The relationship between hospitalization-associated disability and the functional recovery process

This study aimed to investigate the relationship between HAD and functional recovery during hospitalization.

This retrospective study included 195 patients admitted to Ikeda Municipal Hospital between August and October 2022. The inclusion criteria were patients aged ≥ 70 years, hospitalized for acute illness, and who received rehabilitation. Patients with fractures or strokes that directly impaired motor function, those with a pre-admission Barthel index (BI) score of 0, ICU admission, and terminal cancer were excluded. Functional recovery was evaluated based on rehabilitation records, specifically the initiation of sitting at the edge of the bed, wheelchair transfer, and ambulation. Data on age, sex, dementia, BI, length of hospital stay, and discharge to home were extracted from medical records. Statistical comparisons were performed between patients with and without HAD. Logistic regression analysis was used to identify the factors associated with HAD. Additionally, receiver operating characteristic (ROC) curve was used to determine the cut-off values for physiotherapy milestones linked to HAD.

The participants’ mean age was 81.7±7.4 years, with 94 (48.2%) women. The mean length of hospital stay was 18.6±14.3 days. Regarding functional recovery, the mean time to initiation of sitting at the edge of the bed, wheelchair transfers, and ambulation was 1.4±1.2, 1.7±1.6, and 2.0±1.8 days, respectively. HAD occurred in 67 (34.4 %) patients. Significant differences were observed between the HAD and non-HAD groups regarding age, emergency admission status, discharge BI, presence of dementia, length of hospital stay, discharge destination, and recovery process. Logistic regression analysis revealed that after adjusting for age, emergency admission, pre-admission BI, dementia, and length of hospital stay delayed initiation of wheelchair transfer (OR; 1.523, 95% CI; 1.174–1.976) and ambulation (OR; 1.795, 95% CI; 1.385–2.328) were independently associated with HAD. ROC curve analysis demonstrated moderate predictive power for ambulation initiation (AUC=0.717, 95% CI; 0.620–0.815), with a cut-off of 1.5 days for predicting HAD.

HAD was significantly associated with delayed initiation of wheelchair transfer and ambulation during rehabilitation, with early difficulties in ambulation, particularly within day 1, which increased the likelihood of HAD.

This study identified critical cutoff values in the functional recovery process that may help prevent HAD. These findings could inform rehabilitation interventions during hospitalization to reduce the incidence of HAD.