Objective: To quantify the association between accelerometer-assessed physical activity and COVID-19 outcomes. Patients and Methods: Data from 82,253 UK Biobank participants with accelerometer data (measured 2013-2015), complete covariate data, and linked COVID-19 data from 16th March 2020 to 16th March 2021 were included. Two outcomes were investigated: severe COVID-19 (positive test from in-hospital setting or COVID-19 as primary cause of death); non-severe COVID-19 (positive test from community setting). Logistic regressions were used to assess associations with moderate-to-vigorous physical activity (MVPA), total activity, and the intensity gradient. A higher intensity gradient indicates a higher proportion of vigorous activity. Results: Average MVPA was 48.1 (32.7) minutes/day. Physical activity was associated with lower odds of severe COVID-19 (adjusted OR per SD increase: MVPA 0.75[95% CI, 0.67,0.85]; total 0.83[0.74,0.92]; intensity 0.77[0.70,0.86]), with stronger associations in women (MVPA 0.63[0.52,0.77]; total 0.76[0.64,0.90]; intensity 0.63[0.53,0.74]) than men (MVPA (0.84[0.73,0.97]; total 0.88[0.77,1.01]; intensity 0.88 [0.77,1.00]). In contrast, when mutually adjusted, total activity was associated with higher odds of a non-severe infection (1.10[1.04,1.16]), while the intensity gradient was associated with lower odds (0.91[0.86,0.97]). Conclusion: Odds of severe-COVID-19 were ∼25% lower per SD (∼30 minutes/day) MVPA. A greater proportion of vigorous activity was associated with lower odds of severe and non-severe infections. The association between total activity and higher odds of a non-severe infection may be through greater community engagement, thus more exposure to the virus. Results support calls for public health messaging highlighting the potential of MVPA for reducing the odds of severe COVID-19.