In their own voices: Latinas' experiences with a randomized controlled trial.

We conducted exit interviews with a random sample of 39 predominantly Central American immigrant mothers who had completed a longitudinal randomized controlled trial to prevent perinatal depression. We found that rates and levels of perinatal depression in the intervention and control groups were lower than expected and did not differ between groups at 1 year postpartum. Therefore, we conducted extensive semistructured interviews to (a) understand why these high-risk women had such low rates of major depressive episodes and depressive symptoms, and (b) determine if the mechanisms responsible for reductions in depression differed between the intervention and usual care groups. We discovered that the intervention group learned specific mood-management skills from their participation in the intervention, and that the control group experienced their participation in the study as a "low-dose" intervention. Our experience highlights the importance of conducting qualitative studies to understand quantitative outcomes of intervention studies.