Background: During the early days of the COVID-19 global pandemic, nursing homes were identified as ‘hot spots’ for cases and, sadly, deaths of thousands of highly vulnerable older adults in the U.S. Our understanding has evolved, however, what hasn’t been investigated are influential factors (facility and county-level) in specific regions of the country and across different time periods.

Methods: This study focuses on the southeastern U.S. (N=2,948 nursing homes, 823 counties). Data from Centers for Medicare and Medicaid Services for nursing home facilities (i.e., COVID-19 cases, facility ratings, staffing) and 2010 Census/American Community Survey for county-level metrics (i.e., population demographics) were utilized. Two different time periods (May 24, 2020-Sept. 26, 2020 and Sept. 27, 2020-Feb. 6, 2021) were used to identify significant predictors of COVID-19 in nursing homes. Significant predictors of COVID-19 in nursing homes were obtained using stepwise regression. Final regression results were multi-level, accounting for county-level and nursing home-level differences.

Results: In the early period, COVID-19 county-level rates (RR=1.27) were the highest predictor of COVID-19 cases in nursing homes. However, in the later period, changes in nursing home ownership (RR=1.39) became the highest predictor of COVID-19 cases in nursing homes. Significant variables changed across time periods, highlighting the different phases of the pandemic among nursing home populations in the southeastern U.S.

Conclusions: Nursing home administrators and local public health officials should work collaboratively to address the complex issues involved with a global pandemic to better protect and promote the health and safety of their vulnerable older adult populations.