The Association Between PM <sub>2.5</sub> Exposures and Pharmacy Visits Using Mobile Phone and Points of Interests Data in Jiangsu, China

Abstract

Background: Using mobile phone geolocation data, we aim to investigate the short-term effect of PM2·5 exposures on pharmacy visits, a proxy of multiple health outcomes. Methods: Daily visits to 28170 pharmacies in Jiangsu province, China from May 28, 2018 to January 17, 2019 were extracted from anonymized, geographically, and time-referenced call detail records of 136 thousand mobile phone base station macro-cells and the exact locations identified from pharmacy points of interests (pharmacy POIs). For each pharmacy, we used a generalized additive model to estimate the associations of daily visits to pharmacies with daily PM2·5 concentrations estimated on a 10 km grid using a random forest model. The associations between PM2·5 and pharmacy visits at pharmacy POI level were then pooled using a fixed effects meta-analysis at the level of counties, cities, and provinces. Findings: Pharmacy visits increased by 0·339% (95% Confidence Interval (CI): 0·335, 0·342%) in association with a 10 μg/m3 increase in PM2·5 concentrations. The exposure-response for PM2·5 (lag 05) and pharmacy visits was nonlinear with a sharp slope at less than 60 μg/m3 and then became flat. A significant difference between urban (0·457%, 95% CI: 0·453 to 0·462%) and non-urban areas (0·180%, 95% CI: 0·175 to 0·185%) are observed. We observed strong heterogeneities across cities, counties, and even at sub-community scales. Interpretation: PM2·5 pollution may lead to greater demands for pharmacy resources through initiating multiple adverse pathophysiological changes. Funding Statement: This work was supported by the National Key Research and Development Program of China (Grant 2016YFC0207603), Jiangsu Natural Science Foundation (Grant BK20180350), and National Science Foundation of China (Grant 7143307). Declaration of Interests: JSJ served as the Asia Editor of The Lancet in 2017. All other authors declare no competing interests. Ethics Approval Statement: Not required.