Résumé
Psychoactive drug use is a major public health issue in Vietnam, including mental health concerns. Depression can lead to suicide, poorer health outcomes, and remains a burden for people who inject drugs (PWID), their family and society. Better identifying these populations is crucial to develop targeted interventions. Methods: we recruited PWID in a cohort through respondent driven sampling method (DRIVE project). Incident depression was monitored for PWID participating in at least one semi-annual follow-up visit for up to 36 months. PWID were excluded from the analysis when treated for depression. Health professionals assessed depression via a short-form CES-D questionnaire. We selected our variables with a directed acyclic graph (DAG) and used multiple imputation for missing data. A mixed non-linear model was used to identify risk factors. Results: among the 1384 PWID free from depression at baseline, 320 (24%) developed depression. The risk factors of depression were the sex (women), being older, being single, not having stable housing nor health insurance card, having been hospitalized during follow up, using methamphetamine weekly, not being treated with methadone. Years since inclusion were protective. Limitations: the exclusion of missing visits and social desirability could have led to selection and information biases. We used tools for measuring the outcome which might lead to measurement biases. Finally, confusion biases are possible despite our best efforts.Conclusion: identifying risk factors of depression among PWID highlighted the need to develop innovative targeted psychiatric intervention and harm reduction program in situ, with the help of supporting peers.