Using childhood body composition to predict adult disease risk: A systematic review
This is a summary of the following paper: Bander A, Murphy-Alford A, Owino V et al (2022) Childhood BMI and other measures of body composition as a predictor of cardiometabolic non-communicable diseases in adulthood: A systematic review. Public Health Nutrition, 26, 2, 323–350. https://pubmed.ncbi.nlm.nih.gov/36274635/
Non-communicable diseases make up 71% of global deaths and are a huge financial burden for developed and developing countries alike. Childhood malnutrition has recently emerged as a possible risk factor for adult non-communicable disease risk, which could provide a useful marker for estimating the future burden of disease in a population. This systematic review of 29 studies investigated which measures of early life body composition could best predict non-communicable disease risk.
Due to heterogeneity among the included studies – where significant differences were observed in both protocol and outcomes, making it difficult to compare results – the review was presented as a narrative rather than as a quantitative summary. The studies featured a range of sample sizes (between 128 and 34,196 participants), which was a challenge for analysis, although all cohorts were representative of their target populations, which improved the validity of the findings. Most studies (n = 21) were from high-income countries, with five studies from India and only one study each from Guatemala and the Solomon Islands, making it difficult to extrapolate the findings to developing contexts as a whole.
The review featured a robust and systematic identification, screening and selection process, which considered an initial 5,764 studies. However, those included featured a mix of good, adequate and poor-quality studies when considering external and internal validity. It can be concluded that evidence on childhood body composition and later non-communicable disease is severely limited. No studies featured methods such as isotope dilution, plethysmography or dual-energy X-ray absorptiometry, which are more costly than basic anthropometric measures but have greater accuracy and precision. Body mass index (BMI) was the most common proxy measure of body composition used in the studies.
Most studies indicated that childhood BMI is associated with later-life cardiometabolic risk, but that changes in BMI rather than absolute BMI appear to be more important. Several studies showed that infant weight gain (catch-up growth to normal weight-for-age) is protective of non-communicable disease in later life and that low BMI at birth and in infancy are associated with an increased risk of non-communicable disease. Yet, high childhood BMI is also associated with an increased risk of non-communicable disease. The authors do articulate that BMI is a poor measure of adiposity, as it cannot differentiate between fat and fat free mass and is confounded by several factors in children, making it difficult to establish clear links. The lack of adjustment for current body size, which only featured in 11 studies, also impacts the interpretation of these results.
It should be noted that all studies controlled for known confounders to different degrees – which is important, given the broad range of confounding variables that could be present with this type of study – but the studies were observational in nature, which presents its own inherent limitations.