**Background:** Biomass fuel (wood, charcoal, animal dung, crop residues) is a major source of household air pollution, exposing approximately 2.8 billion people globally, with the highest burden in Africa and Southeast Asia where over 60% of households cook with biomass. Prenatal exposure to biomass fuel increases the risk of adverse birth outcomes, including low birth weight. However, few studies in sub-Saharan Africa have examined the association between biomass fuel and birth weight as a continuous variable specifically among term births, which isolates the effect from preterm birth. Nigeria is among the five countries where most preterm and small-for-gestational-age infants are born globally, and biomass cooking fuel is commonly used nationwide.
**Methods:** The primary analysis used data from the Child Electronic Growth Monitoring System (CEGROMS), a cross-sectional study conducted at Barau Dikko Teaching Hospital, Kaduna, northwestern Nigeria. The sample included 1,514 mother-child pairs with singleton term births recruited from May 2019 to March 2021. Exposure was household cooking fuel type categorized as liquefied petroleum gas (LPG), kerosene, or biomass fuel (charcoal, wood, crops/straw, animal dung). Birth weight was measured using digital scales. Linear regression was used to estimate covariate-adjusted associations between birth weight and biomass fuel exposure, adjusting for maternal age, education, parity, BMI at birth, and child sex. Replication analysis used data from 6,975 mother-child pairs from the 2018 Nigerian Demographic and Health Survey (DHS), a nationally representative multistage stratified survey. In the DHS, cooking fuel was categorized as low pollution fuel (electricity, LPG, natural gas), kerosene, or biomass fuel. Linear mixed models with random intercepts for cluster and mother ID were used to account for clustering.
**Key Results:** In CEGROMS, 78.9% of mothers used LPG, 8.9% used kerosene, and 12.3% used biomass fuel. Mothers using biomass fuel were less educated, had higher parity, and lower BMI compared to LPG users. After adjustment, infants of mothers exposed to biomass fuel were on average 113g lighter (95% CI −196 to −29) than those using LPG. No appreciable difference was observed between kerosene and LPG users (−36.6g, 95% CI −121.1 to 48.0). The association between biomass fuel and lower birth weight increased with maternal BMI categories (P_interaction = 0.04): normal weight −33.9g (95% CI −139.3 to 71.6), overweight −138.5g (95% CI −271.7 to −5.4), obese −309g (95% CI −504.5 to −113.6). In the DHS replication, 23.2% used low pollution fuel, 20.2% used kerosene, and 56.6% used biomass fuel. After adjustment for community, household, and individual-level factors, mothers using biomass had infants weighing 50g less at birth (95% CI −103 to 2) compared to low pollution fuel users. Kerosene users showed no significant difference (−14.4g, 95% CI −66.6 to 37.7). In the DHS, the association was significant only among male infants (−70.7g, 95% CI −121.2 to −20.3; P_interaction = 0.04), though this sex interaction was not observed in the northwest region where Kaduna is located (P_interaction = 0.57).
**Clinical Implications:** This study provides evidence that biomass fuel exposure during pregnancy is associated with reduced birth weight among term infants in Nigeria, with a larger effect observed in the Kaduna study (113g reduction) compared to the national sample (50g reduction). The findings suggest that regional disparities in birth outcomes in Nigeria may be partly attributable to differences in cooking fuel use. The authors recommend that pregnant women be asked about cooking fuel use during antenatal care and be provided resources to minimize biomass exposure. They note that intervention trials in low- and middle-income countries have shown that transitioning from biomass to cleaner fuels reduces household PM2.5 and CO and improves birth outcomes. Policy strategies including stove subsidies, fuel subsidies, fuel bans, and behavior change communication are suggested to reduce population-level exposure. The study's strengths include its large sample size, restriction to term births to isolate effects from preterm birth, and replication using a nationally representative dataset with multilevel modeling. Limitations include postnatal assessment of exposure (though recall bias is considered negligible), imprecision in gestational age estimation from maternal self-report, potential bias from home birth weight measurements, and unmeasured confounding from hypertension in pregnancy and malaria.