One-year postcollaborative depression care trial outcomes among predominantly Hispanic diabetes safety net patients

Abstract

Objective: The aim of this study was to determine sustained effectiveness in reducing depression symptoms and improving depression care 1 year following intervention completion.

Method: Of 387 low-income, predominantly Hispanic diabetes patients with major depression symptoms randomized to 12-month socioculturally adapted collaborative care (psychotherapy and/or antidepressants, telephone symptom monitoring/relapse prevention) or enhanced usual care, 264 patients completed 2-year follow-up. Depression symptoms (Symptom Checklist-20 [SCL-20], Patient Health Questionnaire-9 [PHQ-9]), treatment receipt, diabetes symptoms and quality of life were assessed 24 months postenrollment using intent-to-treat analyses.

Results: At 24 months, more intervention patients received ongoing antidepressant treatment (38% vs. 25%, χ(2)=5.11, df=1, P=.02); sustained depression symptom improvement [SCL-20 <0.5 (adjusted odds ratio=2.06, 95% confidence interval=1.09-3.90, P=.03), SCL-20 score (adjusted mean difference -0.22, P=.001) and PHQ-9 ≥50% reduction (adjusted odds ratio=1.87, 95% confidence interval =1.05-3.32, P=.03)]. Over 2 years, improved effects were found in significant study group by time interaction for Short Form-12 mental health, Sheehan Disability Scale (SDS) functional impairment, diabetes symptoms, anxiety and socioeconomic stressors (P=.02 for SDS; P<.0001 for all others); however, group differences narrowed over time and were no longer significant at 24 months.

Conclusions: Socioculturally tailored collaborative care that included maintenance antidepressant medication, ongoing symptom monitoring and behavioral activation relapse prevention was associated with depression improvement over 24 months for predominantly Hispanic patients in primary safety net care.

Trial registration: ClinicalTrials.gov NCT00709150.

Figure 1

MDDP CONSORT Diagram

Figure 2

Intervention vs control differences in mean depression scores Mean depression scores from SCL-20 (range, 0–4) and PHQ-9 (range, 0–27) adjusted for study site, birth place, language, years in US, and baseline SCL-20 score, dysthymia, and number of stressors. Error bars indicate standard errors. Significant group difference in mean SCL-20 was observed at all follow-up points (P≤.001), as well as significant group difference in mean PHQ-9 at 6, 12, and 18 months (P≤.001).