Air Pollutant Exposure and Respiratory Health Among Adolescents in Nairobi, Kenya

Abstract

: Indoor and ambient air pollutants are major modifiable drivers of lung disease in low- and middleincome countries (LMICs). Links between exposure to pollutants and lung disease are reported for multiple outcomes, but remain understudied among adolescents. We hypothesized that indoor and ambient pollutants would be associated with respiratory symptoms and abnormal spirometry among adolescents living with HIV (ALWH) and uninfected adolescents in Nairobi, Kenya, and examined whether these associations differed in ALWH. METHODS: We analyzed data from the BREATHE II study, which includes 343 adolescents (n=168 ALWH) recruited from the Coptic Hope Center for Infectious Diseases and surrounding catchment area in Nairobi. Participants underwent post-bronchodilator spirometry and completed questionnaires that included respiratory symptoms, self-reported indoor fuel use, tobacco smoke exposure, home proximity to highly trafficked roads, and household income. Abnormal spirometry was defined as z-score <-1.64 for FEV1, FVC, or FEV1/FVC (GLI 2012). Logistic regression models adjusted for age, sex, and HIV status determined associations of pollutant exposure with abnormal spirometry and respiratory symptoms. We generated similar models restricted to ALWH. RESULTS: Overall, 53% of participants were male, and median age was 14 years (IQR 12-17; Table). ALWH had well-controlled HIV. The majority (81%) of participants burned fuels indoors; 21% had exposure to secondhand tobacco smoke. Compared to uninfected participants, ALWH had a higher prevalence of prior pneumonia, respiratory symptoms and abnormal spirometry. In adjusted models, the following pollutant exposures were associated with abnormal spirometry: kerosene use (zFEV1/FVC<-1.64 (aOR 5.78 [95%CI 1.45- 23.1], p=0.01)); wood use (zFEV1<-1.64 (3.26 [1.04-10.2], p=0.04)); secondhand smoke exposure (zFEV1<- 1.64 (3.46 [1.27-9.45], p=0.02)). Secondhand smoke exposure was associated with cough (1.79 [1.00-3.20], p=0.05) and activity limitation (3.74 [1.28-11.0], p=0.02). Living <100m from a highly trafficked road was associated with breathlessness (1.79 [1.02-3.17], p=0.04) and chest tightness (1.67 [1.02-2.73], p=0.04). In models restricted to ALWH, associations of kerosene, wood, and secondhand smoke with abnormal spirometry and respiratory symptoms persisted, and although point estimates were generally larger, confidence intervals were wider. CONCLUSION: Air pollutant exposures and respiratory symptoms are highly prevalent among adolescents in Nairobi. Indoor kerosene and wood burning, living with a smoker, and home proximity to heavily trafficked roads were associated with impaired spirometry and respiratory symptoms. Associations persisted in analyses restricted to ALWH. Our findings support that adolescents in LMICs are at increased risk of harm related to unsafe air quality, and ALWH may represent a particularly vulnerable population.