When smoke clouds the mind: Exploring the headache consequences of the 2023 Canadian wildfires

In 2023, Canada experienced one of its worst wildfire seasons to date, with fires raging from British Columbia to Nova Scotia and an estimated 15 million hectares burned.1 Smoke from these wildfires drifted south, leading to poor air quality across the Northeastern United States, a phenomenon that had not occurred previously on this scale on the eastern seaboard. A recent study examined emergency department (ED) visits related to primary headache disorders during periods of increased wildfire smoke in California and identified a positive association between ED visits for tension-type headache and cluster headache; however, no association for migraine was observed.2 Given that weather is one of the most commonly reported triggers of migraine attacks, and there is growing concern that climate change may be causing more severe migraine episodes, we posit that wildfire events such as these may exacerbate migraine and headaches.3 We hypothesized that changes in air quality are associated with increased rates of ED visits for headache during or following peak times of poor air quality. To better understand specifically if changes in air quality index (AQI) during June 2023 impacted the proportion of ED patient visits for headache, we examined ED utilization for a New York City ED, including the weeks of highest AQI (poorest air quality) and comparing this to 1 year prior. We utilized ED data from Montefiore Medical Center in the Bronx, New York, to examine the proportions of headache-related visits to overall ED visits for the weeks of May 28 to June 18, 2023, comparing the same dates 1 year prior (2022). This study was deemed exempt by the institutional review board of the Albert Einstein College of Medicine. ED visit and chief complaint (CC) data were collected, and headache-related visits were identified by ED visit CC from the electronic health record. For this exploratory work, we utilized aggregated data and did not access individual-level health records. For ascertaining environmental air quality during our study period, we utilized measurements collected from New York State Department of Environmental Conservation data that is collected from street-level and rooftop monitors located in the Bronx, New York. This included two commonly used metrics for assessing air quality; airborne particulate matter 2.5 (PM2.5), which represents fine particulate matter that is the result of combustion, and the AQI, which is a measurement of five air pollutants, including ground-level ozone, particle pollution, carbon monoxide, sulfur dioxide, and nitrogen dioxide. US Environmental Protection Agency (EPA) standards for PM2.5 are set at 9.0 micrograms per cubic meter (μg/m3) (annually) or 35 μg/m3 per 24 h. US EPA standards for AQI are AQI 0–50 = good; AQI 51–100 = moderate; AQI 101–150 = unhealthy for sensitive groups; AQI 151–200 = unhealthy; AQI 201–300 = very unhealthy; AQI 301–500 = hazardous. For the statistical analysis, we constructed a Poisson regression model to compare counts of headache-related ED visits to all ED visits between years. The primary predictor was the year (2023 vs. 2022). The primary outcome was the number of headache visits per week. We used the log of total ED visits as an offset to account for variations in overall ED volumes. We calculated the incident rate ratio (IRR) to compare the rate of headache visits between 2023 and 2022, along with 95% confidence intervals (CIs). To ensure the robustness of our models, we assessed for overdispersion. Statistical testing was two tailed and considered significant at p-value <0.05. All analyses were conducted using R version 4.04 (R Core Team). In the 4-week study period in 2023, we identified a total of 1000 patient-level visits for headache among a total of 18,961 ED patient visits. In the 2022 comparison year, we identified a total of 860 visits for headache among a total of 18,578 ED patient visits (Table 1). We found the peak of poor air quality for 2023 occurred between June 4 and June 11, 2023, with AQI >250 and PM2.5 as high as 176 μg/m3 (Figures 1 and 2). For Poisson models comparing the 2023 ED visits to 2022 ED visits for headache (Table 2), we noted an increase in headache visits for the week of June 4 (peak of AQI and PM2.5) with an IRR of 1.19 (95% CI: 0.99–1.43) and for the week of June 11, 2023 (1 week after the peak of high AQI and PM2.5) with an IRR of 1.20 (95% CI: 1.01–1.43). However, this increase was not observed for the week preceding (IRR 1.11, 95% CI: 0.94–1.31) or the week following (IRR 1.04, 95% CI: 0.86–1.26) the peak AQI and PM2.5 levels. For this Poisson analysis, we did not identify overdispersion meeting all model assumptions. In this analysis of the association between ED headache visits and high AQI and PM2.5 secondary to the Canadian wildfires that affected the northeastern United States in 2023, we found an increased IRR for ED visits for headache during the 1 week following the peak compared to 2022. We posit that the effects of air quality triggering headache-related complaints may be delayed, resulting in an increased likelihood of an ED visit following this kind of event. This suggests a possible association between headaches and air quality. Limited recent data exist in this area, but Elser et al. found a positive association between ED visits for tension-type headache and cluster headache.2, 4 In this work, the authors focused on the concentration of wildfire-specific particles measuring less than 2.5 μm/m3 (wildfire PM2.5). Wildfire PM2.5 is suspected to cause headaches because the chemical compositions tend to contain greater levels of oxidative and pro-inflammatory compounds (such as metals and reactive gases).5 The study found that the odds of an ED visit for a primary headache disorder increased by 17% for every 10 μg/m3 increase over a 7-day average concentration but was not significant for migraine headache. There are significant limitations to this research, including the usage of a single medical center, count data, and lack of granular patient-level data and outcomes. Patients were included based on CC rather than specific International Classification of Disease codes, which may introduce a potential for misclassification bias. We did not include other important environmental measures and potential confounders, such as local ambient temperature, humidity, or wind speed. We also acknowledge we did not account for the neighborhood-level variability in exposure across the study area, leading to bias in air quality measurements. Furthermore, patients with milder symptoms may not have presented to the ED; however, we believe this exploratory work provides a framework for future research to explore key demographic and clinical outcomes related to ED visits for headache in the setting of worsening air quality. Weather changes, air pollution, and environmental changes are common triggers for headache and migraine attacks.6 Additionally, natural disasters are suggested to lead to exacerbation of acute headache attacks.7 While our results suggest that the smoke from the 2023 Canadian wildfires was associated with headaches and migraine attacks, further research is warranted. Will we see an increase in migraine attack frequency due to more frequent weather phenomena from climate change? Further analysis of the impact of the 2023 Canadian wildfire smoke is warranted. Fred Cohen: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Brittany Heckel: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Shameeke Taylor: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Ethan Abbott: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Alexis Zebrowski: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—review and editing. Benjamin W. Friedman: Conceptualization; data curation; formal analysis; investigation; methodology; project administration; resources; supervision; validation; visualization; writing—original draft; writing—review and editing. Fred Cohen is an associate editor for Headache. He has no other conflicts of interest to report. Brittany Heckel, Shameeke Taylor, Ethan Abbott, Alexis Zebrowski, and Benjamin W. Friedman declare no conflicts of interest.