What are the predictors of inpatient COVID-19 mortality rates?

By Colin Lightfoot, M.Sc. Infection and ImmunityJan 11 2022Reviewed by Danielle Ellis, B.Sc.

Severe acute respiratory syndrome coronavirus 2 (SARS‑CoV‑2) is a severe infection that spreads swiftly within a population and has put hospitals globally under tremendous strain. As of January 2022, more than 839,000 persons infected with SARS‑CoV‑2 had died in the United States alone. While approximately 80% of those infected with the virus display minor symptoms, approximately 20% develop a "hyperinflammatory reaction" and experience severe respiratory distress. Case fatality estimates vary significantly between nations and even within countries.

Although patient characteristics can help explain some variations, hospital characteristics, such as capacity, and county-level characteristics, such as infection rates, are likely to affect COVID-19 mortality rates. Restricted capacity limits hospitals' ability to respond to the "pandemic-associated surge" and contributes to COVID-19 disease mortality. Data from the pandemic's epicenter demonstrate that mortality rates vary even within the same location.

A new study, conducted by a multidisciplinary team of researchers from Suffolk University, Boston, Massachusetts, examined the relationship between baseline staffing levels of registered nurses (RNs), hospitalists, emergency medicine physicians, and intensivists (staff) and hospital-level COVID-19 mortality rates. The research builds on and extends earlier research that examined a subset of hospital-level characteristics over a shorter time period.

A preprint version of this study, which is yet to undergo peer review, is currently available on the medRxiv* server.

Study: Inpatient COVID-19 Mortality Rates: What are the predictors?. Image Credit: CKA/ Shutterstock

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources

The study

According to the regression model's diagnostic plots, the model's assumptions were substantially met. All variance inflation factors (VIFs) were smaller than 3.30, indicating no evidence of multicollinearity. In addition, registered nurses, hospitalists, and emergency medicine physicians all negatively correlate with risk-standardized event rate (RSER), indicating that a higher staffing level for these three kinds is associated with a lower COVID-19 mortality rate.

On the contrary, an increase in intensivist staffing - the provision of specialized care to critically ill patients - is associated with a rise in mortality. In terms of bed capacity, an increase in hospital beds is associated with an increase in mortality, while increases in cardiac intensive care units and skilled nursing care beds are associated with a decrease in mortality.

The baseline utilization of hospitals is positively connected with the COVID-19 mortality rate, indicating that institutions with higher occupancy rates have a poorer patient survival outcome. However, the regression results suggest that there is no statistically significant difference in death rates across the not-for-profit, for-profit, and non-federal public hospitals. Additionally, affiliation with a system is not a significant predictor of the death rate.

Hospitals with one or more certified programs by the Accreditation Council for Graduate Medical Education (ACGME) have a higher mortality rate than hospitals without such programs.

County-level data indicates that metropolitan counties have a higher rate of COVID-19 mortality than rural counties.

The percentage of people living in poverty and the cumulative COVID case rate relate to higher death risk. When other county-level factors are considered, the % of Black/African American and Hispanic/Latino residents are not found to be statistically significant predictors in this data.

Implications

These samples are restricted to patients whose results are recorded in the UnitedHealth Group Clinical Discovery database, which excludes, for example, Medicaid-insured patients. In addition, since the authors did not have access to reliable race data for commercially insured patients, they assessed the hospital-level distribution of patients by race using residential-level data.

Additionally, the authors investigated hospital-level characteristics using baseline data from 2019. Nonetheless, this is the first study to explore the association between RN and physician staffing levels and hospital-level COVID-19 mortality, and it spans an extended time period and a large geographically diversified population.

This data indicates that staffing levels are a significant predictor of patient outcomes. While policymakers have concentrated their efforts on boosting bed capacity, personal protective equipment, and ventilators, disaster preparation plans and future regulations should also consider RN and hospital-based physician staffing levels. A subsequent review will be required to define minimum staffing levels during pandemics.

In light of these benchmarks, policy and plans should be designed to address potential physician shortages during pandemics, based on the impact staffing levels have on patient outcomes.

This news article was a review of a preliminary scientific report that had not undergone peer-review at the time of publication. Since its initial publication, the scientific report has now been peer reviewed and accepted for publication in a Scientific Journal. Links to the preliminary and peer-reviewed reports are available in the Sources section at the bottom of this article. View Sources