Without a doubt, preparedness for future disasters is primary, but images of deserted streets and highways prompt another take. The air pollution regulators’ dream has been achieved: complete elimination of a significant source of air pollution (traffic) throughout the nation, however briefly.
What else has happened? We’re indoors 24/7; does the outdoor air quality even matter now? While indoors, we’re constantly plugged in; are powerplants working harder? Some factories have shut down, but a few may have increased production.
The epidemiologist in me proposes an intervention study in which exposures to air pollution was essentially instantly changed while everything else stayed the same. Here, “everything else” includes weather, season, smoking habits, diet, proximity to major outdoor pollution sources. Subsequent changes in health status could then confidently be attributed to the difference in air quality if confounding variables were adequately controlled, among them emotional stress level, loss of income, and perhaps exercise and alcohol consumption.
Of course, that if is pretty significant. The underlying assumption is that COVID-19 deaths can be accurately identified and studied separately. The outcome variables would be numbers of other deaths, hospital admissions, or emergency room visits, adjusted for age, race, and gender. The “experiment” would comprise the sheltering period and similar or longer previous and subsequent periods. These data should be analyzed from several large cities, say 50, in various locations here or abroad. These research opportunities are unique, and (hopefully) will not happen again.
This intervention study could compare health effects from three different sources of air pollution: indoors such as smoking, cooking, cleaning; outdoors from local sources like traffic, home heating, small business; outdoors from large regional point sources like power plants, refineries, cement plants. Note that all of these sources emit the tiny particles known as PM2.5 that are of current concern, albeit that they may differ in terms of chemical composition by source.
Here are some outcome scenarios and interpretations for individuals unaffected by COVID-19.
|
Scenarios |
Possible Explanations |
|
No change in health status
|
People were insensitive to local outdoor or indoor air pollution. The benefits of cleaner air are long-term and not reflected in the study period. Cleaner outdoor air did not infiltrate indoors. |
|
Health status improved but only during the period of cleaner outdoor air |
Benefits of infiltrated cleaner air are short-term, primarily affecting previously frail individuals. The health effects of increased exposure to indoor air pollutants are transient. |
|
Health status got worse during the cleaner air period. |
Health effects of increased exposure to indoor air pollutants are harmful, and people were insensitive to traffic-related air pollution. |
|
Health status improved after the clean air period |
The benefits of cleaner outdoor air are long-term and delayed. People were insensitive to any late effects of indoor pollutants. |
Supporting these interpretations, we have prior information that outdoor air does indeed infiltrate indoors except for specially designed buildings [1]. We are reasonably sure that exposure to traffic-related air pollutants is associated with human health effects but are unsure about time scales (short- vs. long-term). We know very little about the health effects of pollutants generated indoors for any time scale. These experiments could tell us whether being confined indoors is more important than having cleaner air outdoors and over what time scales. Such information could radically transform air pollution regulatory control policy here and abroad.
It is reasonable to assume that the results of these experiments would differ among various cities; an additional objective is thus to ascertain the reasons, such as differences in climate, topography, population characteristics, strength, and proximity to large regional point sources. In any event, the more cities studied, the more powerful the statistical analysis. A significant advantage of this proposal is that only public data sources would be involved and that no new measurements would be required.
By contrast, the New York Times reported on a forthcoming paper [2] asserting that improving outdoor air quality (as measured by fine particles) by 10% over 20-y would reduce the COVID-19 death rate by 15%. If this were the case, the improvement in outdoor air quality since 1980 would have reduced mortality by a factor of 5, and having a smoker in the house would increase it by a factor of 16. Their analysis was based on differences among US counties and not on changes over time; it thus refers to the places where the COVID-19 victims lived and not to their personal exposures. Perhaps most importantly, the analysis did not control for race, and recent data show that African-Americans are about twice as likely to die from COVID-19 as Caucasians. The message here is that location does not infer causation.
While none of the above could relieve the pain and suffering of the thousands of unfortunate COVID-19 victims, it presents a unique opportunity to extract useful knowledge for everyone’s long-term benefit.
[1] Lipfert FW. An assessment of air pollution exposure information for health studies. Atmosphere DOI: 10.3390/atmos6111736
[2] Wu X, Nethery RC, Sabath MB et al. Exposure to air pollution and COVID-19 mortality in the United States. medRxiv Preprint
