PROJECT SUMMARY/ABSTRACT
The novel coronavirus disease 2019 (COVID-19) pandemic is currently the most emergent public health
disaster of the entire world. COVID-19 disproportionately affects older adults, people having history of smoking
and comorbidities like diabetes, obesity, and hypertension. There is no vaccine and treatment for COVID-19,
therefore, public health interventions have been taken to control disease transmission. Also, searching
modifiable environmental factors that can help reduce the COVID-19 severity and mortality is crucial. Air
pollution exposure has been shown to have a systemic effect on the human body including lung function
impairment and immune alterations. Recent two ecologic studies from Europe and the United States have
suggested that higher long-term ambient air pollution exposure (PM2.5 and NO2) significantly contributes to the
COVID-19 mortality. These findings from aggregated data of air pollution exposure and total number of deaths
in large geographic areas need to be further verified by cohort study with individual data of air pollution
exposure and COVID-19 case progression and mortality. Also, susceptibility factors such as low social
economic status (SES), race/ethnicity, smoking exposures and comorbidities need to be accounted for in the
analysis. To address the urgent public health question about the role of air pollution exposure in COVID-19
progression, we propose to conduct a retrospective cohort study based on the existing EMR data of all COVID-
19 cases (n>7000) diagnosed at Kaiser Permanente Southern California (KPSC) medical centers. Specific
aims of this proposal have been expanded from our ongoing NIEHS-supported R01 study investigating
prenatal air pollution exposure and children's autism risk among 440,000 KPSC mother-child pairs (APAR
study, 1R01ES029963). In this proposal, the main outcomes of interest are COVID-19 severity assessed by
hospitalization, ICU admission and ventilator use, as well as death. Details dates will be available through
EMR. Natural language processing technologies will be applied to identify the date of first COVID-19 symptom
onset for each case. Thereafter, the earliest date among COVID-19 symptom onset and clinical diagnosis will
be used as the study entry date. Then short- and long-term air pollution exposure will be estimated for each
case by averaging air pollution exposure levels during one-month and one-year before the study entry date.
Both ambient (PM2.5, PM10, NO2 and O3) and traffic-related (line dispersion model estimated NOx) air pollution
exposure will be assessed based on individual residential addresses. Detailed residential history of all KPSC
members have been well-maintained by the KPSC system. Key covariates including age, sex, race/ethnicity,
smoking exposure, body mass index, comorbidities, medication use, and meteorological data will be extracted
to control for confounding and identify susceptible high risk subgroups. With the unique cohort data resource
and world-renowned epidemiologists and exposure scientists, this one-year study will greatly enhance our
knowledge about the effect of air pollution exposure on COVID-19 progression and death.
