# Interaction effect of rehabilitation initiation timing and hospitalization frequency on long-term functional outcomes after stroke in rural China: a retrospective cohort study

## TL;DR

Starting rehabilitation earlier and having more hospitalizations improves long-term recovery for stroke patients in rural China.

## Contribution

This study identifies the interaction between rehabilitation timing and hospitalization frequency on stroke recovery in a rural Chinese population.

## Key findings

- Earlier rehabilitation initiation (within 1–60 days post-stroke) led to greater improvements in ADL at 6 months.
- Each additional inpatient rehabilitation episode within the first month improved outcomes, but not beyond two months.
- Early rehabilitation and higher hospitalization frequency were most beneficial for older patients and those with hemorrhagic stroke.

## Abstract

To investigate the effects of rehabilitation-initiation timing (RIT) and rehabilitation-hospitalization frequency (RHF) on activities of daily living (ADL) evaluated at 6 months post-stroke.

Retrospective cohort study.

Convalescent rehabilitation wards in urban and suburban areas of Xiangyang, China.

A total of 275 patients with ADL impairment following acute or subacute stroke who received inpatient comprehensive rehabilitation between 2021 and 2024.

Participants underwent inpatient multidisciplinary rehabilitation—including physical therapy, occupational therapy, and individualized functional exercises—during each hospitalization, with each inpatient rehabilitation episode lasting for 3 weeks. The main exposures were the timing of rehabilitation initiation and the total number of inpatient rehabilitation episodes within the first 6 months post-stroke.

The primary outcome was the change in ADL, assessed by the Barthel Index (BI), from baseline to the 6-month post-stroke follow-up.

At the 6-month (180-day) follow-up, the mean BI score increased by 12.59 points compared to baseline (95% CI, 5.53–19.65; p < 0.001). Compared to those who started rehabilitation at 61–90 days post-stroke, patients who initiated rehabilitation earlier—at 1–14, 15–30, and 31–60 days—showed greater BI improvements at 6 months, with mean differences of 15.48 (95% CI, 4.90–26.06; p = 0.004), 13.18 (95% CI, 3.85–22.51, p = 0.005), and 8.63 (95% CI, 0.40–16.86, p = 0.04) points, respectively. Among patients who started rehabilitation at 1–14 and 15–30 days, each additional systematic inpatient rehabilitation was associated with a further mean BI increase of 2.24 (95% CI, 0.98–5.46, p = 0.20) and 2.10 (95% CI, 0.87–5.07, p = 0.21) points, respectively, although these differences did not reach statistical significance. Subgroup analysis showed that early rehabilitation significantly improved BI in patients aged ≥65 and those with hemorrhagic stroke. Moreover, higher hospitalization frequency benefited patients with higher education and those with hemorrhagic stroke.

Earlier initiation and greater frequency of inpatient rehabilitation were independently associated with better ADL outcomes at the 6-month mark in rural Chinese stroke survivors. Importantly, the benefit of each additional rehabilitation admission was amplified when therapy began within the first month post-stroke and diminished when initiation was delayed beyond two months, especially among patients with hemorrhagic stroke, aged ≥65 years, women, and those with higher educational attainment.