COVID-19 has caused unprecedented global disruption. Within the US there are over three million cases and over 131,000 deaths. The virus has also caused economic volatility, cancellation and/or suspension of healthcare services, and widespread isolation due to social distancing. COVID-19 has also profoundly impacted health and care for Veterans who are generally older, sicker, more rural, and more economically vulnerable than the overall U.S. population. Veterans may be more likely to contract COVID-19, but they may also face more subtle and longer-lasting risks related to disruptions in care for their chronic health conditions. Access to healthcare is hindered by clinic cancellations/closures, transportation challenges, financial concerns, and perhaps most importantly, self-imposed isolation due to social distancing. These disruptions will also exacerbate known disparities in care for vulnerable Veteran populations including rural residents and racial/ethnic minorities.
Our primary objective is to build capacity for a robust evaluation of the impacts of the pandemic on chronic medical conditions using diabetes as a template. Specifically, using diabetes as a template we will (1) examine the pre-post impact of the pandemic, (2) the extent to which telemedicine use prior to the pandemic impacts the ability of VAMCs to provide care during the pandemic and finally (3) the extent to which neighborhood deprivation as well as racial- ethnic disparities impact the ability of VAMCs to provide care during the pandemic.
We plan to replicate our comprehensive spatio-temporal database that we have built for our ongoing merit (Neelon/Hunt, IIR HX002299-01A2) from 2017 through 2020. For Aim 1, we will use Bayesian hierarchical models to identify hotspots where diabetes outcomes and care are impacted most significantly by the pandemic. For Aim 2, we will use spatial propensity score methods to compare VAMCs with and without extensive telemedicine capabilities before the pandemic to determine to what extent this impacts care during the pandemic12-14. For Aim 3, we will examine interactions between pandemic hotspots, community social deprivation indices, race-ethnic group and diabetes outcomes and quality care metrics.
Results from a model adjusting for patient demographic characteristics and access measures indicated that receipt of primary care during the fourth quarter of fiscal year 2020 varied substantially across VAMC catchment areas, with prevalence estimates ranging from 22% to 60%. Using the same models with in-person and tele-care as the outcomes, our analyses revealed that the probability of having an in-person primary care visit during the same period ranged from 16% to 51% depending on VAMC catchment area, while the probability of having a tele primary care ranged from 14% to 52% depending on VAMC catchment area. Results also indicated that patients living in the most socially vulnerable areas were the most likely to receive any primary care, but were least likely to receive in-person care. Lastly, veterans receiving care at VA medical centers with low rates of tele-heath pre-pandemic were less likely to have received any primary care during the period, with a pronounced reduction in tele-health compared to veterans who had been receiving care at VA medical center with higher rates of tele-health pre-pandemic.
Using diabetes as a template and leveraging our expertise in GIS analysis and advanced spatio-temporal statistics, we have provided an in-depth analysis of the impact of the pandemic on chronic disease care within the VA.
None at this time.
Health Systems, Diabetes and Other Endocrine Disorders, Infectious Diseases
TRL - Applied/Translational, Data Science
Diabetes, Disparities, Telemedicine/Telehealth
None at this time.