Abstract A105: Optimizing Post-Stroke Continuity of Care: Insights from Three Models of Outpatient Follow-Up

Introduction: Outpatient follow-up after stroke is essential to manage vascular risk factors and reduce recurrence, yet adherence is suboptimal. Care models vary, including physician-led, APP rapid-access, and telemedicine clinics. These approaches differ in accessibility and patient engagement, but their relative effectiveness is unclear. Comparing adherence across models and identifying predictors of attendance may inform strategies to reduce stroke recurrence. Objective: To assess adherence and predictors of follow-up in post-stroke patients across varied clinic models. Methods: We conducted a retrospective analysis of patients discharged from a comprehensive stroke center in the U.S. Stroke Belt, evaluating three follow-up models: a physician-led clinic, an APP rapid-access clinic, and a hybrid telemedicine (HTM) clinic, that included both in-person or telehealth visits. Demographic and clinical data were collected from samples of consecutive patients in each model. The primary outcome was adherence to first clinic visit after discharge. Comparisons by encounter type and regression analyses were performed to explore predictors of adherence. Results: Of 1,539 stroke discharges, 1,344 (87.3%) were scheduled for follow-up (physician-led clinic, 259; APP rapid-access clinic, 149; HTM clinic, 936). Overall adherence was 64.8%, comprising 74.8% in-person visits and 25.2% telemedicine visits. The HTM clinic had the highest adherence (p <0.001, Table 1). Across all care models, adherence was associated with younger age, history of tobacco use, reduced wait times, and disposition to home or acute rehabilitation, while non-adherence was associated with hypertension, dementia, and prior stroke (Table 2). Male patients and outside transfers were scheduled more often for telehealth appointments. In regression models, patients with dementia (OR 0.38) and longer wait times (OR 0.99 per day) were independently predictive of poor adherence. Conclusions: In this study of post-stroke clinic models, adherence was highest in the HTM clinic, which was introduced during the COVID-19 pandemic and likely influenced outcomes through unmeasured factors. Adherence was associated with age, comorbidities, disposition, and wait time; non-adherence was linked to prior stroke and hypertension. Given the importance of secondary prevention, these findings may be helpful in identifying patients at risk for poor follow-up and in designing clinical models to improve post-stroke care.