The association between neighborhood social vulnerability and community-based rehabilitation after stroke.

BACKGROUND: Timely rehabilitative care is vital for functional recovery after stroke. Social determinants may influence access to and use of post-stroke care but have been inadequately explored. The study examined the relationship between the Social Vulnerability Index (SVI) and community-based rehabilitation utilization. METHODS: We included 6,843 adults (51.6% female; 75.1% White; mean age 70.1) discharged home after a stroke enrolled in the COMprehensive Post-Acute Stroke Services study, a pragmatic trial conducted in 40 North Carolina hospitals from 2016-2019. Rehabilitation utilization was sourced from administrative claims. Geocoded addresses were linked to 2018 Census tract SVI. Associations between SVI and 90-day rehabilitation use, adjusted for patient's clinical and socio-economic characteristics, were obtained from generalized estimating equations. We also examined the associations of SVI with therapy setting, types of therapy, intensity of visits, and time to first visit. RESULTS: Thirty-five percent of patients had at least one physical (PT) or occupational therapy (OT) visit within 90 days, ranging from 32.4%-38.7% across SVI quintiles. In adjusted analysis, there was no dose-reponse relationship between higher summary SVI, nor most of its sub-domains, and 90-day rehabilitation use. Greater vulnerability in household composition and disability was modestly associated with -0.4% (95% CI -4.1% to 3.4%) to -4.3% (95% CI -0.8% to -7.7%) lower rehabilitation use across SVI quartiles. Greater summary and subdomain SVI was associated with higher odds of receiving therapy in the home versus outpatient clinic (OR = 1.88, 1.58 to 2.17 for Q5 vs Q1 summary SVI) and receiving both PT and OT versus a single-type therapy (1.72, 1.48 to 1.97 for Q5 vs. Q1 summary SVI). No differences were observed for therapy intensity or time to therapy. CONCLUSION: Use of rehabilitation care was low, and largely similar across levels of SVI and most of its subdomains. Individuals residing in areas of high SVI were more likely to receive therapy in the home and to receive dual therapy, possibly reflecting greater need among these individuals. Future studies should evaluate potential mechanisms for these findings and further identify both patient and community factors that may inform strategies to improve rehabilitation use. CLINICAL TRIAL NUMBER: https://www. CLINICALTRIALS: gov/ NCT02588664 [registration date: 2015-10-23].

Duke Scholars

Author Amy Martha Pastva Orthopaedic Surgery, Physical Therapy