Key Takeaways
- In a study of 792,935 Brazilian births, children whose mothers gained less weight than recommended during pregnancy had an 18% higher adjusted hazard of dying before age 5. The adjusted hazard ratio was 1.18 (95% confidence interval, 1.12 to 1.26).
- In raw terms the gap is small: 0.89% of children died in the low-gain group versus 0.69% with recommended gain.
- The link was steepest in women who began pregnancy underweight (adjusted hazard ratio 1.47; 95% confidence interval, 1.10 to 1.96), the smallest group in the study.
- The study is observational. It cannot show that gaining more weight would prevent deaths.
- Earlier US research tied inadequate gain to infant death, and a large meta-analysis tied it to preterm birth and small newborns.
Easy to Measure, Hard to Interpret
A pregnant woman discusses her health with a doctor. Research involving nearly 793,000 births in Brazil links insufficient maternal weight gain during pregnancy to higher childhood mortality.
Weight gain in pregnancy is simple to track and surprisingly hard to read. A Brazilian study of 792,935 births ties a low number to child survival. Whether the weight gain itself matters, or merely signals trouble elsewhere, is the part the data cannot answer.
The study by Santiago-Vieira et al. (2026), published November 20 in JAMA Network Open, drew on the 100 Million Brazilians Cohort. That database links records of low-income families with birth, nutrition-surveillance and death registries. By the authors’ account, the connection between weight gain and child mortality had not been systematically examined in Brazil.
The analysis covers singleton live births from 2008 through 2020 to mothers aged 20 to 49, counting each mother’s first child once. Weight gain meant a third-trimester weight minus the mother’s weight before pregnancy. That starting weight was self-reported, and a weight taken in the first 12 weeks stood in when it was missing. Each woman was then sorted into below, within or above the 2009 US Institute of Medicine range for her body mass index (BMI) before pregnancy. The sorting corrected for how far along she was at the last weigh-in.
Children were followed from birth until death, the fifth birthday or the end of 2020. The models adjusted for maternal age, education, marital status, prenatal visits, prepregnancy BMI and birth year.
Nine in 1,000, Seven in 1,000
By the Institute of Medicine’s ranges, 327,954 mothers (41.4%) gained too little, 231,844 (29.2%) gained the recommended amount and 233,137 (29.4%) gained more. In all, 6,105 children died before age 5.
Among children of mothers who gained too little, 2,928 died, or 0.89%. With recommended gain the rate was 0.69% (1,605 deaths), and with high gain 0.67% (1,572 deaths). Put plainly, about nine in 1,000 against about seven in 1,000.
The headline number is a relative one. After adjustment, the hazard of death before age 5 was 18% higher in the low-gain group (adjusted hazard ratio 1.18; 95% confidence interval, 1.12 to 1.26). A hazard ratio compares how quickly deaths occur in two groups, so a child’s chance of dying did not rise by 18 percentage points. The unadjusted gap is about 0.2 percentage points, or roughly two extra deaths per 1,000 children.
Most deaths came early. Of the 6,105, a total of 5,323 happened in infancy, and 3,299 of those in the first 28 days. Neonatal deaths carried a 21% higher adjusted hazard (1.21; 95% CI, 1.11 to 1.32), and deaths between 28 and 364 days a 15% higher one (1.15; 1.04 to 1.28). Perinatal conditions caused 39.0% of all deaths and congenital malformations 30.2%.
Risk climbed as prepregnancy weight fell
Women who began pregnancy underweight had a 47% higher adjusted hazard when they gained too little (1.47; 95% CI, 1.10 to 1.96). The figure was 1.25 for normal-weight women (1.14 to 1.40) and 1.10 for women with a BMI of 25 or higher (1.01 to 1.21).
That clean gradient comes with a caveat. Only 30,464 women were underweight, which explains the wide interval.
Gaining more than recommended showed no consistent pattern. One borderline result favored high gain for deaths between 28 and 364 days (0.86; 95% CI, 0.76 to 0.98). Everywhere else, high gain bore no relation to mortality.
The Estimate Shrank Under One Test
The authors stress-tested the main result, and the direction held every time. The size did not. Dropping deaths from external causes left the estimate at 1.19 (95% CI, 1.11 to 1.26). Using Brazilian weight-gain charts instead of the US ranges cut it to 1.09 (1.02 to 1.19).
The third test matters most. Among the 116,884 mothers with a recorded prepregnancy weight, the estimate was 1.15 (0.98 to 1.36) and no longer statistically significant. That subgroup is about 15% of the cohort, and its point estimate sits close to the main result. In other words, the interval widened more than the estimate moved.
Other limits are plain. Prepregnancy weight was self-reported or approximated, so some weight-gain values are imprecise. Most families in the cohort are low-income, which may limit how far the findings travel. Women needed weight records at two points, which favored those with more prenatal visits.
A Cause, or Just the Messenger?
Weight gain may work like a dashboard warning light. It signals that something is off without saying what.
Illness can change the reading. Women with hyperemesis gravidarum, severe nausea and vomiting, may lose weight early and gain less later. Weight patterns also differ in women who go on to develop preeclampsia. Either condition can harm a child independently of weight gain.
Hunger is another candidate. Food insecurity during pregnancy is tied to lower weight gain, and micronutrient deficiency and limited dietary variety are common among women of reproductive age in low- and middle-income countries. In a cohort of low-income families, a poor diet could drive both the weight gain and the outcome.
Birth weight is the biggest open question. The researchers chose not to adjust for it because low birth weight may sit on the path between inadequate gain and death, and adjusting could hide a real effect. How much of the association runs through birth weight is unknown. The authors call for formal mediation analyses. Because the study is observational, maternal illness, food insecurity and other factors that adjustment cannot fully remove may account for part of the gap.
Where Other Research Agrees
The US data lean the same way. The Brazilian authors cite an analysis of 8,656,791 American births by Chen and Chauhan (2019). In their summary, women who gained less than recommended had 14% higher neonatal mortality and 12% higher infant mortality.
Davis et al. (2014) in the American Journal of Public Health surveyed 159,244 women with singleton live births in 41 states between 2004 and 2008. Only 34% gained the recommended amount. Underweight women with inadequate gain had sharply higher odds of infant death (odds ratio 6.18; 95% CI, 2.45 to 15.56). So did normal-weight women (1.47; 1.08 to 2.01) and overweight women (2.11; 1.30 to 3.42). Obese women showed no such link (1.01; 0.63 to 1.64).
That survey adjusted for birth weight and gestational age, which the Brazilian team did not, and it relied on mothers’ answers collected about four months after birth. Its strongest effect, in underweight women, mirrors the Brazilian pattern. Odds ratios and hazard ratios measure different things, so the numbers do not line up directly.
Goldstein et al. (2017) in JAMA supplies the birth-outcome evidence. It pooled 23 cohort studies covering 1,309,136 pregnancies. Women who gained too little had higher odds of a baby born small for gestational age (odds ratio 1.53; 95% CI, 1.44 to 1.64). Their odds of preterm birth were higher too (1.70; 1.32 to 2.20), though only four studies, with widely varying results, fed that estimate.
The review also shows that no weight-gain range is free of cost. Gaining too little went with lower odds of large-for-gestational-age babies (0.59) and macrosomia (0.60). Gaining too much went with more large babies (1.85) and more cesarean deliveries (1.30).
Birth outcomes are not deaths, so this review cannot explain the Brazilian results. It does point to a plausible route, since preterm birth and low birth weight are among the leading causes of infant death. What Santiago-Vieira’s team adds is follow-up to age 5 in a low-income population, with 6,105 deaths to analyze.
What It Means in the Prenatal Clinic
Santiago-Vieira and colleagues suggest that tracking weight gain during prenatal care may help flag women who need nutritional or clinical support. They also say policies should address food security, maternal nutrition and access to prenatal care, with particular attention to underweight women.
Nothing here shows that pushing weight gain upward would save lives, and gain above guidelines carries its own risks in other research. Pregnant women can ask their prenatal care team what range applies to their own prepregnancy weight.
A slow-gaining woman who started out underweight is the patient this study says deserves a closer look.
Funding and Disclosures
The Wellcome Trust supported Enny S. Paixão. A Brazilian Ministry of Health science and technology department supported Rita de Cássia Ribeiro-Silva. The study also used resources from the Centre for Data and Knowledge Integration for Health, which is funded by the Bill & Melinda Gates Foundation, the Wellcome Trust, the Health Surveillance Secretariat of the Ministry of Health and the Secretariat of Science and Technology of the State of Bahia.
Thais Rangel Bousquet Carrilho reported grants from the Canadian Institutes of Health Research and Michael Smith Health Research BC during the study. Paixão reported Wellcome Trust grants outside the submitted work. No other disclosures were reported.
The funders had no role in the study’s design, data collection, management, analysis or interpretation, in preparing, reviewing or approving the manuscript, or in the decision to submit it.
References
Chen, H. Y., & Chauhan, S. P. (2019). Association between gestational weight gain adequacy and adverse maternal and neonatal outcomes. American Journal of Perinatology, 36(6), 615–623. https://doi.org/10.1055/s-0038-1672196
Davis, R. R., Hofferth, S. L., & Shenassa, E. D. (2014). Gestational weight gain and risk of infant death in the United States. American Journal of Public Health, 104(S1), S90–S95. https://doi.org/10.2105/AJPH.2013.301425
Goldstein, R. F., Abell, S. K., Ranasinha, S., Misso, M., Boyle, J. A., Black, M. H., Li, N., Hu, G., Corrado, F., Rode, L., Kim, Y. J., Haugen, M., Song, W. O., Kim, M. H., Bogaerts, A., Devlieger, R., Chung, J. H., & Teede, H. J. (2017). Association of gestational weight gain with maternal and infant outcomes: A systematic review and meta-analysis. JAMA, 317(21), 2207–2225. https://doi.org/10.1001/jama.2017.3635
Santiago-Vieira, C., de Farias Costa, P. R., Carrilho, T. R. B., Cerqueira-Silva, T., Paixão, E. S., Fiaccone, R. L., Li, L., Tedde, J. G. G., Barreto, M. L., & Ribeiro-Silva, R. de C. (2026). Gestational weight gain and mortality among children in Brazil. JAMA Network Open, 9(11), e2645176. https://doi.org/10.1001/jamanetworkopen.2026.45176




