Abstract/Summary

Although maltreatment experiences in childhood increase the risk for depression, not all maltreated children become depressed. This review aims to systematically examine the existing literature to identify modifiable factors that increase vulnerability to, or act as a buffer against, depression, and could therefore inform the development of targeted interventions. Thirteen databases (including Medline, PsychINFO, SCOPUS) were searched (between 1984 and 2014) for prospective, longitudinal studies published in English that included at least 300 participants and assessed associations between childhood maltreatment and later depression. The study quality was assessed using an adapted Newcastle-Ottawa Scale checklist. Meta-analyses (random effects models) were performed on combined data to estimate the effect size of the association between maltreatment and depression. Meta-regressions were used to explore effects of study size and quality. We identified 22 eligible articles (N=12 210 participants), of which 6 examined potential modifiable predictors of depression following maltreatment. No more than two studies examined the same modifiable predictor; therefore, it was not possible to examine combined effects of modifiable predictors with meta-regression. It is thus difficult to draw firm conclusions from this study, but initial findings indicate that interpersonal relationships, cognitive vulnerabilities and behavioral difficulties may be modifiable predictors of depression following maltreatment. There is a lack of well-designed, prospective studies on modifiable predictors of depression following maltreatment. A small amount of initial research suggests that modifiable predictors of depression may be specific to maltreatment subtypes and gender. Corroboration and further investigation of causal mechanisms is required to identify novel targets for intervention, and to inform guidelines for the effective treatment of maltreated children.