographic and clinical characteristics were analyzed using an independent 2-sample t-test for interval variables and a chi-square test for categorical variables. Significance was defined as P < .05.

Results

First-Start Time

First-start time was analyzed in 426 pre-COVID-19, 357 peak-COVID-19, and 2304 post-peak-COVID-19 cases. The unadjusted mean delay length was significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004) (Table 1).

The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different, but they have been slightly higher since the onset of COVID-19. The proportion of cases that have started early, as well as significantly early past a 15-minute threshold, have also trended down since the onset of the COVID-19 pandemic, but this difference was again not significant. The temporal relationship of first-start delay, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 1. The trend of increasing delay is loosely associated with the COVID-19 burden experienced by our hospital. The start of COVID-19 as well as both COVID-19 peaks have been associated with increased delays in our hospital.

Turnover Time

Turnover time was assessed in 437 pre-COVID-19, 278 peak-restrictions, and 2411 post-peak-restrictions cases. Turnover time during peak restrictions was not significantly different from pre-COVID-19 (88 [100] vs 85 [95]) and has since remained relatively unchanged (83 [87], P = .78). A similar trend held for comparisons of proportion of cases with turnover time past 90 minutes and average times past the 90-minute threshold (Table 2). The temporal relationship between COVID-19 burden and turnover time, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 2. Both figures demonstrate a slight initial increase in turnover time delay at the start of COVID-19, which stabilized with little variation thereafter.

We analyzed the OR efficiency metrics of first-start and turnover time during the 90-day period before COVID-19 (pre-COVID-19), the 90 days following Tennessee declaring a state of emergency (peak COVID-19), and the time following this period (post-COVID-19) for all neurosurgical and neuroendovascular cases at Vanderbilt University Medical Center (VUMC). We found a significant difference in unadjusted mean delay length in first-start time between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes for pre-COVID-19, peak-COVID-19, and post-COVID-19: 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). No significant increase in turnover time between cases was found between these 3 time periods. Based on metrics from first-start delay and turnover time, our center was able to maintain OR efficiency before, during, and after peak COVID-19.

After the Centers for Disease Control and Prevention released guidelines recommending deferring elective procedures to conserve beds and PPE, VUMC made the decision to suspend all elective surgical procedures from March 18 to April 24, 2020. Prior research conducted during the COVID-19 pandemic has demonstrated more than 400 types of surgical procedures with negatively impacted outcomes when compared to surgical outcomes from the same time frame in 2018 and 2019.⁴ For more than 20 of these types of procedures, there was a significant association between procedure delay and adverse patient outcomes.⁴ Testing protocols for patients prior to surgery varied throughout the pandemic based on vaccination status and type of procedure. Before vaccines became widely available, all patients were required to obtain a PCR SARS-CoV-2 test within 48 to 72 hours of the scheduled procedure. If the patient’s procedure was urgent and testing was not feasible, the patient was treated as a SARS-CoV-2–positive patient, which required all health care workers involved in the case to wear gowns, gloves, surgical masks, and eye protection. Testing patients preoperatively likely helped to maintain OR efficiency since not all patients received test results prior to the scheduled procedure, leading to cancellations of cases and therefore more staff available for fewer cases.

After vaccines became widely available to the public, testing requirements for patients preoperatively were relaxed, and only patients who were not fully vaccinated or severely immunocompromised were required to test prior to procedures. However, approximately 40% of the population in Tennessee was fully vaccinated in 2021, which reflects the patient population of VUMC.⁵ This means that many patients who received care at VUMC were still tested prior to procedures.

Adopting adequate safety protocols was found to be key for OR efficiency during the COVID-19 pandemic since performing surgery increased the risk of infection for each health care worker in the OR.⁶ VUMC protocols identified procedures that required enhanced safety measures to prevent infection of health care workers and avoid staffing shortages, which would decrease OR efficiency. Protocols mandated that only anesthesia team members were allowed to be in the OR during intubation and extubation of patients, which could be one factor leading to increased delays and decreased efficiency for some institutions. Methods for neurosurgeons to decrease risk of infection in the OR include postponing all nonurgent cases, reappraising the necessity for general anesthesia and endotracheal intubation, considering alternative surgical approaches that avoid the respiratory tract, and limiting the use of aerosol-generating instruments.^7,8 VUMC’s success in implementing these protocols likely explains why our center was able to maintain OR efficiency throughout the COVID-19 pandemic.

A study conducted by Andreata et al showed a significantly increased mean first-case delay and a nonsignificant increased turnover time in orthopedic surgeries in Northern Italy when comparing surgeries performed during the COVID-19 pandemic to those performed prior to COVID-19.² Other studies have indicated a similar trend in decreased OR efficiency during COVID-19 in other areas around the world.^9,10 These findings are not consistent with our own findings for neurosurgical and neuroendovascular surgeries at VUMC, and any change at our institution was relatively immaterial. Factors that threatened to change OR efficiency—but did not result in meaningful changes in our institutional experience—include delays due to pending COVID-19 test results, safety procedures such as PPE donning, and planning difficulties to ensure the existence of teams with non-overlapping providers in the case of a surgeon being infected.^2,11-13

Globally, many surgery centers halted all elective surgeries during the initial COVID-19 spike to prevent a PPE shortage and mitigate risk of infection of patients and health care workers.^8,12,14 However, there is no centralized definition of which neurosurgical procedures are elective, so that decision was made on a surgeon or center level, which could lead to variability in efficiency trends.¹⁴ One study on neurosurgical procedures during COVID-19 found a 30% decline in all cases and a 23% decline in emergent procedures, showing that the decrease in volume was not only due to cancellation of elective procedures.¹⁵ This decrease in elective and emergent surgeries created a backlog of surgeries as well as a loss in health care revenue, and caused many patients to go without adequate health care.¹⁰ Looking forward, it is imperative that surgical centers study trends in OR efficiency from COVID-19 and learn how to better maintain OR efficiency during future pandemic conditions to prevent a backlog of cases, loss of health care revenue, and decreased health care access.

Limitations

Our data are from a single center and therefore may not be representative of experiences of other hospitals due to different populations and different impacts from COVID-19. However, given our center’s high volume and diverse patient population, we believe our analysis highlights important trends in neurosurgery practice. Notably, data for patient and OR timing are digitally generated and are entered manually by nurses in the electronic medical record, making it prone to errors and variability. This is in our experience, and if any error is present, we believe it is minimal.

Conclusion

The COVID-19 pandemic has had far-reaching effects on health care worldwide, including neurosurgical care. OR efficiency across the United States generally worsened given the stresses of supply chain issues, staffing shortages, and cancellations. At our institution, we were able to maintain OR efficiency during the known COVID-19 peaks until October 2021. Continually functional neurosurgical ORs are important in preventing delays in care and maintaining a steady revenue in order for hospitals and other health care entities to remain solvent. Further study of OR efficiency is needed for health care systems to prepare for future pandemics and other resource-straining events in order to provide optimal patient care.

Corresponding author: Campbell Liles, MD, Vanderbilt University Medical Center, Department of Neurological Surgery, 1161 21st Ave. South, T4224 Medical Center North, Nashville, TN 37232-2380; david.c.liles.1@vumc.org

Disclosures: None reported.

From the Department of Neurological Surgery, Vanderbilt University Medical Center, Nashville, TN (Stefan W. Koester, Puja Jagasia, and Drs. Liles, Dambrino IV, Feldman, and Chambless), and the Department of Anesthesiology, Vanderbilt University Medical Center, Nashville, TN (Drs. Mathews and Tiwari).

ABSTRACT

Background: The COVID-19 pandemic has had broad effects on surgical care, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and newly implemented anti-infective measures. Our aim was to assess neurosurgery OR efficiency before the COVID-19 pandemic, during peak COVID-19, and during current times.

Methods: Institutional perioperative databases at a single, high-volume neurosurgical center were queried for operations performed from December 2019 until October 2021. March 12, 2020, the day that the state of Tennessee declared a state of emergency, was chosen as the onset of the COVID-19 pandemic. The 90-day periods before and after this day were used to define the pre-COVID-19, peak-COVID-19, and post-peak restrictions time periods for comparative analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover). Univariate analysis used Wilcoxon rank-sum test for continuous outcomes, while chi-square test and Fisher’s exact test were used for categorical comparisons. Significance was defined as P < .05.

Results: First-start time was analyzed in 426 pre-COVID-19, 357 peak-restrictions, and 2304 post-peak-restrictions cases. The unadjusted mean delay length was found to be significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different. The proportion of cases that started early, as well as significantly early past a 15-minute threshold, have not been impacted. There was no significant change in turnover time during peak restrictions relative to the pre-COVID-19 period (88 [100] minutes vs 85 [95] minutes), and turnover time has since remained unchanged (83 [87] minutes).

Conclusion: Our center was able to maintain OR efficiency before, during, and after peak restrictions even while instituting advanced infection-control strategies. While there were significant changes, delays were relatively small in magnitude.

Keywords: operating room timing, hospital efficiency, socioeconomics, pandemic.

The COVID-19 pandemic has led to major changes in patient care both from a surgical perspective and in regard to inpatient hospital course. Safety protocols nationwide have been implemented to protect both patients and providers. Some elements of surgical care have drastically changed, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and increased sterilization measures. Furloughs, layoffs, and reassignments due to the focus on nonelective and COVID-19–related cases challenged OR staffing and efficiency. Operating room staff with COVID-19 exposures or COVID-19 infections also caused last-minute changes in staffing. All of these scenarios can cause issues due to actual understaffing or due to staff members being pushed into highly specialized areas, such as neurosurgery, in which they have very little experience. A further obstacle to OR efficiency included policy changes involving PPE utilization, sterilization measures, and supply chain shortages of necessary resources such as PPE.

Neurosurgery in particular has been susceptible to COVID-19–related system-wide changes given operator proximity to the patient’s respiratory passages, frequency of emergent cases, and varying anesthetic needs, as well as the high level of specialization needed to perform neurosurgical care. Previous studies have shown a change in the makeup of neurosurgical patients seeking care, as well as in the acuity of neurological consult of these patients.¹ A study in orthopedic surgery by Andreata et al demonstrated worsened OR efficiency, with significantly increased first-start and turnover times.² In the COVID-19 era, OR quality and safety are crucially important to both patients and providers. Providing this safe and effective care in an efficient manner is important for optimal neurosurgical management in the long term.³ Moreover, the financial burden of implementing new protocols and standards can be compounded by additional financial losses due to reduced OR efficiency.

Methods

To analyze the effect of COVID-19 on neurosurgical OR efficiency, institutional perioperative databases at a single high-volume center were queried for operations performed from December 2019 until October 2021. March 12, 2020, was chosen as the onset of COVID-19 for analytic purposes, as this was the date when the state of Tennessee declared a state of emergency. The 90-day periods before and after this date were used for comparative analysis for pre-COVID-19, peak COVID-19, and post-peak-restrictions time periods. The peak COVID-19 period was defined as the 90-day period following the initial onset of COVID-19 and the surge of cases. For comparison purposes, post-peak COVID-19 was defined as the months following the first peak until October 2021 (approximately 17 months). COVID-19 burden was determined using a COVID-19 single-institution census of confirmed cases by polymerase chain reaction (PCR) for which the average number of cases of COVID-19 during a given month was determined. This number is a scaled trend, and a true number of COVID-19 cases in our hospital was not reported.

Neurosurgical and neuroendovascular cases were included in the analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases, defined as the time from the patient leaving the room until the next patient entered the room. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover, which is a standard for our single-institution perioperative center). Statistical analyses, including data aggregation, were performed using R, version 4.0.1 (R Foundation for Statistical Computing). Patients’ demographic and clinical characteristics were analyzed using an independent 2-sample t-test for interval variables and a chi-square test for categorical variables. Significance was defined as P < .05.

Results

First-Start Time

First-start time was analyzed in 426 pre-COVID-19, 357 peak-COVID-19, and 2304 post-peak-COVID-19 cases. The unadjusted mean delay length was significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004) (Table 1).

The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different, but they have been slightly higher since the onset of COVID-19. The proportion of cases that have started early, as well as significantly early past a 15-minute threshold, have also trended down since the onset of the COVID-19 pandemic, but this difference was again not significant. The temporal relationship of first-start delay, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 1. The trend of increasing delay is loosely associated with the COVID-19 burden experienced by our hospital. The start of COVID-19 as well as both COVID-19 peaks have been associated with increased delays in our hospital.

Turnover Time

Turnover time was assessed in 437 pre-COVID-19, 278 peak-restrictions, and 2411 post-peak-restrictions cases. Turnover time during peak restrictions was not significantly different from pre-COVID-19 (88 [100] vs 85 [95]) and has since remained relatively unchanged (83 [87], P = .78). A similar trend held for comparisons of proportion of cases with turnover time past 90 minutes and average times past the 90-minute threshold (Table 2). The temporal relationship between COVID-19 burden and turnover time, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 2. Both figures demonstrate a slight initial increase in turnover time delay at the start of COVID-19, which stabilized with little variation thereafter.

We analyzed the OR efficiency metrics of first-start and turnover time during the 90-day period before COVID-19 (pre-COVID-19), the 90 days following Tennessee declaring a state of emergency (peak COVID-19), and the time following this period (post-COVID-19) for all neurosurgical and neuroendovascular cases at Vanderbilt University Medical Center (VUMC). We found a significant difference in unadjusted mean delay length in first-start time between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes for pre-COVID-19, peak-COVID-19, and post-COVID-19: 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). No significant increase in turnover time between cases was found between these 3 time periods. Based on metrics from first-start delay and turnover time, our center was able to maintain OR efficiency before, during, and after peak COVID-19.

After the Centers for Disease Control and Prevention released guidelines recommending deferring elective procedures to conserve beds and PPE, VUMC made the decision to suspend all elective surgical procedures from March 18 to April 24, 2020. Prior research conducted during the COVID-19 pandemic has demonstrated more than 400 types of surgical procedures with negatively impacted outcomes when compared to surgical outcomes from the same time frame in 2018 and 2019.⁴ For more than 20 of these types of procedures, there was a significant association between procedure delay and adverse patient outcomes.⁴ Testing protocols for patients prior to surgery varied throughout the pandemic based on vaccination status and type of procedure. Before vaccines became widely available, all patients were required to obtain a PCR SARS-CoV-2 test within 48 to 72 hours of the scheduled procedure. If the patient’s procedure was urgent and testing was not feasible, the patient was treated as a SARS-CoV-2–positive patient, which required all health care workers involved in the case to wear gowns, gloves, surgical masks, and eye protection. Testing patients preoperatively likely helped to maintain OR efficiency since not all patients received test results prior to the scheduled procedure, leading to cancellations of cases and therefore more staff available for fewer cases.

After vaccines became widely available to the public, testing requirements for patients preoperatively were relaxed, and only patients who were not fully vaccinated or severely immunocompromised were required to test prior to procedures. However, approximately 40% of the population in Tennessee was fully vaccinated in 2021, which reflects the patient population of VUMC.⁵ This means that many patients who received care at VUMC were still tested prior to procedures.

Adopting adequate safety protocols was found to be key for OR efficiency during the COVID-19 pandemic since performing surgery increased the risk of infection for each health care worker in the OR.⁶ VUMC protocols identified procedures that required enhanced safety measures to prevent infection of health care workers and avoid staffing shortages, which would decrease OR efficiency. Protocols mandated that only anesthesia team members were allowed to be in the OR during intubation and extubation of patients, which could be one factor leading to increased delays and decreased efficiency for some institutions. Methods for neurosurgeons to decrease risk of infection in the OR include postponing all nonurgent cases, reappraising the necessity for general anesthesia and endotracheal intubation, considering alternative surgical approaches that avoid the respiratory tract, and limiting the use of aerosol-generating instruments.^7,8 VUMC’s success in implementing these protocols likely explains why our center was able to maintain OR efficiency throughout the COVID-19 pandemic.

A study conducted by Andreata et al showed a significantly increased mean first-case delay and a nonsignificant increased turnover time in orthopedic surgeries in Northern Italy when comparing surgeries performed during the COVID-19 pandemic to those performed prior to COVID-19.² Other studies have indicated a similar trend in decreased OR efficiency during COVID-19 in other areas around the world.^9,10 These findings are not consistent with our own findings for neurosurgical and neuroendovascular surgeries at VUMC, and any change at our institution was relatively immaterial. Factors that threatened to change OR efficiency—but did not result in meaningful changes in our institutional experience—include delays due to pending COVID-19 test results, safety procedures such as PPE donning, and planning difficulties to ensure the existence of teams with non-overlapping providers in the case of a surgeon being infected.^2,11-13

Globally, many surgery centers halted all elective surgeries during the initial COVID-19 spike to prevent a PPE shortage and mitigate risk of infection of patients and health care workers.^8,12,14 However, there is no centralized definition of which neurosurgical procedures are elective, so that decision was made on a surgeon or center level, which could lead to variability in efficiency trends.¹⁴ One study on neurosurgical procedures during COVID-19 found a 30% decline in all cases and a 23% decline in emergent procedures, showing that the decrease in volume was not only due to cancellation of elective procedures.¹⁵ This decrease in elective and emergent surgeries created a backlog of surgeries as well as a loss in health care revenue, and caused many patients to go without adequate health care.¹⁰ Looking forward, it is imperative that surgical centers study trends in OR efficiency from COVID-19 and learn how to better maintain OR efficiency during future pandemic conditions to prevent a backlog of cases, loss of health care revenue, and decreased health care access.

Limitations

Our data are from a single center and therefore may not be representative of experiences of other hospitals due to different populations and different impacts from COVID-19. However, given our center’s high volume and diverse patient population, we believe our analysis highlights important trends in neurosurgery practice. Notably, data for patient and OR timing are digitally generated and are entered manually by nurses in the electronic medical record, making it prone to errors and variability. This is in our experience, and if any error is present, we believe it is minimal.

Conclusion

The COVID-19 pandemic has had far-reaching effects on health care worldwide, including neurosurgical care. OR efficiency across the United States generally worsened given the stresses of supply chain issues, staffing shortages, and cancellations. At our institution, we were able to maintain OR efficiency during the known COVID-19 peaks until October 2021. Continually functional neurosurgical ORs are important in preventing delays in care and maintaining a steady revenue in order for hospitals and other health care entities to remain solvent. Further study of OR efficiency is needed for health care systems to prepare for future pandemics and other resource-straining events in order to provide optimal patient care.

Corresponding author: Campbell Liles, MD, Vanderbilt University Medical Center, Department of Neurological Surgery, 1161 21st Ave. South, T4224 Medical Center North, Nashville, TN 37232-2380; david.c.liles.1@vumc.org

Disclosures: None reported.

From the Department of Neurological Surgery, Vanderbilt University Medical Center, Nashville, TN (Stefan W. Koester, Puja Jagasia, and Drs. Liles, Dambrino IV, Feldman, and Chambless), and the Department of Anesthesiology, Vanderbilt University Medical Center, Nashville, TN (Drs. Mathews and Tiwari).

ABSTRACT

Background: The COVID-19 pandemic has had broad effects on surgical care, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and newly implemented anti-infective measures. Our aim was to assess neurosurgery OR efficiency before the COVID-19 pandemic, during peak COVID-19, and during current times.

Methods: Institutional perioperative databases at a single, high-volume neurosurgical center were queried for operations performed from December 2019 until October 2021. March 12, 2020, the day that the state of Tennessee declared a state of emergency, was chosen as the onset of the COVID-19 pandemic. The 90-day periods before and after this day were used to define the pre-COVID-19, peak-COVID-19, and post-peak restrictions time periods for comparative analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover). Univariate analysis used Wilcoxon rank-sum test for continuous outcomes, while chi-square test and Fisher’s exact test were used for categorical comparisons. Significance was defined as P < .05.

Results: First-start time was analyzed in 426 pre-COVID-19, 357 peak-restrictions, and 2304 post-peak-restrictions cases. The unadjusted mean delay length was found to be significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different. The proportion of cases that started early, as well as significantly early past a 15-minute threshold, have not been impacted. There was no significant change in turnover time during peak restrictions relative to the pre-COVID-19 period (88 [100] minutes vs 85 [95] minutes), and turnover time has since remained unchanged (83 [87] minutes).

Conclusion: Our center was able to maintain OR efficiency before, during, and after peak restrictions even while instituting advanced infection-control strategies. While there were significant changes, delays were relatively small in magnitude.

Keywords: operating room timing, hospital efficiency, socioeconomics, pandemic.

The COVID-19 pandemic has led to major changes in patient care both from a surgical perspective and in regard to inpatient hospital course. Safety protocols nationwide have been implemented to protect both patients and providers. Some elements of surgical care have drastically changed, including operating room (OR) staffing, personal protective equipment (PPE) utilization, and increased sterilization measures. Furloughs, layoffs, and reassignments due to the focus on nonelective and COVID-19–related cases challenged OR staffing and efficiency. Operating room staff with COVID-19 exposures or COVID-19 infections also caused last-minute changes in staffing. All of these scenarios can cause issues due to actual understaffing or due to staff members being pushed into highly specialized areas, such as neurosurgery, in which they have very little experience. A further obstacle to OR efficiency included policy changes involving PPE utilization, sterilization measures, and supply chain shortages of necessary resources such as PPE.

Neurosurgery in particular has been susceptible to COVID-19–related system-wide changes given operator proximity to the patient’s respiratory passages, frequency of emergent cases, and varying anesthetic needs, as well as the high level of specialization needed to perform neurosurgical care. Previous studies have shown a change in the makeup of neurosurgical patients seeking care, as well as in the acuity of neurological consult of these patients.¹ A study in orthopedic surgery by Andreata et al demonstrated worsened OR efficiency, with significantly increased first-start and turnover times.² In the COVID-19 era, OR quality and safety are crucially important to both patients and providers. Providing this safe and effective care in an efficient manner is important for optimal neurosurgical management in the long term.³ Moreover, the financial burden of implementing new protocols and standards can be compounded by additional financial losses due to reduced OR efficiency.

Methods

To analyze the effect of COVID-19 on neurosurgical OR efficiency, institutional perioperative databases at a single high-volume center were queried for operations performed from December 2019 until October 2021. March 12, 2020, was chosen as the onset of COVID-19 for analytic purposes, as this was the date when the state of Tennessee declared a state of emergency. The 90-day periods before and after this date were used for comparative analysis for pre-COVID-19, peak COVID-19, and post-peak-restrictions time periods. The peak COVID-19 period was defined as the 90-day period following the initial onset of COVID-19 and the surge of cases. For comparison purposes, post-peak COVID-19 was defined as the months following the first peak until October 2021 (approximately 17 months). COVID-19 burden was determined using a COVID-19 single-institution census of confirmed cases by polymerase chain reaction (PCR) for which the average number of cases of COVID-19 during a given month was determined. This number is a scaled trend, and a true number of COVID-19 cases in our hospital was not reported.

Neurosurgical and neuroendovascular cases were included in the analysis. Outcomes included delay in first-start and OR turnover time between neurosurgical cases, defined as the time from the patient leaving the room until the next patient entered the room. Preset threshold times were used in analyses to adjust for normal leniency in OR scheduling (15 minutes for first start and 90 minutes for turnover, which is a standard for our single-institution perioperative center). Statistical analyses, including data aggregation, were performed using R, version 4.0.1 (R Foundation for Statistical Computing). Patients’ demographic and clinical characteristics were analyzed using an independent 2-sample t-test for interval variables and a chi-square test for categorical variables. Significance was defined as P < .05.

Results

First-Start Time

First-start time was analyzed in 426 pre-COVID-19, 357 peak-COVID-19, and 2304 post-peak-COVID-19 cases. The unadjusted mean delay length was significantly different between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes, 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004) (Table 1).

The adjusted average delay length and proportion of cases delayed beyond the 15-minute threshold were not significantly different, but they have been slightly higher since the onset of COVID-19. The proportion of cases that have started early, as well as significantly early past a 15-minute threshold, have also trended down since the onset of the COVID-19 pandemic, but this difference was again not significant. The temporal relationship of first-start delay, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 1. The trend of increasing delay is loosely associated with the COVID-19 burden experienced by our hospital. The start of COVID-19 as well as both COVID-19 peaks have been associated with increased delays in our hospital.

Turnover Time

Turnover time was assessed in 437 pre-COVID-19, 278 peak-restrictions, and 2411 post-peak-restrictions cases. Turnover time during peak restrictions was not significantly different from pre-COVID-19 (88 [100] vs 85 [95]) and has since remained relatively unchanged (83 [87], P = .78). A similar trend held for comparisons of proportion of cases with turnover time past 90 minutes and average times past the 90-minute threshold (Table 2). The temporal relationship between COVID-19 burden and turnover time, both unadjusted and adjusted, from December 2019 to October 2021 is shown in Figure 2. Both figures demonstrate a slight initial increase in turnover time delay at the start of COVID-19, which stabilized with little variation thereafter.

We analyzed the OR efficiency metrics of first-start and turnover time during the 90-day period before COVID-19 (pre-COVID-19), the 90 days following Tennessee declaring a state of emergency (peak COVID-19), and the time following this period (post-COVID-19) for all neurosurgical and neuroendovascular cases at Vanderbilt University Medical Center (VUMC). We found a significant difference in unadjusted mean delay length in first-start time between the time periods, but the magnitude of increase in minutes was immaterial (mean [SD] minutes for pre-COVID-19, peak-COVID-19, and post-COVID-19: 6 [18] vs 10 [21] vs 8 [20], respectively; P = .004). No significant increase in turnover time between cases was found between these 3 time periods. Based on metrics from first-start delay and turnover time, our center was able to maintain OR efficiency before, during, and after peak COVID-19.

After the Centers for Disease Control and Prevention released guidelines recommending deferring elective procedures to conserve beds and PPE, VUMC made the decision to suspend all elective surgical procedures from March 18 to April 24, 2020. Prior research conducted during the COVID-19 pandemic has demonstrated more than 400 types of surgical procedures with negatively impacted outcomes when compared to surgical outcomes from the same time frame in 2018 and 2019.⁴ For more than 20 of these types of procedures, there was a significant association between procedure delay and adverse patient outcomes.⁴ Testing protocols for patients prior to surgery varied throughout the pandemic based on vaccination status and type of procedure. Before vaccines became widely available, all patients were required to obtain a PCR SARS-CoV-2 test within 48 to 72 hours of the scheduled procedure. If the patient’s procedure was urgent and testing was not feasible, the patient was treated as a SARS-CoV-2–positive patient, which required all health care workers involved in the case to wear gowns, gloves, surgical masks, and eye protection. Testing patients preoperatively likely helped to maintain OR efficiency since not all patients received test results prior to the scheduled procedure, leading to cancellations of cases and therefore more staff available for fewer cases.

After vaccines became widely available to the public, testing requirements for patients preoperatively were relaxed, and only patients who were not fully vaccinated or severely immunocompromised were required to test prior to procedures. However, approximately 40% of the population in Tennessee was fully vaccinated in 2021, which reflects the patient population of VUMC.⁵ This means that many patients who received care at VUMC were still tested prior to procedures.

Adopting adequate safety protocols was found to be key for OR efficiency during the COVID-19 pandemic since performing surgery increased the risk of infection for each health care worker in the OR.⁶ VUMC protocols identified procedures that required enhanced safety measures to prevent infection of health care workers and avoid staffing shortages, which would decrease OR efficiency. Protocols mandated that only anesthesia team members were allowed to be in the OR during intubation and extubation of patients, which could be one factor leading to increased delays and decreased efficiency for some institutions. Methods for neurosurgeons to decrease risk of infection in the OR include postponing all nonurgent cases, reappraising the necessity for general anesthesia and endotracheal intubation, considering alternative surgical approaches that avoid the respiratory tract, and limiting the use of aerosol-generating instruments.^7,8 VUMC’s success in implementing these protocols likely explains why our center was able to maintain OR efficiency throughout the COVID-19 pandemic.

A study conducted by Andreata et al showed a significantly increased mean first-case delay and a nonsignificant increased turnover time in orthopedic surgeries in Northern Italy when comparing surgeries performed during the COVID-19 pandemic to those performed prior to COVID-19.² Other studies have indicated a similar trend in decreased OR efficiency during COVID-19 in other areas around the world.^9,10 These findings are not consistent with our own findings for neurosurgical and neuroendovascular surgeries at VUMC, and any change at our institution was relatively immaterial. Factors that threatened to change OR efficiency—but did not result in meaningful changes in our institutional experience—include delays due to pending COVID-19 test results, safety procedures such as PPE donning, and planning difficulties to ensure the existence of teams with non-overlapping providers in the case of a surgeon being infected.^2,11-13

Globally, many surgery centers halted all elective surgeries during the initial COVID-19 spike to prevent a PPE shortage and mitigate risk of infection of patients and health care workers.^8,12,14 However, there is no centralized definition of which neurosurgical procedures are elective, so that decision was made on a surgeon or center level, which could lead to variability in efficiency trends.¹⁴ One study on neurosurgical procedures during COVID-19 found a 30% decline in all cases and a 23% decline in emergent procedures, showing that the decrease in volume was not only due to cancellation of elective procedures.¹⁵ This decrease in elective and emergent surgeries created a backlog of surgeries as well as a loss in health care revenue, and caused many patients to go without adequate health care.¹⁰ Looking forward, it is imperative that surgical centers study trends in OR efficiency from COVID-19 and learn how to better maintain OR efficiency during future pandemic conditions to prevent a backlog of cases, loss of health care revenue, and decreased health care access.

Limitations

Our data are from a single center and therefore may not be representative of experiences of other hospitals due to different populations and different impacts from COVID-19. However, given our center’s high volume and diverse patient population, we believe our analysis highlights important trends in neurosurgery practice. Notably, data for patient and OR timing are digitally generated and are entered manually by nurses in the electronic medical record, making it prone to errors and variability. This is in our experience, and if any error is present, we believe it is minimal.

Conclusion

The COVID-19 pandemic has had far-reaching effects on health care worldwide, including neurosurgical care. OR efficiency across the United States generally worsened given the stresses of supply chain issues, staffing shortages, and cancellations. At our institution, we were able to maintain OR efficiency during the known COVID-19 peaks until October 2021. Continually functional neurosurgical ORs are important in preventing delays in care and maintaining a steady revenue in order for hospitals and other health care entities to remain solvent. Further study of OR efficiency is needed for health care systems to prepare for future pandemics and other resource-straining events in order to provide optimal patient care.

Corresponding author: Campbell Liles, MD, Vanderbilt University Medical Center, Department of Neurological Surgery, 1161 21st Ave. South, T4224 Medical Center North, Nashville, TN 37232-2380; david.c.liles.1@vumc.org

Disclosures: None reported.