Thesis (M.S.H.P.) / BACKGROUND: Dual eligible (DE) patients qualify for Medicare and Medicaid. There are approximately nine million DE patients in the United States, and healthcare costs for this population totaled 319.5 billion dollars in 2011. Behavioral health illness (BHI) is a risk factor for increased healthcare service utilization. The healthcare utilization of adult DE patients <65 years of age with BHI has been studied sparsely. This study sought to describe the adult DE patient population <65 years of age at an urban academic safety net health center and compare hospital and emergency department (ED) utilization of those with and without BHI. METHODS: The study was a secondary analysis of hospital administrative data. Inclusion criteria were patients with Medicare and Medicaid between ages 18 and 65 years, who utilized Boston Medical Center between 1/1/2011 and 1/1/2012. The independent variable was diagnosis of BHI, and the dependent variables were hospital admission and ED utilization. Chi-square and Wilcoxon rank-sum tests were used for descriptive statistics on categorical and continuous variables, respectively. Greedy propensity-score matching without replacement with a caliper distance of half of a standard deviation was used to control for confounding factors. Rate ratios (RR) and confidence intervals (CI) were determined after matching and after adjusting for those variables that remained significantly different after matching. RESULTS: Pre-propensity-score matched data showed significant differences in age, sex, race/ethnicity, marital status, education, employment, physical comorbidities, and Charlson Comorbidity Index score. Post-propensity-score matched analysis found significant differences in sex, Hispanic race, and other education and employment status. As compared to those patients without BHI, patients with BHI had RR 2.07 (CI: 1.81- 2.38) (p<.0001) of hospital admission and a RR 1.61 (CI:1.46-1.77) (p<.0001) of ED utilization. After adjustment, RR for hospital admission and ED utilization remained significantly different and even increased slightly, RR 2.14 (CI: 1.87-2.46) (p<.0001) and RR 1.64 (CI:1.49-1.81) (p<.0001), respectively. CONCLUSION: As compared to DE patients without BHI, those with BHI had significantly more hospital admission and ED utilization, even after controlling for confounding factors. Results suggest interventions for decreasing healthcare services utilization in this population should focus on those DE patients with mental health illness. / 2031-01-01