Clinical question: Is the observed drop in COVID-19 mortality caused by changing demographics or improvements in patient care?
Background: At the start of the pandemic, COVID-19 had high mortality rates (6.9% in April according to the U.S. Centers for Disease Control and Prevention). More recently, the mortality rate had decreased to 1.9% of diagnosed cases at the end of September 2020. Concurrently, the median age of confirmed cases dropped as well, from 46 to 38 years, and availability of testing has expanded. It remains unclear whether the drop in mortality rate is because of affected patients with fewer comorbidities, less symptomatic patients, or improvements in clinical care.
Study design: Retrospective cohort study.
Setting: Large urban academic health system.
Synopsis: Researchers analyzed admissions from March 1 through Aug. 31, 2020, to NYU Langone Health System in New York of patients older than 18 years with laboratory-confirmed infection with SARS-CoV-2 during the hospitalization or in the preceding 2 weeks. In total, 5,118 patients qualified for analysis, of which 53% were hospitalized during March and April. Two separate multivariate logistic regression models for mortality were created based on patient demographics, comorbidities, and presenting vital signs and lab-result abnormalities. In the first model, the month of admission was not included, and the standardized mortality ratio (SMR) – the ratio of the sum of observed and expected deaths – for each month was obtained. In the second model, the month of admission was included as a covariate, and the average marginal effect (AME) – the difference in probability of death or discharge to hospice between March and a subsequent time period for equivalent patients – was calculated. The SMR declined progressively over time from 1.26 (95% confidence interval, 1.15-1.39) in March to 0.38 (95% CI, 0.12-0.88) in August. When accounting for demographic and clinical severity changes, the adjusted AME declined every subsequent month after March reaching a maximum of 18.2 (95% CI, 12.0-24.4) percentage point decrease in probability of death in August.
Bottom line: Mortality from SARS-CoV-2 was significantly lower at the end of the 6-month period when adjusted for demographic and clinical factors for patients admitted to a single health care system in the United States.
Citation: Horwitz LI et al. Trends in COVID-19 Risk-Adjusted Mortality Rates. J Hosp Med. 2020 Oct 23. doi: 10.12788/jhm.3552
Dr. Halpern is a med-peds hospitalist at Brigham and Women’s Hospital in Boston.
