Does Wildfire Smoke-Induced Air Pollution Impact Health Outcomes Among Older Adults?

This study investigates how acute air pollution from wildfire smoke impacts health outcomes in older adults. Using Medicare records from 2007–2019, the research examines daily wildfire smoke exposure, associated PM2.5 levels, and subsequent mortality and ER visits among U.S. adults aged 65 and older. Findings reveal that wildfire smoke is responsible for 18% of PM2.5 levels, 0.42% of deaths, and 0.69% of ER visits in this population. Smaller pollution shocks have outsized health effects, suggesting substantial benefits from air quality improvements even in areas meeting regulatory standards. The study underscores the nonlinear, concave relationship between pollution and health.

Citation

Miller, Nolan H., Molitor, David, & Zou, Eric. The Nonlinear Effects of Air Pollution on Health: Evidence from Wildfire Smoke. NBER Working Paper No. 32924, September 2024. Available at: http://www.nber.org/papers/w32924​.

Extended Summary

Central Research Question

The study primarily investigates the health effects of air pollution stemming from wildfire smoke, specifically exploring how varying levels of pollution impact mortality and emergency room (ER) visits among older adults in the U.S. The researchers focus on understanding the nonlinear, or concave, relationship between pollution exposure and health, aiming to determine if smaller pollution increases have proportionally larger health effects than larger pollution spikes. This research has significant implications for public health policies and air quality standards, especially as wildfire smoke becomes a more pervasive issue.

Previous Literature

Previous studies have recognized the link between air pollution and adverse health outcomes, with regulatory standards often reflecting an assumption that health risks rise more steeply at higher pollution levels. Most literature assumes a convex relationship between pollution and health, where higher pollution levels result in disproportionately large health impacts. Regulatory frameworks like the U.S. Clean Air Act are based on this assumption, suggesting that areas with lower pollution levels need less regulation. However, empirical evidence on the shape of this relationship—particularly whether it is indeed convex or could be concave—is scarce. Some recent studies hint at the potential concavity in pollution-health relationships, but few have thoroughly examined this using wildfire smoke, which provides a unique, variable pollution source for study.

Data

The study utilizes a robust dataset comprising administrative Medicare records for older adults (aged 65 and over) across the United States from 2007 to 2019. The data includes daily health outcomes such as mortality and ER visits, alongside wildfire smoke exposure and ambient pollution levels. Wildfire smoke data is sourced from the National Oceanic and Atmospheric Administration’s Hazard Mapping System, which uses satellite imagery to track smoke plumes' density and proximity to counties. PM2.5 (particulate matter less than 2.5 micrometers in diameter) levels—widely regarded as the most harmful component of wildfire smoke for human health—were collected from the Environmental Protection Agency’s Air Quality System. This large, spatially granular dataset allows the researchers to trace the health impacts of varying intensities of smoke exposure across different regions and days.

Methods

To address the study’s research question, the authors employ an event-study methodology and a concentration–response (C–R) analysis. This method capitalizes on the natural variation in wildfire smoke exposure as it drifts across counties, enabling a quasi-experimental design that estimates the causal impact of smoke-induced PM2.5 on health outcomes. Specifically, they track health outcomes on days with varying smoke exposure levels, using county-level fixed effects and controls for day-specific factors to isolate the causal effects of smoke. The event-study portion looks at the impact of a single smoke exposure event on PM2.5 levels, mortality, and ER visits up to 20 days before and after the event. The concentration–response analysis further examines how health impacts scale with different smoke shock intensities, creating a visual and quantitative representation of the relationship between PM2.5 levels and health outcomes. This approach reveals whether health impacts rise steeply at low levels of pollution or level off with increased exposure, challenging traditional convex assumptions.

Findings/Size Effects

The analysis finds that wildfire smoke is responsible for about 18% of ambient PM2.5 levels in the U.S., contributing significantly to mortality and ER visits among older adults. The study estimates that wildfire smoke exposure accounts for 0.42% of deaths and 0.69% of ER visits among this population. When scaled to the U.S. population aged 65 and over, this translates to approximately 10,070 premature deaths and 191,541 excess ER visits annually. These findings underscore that even moderate increases in PM2.5 from wildfire smoke can have substantial health impacts. One key finding is the nonlinear, concave relationship between smoke-induced PM2.5 levels and health outcomes: smaller pollution shocks have a disproportionately high health impact per unit of PM2.5, whereas larger shocks yield diminishing marginal effects. For example, counties directly beneath a smoke plume experience a sharp rise in PM2.5 levels and associated health impacts, which gradually diminish as exposure levels rise beyond a certain threshold (around 6 µg/m³ of PM2.5). This pattern suggests that reducing smaller, more frequent pollution shocks could yield substantial health benefits, even in regions that meet current air quality standards.

Conclusion

The study challenges prevailing assumptions about the pollution-health relationship, indicating that air quality regulations might need to shift focus to account for significant health impacts even at relatively low pollution levels. The findings suggest that additional health benefits can be achieved by tightening air quality standards, especially for vulnerable populations such as older adults. The concave relationship implies that moderate improvements in air quality could produce substantial public health gains, underscoring the need for adaptive policies as wildfire smoke events become more frequent and intense. By illustrating that smaller pollution increases can have disproportionately large health effects, the study advocates for re-evaluating regulatory approaches and potentially lowering allowable PM2.5 thresholds to protect public health more effectively. Future research could expand on these findings by exploring long-term health impacts and adaptation behaviors that may mitigate health risks from frequent pollution shocks.