BETHESDA, MD. – A novel, peer- and nurse-led intervention in a primary care setting for type 2 diabetes in people with serious mental illness was associated with improvements in depression symptoms, global psychopathology, and overall health, a study has shown.
“The intervention really is patient self-management. It could be a nice complement to team-based, multidisciplinary care,” said Martha Sajatovic, MD, who presented the data in a poster at a National Institute of Mental Health conference on mental health services research. Dr. Sajatovic is the Willard W. Brown Chair and director of the Neurological & Behavioral Outcomes Center at University Hospitals Neurological Institute in Cleveland.
People with serious mental illness (SMI) have a significantly higher risk of premature death than do those in the general population, in part because this cohort experiences higher rates of metabolic disease, often exacerbated by higher rates of smoking, poor diet, substance abuse, and lack of exercise. However, in a 60-week randomized controlled trial of 200 people with SMI and comorbid type 2 diabetes, which was conducted in a primary care setting, those who were taught better self-care fared better than did those who received treatment as usual.
The group-based, psychosocial intervention – called “targeted training in illness” – blended psychoeducation, problem identification, goal setting, behavioral modeling, and care coordination around SMI and diabetes. In the first 12 weeks, groups of 6-10 people met in weekly, hour-long sessions co-led by a peer and nurse educator. Group discussions focused on self-management of diabetes through proper eating habits, regular exercise, tobacco cessation, and other forms of behavior modification.
Meeting as a group helped to “combat some of the social isolation that you see in this population,” Dr. Sajatovic said in an interview. “The peer leadership is really critical, too, because it empowers [the participants]. I believe peer support gives resilience ... and helps [the group] see you don’t have to be perfect to make progress.” In the study, the 3 months of group sessions were followed by weekly telephone maintenance sessions with either the peer or nurse educator for 48 weeks.
Half of the study’s participants – two-thirds of whom were women, and just over half of whom were black – had had a diagnosis of diabetes for at least 10 years; half of all participants used insulin. All had either schizophrenia, schizoaffective disorder, bipolar disorder, or major depressive disorder. Baseline rates of depression were high, and psychotic symptoms were minimal.
After assessments at baseline, 13, 30, and 60 weeks, the study arm was found to have improvements in depression, global psychopathology, and functional status, which Dr. Sajatovic said could be attributable to the group’s significantly improved knowledge about diabetes (P less than .01).
Glycemic control improved generally, a surprising finding that Dr. Sajatovic said could have been tied to the expansion of Medicaid in Ohio, where the study was done, and a “real concerted effort” to provide treatment by medical homes at this time.
While no significant difference between the groups was found overall, a post hoc analysis showed a difference in the 53% of the entire sample who had good to fair glycemic control (hemoglobin A1c equal to or less than 7.5) at baseline: At 60 weeks, those in the treatment arm achieved stable, long-term control compared with controls, whose values had worsened slightly (P = .024). Those people tended to be older, more likely to have schizophrenia, and less likely to be on insulin, and to have a shorter history of diabetes, said Dr. Sajatovic, professor of psychiatry and of neurology at Case Western Reserve University, Cleveland.
Compared with controls, the study arm had greater improvement at 60 weeks in Clinical Global Impression scores (P = .0008); Montgomery-Åsberg Depression Rating Scale scores (P = .0156); Global Assessment of Functioning scores (P = .0031); and knowledge of diabetes (P less than .0002), as well as an improvement trend in Sheehan Disability Scale scores (P = .0863). There was no difference between the groups on the Brief Psychiatric Rating Scale, the Short Form–36 or HbA1c values. By study’s end, Dr. Sajatovic said about a quarter had been lost to follow-up.
The intervention meets three important criteria, she said. “First, people need to know what to do. Then, they need to have confidence, or self-efficacy. The third thing is that the person has to believe in a given outcome based on a given behavior.”
Dr. Sajatovic did not have any relevant disclosures. The National Institutes of Health funded the study.