LAUSR

OBJECTIVE Both slow gait speed (GS) and higher levels of frailty are associated with adverse outcomes in community-dwelling older people. However these measures are not routinely utilized to stratify risk status in the hospital setting. Here we assessed their predictive validity in older inpatients. DESIGN A prospective cohort study. SETTING Inpatient rehabilitation wards of a tertiary hospital. PARTICIPANTS Adults 65 years and older (N=258). INTERVENTIONS A frailty index (FI) was calculated from routinely collected data and GS was determined from a timed 10-meter walk test. MAIN OUTCOME MEASURES Adverse outcomes were longer length of stay (≥75th percentile), poor discharge outcome (discharge to a higher level of care or inpatient mortality), and inpatient delirium and falls. RESULTS Mean age ± SD was 79±8 years and 54% were women. Mean FI ± SD on admission was 0.42±0.13 and an FI could be derived in all participants. Mean GS ± SD was 0.26±0.33 m/sec. Those unable to complete a timed walk on admission (50%) were allocated a GS of 0. There was a weak but significant inverse relationship between FI and GS (correlation coefficient -0.396). Both parameters were significantly associated with longer length of stay (P<.001), poor discharge outcome (P≤.001), and delirium (P<.05).The prevalence of adverse outcomes was highest in the cohort who were more frail and unable to mobilize at admission to rehabilitation. CONCLUSIONS FI and GS each showed predictive validity for adverse outcomes. In a geriatric rehabilitation setting, they measure different aspects of vulnerability and combining the 2 may add value in identifying patients most at risk.