Factors associated with return to pre-admission functional category at hospital discharge among acute inpatients receiving physical therapy and/or occupational therapy: A retrospective cohort study

BACKGROUND: Functional decline associated with acute hospitalization may persist at hospital discharge. This study examined associations between routinely documented admission-time characteristics and return to the same pre-admission functional category among acute inpatients receiving physical therapy and/or occupational therapy. METHODS: This exploratory single-center retrospective cohort study included 444 unique adults who received acute inpatient physical therapy and/or occupational therapy at a tertiary hospital. Ninety patients classified as bedridden before admission were excluded because no outcome variation was observed in that stratum. Modified Poisson regression with robust variance was used to estimate adjusted risk ratios (aRRs) for return to the same pre-admission documentation-derived functional category at discharge. RESULTS: The primary cohort included 354 patients, of whom 146 (41.2%) returned to the same pre-admission documentation-derived functional category at discharge. Each one-level worsening in admission functional severity was associated with a lower probability of return (aRR 0.59, 95% confidence interval [CI] 0.47-0.75). In the exploratory subgroup of 164 patients independently ambulatory before admission, worse admission functional severity was associated with a higher risk of discharge below the pre-admission category (aRR 1.17, 95% CI 1.03-1.32). Associations with premorbid functional category reflected different premorbid functional targets and were not interpreted as comparative recovery probabilities. CONCLUSIONS: Routinely documented premorbid and admission functional information may help identify acute inpatients receiving physical therapy and/or occupational therapy who may be at risk of leaving hospital below their pre-admission functional category. Because the endpoint represents no net decline relative to premorbid function and reflects different targets across baseline strata, findings should not be interpreted as evidence that patients with greater premorbid dependency recover better.