Portable Air Purifiers to Mitigate the Harms of Wildfire Smoke for People with Asthma

Wildfires are increasing in frequency and intensity across the world. Changing temperatures, drought, and vegetation stemming from anthropogenic climate and land-use change have interacted to boost conditions for their development and spread (1). Wildfires produce harmful emissions and are a significant source of environmental air pollution, including fine particulate matter (i.e., particles <2.5 µm in diameter [PM_2.5]), ozone, and carbon monoxide; as much as one fourth of ambient PM_2.5 in the United States can be attributed to wildfire smoke, a trend that is projected to increase (2).

There is a clear relationship between short-term rise in PM_2.5 and adverse respiratory (and many other) health outcomes (3). This is especially relevant for individuals with asthma, in whom the link between short-term PM_2.5 exposure with worse of symptoms and higher risk of exacerbation is well established, particularly for children (4, 5). For these individuals, air filtration has been proposed as an option to reduce personal exposure. High-efficiency particulate air (HEPA) filters remove 99.97% of particles with a size of 0.3 µm, and, perhaps contrary to popular belief, capture a greater percentage of particles both larger and smaller than this worst-case size. HEPA purifiers can reduce indoor PM_2.5 concentrations by approximately 50–80%, even in countries with relatively high ambient pollution levels, suggesting that they are effective in a wide range of real-world conditions (6). Government programs that encourage the purchase of portable air filters to mitigate poor air quality caused by wildfires are being tried in some jurisdictions, but many barriers exist for their widespread adoption. Chief among them is uncertainty regarding their cost and value, especially when applied as a matter of large-scale public policy.

In this issue of the Journal, Adibi and colleagues (pp. 175–184) take up this question for individuals living with asthma in British Columbia, Canada (7). Using estimated PM_2.5 levels from 2018 to 2022, the authors investigated the cost-effectiveness of a government program that rebates the cost of HEPA purifiers for people with asthma. They used a Markov model to simulate the interaction between PM_2.5 and asthma outcomes and calculated the health differences and resultant changes in cost and quality of life that may be plausibly expected. They found that a $100 rebate would have been likely to be cost-effective to healthcare systems across British Columbia had it had been implemented during this period.

Cost-effectiveness analysis weighs the costs and benefits of an intervention, and its central measure is the incremental cost-effectiveness ratio (ICER). A typical representation of the benefit is the increase in quality-adjusted life-years (QALYs), which is the gained life expectancy discounted by the perceived quality of that life. For general asthma, in which baseline mortality rates are already low, interventions that increase QALYs mainly do so through improved quality of life. The ICER is calculated as the cost of an intervention divided by its benefit, and it helps decision-makers infer whether an intervention is cost-effective based on its relationship to their willingness to pay. The authors used a willingness-to-pay threshold of $50,000 per QALY, which is a typical but debated value when assessing technological interventions in Western economies (8). Although thresholds are useful as yardsticks, decisions of cost-effectiveness are complex and not made based on this metric alone.

The Markov model was well designed. The probabilities of transition between different asthma control states and their relative differences in costs and benefits were logical, derived from appropriate literature, and transparently reported. As models, abstractions and simplifications are always necessary, and one deserves particular attention.

As identified by the authors, the greatest factor that affected the conclusions was the strength of the exposure–response relationship between PM_2.5 and rescue inhaler prescriptions, which the model incorporated as a signal for worse asthma control. Although asthma exacerbations are costly in monetary and quality-of-life terms, they were relatively uncommon, causing the parameters related to nonexacerbated asthma to have a stronger influence. A review of the sensitivity analyses hints that the model’s calculus tips on whether a 10-μg/m³ increase in PM_2.5 corresponds to the empirically determined 4% increase in the risk of worsened asthma. The dimensionality of the exposure–response relationship has been variably reported (for example, some logarithmically transform PM_2.5); if it is nonlinear at the stochastically high concentrations of PM_2.5 expected during wildfire events, an air purifier may have substantially greater or lesser benefit than estimated (9).

Similarly, the program’s cost-effectiveness was diminished to a greater proportion than might be expected in a scenario in which filters were operated during only days with poor air quality, even if these days are when wildfire smoke would be overrepresented. This can at least be partially explained by the model assumption that a reduction in non–wildfire smoke PM_2.5 would also benefit asthma to the same degree as a reduction in wildfire smoke PM_2.5. Wildfire smoke PM_2.5 plausibly has greater toxicity than PM_2.5 from other sources, suggesting that this alternative scenario may be more cost-effective than estimated (10).

This study raises relevant questions regarding the acceptability of this program from the patient perspective. How many of the eligible individuals with asthma would adopt it? Of those who do, would they operate their filters continuously, even during days without wildfire smoke, and bear the costs of replacing the filters to keep them operating optimally as assumed? Although empirical estimates of air filter effectiveness incorporate adherence to some extent, rates of adoption outside of research settings in which participants are not incentivized may be lower. A 2019 survey of Chinese households found that most that had air purifiers did not use them at all, and those that did operated them irregularly or at low settings (11). These are opportunities for future research.

The United Nations Environment Program urges governments to better manage and mitigate the risk of expanding wildfires to human health (12). Making portable air purifiers more accessible to vulnerable populations, including individuals with asthma, could be a cost-effective tool in this armamentarium. By creating a well-researched cost-effectiveness model that may be localized to other jurisdictions, the work of Adibi and colleagues has provided valuable information to advance this policy conversation worldwide.