Unanticipated effect of a randomized peer network intervention on depressive symptoms among young methamphetamine users in Thailand

We examined the effect on depressive symptoms of a peer network-oriented intervention effective in reducing sexual risk behavior and methamphetamine (MA) use. Current Thai MA users aged 18-25 years and their drug and/or sex network members enrolled in a randomized controlled trial with 4 follow-ups over 12 months. A total of 415 index participants recruited 568 network members. Linear repeated measures models were fit with depressive symptoms (Center for Epidemiologic Studies Depression Scale [CES-D]) as the outcome. Among indexes in the intervention condition, mean CES-D decreased from 20.0 to 15.7 (p < 0.0001) over follow-up. Controlling for covariates and changes in MA and alcohol use, CES-D in this group decreased by 0.35 points per month (95% confidence interval, 0.45, -0.25). All other groups showed minimal changes in CES-D score. The peer-oriented, community-based intervention designed to reduce MA use and sexual risk behavior also resulted in substantial reduction in depressive symptoms, independent of changes in MA use.