BackgroundAmbient air pollution and malnutrition, particularly anemia, are risk factors for pneumonia, a leading cause of death in children under five. We simultaneously assessed these risk factors in Quito, Ecuador.MethodsIn 2005, we studied two socioeconomically similar neighborhoods in Quito: Lucha de los Pobres (LP) and Jaime Roldos (JR). LP had relatively high levels of air pollution (annual median PM2.5 = 20.4 μg/m3; NO2 = 29.5 μg/m3) compared to JR (annual median PM2.5 = 15.3 μg/m3; NO2 = 16.6 μg/m3). We enrolled 408 children from LP (more polluted) and 413 children from JR (less polluted). All subjects were aged 18-42 months. We obtained medical histories of prior physician visits and hospitalizations during the previous year, anthropometric nutrition data, hemoglobin levels, and hemoglobin oxygen saturation via oximetry.ResultsIn anemic children, higher pollution exposure was significantly associated with pneumonia hospitalization (OR = 6.82, 95%CI = 1.45-32.00; P = 0.015). In non-anemic children, no difference in hospitalizations by pollution exposure status was detected (OR = 1.04, NS). Children exposed to higher levels of air pollution had more pneumonia hospitalizations (OR = 3.68, 1.09-12.44; P = 0.036), total respiratory illness (OR = 2.93, 95% CI 1.92-4.47; P < 0.001), stunting (OR = 1.88, 1.36-2.60; P < 0.001) and anemia (OR = 1.45, 1.09-1.93; P = 0.013) compared to children exposed to lower levels of air pollution. Also, children exposed to higher levels of air pollution had significantly lower oxygen saturation (92.2% ± 2.6% vs. 95.8% ± 2.2%; P < 0.0001), consistent with air pollution related dyshemoglobinemia.ConclusionsAmbient air pollution is associated with rates of hospitalization for pneumonia and with physician's consultations for acute respiratory infections. Anemia may interact with air pollution to increase pneumonia hospitalizations. If confirmed in larger studies, improving nutrition-related anemia, as well as decreasing the levels of air pollution in Quito, may reduce pneumonia incidence.

No abstract