Partially Virtual vs In-Person Care in Cardiovascular Conditions—Time Trends, Patient Characteristics, and Medication Management: Cohort Study

Background: Despite growing initiatives to promote partially virtual care, there is still a lack of comprehensive data describing its use, patient characteristics, and medication management practices. Objective: This study aimed to examine the use of partially virtual care in the management of cardiovascular disease. Methods: We used nationwide Danish registries to identify all patients with a first-time diagnosis of atrial fibrillation (AF), heart failure (HF), pulmonary embolism (PE), or acute coronary syndrome (ACS) between 2019 and 2023. Patients were categorized as receiving partially virtual care if followed up at least once virtually or in-person care if seen solely in-person. We evaluated temporal changes in the use of partially virtual care and calculated the odds of receiving partially virtual care vs in-person care and the prevalence of initiating guideline-directed medical therapy (GDMT). For GDMT initiation, the exposure was instead the type of the patient's first follow-up visit. Multivariable logistic regression was used to estimate the odds of receiving partially virtual care, adjusted for frailty score, residence, and period of diagnosis. Separate multivariable logistic regression models were then used to estimate adjusted average differences in the prevalence of GDMT initiation between virtual and in-person follow-up among patients not already on the medication, accounting for frailty score, residence, and period of diagnosis. Results: In total, 45,919 patients with AF, 30,482 with HF, 12,451 with PE, and 12,435 with ACS were followed up. Among these, 27.3% (AF: 12,517/45,919), 51.7% (HF: 15,749/30,482), 47.5% (PE: 5918/12,451), and 37.2% (ACS: 4627/12,435) had partially virtual follow-up. The use of partially virtual follow-up increased during the COVID-19 pandemic and remained high afterward. In 2023, 17.4% (475/2731) to 35.3% (876/2481) of patients' first follow-up visits were virtual, depending on the condition. Patient-specific factors significantly associated with partially virtual care were patients residing in cities for those with AF (odds ratio [OR] 1.27, 95% CI 1.21-1.34), HF (OR 1.38, 95% CI 1.31-1.46), and PE (OR 1.14, 95% CI 1.04-1.24), but not for patients with ACS, and "high frailty" for those with AF (OR 1.63, 95% CI 1.48-1.80) and PE (OR 1.53, 95% CI 1.29-1.81). A large percentage of patients were on GDMT prior to the first follow-up visit. Differences in new initiation of GDMT were small in absolute terms across conditions, regardless of virtual or in-person follow-up; for example, for patients with HF, 2.23% initiated sodium-glucose cotransporter-2 inhibitors following a virtual visit, vs 4.47% after an in-person visit (prevalence difference -2.24%, 95% CI, -2.61 to -1.87). Conclusions: Among patients who received follow-up, a considerable proportion of patients with cardiovascular disease (from 12,517/45,919, 27.3% to 15,749/30,482, 51.7%) received partially virtual care. The odds of receiving partially virtual care varied significantly based on patient characteristics, particularly place of residence and frailty score. Finally, small absolute differences were found in the initiation of new GDMT.