Association of Physical Activity With SARS-CoV-2 Infection and Severe Clinical Outcomes Among Patients in South Korea

IMPORTANCE The association of moderate to vigorous physical activity (MVPA) with COVID-19 outcomes is unclear and needs to be investigated. OBJECTIVE To identify the association of longitudinal changes in MVPA with SARS-CoV-2 infection and severe COVID-19 outcomes. DESIGN, SETTING, AND PARTICIPANTS This nested case-control study used data from 6 396 500 adult patients in South Korean who participated in National Health Insurance Service (NHIS) biennial health screenings from period 1 (2017-2018) to period 2 (2019-2020). Patients were followed from October 8, 2020, until the diagnosis of COVID-19 or December 31, 2021. EXPOSURE Moderate to vigorous physical activity was measured by self-report on questionnaires during both NHIS health screenings and calculated by adding the frequency (times per week) of each moderate (similar to 30 minutes per day) and vigorous (similar to 20 minutes per day) physical activity. MAIN OUTCOMES AND MEASURES The main outcomes were a positive diagnosis of SARS-CoV-2 infection and severe COVID-19 clinical events. Adjusted odds ratio (aORs) and 99% CIs were calculated using multivariable logistic regression analysis. RESULTS A total of 183 350 patients with COVID-19 (mean [SD] age, 51.9 [13.8] years; female, 89 369 [48.7%]; male, 93 981 [51.3%]) among 2 110 268 participants were identified. For participants with vs without COVID-19, the proportion of MVPA frequency at period 2 was 35.8% vs 35.9% for physically inactive, 18.9% vs 18.9% for 1 to 2 times perweek, 17.7% vs 17.7% for 3 to 4 times perweek, and 27.5% vs 27.4% for 5 or more times perweek. Among unvaccinated, physically inactive patients at period 1, the odds for infection increased when engaged in MVPA 1 to 2 times per week (aOR, 1.08; 99% CI, 1.01-1.15), 3 to 4 times per week (aOR, 1.09; 99% CI, 1.03-1.16), or 5 or more times per week (aOR, 1.10; 99% CI, 1.04-1.17) at period 2. Conversely, among unvaccinated patients with MVPA of 5 or more times per week at period 1, the odds for infection decreased when engaged 1 to 2 times perweek (aOR, 0.90; 99% CI, 0.81-0.98) or physically inactive (aOR, 0.80; 99% CI, 0.73-0.87) at period 2. The trend of MVPA and incident infection was mitigated when participants were fully vaccinated. Furthermore, the odds for severe COVID-19 showed significant but limited associations with MVPA. CONCLUSIONS AND RELEVANCE The findings of this nested case-control study show a direct association of MVPA with risk of SARS-CoV-2 infection, which was mitigated after completion of the COVID-19 vaccination primary series. In addition, higher levels of MVPA were associated with a lower risk of severe COVID-19 outcomes to limited proportions.