Obesity is a prevalent global-health problem associated with substantial morbidity, impairment and economic burden. Because most readily available forms of treatment are ineffective in the long term, it is essential to advance knowledge of obesity prevention by identifying potentially modifiable risk factors. Findings from experimental studies in non-human primates suggest that adverse childhood experiences may influence obesity risk. However, observations from human studies showed heterogeneous results. To address these inconsistencies, we performed Medline, PsycInfo and Embase searches till 1 August 2012 for articles examining the association between childhood maltreatment and obesity. We then conducted a meta-analysis of the identified studies and explored the effects of various possible sources of bias. A meta-analysis of 41 studies (190 285 participants) revealed that childhood maltreatment was associated with elevated risk of developing obesity over the life-course (odds ratio=1.36; 95% confidence interval=1.26-1.47). Results were not explained by publication bias or undue influence of individual studies. Overall, results were not significantly affected by the measures or definitions used for maltreatment or obesity, nor by confounding by childhood or adult socioeconomic status, current smoking, alcohol intake or physical activity. However, the association was not statistically significant in studies of children and adolescents, focusing on emotional neglect, or adjusting for current depression. Furthermore, the association was stronger in samples including more women and whites, but was not influenced by study quality. Child maltreatment is a potentially modifiable risk factor for obesity. Future research should clarify the mechanisms through which child maltreatment affects obesity risk and explore methods to remediate this effect.