Air pollution from wildfire smoke
For my PhD thesis, I proposed a methodological framework for environmental public health surveillance, and applied it to wildfire smoke surveillance in British Columbia.
Wildfire smoke is a surprisingly large contributor to poor air quality in a lot of the world. The acute air quality that any Canadian will experience is likely to be caused by wildfires, and the smoke can travel far from the origin. The WHO estimates that landscape fires account for approximately 339,000 deaths per year globally1. Air pollution in general, including wildfire smoke, is known to be harmful in either acute or chronic exposure. For asthmatics and others with lung problems, breathing poor quality air is much more dangerous.
Environmental public health surveillance is a newer area of research and practice as compared to infectious disease surveillance, which is very established. The goals are similar - any public health surveillance system uses available data to monitor population health, provide information and context for decision-making. In an infectious disease context, a big part of a surveillance system is often detection of outbreaks. In an environmental health context, the goals are similar but the situation is slightly different. There needs to be regular monitoring of both the exposure (e.g., fine matter air particulates) as well as a potential health response from the population under surveillance (e.g., are people having more asthma attacks?).
The exposure is challenging (maybe impossible) to measure perfectly, and there are many health impacts that could be monitored. Healthcare utilization data is most commonly used, such as data on physician visits, emergency department visits, and hospitalizations. In my thesis I looked at two health outcomes: asthma-related primary care physician visits, and asthma medication dispensations. Each of these datasets are unusually rich in the province of British Columbia.
My partner on this research was the BC Centre for Disease Control, where my co-supervisor Dr. Sarah Henderson is a senior environmental health scientist. You can read more about my doctoral research in my first manuscript which has been published2; the later two are still in the publication process. The BCCDC is implementing the methods I developed in my dissertation to be used for wildfire smoke surveillance starting summer 2017. On my supervisory committee were Dr. David Buckeridge and Dr. Gavin Shaddick. Gavin is a statistician at the University of Bath who specializes in Bayesian spatio-temporal modelling and environmental statistics. He was the primary supervisor for my first manuscript.
1 Johnston FH, Henderson SB, Chen Y, Randerson JT, Marlier M, DeFries RS, Kinney P, Bowman DM, Brauer M. 2012. Estimated global mortality attributable to smoke from landscape fires. Environmental Health Perspectives. 120(5): 695-701.
2 Morrison K, Shaddick G, Henderson SB, Buckeridge DL. 2016. A latent process model for forecasting multiple time series in environmental public health surveillance. Statistics in Medicine, 35(18): 3085–3100.