The convergence of the COVID-19 pandemic and widespread wildfires in the Western United States during the latter half of 2020 presented a unique opportunity to study the potential impact of air pollution on the spread of SARS-CoV-2. Wildfires generate significant amounts of PM2.5, a known respiratory irritant and immune system modulator. Prior research suggested a possible link between air pollution and increased COVID-19 severity, with PM2.5 potentially increasing susceptibility to respiratory viruses and enhancing the survival and spread of viral bioaerosols. Washoe County, Nevada, experienced both a surge in COVID-19 cases and substantial exposure to wildfire smoke. This study aimed to determine if there was an association between elevated PM2.5 levels from wildfires and increased SARS-CoV-2 infection rates in Reno, Nevada, a city situated in an intermountain valley that limits pollutant dispersion, thereby potentially increasing exposure for its residents. The study utilized data from Renown Health, a major healthcare system in Washoe County, which accounted for approximately 25% of the county's laboratory-confirmed COVID-19 cases. The researchers hypothesized that elevated levels of PM2.5 from wildfire smoke would be associated with a higher SARS-CoV-2 test positivity rate in Reno.

Existing literature established the detrimental effects of air pollution, particularly PM2.5, on respiratory health. Studies demonstrated that PM2.5 can modify immune responses, causing inflammation and increasing susceptibility to respiratory viruses. The potential for PM2.5 to facilitate the spread and survival of various bioaerosols, including those containing SARS-CoV-2, has also been documented. Some prior studies had indicated a correlation between particulate matter and COVID-19 mortality, but further investigation was needed to establish a definitive causal link. Previous research had also shown the detrimental effects of wildfire smoke on respiratory health, and researchers had predicted that the concurrence of wildfire smoke events with the COVID-19 pandemic could significantly worsen the pandemic’s impact. This study builds upon this existing research by focusing specifically on the effect of wildfire PM2.5 on SARS-CoV-2 infection rates in Reno, Nevada, controlling for factors not included in previous research.

The study employed a time-series analysis using generalized additive models (GAMs) with a negative binomial distribution (to account for overdispersion in the count data). The data encompassed a 159-day period from May 15th to October 20th, 2020. Daily PM2.5 concentrations were obtained from four air quality monitors in Reno and Sparks, Nevada, using a weighted average based on the proximity of Renown Health patients to each monitor. Missing PM2.5 data were supplemented using data from the Washoe County air quality management division. Daily temperature and humidity data were obtained from a nearby weather station. SARS-CoV-2 test results (NAA tests) and patient demographic data were provided by Renown Health. Only a patient's initial positive test was included in the daily positive case count, and for calculating the daily number of tests administered, only one negative test was included per patient per day. The researchers selected May 15th, 2020 as the start of the study period because testing seemed to have stabilized by that point. The model included the count of positive COVID-19 cases as the response variable, with the total number of tests administered included as an offset, thus modelling the positivity rate rather than the raw number of positive tests. The model controlled for several potential confounders, including 7-day average temperature (to account for seasonal changes), the previous day’s positive case count (to account for autocorrelation), and day of the week effects. A smooth function of time was included to account for unmeasured confounding factors, such as the overall prevalence of the virus and changes in human behavior. The researchers explored the relationship between the positivity rate and PM2.5 using various lags (0, 7, and 14 days for daily mean and 7-day average PM2.5, and additional lags for daily mean PM2.5 and 3-day average PM2.5). They refined their analysis using distributed lag models (DLMs) to estimate lag effects constrained to follow a quadratic or cubic function, ensuring realistic lag associations. To estimate excess cases resulting from wildfire smoke, they used the model with the 7-day average PM2.5 as a predictor, generating fitted values for the period of August 16th to October 10th (the time of heaviest wildfire smoke impact), comparing them to fitted values assuming 2019 average PM2.5 levels for that period. A resampling approach was used to calculate confidence intervals for the estimated excess cases. Sensitivity analyses were conducted to assess the robustness of the findings by modifying model parameters such as the degrees of freedom for the time smooth, adding relative humidity as a predictor, and changing the temperature averaging period. The statistical analyses were conducted using R version 3.6.0 with packages mgcv, mgcv.helper, and ggmap.

The study found a statistically significant positive association between PM2.5 and the SARS-CoV-2 test positivity rate in Reno, Nevada. Specifically, a 10 µg/m³ increase in the 7-day average PM2.5 concentration was associated with a 6.3% relative increase in the positivity rate (95% CI: 2.5% to 10.3%). Analyses exploring different PM2.5 lags revealed that the strongest association occurred with PM2.5 concentrations 2–6 days prior to the observed increase in positivity rate. Distributed lag models, incorporating quadratic and cubic functions to model the lag effects, yielded similar results indicating that intermediate lags had the largest positive associations. The researchers estimated that the wildfire smoke exposure accounted for an additional 178 (CI: 149-98) positive COVID-19 cases at Renown Health between August 16th and October 10th, representing a 17.7% (CI: 14.4%-20.1%) increase in cases during that period. Sensitivity analyses, which involved altering several model components (i.e. degrees of freedom for the smoothing term, inclusion of humidity as a predictor, and the timeframe for temperature averaging), did not substantially alter these key results. The demographic data showed that the rate of positive tests was highest among patients aged 18-29 (11.3%), followed by the 30-49 age group (9.6%), and the 50-69 age group (7.4%). The rate of positive tests among those aged less than 18 years old and those aged 70 years old and above were considerably lower at 6.0% and 4.9%, respectively. Importantly, the overall positivity rate for all patients during the study period was 8.0%.

The study's findings strongly suggest that exposure to wildfire smoke, specifically elevated PM2.5 levels, contributed substantially to the increased number of COVID-19 cases in Reno, Nevada during the 2020 wildfire season. This supports previous research indicating that air pollution can exacerbate the pandemic. The observed lag of 2-6 days between PM2.5 exposure and increased infection rates suggests a short-term, cumulative effect, potentially due to cellular changes induced by PM2.5, increased viral infectivity, or PM2.5 particles acting as vectors of spread. The study controlled for several factors not included in previous studies, such as overall virus prevalence, temperature, and the number of tests administered, strengthening the evidence for an association between wildfire smoke and COVID-19 incidence. The results align with other research showing associations between air pollution and higher COVID-19 incidence and mortality rates in various regions of the world. The possibility of non-biological factors is also considered, including how air quality monitoring applications and public policy decisions might affect human behavior and thus the spread of SARS-CoV-2 during periods of poor air quality. While the study focused solely on PM2.5, the authors acknowledged the potential for other wildfire smoke components to contribute to observed associations. The observed increased case numbers due to wildfire smoke are assumed to have resulted in excess mortality. This warrants further investigation. The implications of the study extend beyond Reno, highlighting the need for preparedness strategies to manage the risks of both wildfires and pandemics concurrently.

This study provides robust evidence linking wildfire smoke (PM2.5) exposure to increased SARS-CoV-2 infection rates. The significant increase in COVID-19 cases attributable to wildfire smoke underscores the importance of considering air quality when planning pandemic preparedness and response efforts. Future research should examine the precise mechanisms by which PM2.5 exposure increases SARS-CoV-2 infection risk and evaluate the impact of other wildfire smoke components. Furthermore, studies that explore strategies to mitigate the combined risks of wildfires and pandemics, including the development of effective public health interventions, are vital.

The study's reliance on observational data inherently limits its ability to definitively establish causality. While the study controlled for several factors, unmeasured confounders could still exist. The method used for estimating PM2.5 exposure represented population-level exposure and did not account for individual variations in exposure due to factors such as occupation, lifestyle, or socioeconomic status. The focus on Renown Health data limits the generalizability of findings to the broader Reno population and potentially other populations with different demographics or healthcare seeking behaviors. The study concentrated on PM2.5, leaving out the potential influence of other pollutants present in wildfire smoke. Future research can address these limitations through the utilization of more sophisticated exposure assessment methods and a broader study population.