Construct validity, responsiveness and minimal important difference of the cumulated ambulation score in older adults with hip fracture in sub-acute rehabilitation facility.

To assess the construct validity, responsiveness and minimal important difference of the cumulated ambulation score in patients with hip fracture in sub-acute rehabilitation facility. Observational, prospective, monocenter, cohort study. Rehabilitation Institute. 456 older adults with hip fracture (≥65 years) admitted for inpatient rehabilitation. Cumulated ambulation score, functional independence measure and functional ambulation category were collected at admission and discharge. Construct validity and responsiveness were assessed through hypothesis testing and minimal important difference was determined using the anchor-based method; floor and ceiling effects were also assessed. The cumulated ambulation score showed strong correlations with the functional independence measure and functional ambulation category scores at both admission and discharge, satisfying all the hypotheses for construct validity. The effect size of cumulated ambulation score was 1.63. Changes in cumulated ambulation score had a moderate-to-strong correlation with changes of other instruments and were able to discriminate patients improved from those not-improved, and patients classified as independent ambulators from those dependent. A ceiling effect was found only at discharge. The estimated minimal important difference was 2 points. The cumulated ambulation score showed high levels of construct validity and responsiveness according to the hypothesis testing. A two points improvement at the end of rehabilitation was found to be clinically important in people with hip fracture in the sub-acute phase. The ceiling effect found at discharge suggested the limitation of the scale in assessing people with a partially recovered autonomy in performing postural changes and gait.

Déclaration de conflit d'intérêts

Declaration of Conflicting InterestsThe authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.