RESUMO
BACKGROUND: Anemia is common in low- and middle-income countries (LMICs), causing significant health issues and social burdens. Exposure to household air pollution from using biomass fuels for cooking and heating has been associated with anemia, but the exposure-response association has not been studied. OBJECTIVES: We evaluated the associations between personal exposure to air pollution and both hemoglobin levels and anemia prevalence among pregnant women in a multi-country randomized controlled trial. METHODS: We studied 3,163 pregnant women aged 18-35 years with 9-20 weeks of gestation, recruited as part of the Household Air Pollution Intervention Network (HAPIN) randomized controlled trial in Guatemala, India, Peru, and Rwanda. We assessed 24-hour personal exposures to fine particulate matter (PM2.5), black carbon (BC), and carbon monoxide (CO), and measured hemoglobin levels at baseline (15 ± 3 weeks gestation). Linear and logistic regression models were used to examine the associations of measured pollutants with hemoglobin levels and anemia prevalence, adjusting for confounding. RESULTS: Single-pollutant models showed associations of CO with higher hemoglobin levels and lower anemia prevalence. Bipollutant models involving CO and PM2.5 also revealed that an interquartile range (IQR) increase in CO concentrations (2.26 ppm) was associated with higher hemoglobin levels [ß = 0.04; 95 % confidence interval (CI): 0.01, 0.07], and a lower odds of anemia prevalence [odds ratios (OR) = 0.90; 95 % CI: 0.83, 0.98]. PM2.5 was inversely related to hemoglobin and positively associated with anemia, but results were not statistically significant at the 0.05 alpha level. County-specific results showed that 3 of 4 countries showed a similar association between CO and hemoglobin. We found no association of BC levels with hemoglobin levels or with anemia prevalence. CONCLUSION: Our findings suggest that exposure to CO is associated with higher hemoglobin and lower anemia prevalence among pregnant women, whereas PM2.5 showed the opposite associations.