My primary research interest is in the development and application of statistical methods for analyzing complex spatial-temporal environmental exposure and health data. Our current methodological research focuses on three specific areas of environmental epidemiology. 
  • Ambient air pollution exposure assessment
  • Health effect estimation leveraging large administrative databases 
  • Health impacts of climate change
I collaborate extensively with researchers from the Department of Environmental Health, the Georgia Institute of Technology, and the US CDC. 

In air pollution epidemiology, population studies routinely utilize air quality measurements from fixed-location outdoor monitoring networks to assign exposures. However, these monitors have limited spatial coverage and are preferentially located in urban areas. Moreover, ambient levels may not reflect human exposure to pollution from outdoor sources since individuals spend the majority of their time indoors. In order to improve the availability and resolution of air pollution data, one approach is to supplement monitoring measurements with additional data products that reflect pollution levels. We are currently developing methods and supporting applications for using three particular data sources: (1) numerical simulations from computer models, (2) remotely-sensed satellite imagery, and (3) stochastic simulations of personal exposures.
  • Chang HH, Hu X, Liu Y (2014).  Calibrating MODIS aerosol optical depth for predicting daily PM2.5 concentrations via statistical downscaling. Journal of Exposure Science and Environmental Epidemiology, 24, 398-404.  [Link]
  • Balachandran S,  Chang HH, Pach JE, Holmes HA, Mulholland JA, Russell AG (2013). Bayesian-based ensemble technique for source apportionment of PM2.5. Environmental Science & Technology. 47, 13511-13518. [Link]
  • Chang HH, Fuentes M, and Frey HC (2012). Time series analysis of personal exposure to ambient PM2.5 and mortality using an exposure simulator. Journal of Exposure Science and Environmental Epidemiology. 22(5), 483-488.  [Link]
Grant Support: EPA R834799 and  NIH R21ES022795.


Health Effect Estimation

Environmental epidemiology has benefited greatly from our ability to leverage health databases developed for administrative or billing purposes. These databases allow researchers to conduct large-scale population-based studies that have played essential roles in setting regulatory standards and protecting public health. Our current work has two focuses. 
  1. Assess the impacts of exposure measurement error in health effect estimation, and to develop statistical methods to account of the error. We are particularly interested in situations where novel air quality estimates are used for health effect estimation. 
  2. Develop statistical methods for estimating associations between adverse birth outcomes and environmental factors. 
  • Chang HH, Warren JL, Darrow LA, Reich BJ, Waller LA. Assessment of critical exposure and outcome windows in time-to-event analysis with application to air pollution and preterm birth. Biostatistics doi: 10.1093/biostatistics/kxu060. [Link]
  • Dionisio KL, Baxter LK, Chang HH (2014). An empirical assessment of exposure measurement error and effect attenuation in bipollutant epidemiologic models. Environmental Health Perspectives, 122, 1216-1224. [Link]
  • Chang HH, Peng RD, and Dominici F (2011). Estimating the acute health effects of coarse particulate matter accounting for exposure measurement error. Biostatistics. 12(4):637-653. [Link]
Grant Support: EPA R834799 and  NIH R21ES022795.


Climate Change

One essential component of climate change science is to quantify future health impact projections that can be attributed to different future emission scenarios. Timely knowledge on the health impacts of climate change can play an important role in maintaining environmental sustainability.  We have been developing statistical methods for incorporating and quantifying various sources of uncertainties in health impact projections.
  • Chang HH, Hao H, Sarnat SE (2014). A statistical modeling framework for projecting future ambient ozone and its health impact due to climate change. Atmospheric Environment, 89, 290-297. [Link]
  • Zhou J, Chang HH, Fuentes M (2012).  Estimating the health impacts of climate change with calibrated model output. Journal of Agricultural, Biological, and Environmental Statistics. 17(3), 377-394. [Link]
  • Chang HH, Zhou J, and Fuentes M (2010). Impact of climate change on ambient ozone level and mortality in Southeastern United States.  International Journal of Environmental Research and Public Health. 7(7):2866-2880. [Link]
Grant Support:  NIH R21ES023763 and NSF 1360330