Background Depression commonly co-exists with human immunodeficiency virus (HIV), but in low- and middle-income countries (LMICs), where the HIV burden is greatest, mental health resources are limited. These settings may benefit from psychosocial interventions delivered to people living with HIV/AIDS (PLWH) by non-mental health specialists. We aimed to systematically review randomised controlled trials (RCTs) that investigated the effectiveness of psychosocial interventions delivered by non-mental health specialists to prevent depression in PLWH in LMICs. Methods We used a comprehensive electronic search strategy to identify RCTs of any stage, including pilot studies, which reported on the effectiveness of a psychosocial intervention on depression among adults living with HIV/AIDS in a LMIC setting. Screening, study selection and data extraction was completed independently by two authors. We assessed risk of bias using the Cochrane risk of bias (RoB) tool and performed a narrative synthesis. Results We identified 3431 studies, from which we included 15 studies corresponding to 14 RCTs and a total of 3997 PLWH. Eleven studies were parallel RCTs, one was a stepped-wedged RCT, one was a full factorial RCT, one was a three-arm RCT and four were pilot studies. Studies were generally small, with eight including depression as a primary outcome. All but four trials included men and women and most studies followed participants for less than one year. Twelve trials had at least one domain in which there was a high risk of bias, with the remaining two trials having at least one domain of concern, due to lack of reporting of items. In 12 studies people in the intervention arm had statistically significantly ( P < 0.05) lower or more reduced depressive symptom scores, or were less likely to have major depression, at final follow-up than people in the control group. Conclusions Psychosocial interventions delivered by non-specialist mental health workers may be effective in preventing or reducing depression in PLWH in LMICs. However, existing studies are small with a relatively short follow-up period and have methodological limitations. Future trials should address these shortcomings, establish whether intervention effects are clinically meaningful and investigate cost-effectiveness.