Lead/Presenter: Rebecca Tisdale, COIN - Palo Alto
All Authors: Tisdale RL (VA Palo Alto / Center for Innovation to Implementation (Ci2i)), Der-Martirosian C (Greater LA VA / Center for the Study of Healthcare Innovation, Implementation & Policy) Yoo CK (Greater LA VA / Center for the Study of Healthcare Innovation, Implementation & Policy) Chu K (Greater LA VA / Center for the Study of Healthcare Innovation, Implementation & Policy; Veterans Emergency Management Evaluation Center, North Hills, CA) Zulman D (VA Palo Alto / Center for Innovation to Implementation (Ci2i)) Leung LB (Greater LA VA / Center for the Study of Healthcare Innovation, Implementation & Policy; Division of General Internal Medicine-Health Services Research, David Geffen School of Medicine, University of California Los Angeles)
Video care expanded rapidly in the Veterans Health Administration (VA) at the onset of the COVID-19 pandemic and remains a significant proportion of all VA care. Prevalence of cardiovascular disease (CVD) in Veterans is high, and ensuring access to care for CVD will likely require continued virtual care use. However, there remains a lack of evidence regarding which patients with these CVD conditions are more likely to receive video care. We sought to characterize use of video care for Veterans with two common cardiovascular diseases, heart failure and hypertension.
This retrospective cohort study included Veterans established in VA primary care with diagnoses of heart failure and/or hypertension between 3/11/2019 and 3/10/2022, i.e., in the calendar year prior to the novel Coronavirus (COVID-19) pandemic and for the first two pandemic years. We identified individual-level predictors of one or more video-based visits, accounting for patient- and site-level clustering with a two-level mixed-effects logistic regression model adjusted for sociodemographic and clinical covariates and time.
Our analytic cohort comprised 3,807,820 Veterans with diagnoses of heart failure, hypertension, or both with 52 million visits. 456,901 Veterans had both heart failure and hypertension, 50,753 had heart failure only, and 3,300,166 had hypertension only. Veterans with heart failure and hypertension had an average baseline age of 71.6 years and mean Charlson Comorbidity Index of 3.0. 2.9% were female at birth, and 34.8% lived in a rural or highly rural setting. In our multi-level logistic regression model, odds of using video care were highest during the first year of the pandemic, then declined in the second (AOR 15.3, 95% CI 15.1-15.4 and 11.5, 95% CI 11.3-11.6, respectively, compared to the pre-pandemic year). Male patients had lower odds of ever using video care than female patients (adjusted odds ratio [AOR] 0.73, 95% confidence interval [CI] 0.72-0.74). Age showed a gradient: patients 75 years or older had an AOR of 0.38 compared to those aged 18-44 years (95% CI 0.38-0.39). Rural-dwelling Veterans had lower odds of using video care than urban-dwellers (AOR 0.71, 95% CI 0.70-0.71). Veterans with heart failure had slightly higher odds of video care use than those with hypertension only (AOR 1.06, 95% CI 1.05-1.07).
Veterans with CVD had a 15-fold increase in odds of use of video-based care in the first year of the COVID-19 pandemic. Male, older, and rural-dwelling Veterans had lower odds of using video care than their respective reference groups.
Video care remains an important proportion of care delivered to Veterans with cardiovascular disease. Given lower odds of video care among certain veteran groups, continued expansion of video care could make CVD services increasingly inequitable. As VA expands virtual care for CVD, these insights can inform equitable and effective triage of patients to virtual versus in-person care.