Birth weight in relation to health and disease in later life: an umbrella review of systematic reviews and meta-analyses

Background: Birth weight, a marker of the intrauterine environment, has been extensively studied in epidemiological research in relation to subsequent health and disease. Although numerous meta-analyses have been published examining the association between birth weight and subsequent health-related outcomes, the epidemiological credibility of these associations has not been thoroughly assessed. The objective of this study is to map the diverse health outcomes associated with birth weight and evaluate the credibility and presence of biases in the reported associations. Methods: An umbrella review was performed to identify systematic reviews and meta-analyses of observational studies investigating the association between birth weight and subsequent health outcomes and traits. For each association, we estimated the summary effect size by random-effects and fixed-effects models, the 95 % confidence interval, and the 95 % prediction interval. We also assessed the between-study heterogeneity, evidence for small study effects and excess significance bias. We further applied standardized methodological criteria to evaluate the epidemiological credibility of the statistically significant associations. Results: Thirty-nine articles including 78 associations between birth weight and diverse outcomes met the eligibility criteria. A wide range of health outcomes has been studied, ranging from anthropometry and metabolic diseases, cardiovascular diseases and cardiovascular risk factors, various cancers, respiratory diseases and allergies, musculoskeletal traits and perinatal outcomes. Forty-seven of 78 associations presented a nominally significant summary effect and 21 associations remained statistically significant at P < 1 x 10(-6). Thirty associations presented large or very large between study heterogeneity. Evidence for small-study effects and excess significance bias was present in 13 and 16 associations, respectively. One association with low birth weight (increased risk for all-cause mortality), two dose-response associations with birth weight (higher bone mineral concentration in hip and lower risk for mortality from cardiovascular diseases per 1 kg increase in birth weight) and one association with small-for-gestational age infants with normal birth weight (increased risk for childhood stunting) presented convincing evidence. Eleven additional associations had highly suggestive evidence. Conclusions: The range of outcomes convincingly associated with birth weight might be narrower than originally described under the "fetal origin hypothesis" of disease. There is weak evidence that birth weight constitutes an effective public health intervention marker.