Background And Aims Little is known about long-term outcomes associated with access to inpatient rehabilitation after stroke. Aim: to compare long-term outcomes for patients discharged directly home with those discharged to inpatient rehabilitation after stroke.

Methods Data collected in the Australian Stroke Clinical Registry (2010–2013) were linked to hospital admission records and the national death index as part of the Stroke123 study. Multilevel multivariable regression analyses were conducted, adjusting for patient and hospital factors. Outcomes were death and hospital readmissions up to 365-days and health-related quality of life (HRQoL) at 90–180days.

Results 7,847 patients were included (median age 71 years, 59% male, 83% ischemic stroke); 4405 (56%) were discharged home and 3,442 (44%) to inpatient rehabilitation. Patients discharged directly home were more likely to be aged under 65 years and to walk independently at hospital admission. Patients discharged to inpatient rehabilitation had more comorbidities and greater odds of having dementia. Hazard of death was not significantly different between the two groups. Patients discharged directly home were more likely to be readmitted to hospital than patients discharged to inpatient rehabilitation (adjusted subhazard ratio [aSHR]:90-days 0.54, 95%CI 0.49, 0.61; aSHR:180-days 0.74, 95%CI 0.67, 0.82; aSHR:365-days 0.85, 95%CI 0.78, 0.93). Patients discharged directly home had higher HRQoL scores, but were more likely to report mobility problems (adjusted OR 0.54, 95%CI 0.47, 0.63).

Conclusions Patients discharged directly home have higher risks of readmission and self-reported mobility problems than patients discharged to inpatient rehabilitation. To reduce this disadvantage, better support is indicated for people who are discharged directly home after stroke.