The Association between Technology Capacity and Prenatal Care during COVID ‐19 Pandemic: The Modifying Role of Social Support

Abstract

included an average response rate of 71% per week. Our secondary data source was the 2019 Uniform Data System. Using generalized estimating equations with exchangeable correlation structures, we evaluated temporal week-to-week rates of telehealth visits (% of all visits that were virtual) across urban vs. rural areas; racial/ethnic groups; linguistic groups; and states. Models adjusted for the percent change in average weekly visits relative to the pre-COVID time period, used state fixed effects, and clustered errors at the FQHC-level. Population Studied: 100% sample of FQHCs across the US (N = 1349), serving >28 million low-income patients. Our analytic sample included 11,169 FQHC-weeks. Principal Findings: From April through June of 2020, 96% of FQHCs across the US were using some telehealth and 47% of all FQHC visits were virtual-from 52% of visits in early April to 37% of visits by the end of June. However, there was significant heterogeneity in these trends. For instance, while urban and rural FQHCs had statistically similar rates of virtual visits in April 2020, over time, rural areas were increasingly less likely to be using telehealth (39% of visits in urban areas vs. 29% of visits in rural areas by late June 2020, p < 0.001). FQHCs serving a high proportion (>25%) of Black patients, Hispanic patients, or non-English speaking patients experienced a significantly lesser rate of decline in virtual visits over time relative to all other FQHCs (p < 0.001 for all). Finally, telehealth utilization varied widely by state over our study period, including highest rates of use in CT (80%), RI (78%), and MA (76%) and lowest rates of use in SD (14%), KS (18%), GA (21%), AR (23%), TN (24%), ID (26%), and SC (26%). Conclusions: As FQHCs across the US reach a steady state of integrating telehealth into care delivery, there are significant differences in telehealth utilization that may be mitigating existing inequities in health care access, particularly for FQHCs that disproportionately serve Black, Hispanic, or non-English speaking patients. However, geographic inequities in telehealth utilization are vast. Implications for Policy or Practice: Permanently expanding coverage and reimbursement of telehealth services may help to address racial/ ethnic and linguistic inequities in access to care. Wide across-state variation in telehealth highlights the potential importance of state-level policy, leadership, and investment in telehealth in enabling, or hindering, its implementation and reach. Research Objective: The Centers for Disease Control and Prevention called for telehealth uptakes to avoid disruption of routine prenatal care (PNC) during the COVID-19 pandemic. Yet, lack of technology capacity may hinder telehealth access and disrupt PNC. Social support could serve as a protective factor against technology barriers and moderate the association between technology capacity and attending PNC. To date, little data are available regarding the associations between technology capacity, social support, and PNC. This study assessed the associations of technology capacity with PNC provider continuity and overall PNC experience and to examine whether social support modifies such associations. Study Design: A nationwide survey was completed from 5/4/2020-5/6/2020 by 668 U.S. pregnant women, 18-44 years of age and being at least 8-week gestation. Two aspects of PNC were studies: 1) pregnant women's overall rating of PNC experience, ranging from very poor (1) to outstanding (5), and 2) PNC provider continuity , defined as whether a woman stayed with the same PNC provider during the pandemic. Technology capacity was measured as the total scores of participants' confidence and necessary skills for telehealth use (range: 2-10). Social support was assessed using the modified Medical Outcomes Study Social Support Survey and categorized to three levels (low, medium and high) with mean +/À one standard deviation. Multivariable linear regressions and logistic regression were used to examine the associations between technology capacity and PNC outcomes and the modifying roles of social support. Population Studied: 668 women who were > 8 weeks pregnant, aged 18-44 years, living and planning to give birth in the United States, and having initiated PNC were included in the analysis. Principal Findings: A total of 668 pregnant women have initiated PNC by 5/6/2020. They were mostly aged 25-34 years (91.8%), diverse (44.3% non-Hispanic Black, 8.5% Hispanic, 44.3% non-Hispanic White and 6.6% other non-Hispanic race), urban residents (86.4%), and full-time employed (63.9%). Technology capacity and social support were positively associated with PNC experience and provider continuity (Ps < 0.01). However, technology capacity was positively related to PNC experience (β = 0.31, 95% CI 0.16 to 0.45; P < 0.001) and PNC provider continuity (OR = 1.80, 95% CI 1.52 to 2.14; P < 0.001) only among women with low social support, but not among women with high social support.