Effect of early mobilization in patients hospitalized for acute stroke with premorbid disability: a target trial emulation using instrumental variable analysis.

Abstract

To evaluate the per-protocol effect of early mobilization on in-hospital adverse events among patients with acute stroke who had premorbid disability, using a target trial emulation framework. This retrospective observational study conducted at a single center in Japan included patients with acute stroke, all of whom had a premorbid disability. Early mobilization was defined as mobilization within 2 days of hospital admission. The primary outcome was the occurrence of any adverse event-such as death or neurological or non-neurological adverse events-between day 4 and day 30 following admission. Sunday to Thursday admission was used as an instrumental variable (IV) to estimate the causal effect of early mobilization, based on reduced rehabilitation staffing on weekends. A two-stage least squares regression model was applied, adjusting for relevant clinical covariates. Among 165 patients, 58% received early mobilization. Admission from Sunday to Thursday increased the probability of early mobilization by a mean of 0.53 points (F-statistic = 44.79). IV analysis indicated that early mobilization reduced the probability of adverse events by 28 percentage points (absolute risk difference) [95% confidence interval: -55 to -2 percentage points]. Early mobilization appeared to reduce adverse events in patients with acute stroke and premorbid disability.