1104 — What is the right mix? Variation in the proportion of primary care delivered via video for Veterans with common chronic conditions
Lead/Presenter: Jacqueline Ferguson,
COIN - Palo Alto
All Authors: Ferguson JM (Center for Innovation to Implementation, VA Palo Alto), Wray C (Section of Hospital Medicine, VA San Francisco Health Care System) Greene L (Center for Innovation to Implementation, VA Palo Alto) Van Campen J (Center for Innovation to Implementation, VA Palo Alto) Zulman DM (Center for Innovation to Implementation, VA Palo Alto)
Veteransâ€™ sociodemographic characteristics such as age, gender, and urbanicity predict their likelihood to engage in video-based care. However, it is not known whether their specific clinical (health) conditions also predict the likelihood of video-based care utilization. In this evaluation, we characterized the proportion of VA patientsâ€™ primary care that is delivered via video, and explored variation in patterns by chronic conditions assessed in each encounter.
We examined over 18.5 million outpatient encounters nationwide within VA occurring between January 2020 and May 2022. We classified each encounter by care modality (in-person, video, phone, remote monitoring) and by 40 chronic conditions using ICD-10 diagnoses codes associated with each visit. We calculated the proportion of video encounters (number of video encounters/total number of encounters) for each chronic condition and each month. For each chronic condition, we examined temporal trends to identify 1) sustained and 2) limited applications of video care following the shift to virtual care early in the COVID-19 pandemic. We evaluated chronic conditions with atypical proportions of care using z-scores (i.e., number of standard deviations a given chronic condition is above/below the mean proportion of video care across all conditions).
On average, 10.2% of encounters were video-based. Among the most prevalent conditions managed in primary care settings, the proportion of care delivered via video ranged from 7.1% (COPD) to 16.8% (Migraines). While all conditions saw a marked increase in the proportion of video care following the pandemic-driven shift to virtual care in March 2020, there were differences in the magnitude and duration of the increase. The proportion of visits delivered via video was higher than average (z-score > 1) for conditions such as obesity, lower back pain, migraines, anxiety/depression/PTSD, and asthma, while the proportion of visits was lower than average (z-score < -1) among conditions like atrial fibrillation, heart failure, COPD, and ischemic heart disease.
Using a novel measure of virtual care delivery, we found variation in the proportion of visits delivered via video across common chronic conditions. Our findings suggest a Veteranâ€™s chronic conditions may impact whether they receive video care. Explanations for this pattern are multifaceted and may be confounded by facility-level differences in virtual care uptake, coding differences among providers, patient characteristics, and preferences around modality of care. Future work should examine whether patterns are related to condition characteristics, such as chronicity of symptoms and the need for a physical exam, which could underlie some of the observed patterns.
As VA continues to expand and optimize its use of video-based care, our findings may help identify the types of chronic conditions that are most commonly seen in virtual settings. Successful implementation of video care among chronic conditions with high levels of video care may be leveraged to support providers managing chronic conditions with lower levels of video care applications. Furthermore, the proportion of care delivered via video may be a valuable measure for studies that seek to identify the optimal mix of virtual and in-person for various clinical scenarios.