Articles www.thelancet.com Vol 380 October 13, 2012 1325 Lancet 2012; 380: 1325–30 Published Online August 9, 2012 http://dx.doi.org/10.1016/ S0140-6736(12)60869-1 See Comment page 1292 Faculty of Epidemiology and Population Health, London School of Hygiene and Tropical Medicine, London, UK (J A Cresswell PhD, Prof O M R Campbell PhD, M J De Silva PhD, V Filippi PhD) Correspondence to: Dr Jenny Cresswell, London School of Hygiene and Tropical Medicine, London WC1E 7HT, UK jenny.cresswell@lshtm.ac.uk Effect of maternal obesity on neonatal death in sub-Saharan Africa: multivariable analysis of 27 national datasets Jenny A Cresswell , Oona M R Campbell, Mary J De Silva, Véronique Filippi Summary Background Rates of obesity are increasing worldwide, including in sub-Saharan Africa. Neonates born to obese mothers in low-income settings are at increased risk of complications including admission to neonatal intensive care, macrosomia, low Apgar scores, and perinatal death. We investigated whether maternal obesity is a risk factor for neonatal death in sub-Saharan Africa and the effect on the detailed timing of death within the neonatal period. Methods Cross-sectional Demographic and Health Surveys from 27 sub-Saharan countries (2003–09) were pooled. We used multivariable logistic regression to assess the risk of neonatal death (in women’s most recent singleton livebirth in the 5 years preceding the survey) by maternal body-mass index (BMI) category (measured during the survey). Timing of death was investigated with a discrete-time survival model. Findings 15 518 of 81 126 eligible women were overweight (4266 were obese), 52 006 had an optimum BMI, and 13 602 were underweight. Maternal obesity was associated with an increased odds of neonatal death after adjustment for confounding factors (adjusted odds ratio 1·46, 95% CI 1·11–1·91). Maternal obesity was a significant risk factor for neonatal deaths occurring during the first 2 days of life (1·62, 1·11–2·37). We noted no statistically significant relation later in the neonatal period (days 2–6 1·36, 0·84–2·21; days 7–27 1·19, 0·65–2·18), possibly because of low statistical power. Interpretation Maternal obesity in sub-Saharan Africa is associated with increased risk of early neonatal death. Potential mechanisms include prematurity, intrapartum events, or infections. Strategies to prevent and reduce obesity need to be considered; obese women should be advised to deliver in a health-care facility that can provide emergency obstetric and neonatal care. Funding Economic and Social Research Council. Introduction For many people, low-income countries are associated with images of starvation and hunger, yet obesity is a rapidly growing problem. 1–3 Ongoing global shifts in diet towards the consumption of high-fat and high-sugar foods, coupled with urbanisation and an increasingly sedentary lifestyle for many populations, are generating an upwards shift in body-mass index (BMI). In most low-income and middle-income countries, more women are overweight or obese (BMI≥25 kg/m²) than are under- weight (BMI<18·5 kg/m²). 4 If secular trends continue, by 2030 there will be 185·8 million overweight or obese adults living in sub-Saharan Africa (28·7%), and 17·5% of adults will be obese. 5 Neonates born to obese mothers are at increased risk of complications including admission to neonatal intensive care, 6 macrosomia, 7 low Apgar scores, 8 and perinatal death. 9 We know of no previous studies that have investigated the effect of maternal obesity on neonatal mortality in low-income countries, where the burden of neonatal mortality is greatest. 10 In such settings, the facilities available to manage high-risk births, such as the complications that commonly arise in babies of obese mothers, are often of poor quality or non-existent. As such, many complications that are treatable in a well equipped unit can often result in the death of the infant. Much of the improvement in neonatal mortality reported so far has been achieved by reduction of neonatal tetanus, rather than prevention of intrapartum-related deaths. 11 Our study aimed to investigate whether maternal obesity is a risk factor for neonatal death within the sub-Saharan context, and to investigate the association between maternal obesity and the timing of neonatal death. Methods Dataset The Demographic and Health Surveys (DHS) are nationally representative cross-sectional household sur- veys that use a standardised questionnaire to facilitate between-country comparisons. DHS have been shown to be of high data quality, particularly for low-income settings. 12 To generate a large dataset with sufficient statis- tical power to investigate the effect of maternal obesity on the timing of neonatal death, we pooled data from the most recent survey in 27 countries in sub-Saharan Africa since 2000. The response rate was more than 90%. The outcome of the most recent livebirth for each woman within the 5 years preceding the survey was considered in the analysis (index birth). Multiple births were excluded. The DHS collect detailed birth histories on livebirths for each woman; respondents were asked about the month and year of each birth, any multiple gestations, each