Patients with cirrhosis treated at hospitals with the highest safety-net burden, defined by their proportion of Medicaid or uninsured patients, had a 5% higher mortality rate than patients who were treated at hospitals with the lowest burden, according to a study of over 300,000 patients.
The study, which was published in theanalyzed inpatient data from the National Inpatient Sample (NIS) database focusing on a 4-year time span between 2012 and 2016. The hospitals were categorized by safety-net burden, which was defined as having either a high, medium, or low number of uninsured patients or patients with Medicaid.
This is the first-known study to evaluate the impact of a hospital’s safety-net burden on hospitalization outcomes in cirrhosis patients, wrote authors, of Stanford (Calif.) University and Grishma Hirode, MAS, of the University of Toronto. Previous studies have shown that safety-net hospitals, especially those with a high safety-net burden, have poorer patient outcomes. These hospitals also serve a patient population that is at high risk for chronic liver disease and cirrhosis.
The new analysis included 322,944 individual hospitalizations of patients with cirrhosis. Of these, 57.8% were male, 63.7% were White, 9.9% were Black, and 15.6% were Hispanic. In terms of safety-net burden, 107,446 hospitalizations were at high-burden hospitals, 103,508 were at medium-burden hospitals, and 111,990 hospitalizations were at low-burden hospitals.
Overall, cirrhosis-related hospitalizations in hospitals with the highest burden were found to have significantly greater odds of in-hospital mortality than the lowest tertile hospitals (odds ratio, 1.05, P = .044). The patients were also younger (mean age, 56.7 years vs. 59.8 years in low-burden hospitals). They also had a higher proportion of male patients, minority patients, Hispanic patients, and patients with Medicaid or no insurance.
The odds of hospitalization in the highest tertile hospitals were found to be significantly higher, compared with the middle and lowest tertiles for Blacks and Hispanics, compared with Whites (OR 1.26 and OR 1.63, respectively). Black patients (OR, 1.26; 95%CI, 1.17-1.35; P < .001) and Hispanic patients (OR, 1.63; 95% CI, 1.50-1.78; P< .001) were more likely to be admitted for care at high-burden hospitals (26% to 54%). In-hospital mortality rates among all hospitalizations were 5.95% and the rate did not significantly differ by hospital burden status.
“Despite adjusting for safety-net burden, our study continued to demonstrate ethnic disparities in in-hospital mortality among cirrhosis-related hospitalizations,” the researchers wrote. Overall, the odds of in-hospital mortality were 27% higher in Black patients as compared with White patients.
However, significantly lower mortality was observed in Hispanic patients as compared with White patients (4.9% vs. 6.0%, P < .001), but why this occurred was not entirely clear. “Hispanic patients may be more likely to have NASH [nonalcoholic steatohepatitis]-related cirrhosis, which generally has a slower disease progression, compared with [hepatitis C virus] or alcoholic cirrhosis. As such, it is likely that NASH-cirrhosis Hispanic patients had less severe disease at presentation,” the researchers wrote.
Study design has limitations, but shows concerning trends
The study findings were limited by several factors including the inability to show causality based on the observational study design and cross-sectional nature of the database, the researchers said. The NIS database records individual hospitalizations, not individual patient data which means that it may include repeat hospitalizations from the same patient. In addition, the study was limited by a lack of data on outpatient cirrhosis outcomes and non–liver-related comorbidities.
However, the finding that ethnic minorities with cirrhosis were significantly more likely to be hospitalized in high safety-net hospitals than White patients is concerning, and more research is needed, they said.
“These observations highlight that, while disparities in resources and health care delivery inherent to safety-net health systems may partly explain and provide opportunities to improve cirrhosis hospitalization care, they alone do not explain all of the ethnic disparities in cirrhosis outcomes observed,” they concluded.
The current study was important to conduct at this time because rates of cirrhosis are on the rise, Michael Volk, MD, of Loma Linda (Calif.) University Health, said in an interview. “Millions of patients receive care in safety-net hospitals across the country.”
Dr. Volk said that he was not surprised by the overall outcomes. “Unfortunately, I expected that patient outcomes would be worse at safety-net hospitals than wealthier hospitals. However, I was surprised that Blacks had higher in-hospital mortality than Whites, even after adjusting for the hospital.”
Dr. Volk echoed the study’s stated limitation of the lack of data to address disparities.
“Additional research is needed to determine whether the higher in-hospital mortality among Blacks is related to biological differences such as differential rates of disease progression, or social differences such as access to outpatient care,” he said.
The study received no outside funding. The researchers had no financial conflicts to disclose. Dr. Volk had no relevant financial conflicts to disclose.