Myocardial Infarction across COVID-19 Pandemic Phases: Insights from the Veterans Health Affairs System (preprint)

Background Cardiovascular procedural treatments were deferred at scale during the COVID-19 pandemic, with unclear impact on patients presenting with Non-ST Elevation Myocardial Infarction (NSTEMI). Methods In a retrospective cohort study of all patients diagnosed with NSTEMI in the U.S. Veterans Affairs Healthcare System from 1/1/19 to 10/30/22 (n=67,125), procedural treatments and outcomes were compared between the pre-pandemic period and six unique pandemic phases (1: Acute phase, 2: Community spread, 3: First Peak, 4: Post-Vaccine, 5. Second Peak, 6. Recovery). Multivariable regression analysis was performed to assess association between pandemic phases and 30-day mortality. Results NSTEMI volumes dropped significantly with the pandemic onset (62.7% of pre-pandemic peak) and did not revert to pre-pandemic levels in subsequent phases, even after vaccine availability. Percutaneous coronary intervention (PCI) and/or coronary artery bypass grafting (CABG) volumes declined proportionally. Compared to the pre-pandemic period, NSTEMI patients experienced higher 30-day mortality during Phase 2 and 3, even after adjustment for COVID-19 positive status, demographics, baseline comorbidities, and receipt of procedural treatment (adjusted OR for Phase 2-3 combined: 1.26 [95% CI 1.13-1.43], p<0.01). Patients receiving VA-paid community care had a higher adjusted risk of 30-day mortality compared to those at VA hospitals across all six pandemic phases. Conclusions Higher mortality after NSTEMI occurred during the initial spread and first peak of the pandemic, but resolved before the second, higher peak - suggesting effective adaptation of care delivery but a costly delay to implementation. Investigation into the vulnerabilities of the early pandemic spread are vital to informing future resource-constrained practices.

Excess mortality in the general population versus Veterans Healthcare System during the first year of the COVID-19 pandemic in the United States (preprint)

Importance The COVID-19 pandemic had a substantial impact on the overall rate of death in the United States during the first year. It is unclear whether access to comprehensive medical care, such as through the VA healthcare system, altered death rates compared to the US population. Objective: Quantify the increase in death rates during the first year of the COVID-19 pandemic in the general US population and among individuals who receive comprehensive medical care through the Department of Veterans Affairs (VA). Design: Analysis of changes in all-cause death rates by quarter, stratified by age, sex race/ethnicity, and region, based on individual-level data. Hierarchical regression models were fit in a Bayesian setting. Standardized rates were used for comparison between populations. Setting and participants: General population of the United States, enrollees in the VA, and active users of VA healthcare. Exposure and main outcome: Changes in rates of death from any cause during the COVID-19 pandemic in 2020 compared to previous years. Results Sharp increases were apparent across all of the adult age groups (25 years and older) in both the general US population and the VA populations. Across all of 2020, the relative increase in death rates was similar in the general US population (RR: 1.20 (95% CI: 1.17, 1.22)), VA enrollees (RR: 1.20 (95% CI: 1.14, 1.29)), and VA active users (RR: 1.19 (95% CI: 1.14, 1.26)). Because the pre-pandemic standardized mortality rates were higher in the VA populations prior to the pandemic, the absolute rates of excess mortality were higher in the VA populations. Conclusions and Relevance: Despite access to comprehensive medical care, active users of the VA had similar relative mortality increases from all causes compared with the general US population. Factors that influenced baseline rates of death and that mitigated viral transmission in the community are more likely to have influenced the impact of the pandemic.

The COVID-19 pandemic shifted the Veterans Affairs System toward being a payer and virtual care provider: is it sustainable? (preprint)

Objective: To examine how VA shifted care delivery methods one year into the pandemic. Study Setting: All encounters paid or provided by VA between January 1, 2019 and February 27, 2021. Study Design: We aggregated all VA paid or provided encounters and classified them into community (non-VA) acute and non-acute visits, VA acute and non-acute visits, and VA virtual visits. We then compared the number of encounters by week over time to pre-pandemic levels. Data Extraction Methods: Aggregation of administrative VA claims and health records. Principal Findings: VA has experienced a dramatic and persistent shift to providing virtual care and purchasing care from non-VA providers. Before the pandemic, a majority (63%) of VA care was provided in-person at a VA facility. One year into the pandemic, in-person care at VA's constituted just 33% of all visits. Most of the difference made up by large expansions of virtual care; total VA provided visits (in person and virtual) declined (4.9 million to 4.2 million) while total visits of all types declined only 3.5%. Community provided visits exceeded prepandemic levels (2.3 million to 2.9 million, +26%). Conclusion: Unlike private health care, VA has resumed in-person care slowly at its own facilities, and more rapidly in purchased care with different financial incentives a likely driver. The very large expansion of virtual care nearly made up the difference. With a widespread physical presence across the U.S., this has important implications for access to care and future allocation of medical personnel, facilities, and resources.