Incidence and risk of diabetes mellitus associated with depressive symptoms in adults: Evidence from longitudinal studies

AimsWe estimated the incidence and risk of diabetes associated with depressive symptoms using data from longitudinal studies.Materials and methodsDatabases were systematically searched for relevant studies. Incidence of diabetes is presented as cumulative incident proportion (CIP). Pooled effect sizes were calculated using random-effects model. The data were reconstructed to compute relative risk (RR).ResultsThe 16 studies selected for review generated 16 datasets of which 8 studies reporting binary estimates (RR) and 8 studies reporting time-to-event estimates (hazard ratio (HR)). Both RR and HR were significant at 1.67 (95% CI: 1.30–2.15) and 1.45 (95% CI: 1.12–1.87) for incident diabetes associated with depressive symptoms.ConclusionOur observations revealed greater cumulative incidence of diabetes in depressed than in non depressed groups. Depression should be included among risk factors that required regular screening for diabetes.