Air Pollution and Migraine: Refining Estimates through Innovative Analysis, Exposure and Outcome Measurement

Background: Migraine affects over 1 billion people worldwide, including over 1 million Ontarians. While avoiding triggers is an important migraine prevention strategy, surprisingly little is known about potential environmental triggers. Air pollution is a possible a trigger, but most studies have used blunt exposure estimates, single-pollutant models, and/or used emergency room records as outcomes, all of which might bias the exposure-outcome relationship. Objectives: This thesis’s objectives are: (1) to describe the state of knowledge of associations between air pollution and migraine; (2) to examine the influence of age, sex, and jurisdiction on the probability of self-reported emergency room visits for migraine, and the attack recording habits of smartphone app users; and (3) estimate the association between multiple air pollutants and migraine attack onset using novel data sources and analytical methods. Methods: To achieve these objectives I (1) systematically reviewed previous evidence. I (2) designed and conducted an online survey of 389 people living with migraine in Canada and the USA and collected information on demographic and geographic measures and on emergency room visits during a migraine attack. Finally, (3) using a case time series distributed-lag non-linear models, I estimated the association of ambient nitrogen dioxide (NO2), ozone (O3), and particulate matter 2.5 µm or smaller in diameter (PM2.5) with migraine attacks recorded using an established smartphone app, Migraine Buddy. Results: (1) My systematic review showed that for associations between migraine and carbon monoxide (CO), nitrogen dioxide, ozone, and particulate matter, effect estimates were mainly positive. (2) My survey results demonstrated that young people and people in Canada were more likely to attend ER for a migraine attack compared to those living in the USA, suggesting that ER information is not a random subsample of migraine events. The study also suggested that people with migraine may record attacks in a more consistent temporal manner compared creation of attack records via ER visits for migraine. (3) In multiple pollutant models, individuals were more likely to report migraine attacks on days with relatively higher NO2 and O3. In contrast to most previous studies, there was no association between days with higher PM2.5 and migraine attacks. Conclusions: Overall, this work both describes the state of knowledge of air pollution and migraine, and builds upon it via a targeted survey, novel exposure and outcome measures, and statistical methods. The results of the empirical papers, in particular paper three, provide an improved understanding of how ambient pollution impacts migraine. These results can inform individuals, clinicians, and policy makers regarding how ongoing levels of air pollution, and associated mitigation efforts, impact the risk of migraine at the population level.

Department dc:contributor.department

Dalla Lana School of Public Health

Year dc:date.issued

2026

Author dc:creator

• Portt, Andrea Elizabeth

Advisor dc:contributor.advisor

• Smith, Peter M

• air pollution
• emergency room
• measurement
• migraine
• smartphone app data
• systematic review

Statement dc:rights

• Attribution 4.0 International

Licence dc:rights.uri

• http://creativecommons.org/licenses/by/4.0/

Handle dc:identifier.uri

https://hdl.handle.net/1807/153503

OAI record oai:identifier

https://utoronto.scholaris.ca/server/oai/request?verb=GetRecord&metadataPrefix=dim&identifier=oai%3Autoronto.scholaris.ca%3A1807%2F153503