Effect of early intensive rehabilitation on the clinical outcomes of patients with acute stroke

Aim: Intensive rehabilitation effectively improves physical functions in patients with acute stroke, but the frequency of intervention and its cost-effectiveness are poorly studied. This study aimed to examine the effect of early high-frequency rehabilitation intervention on inpatient outcomes and medical expenses of patients with stroke.

Methods: The study retrospectively included 1759 patients with acute stroke admitted to the Kobe City Medical Center General Hospital between 2013 and 2016. Patients with a transient ischemic attack, subarachnoid hemorrhage, and those who underwent urgent surgery were excluded. Patients were divided into two groups according to the frequency of rehabilitation intervention: the high-frequency intervention group (>2 times/day, n = 1105) and normal-frequency intervention group (<2 times/day, n = 654). A modified Rankin scale score ≤2 at discharge, immobility-related complications and medical expenses were compared between the groups.

Results: The high-frequency intervention group had a significantly shorter time to first rehabilitation (median [interquartile range], 19.0 h [13.1-38.4] vs. 24.7 h [16.1-49.4], P < 0.001) and time to first mobilization (23.3 h [8.7-47.2] vs. 22.8 h [5.7-62.3], P = 0.65) than the normal-frequency intervention group. Despite higher disease severity, the high-frequency intervention group exhibited favorable outcomes at discharge (modified Rankin scale, ≤2; adjusted odds ratio, 1.89; 95% confidence interval, 1.25-2.85; P = 0.002). No significant differences were observed between the two groups concerning the rate of immobility-related complications and total medical expenses during hospitalization.

Conclusions: High-frequency intervention was associated with improved outcomes and decreased medical expenses in patients with stroke. Our results may contribute to reducing medical expenses by increasing the efficiency of care delivery. Geriatr Gerontol Int 2021; 21: 623-628.