OBJECTIVES: The Texas DMIE study examines whether participants with potentially disabling physical and/or behavioral conditions receiving a coordinated set of health and employment supports can avoid dependence on federal disability benefits.  This study examines differences in antidepressant adherence and persistence between intervention and control participants enrolled in the Texas DMIE study. METHODS: DMIE is a randomized controlled trial with 1616 participants aged 21-60; working a minimum average of 40 hours per month; diagnosed with behavioral health and/or physical health conditions; and who are not currently receiving disability.  A subgroup with a major depression diagnosis and prescribed antidepressant medication 12-months prior and 12-months post enrollment was examined.  Drug adherence was measured using proportion of days covered (PDC) for each patient during the 365-day pre- and post-enrollment observation periods.  Medication persistence measured the duration of time from initiation to discontinuation of antidepressants based on a ≥ 35-day refill supply gap.  Covariates included overall health morbidity, age, race/ethnicity, gender, occupation, serious mental illness status, and recruitment method (mail/telephone versus in-person). Findings were based on analysis of covariance for adherence and survival analysis for persistence. RESULTS: This study sample included 166 DMIE participants (Intervention n= 101 and Control n= 65), with a mean age of 47.8 years. The model showed (F[1148]= 4.47, p<0.05) intervention participants recruited in-person at community health clinics with a higher mean PDC (70%) compared with control group participants (54%).  Variation in persistence were statistically significant among older (HR= .97, p<0.05) and higher education (HR= .71, p<0.01) participants, however, group differences were not found between the cohorts. CONCLUSIONS: Adherence rates differed between intervention and control based on recruitment cohort which may reflect lower motivation level on the mail/telephone recruited participants.  Findings may help guide implementation of healthcare reform targeted toward low-income working adults lacking health insurance.