In a recent article published in JAMA Network Open, researchers performed a retrospective cohort study among 64,856 Veterans Health Administration (VHA) enrollees aged ≥65 years to describe VHA characteristics and community hospitals delivering care for veteran coronavirus disease 2019 (COVID-19) patients.
Additionally, they compared COVID-19-related deaths and re-admission rates in VHA vs. community hospitals.
Study: Mortality Among US Veterans Admitted to Community vs Veterans Health Administration Hospitals for COVID-19. Image Credit: TylerOlson/Shutterstock.com
Background
In the United States of America (USA), the VHA runs 123 acute care hospitals in rural and urban settings, where elderly patients with severe COVID-19 receive primary care.
However, these VHA hospitals were more than one hour drive for over one-third of VHA enrollees, whereas community hospitals with acute care facilities were relatively near.
Due to poor access to VHA hospitals, community hospitals played a crucial role in caring for veterans with severe COVID-19, especially in rural settings.
To deliver accessible and high-quality care for veterans during future pandemics or COVID-19 surges, VHA needs to understand whether a hospital's geographical location impacted the outcome of primary care among veterans with COVID-19. Also, VHA needs to know the frequency or care outcomes in VHA vs. community hospitals among veterans with severe COVID-19.
About the study
In the present study, researchers combined data from five sources, such as VHA, Medicare, and American Hospital Association (AHA) surveys, to describe the outcomes of COVID-19-related hospitalizations among VHA enrollees aged ≥65 years, stratified by location of their hospital admission.
They covered admissions to VHA and community hospitals, accessed via fee-for-service (FFS) Medicare and Care in the Community (CITC) program.
The primary exposure variable of this study was an indicator of admission to a VHA or community hospital. The team identified all hospital admissions with a COVID-19 diagnosis based on the International Classification of Diseases (ICD)-10-clinical modification (CM) code. They categorized the hospital based on where a patient finally received primary care.
Further, they developed an analytic cohort to compare outcomes in VHA vs. community hospitals. The dependent variable for primary analysis was hospitalization in the VHA hospital. In contrast, the independent variables included patient age, comorbidities, race/ethnicity, gender, residence, social vulnerability index, distance to nearest VHA or community hospital, hospital admission date, and acuity, measured as a need for mechanical ventilation at the time of admission.
In the secondary analyses, the researchers examined outcomes separately for community hospital admissions paid by FFS Medicare and VHA's CITC program.
The two primary study outcomes were mortality within 30 days of hospitalization and re-admission to a hospital in 30 days after getting discharged. They used logistic regression to estimate the propensity for VHA admission rather than a community hospital.
In statistical analyses, the researchers compared the patient characteristics of all patients admitted to VHA or community hospitals. Likewise, they compared hospital characteristics using the hospital as an analysis unit.
The team used two-sided χ2 and rank sum tests for comparing categorical and continuous variables, respectively, and defined significance based on a p-value of less than 0.01.
Results
During the study, the researchers noted 127,156 COVID-19-related hospitalizations of VHA enrollees aged ≥65 years.
The analytic cohort of 64,856 veterans had an average age of 77.6 years. They received care for COVID-19 in 121 VHA hospitals and 4,369 community hospitals. Of 63,562 men enrolled in the VHA, 17,035 (26.3%) sought admissions in VHA hospitals, 36,362 (56.1%) in community hospitals via FFS Medicare, and 11,459 (17.7%) in community hospitals via the CITC program.
Compared with VHA enrollees admitted to VHA hospitals, those admitted to community hospitals were likely White, older, living in rural areas, and less socially vulnerable. The mean age of these people was 78.2 years.
Further, among VHA enrollees admitted to community hospitals via Medicare or the CITC program, the latter was younger, with a mean age of 75.7 years vs. 79 years of the mean age of the former cohort's people.
Most lived in rural settings, hence, were distant from the nearest VHA hospital, with a median distance to a VHA hospital equal to 132 km. With time, hospital admissions via the CITC program became more frequent.
Accordingly, during the last four months of the study, CITC admissions surged to 27.4% (3,135/11,459) compared to Medicare admissions (6,421/36,362, i.e., 17.7%).
Community hospitals were less likely to be in a city than a VHA hospital (108 vs. 2352) and had fewer acute care medical and surgical beds. Nearly 50% of the community hospitals that admitted VHA enrollees with severe COVID-19 were in rural areas, and 24.7% were Critical Access Hospitals.
In the USA, most rural hospitals lack the resources to maintain their financial viability. Thus, they are shutting down at high rates.
Conclusions
The present study showed that veterans experienced markedly higher risk-adjusted mortality in community hospitals than VHA hospitals. In contrast to mortality, re-admission rates were slightly higher after admission to VHA hospitals than community hospitals.
Future studies should evaluate whether this reflected an undesired outcome or was necessary to improve the accessibility of primary care during care transitions.
So that VHA gives proper care to rural veterans enrolled with them, they need to provide more support to rural community hospitals providing acute care overall, especially during surges in demand for care during pandemics.
Mild COVID-19 disrupts brain connectivity and reduces memory function in adolescents and young adults