Predictors of Poor Health Outcomes in Midlife and Older Adults with Diabetes in the 2018-2020 Health and Retirement Study

<h3>Context:</h3> Adults with diabetes have higher premature death rates, functional disability, and comorbidities than adults without diabetes. Diabetes co-occurs with other chronic conditions (CC), functional limitations (FL), and geriatric syndromes (GS). However, the effects of these three combined on health outcomes have not been examined in midlife and older adults with diabetes. <h3>Objective:</h3> We aimed to identify the combinations of CC, FL, and GS that predict poor health outcomes in midlife and older adults with diabetes. <h3>Study Design and Analysis:</h3> Cross-sectional study. We conducted Classification and Regression Tree (CART) analysis to identify the specific combinations of CC, FL, and GS conditions and sociodemographic characteristics that predict each outcome overall, and by age and race. <h3>Setting or Dataset:</h3> Respondents from the 2018 and 2020 Health and Retirement Study (HRS). <h3>Population studied:</h3> Respondents with self-reported diabetes 50 years or older. <h3>Intervention/Instrument:</h3> Observational study. <h3>Outcome Measures:</h3> Self-reported fair/poor health in 2018, self-reported worse health at two years, and two-year mortality. <h3>Results:</h3> Of the 4,019 older adults with diabetes, 42% reported fair/poor health, 24% reported worse health at two years, and 7% died in two years. We found that the specific combinations of conditions and sociodemographic characteristics differed based on outcome and stratification. The highest percentage of individuals reporting fair/poor health (80%) was observed among individuals with visual impairment and incomes below 300% of the Federal Poverty Level. Individuals ≥ 58.5 years with psychiatric conditions, limitations in activities and instrumental activities of daily living (IADL), and lower body limitations had the highest percentage of self-reported worse health at two years (62%). For two-year mortality, the highest percentage (47%) was observed among individuals aged ≥ 85.5 years with IADL limitations. <h3>Conclusions:</h3> The significant association between C-MM and health outcomes shows that accounting for FL and GS in addition to CC can help identify midlife and older adults with diabetes at an increased risk of adverse health outcomes. Additionally, identifying individuals with co-occurrence of CC, FL, and GS can be more informative than only the co-occurrence of CC in predicting health outcomes.