We systematically reviewed quantitative research that compared developmental health or wellbeing outcomes for maltreated children placed in out‐of‐home care (OoHC) with those cared for in their home. Randomised, quasi‐randomised and cohort studies that met detailed criteria were eligible. Study results and bias risk information were extracted by two authors using predefined data fields. Narrative synthesis was used because of the diversity of studies. No randomised studies were found. Thirty‐one papers from 11 cohort studies were included. The studies covered 15 developmental health or wellbeing outcomes (63 distinct comparisons). Of 40 significance‐tested comparisons, 29 were consistent with no evidence of benefit or harm of OoHC, seven were consistent with harm and four with benefit. Three studies with low risk of selection bias showed no evidence of significant differences, or found worse outcomes for OoHC. Overall, evidence from cohort studies shows limited evidence of improved outcomes, and some evidence of worse outcomes associated with OoHC. These results may be explained by selection bias favouring placement of children likely to have worse outcomes in OoHC rather than in‐home care. Further research, including randomised controlled trials and carefully matched data‐sets, is needed to determine the effectiveness of OoHC. Copyright © 2016 John Wiley & Sons, Ltd. ‘Studies covered 15 developmental health or wellbeing outcomes’ Key Practitioner Messages Most research has high bias risk, as children who enter OoHC have more disadvantaged backgrounds and higher risk for poor outcomes than other maltreated children. Even allowing for significant bias risk in the studies, there are mixed results for children entering OoHC, and the three highest-quality studies showed no difference or worse outcomes for OoHC.