RESUMO
AIM: To update a systematic review and meta-analysis of the association of breastfeeding with overweight or obesity that had been commissioned by the World Health Organization. We also assessed the likelihood of residual confounding. METHODS: Two independent reviewers searched MEDLINE, LILACS and Web of Science for manuscripts published between August 2014 and May 2021. Studies that only evaluated infants were excluded. Random-effects models were used to pool the estimates. RESULTS: The review comprised 159 studies with 169 estimates on the association of breastfeeding with overweight or obesity, and most of the studies were carried out among individuals aged 1-9 years (n = 130). Breastfeeding protected against overweight or obesity (pooled odds ratio:0.73, 95% confidence interval:0.71; 0.76). And, even among the 19 studies that were less susceptible to publication bias, residual confounding and misclassification, a benefit was observed (pooled odds ratio:0.85, 95% confidence interval:0.77; 0.93). Among those studies that were clearly susceptible to positive confounding by socioeconomic status, a benefit of breastfeeding was observed even after adjusting for socioeconomic status (pooled odds ratio:0.76, 95% confidence interval: 0.69; 0.83). CONCLUSION: Breastfeeding reduced the odds of overweight or obesity, and this association was unlikely to be due to publication bias and residual confounding.