Correlation of COVID-19 Mortality with Clinical Parameters in an Urban and Suburban Nursing Home Population

Kirby, R. S.; Kirby, J. A.

Importance and Objective: COVID-19 has a high mortality rate amongst nursing home populations (26.4% nationally and 28.3% in New Jersey). Identification of factors influencing mortality in COVID-19 positive nursing home populations may help direct physicians towards appropriate glycemic, blood pressure, weight, kidney function, lipid, thyroid, and hematologic management to reduce COVID-19 mortality. Design, Setting, and Participants: Retrospective cross-sectional study of patients in two nursing home facilities (one urban, one suburban) from 3/16/2020 to 7/13/2020 with positive COVID-19 PCR assays. Age, race, sex, lipids, hematologic parameters, body mass index, blood pressure, thyroid function, albumin, blood urea nitrogen, creatinine, and hemoglobin A1c were correlated with COVID-19 mortality by chi-squared analysis. Main Outcome and Results: 56 patients met the inclusion criteria for the study. Mortality was 14.3% while the New Jersey nursing home average mortality rate was 28.3% as of August 2020. Our patient cohort had a 49.5% reduction in mortality compared to the state average. In our overall cohort, none of the clinical parameters correlated with COVID-19 mortality using chi-squared analysis. In the 56 patient cohort, average clinical and laboratory findings were 74.0 years, 62.5% female, 28.5% uncontrolled hypertension, BMI 25.6, hemoglobin A1c 6.4, TSH 2.4, vitamin B12 568.3, folate 12.4, iron 47.8, total iron binding capacity 271.8, hemoglobin 11.6, albumin 3.5, triglycerides 100.3, total cholesterol 133.5, HDL 40.9, and BUN to Creatinine ratio 22.2:1. Logistic multivariate regression analyses failed to demonstrate clinically significant correlation with COVID-19 mortality. In the urban nursing home, BUN to creatinine ratio exceeding 20:1 was the only factor that showed statistical significance to COVID-19 mortality (p = 0.03). In the suburban nursing home, age over 80 was the only clinical factor demonstrating statistical significance to COVID-19 mortality (p = 0.003). Conclusions and Relevance: In our COVID-19 positive nursing home patients, no one parameter was clinically significant in the overall 56-patient cohort; however, mortality in our population was 14.3% compared to New Jerseys 28.3%, a 49.5% reduction in mortality. Rigorous control of the aforementioned clinical parameters may have contributed to this reduction in mortality. Further research requires analysis of more nursing home patients to determine whether rigorous control of clinical parameters decreases mortality from COVID-19.