Background: Timely rehabilitation after stroke is essential for optimizing recovery. Patients discharged home can experience unnecessary service delays and gaps in care. Our aim was to examine sociodemographic characteristics associated with receipt of rehabilitation services within 30 days after discharge home in stroke or transient ischemic attack (TIA) patients.
Methods: COMPASS is a cluster-randomized pragmatic trial measuring the effectiveness of the COMPASS model of care vs. usual care on functional status 90 days after stroke or TIA for patients discharged home. We analyzed data from 369 participants who enrolled in the intervention arm of the COMPASS trial, completed the post-acute clinic visit within 14 days, and the 30-day call between July 2016 and May 15, 2017. For those who were referred to rehabilitation services at hospital discharge or at the clinic visit, receipt of home health (HH) and outpatient (OP) rehabilitation services was reported by the participant during the 30-day call. Sociodemographic differences between those who did and did not receive these services were evaluated. We computed adjusted odds ratios with 95% confidence intervals using logistic regression to identify factors associated with receipt of rehabilitation services.
Results: Of the 369 patients, 176 (47.7%) had ascertainment of receipt of services. Of the 115 patients referred to HH, 50 (43.5%) received it, and of 85 patients referred to OP therapy, 29 (34.1%) received it. There were no statistically significant differences in sociodemographic characteristics related to receipt of HH, but non-whites were less likely (3 of 20, or 15%) than whites (25 of 64, 34.9%) to receive OP therapy. After adjustment for NIHSS, ambulatory status at hospital admission, and age, non-white participants had a 78% decreased odds (OR 0.22; 95% CI 0.05, 0.95; p=0.04) of receiving OP therapy.
Conclusion: These findings indicate that a significant gap exists between referral to and receipt of recommended post-acute stroke rehabilitation services. In addition, despite the small sample size, our results suggest that there may be a disparity in receipt of OP therapy for non-white patients relative to white patients.