Weiran Yuchi1, Enkhjargal Gombojav2, Buyantushig Boldbaatar2, Jargalsaikhan Galsuren2, Sarangerel Enkhmaa3, Bolor Beejin4, Gerel Naidan2, Chimedsuren Ochir2, Bayarkhuu Legtseg5, Tsogtbaatar Byambaa6, Prabjit Barn1, Sarah B Henderson7, Craig R Janes8, Bruce P Lanphear1, Lawrence C McCandless1, Tim K Takaro1, Scott A Venners1, Glenys M Webster1, Ryan W Allen9. 1. Faculty of Health Sciences, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada. 2. School of Public Health, Mongolian National University of Medical Sciences, Zorig Street, Ulaanbaatar, 14210, Mongolia. 3. Institute of Meteorology and Environmental Monitoring, Ministry of Environment of Mongolia, Mongolia. 4. Mongolian National Center for Public Health, Olympic Street 2, Ulaanbaatar, Mongolia. 5. Sukhbaatar District Health Center, 11 Horoo, Tsagdaagiin Gudamj, Sukhbaatar District, Ulaanbaatar, Mongolia. 6. Ministry of Health of Mongolia, Olympic Street-2, Government Building VIII, Sukhbaatar District, Ulaanbaatar, Mongolia. 7. Environmental Health Services, British Columbia Centre for Disease Control, 655 W. 12th Ave, Vancouver, BC, V5T 4R4, Canada. 8. School of Public Health and Health Systems, University of Waterloo, 200 University Avenue West, Waterloo, ON, N2L 3G1, Canada. 9. Faculty of Health Sciences, Simon Fraser University, 8888 University Drive, Burnaby, BC, V5A 1S6, Canada. Electronic address: allenr@sfu.ca.
Abstract
BACKGROUND: Indoor and outdoor fine particulate matter (PM2.5) are both leading risk factors for death and disease, but making indoor measurements is often infeasible for large study populations. METHODS: We developed models to predict indoor PM2.5 concentrations for pregnant women who were part of a randomized controlled trial of portable air cleaners in Ulaanbaatar, Mongolia. We used multiple linear regression (MLR) and random forest regression (RFR) to model indoor PM2.5 concentrations with 447 independent 7-day PM2.5 measurements and 87 potential predictor variables obtained from outdoor monitoring data, questionnaires, home assessments, and geographic data sets. We also developed blended models that combined the MLR and RFR approaches. All models were evaluated in a 10-fold cross-validation. RESULTS: The predictors in the MLR model were season, outdoor PM2.5 concentration, the number of air cleaners deployed, and the density of gers (traditional felt-lined yurts) surrounding the apartments. MLR and RFR had similar performance in cross-validation (R2 = 50.2%, R2 = 48.9% respectively). The blended MLR model that included RFR predictions had the best performance (cross validation R2 = 81.5%). Intervention status alone explained only 6.0% of the variation in indoor PM2.5 concentrations. CONCLUSIONS: We predicted a moderate amount of variation in indoor PM2.5 concentrations using easily obtained predictor variables and the models explained substantially more variation than intervention status alone. While RFR shows promise for modelling indoor concentrations, our results highlight the importance of out-of-sample validation when evaluating model performance. We also demonstrate the improved performance of blended MLR/RFR models in predicting indoor air pollution.
BACKGROUND: Indoor and outdoor fine particulate matter (PM2.5) are both leading risk factors for death and disease, but making indoor measurements is often infeasible for large study populations. METHODS: We developed models to predict indoor PM2.5 concentrations for pregnant women who were part of a randomized controlled trial of portable air cleaners in Ulaanbaatar, Mongolia. We used multiple linear regression (MLR) and random forest regression (RFR) to model indoor PM2.5 concentrations with 447 independent 7-day PM2.5 measurements and 87 potential predictor variables obtained from outdoor monitoring data, questionnaires, home assessments, and geographic data sets. We also developed blended models that combined the MLR and RFR approaches. All models were evaluated in a 10-fold cross-validation. RESULTS: The predictors in the MLR model were season, outdoor PM2.5 concentration, the number of air cleaners deployed, and the density of gers (traditional felt-lined yurts) surrounding the apartments. MLR and RFR had similar performance in cross-validation (R2 = 50.2%, R2 = 48.9% respectively). The blended MLR model that included RFR predictions had the best performance (cross validation R2 = 81.5%). Intervention status alone explained only 6.0% of the variation in indoor PM2.5 concentrations. CONCLUSIONS: We predicted a moderate amount of variation in indoor PM2.5 concentrations using easily obtained predictor variables and the models explained substantially more variation than intervention status alone. While RFR shows promise for modelling indoor concentrations, our results highlight the importance of out-of-sample validation when evaluating model performance. We also demonstrate the improved performance of blended MLR/RFR models in predicting indoor air pollution.
Authors: Undarmaa Enkhbat; Enkhjargal Gombojav; Chimeglkham Banzrai; Sarangerel Batsukh; Buyantushig Boldbaatar; Enkhtuul Enkhtuya; Chimedsuren Ochir; David C Bellinger; Bruce P Lanphear; Lawrence C McCandless; Ryan W Allen Journal: Environ Health Date: 2021-07-05 Impact factor: 5.984