Assessing COVID-19’s Impact on ED Visits Among the Elderly
Identifying Predictors of Hospital Admission from the ED Among the Elderly
Healthcare ChallengeThe emergency department (ED) is the primary source for hospital admissions in the United States. Over 70% of hospital admissions among Medicare patients originate from the ED at an extraordinary cost to Medicare. Rates of hospital admission from the ED vary widely for a number of conditions across regions and hospitals and even across physicians within hospitals. However, the amount of variation at each of these levels and the extent to which such variation may correlate across domains of interest is unknown.
A critical need exists to determine the extent to which factors related to the COVID-19 pandemic have altered physician decision-making in the ED and whether these decisions are associated with greater harm for patients or if the post-pandemic admission rates are more optimal.
How we’re meeting it
In a supplement to an ongoing research project, Dartmouth Institute Professor James O’Malley, PhD, in collaboration with researchers at Harvard University, is working to determine whether practice patterns in the ED (primarily the decision to admit) are being impacted by the COVID-19 pandemic, the extent to which these changes depend on the local impact of COVID-19, and to determine whether changes in patterns of admission are associated with worse clinical outcomes for non-COVID-19 related conditions. The specific aims are:
Aim 1: To examine how the COVID-19 pandemic has impacted ED visit rates for specific diagnoses in parts of the country more and less effected by the pandemic.
The team will leverage national data on ED use by traditional Medicare beneficiaries across the country, focusing on rates of presentation for common medical and surgical conditions in areas more and less impacted by COVID-19.
Aim 2: To determine whether rates of admission to the hospital differ before, during, and after COVID-19 surges across areas more or less impacted by COVID-19.
The team will use a quasi-experimental difference-in-difference approach to measure whether the intensity of COVID-19 at the zip code level (measured by fatality rates) is related to the degree of change in rates of admission.
Aim 3: To determine the relationship between changes in rates of admission (from Aim 2) and patient harm, as defined by repeat ED visits, hospitalizations, and deaths within 7 and 30 days.
This study is part of a larger research effort by O’Malley and his colleagues to better understand the factors that contribute to variation in hospital ED admission rates among the elderly and to identify how these factors may help to predict admissions in the future.
In conducting the statistical modeling and analysis, an important problem involving the comparison of provider effects across population strata was encountered that led to O’Malley developing a methods-focused manuscript that details the solution. He is in the process of submitting this work for publication.