Recent research implicates antibiotic use as a potential contributor to child obesity risk. In this narrative review, we examine current observational evidence on the relation between antibiotic use in early childhood and subsequent measures of child body mass.

We searched PubMed, Web of Science and the Cochrane Library to identify studies that assessed antibiotic exposure before 3 years of age and subsequent measures of body mass or risk of overweight or obesity in childhood.

We identified 13 studies published before October 2017, based on a total of 6 81 332 individuals, which examined the relation between early life antibiotic exposure and measures of child body mass. Most studies did not appropriately account for confounding by indication for antibiotic use. Overall, we found no consistent and conclusive evidence of associations between early life antibiotic use and later child body mass [minimum overall adjusted odds ratio (aOR) reported: 1.01, 95% confidence interval (95% CI) 0.98–1.04, N = 2 60 556; maximum overall aOR reported: 2.56, 95% CI 1.36–4.79, N = 616], with no clinically meaningful increases in weight reported (maximum increase: 1.50 kg at 15 years of age). Notable methodological differences between studies, including variable measures of association and inclusion of confounders, limited more comprehensive interpretations.

Evidence to date is insufficient to indicate that antibiotic use is an important risk factor for child obesity, or leads to clinically important differences in weight. Further comparable studies using routine clinical data may help clarify this association.

To examine trends in stunting and overweight in Peruvian children, using 2006 WHO Multicentre Growth Reference Study criteria.

Trend analyses using nationally representative cross-sectional surveys from Demographic and Health Surveys (1991–2011). We performed logistic regression analyses of stunting and overweight trends in sociodemographic groups (sex, age, urban–rural residence, region, maternal education and household wealth), adjusted for sampling design effects (strata, clusters and sampling weights).

Peru.

Children aged 0–59 months surveyed in 1991–92 (n 7999), 1996 (n 14 877), 2000 (n 11 754), 2007–08 (n 8232) and 2011 (n 8186).

Child stunting declined (F(1, 5149) = 174·8, P ≤ 0·00) and child overweight was stable in the period 1991–2011 (F(1, 5147) = 0·4, P ≤ 0·54). Over the study period, levels of stunting were highest in rural compared with urban areas, the Andean and Amazon regions compared with the Coast, among children of low-educated mothers and among children living in households in the poorest wealth quintile. The trend in overweight rose among males in coastal areas (F(1, 2250) = 4·779, P ≤ 0·029) and among males in the richest wealth quintile (F(1, 1730) = 5·458, P ≤ 0·020).

The 2011 levels of stunting and overweight were eight times and three and a half times higher, respectively, than the expected levels from the 2006 WHO growth standards. The trend over the study period in stunting declined in most sociodemographic subgroups. The trend in overweight was stable in most sociodemographic subgroups.

The prevalence of child overweight in the Czech Republic is substantially lower than that in the USA. The objective of the present pilot study was to explore dietary intakes, frequency of dining in fast-food establishments, and the amount and intensity of physical activity between a sample of American and Czech children.

A cross-sectional correlational pilot study.

Four public schools in the USA and four public schools in the Czech Republic.

Ninety-five Czech and forty-four American 4–6th graders from urban public schools participated in the study. Dietary intake and number of fast-food visits were evaluated using two multiple-pass 24 h recalls. Physical activity was measured using the modified Self-Administered Physical Activity Checklist.

American children (mean age 10·8 (se 0·2) years) consumed more energy and fat, less fruits and vegetables, more soft drinks, and visited fast-food establishments more often than Czech children (mean age 11·0 (se 0·1) years). Although no differences were found in vigorous activity by nationality, Czech children spent significantly more time in moderate physical activities than American children.

Despite the influx of some negative Western dietary trends into the country, Czech children had a healthier diet and were more physically active than American children. Further research is warranted to determine whether the same differences in dietary intakes, physical activity and fast-food visits exist between nationally representative samples of American and Czech children.