Cognitive assessment to optimize prediction of functional outcome in subacute hip fracture: a short-term prospective study.

Background: Cognitive impairment is a long-known negative prognostic factor after hip fracture. Cognition is usually screened by a single easy-to-administer bedside tool, but recent studies have shown that screening tests may be not enough to rule out cognitive impairment with an unfavorable prognostic role. Unfortunately, data on outcome prediction by further cognitive assessments is sparse.

Aim: We focused on patients with subacute hip fracture defined cognitively intact or mildly impaired on the screening evaluation performed by the Short Portable Mental Status Questionnaire (SPMSQ). We hypothesized that each of 3 further cognitive tests could independently predict activities of daily living, with optimal prediction of function obtained by performing all three the tests.

Design: Short-term prospective study.

Setting: Rehabilitation ward.

Population: Inpatients with subacute hip-fracture.

Methods: Three cognitive tests were performed on admission to rehabilitation in the patients who made ≤4 errors on the SPMSQ: Montreal Cognitive Assessment (MoCA), Rey Auditory Verbal Learning Test (RAVLT, immediate and delayed recall) and Frontal Assessment Battery (FAB). We assessed activities of daily living by the Barthel index. Successful rehabilitation was defined with a Barthel Index Score ≥85.

Results: Each of the three cognitive tests assessed before rehabilitation significantly predicted the Barthel index scores measured at the end of the rehabilitation course in our sample of 280 inpatients. However, only the MoCA score retained its significant predictive role when the scores from the three tests were included together as independent variables in a multiple regression model, with adjustments for a panel of potential confounders (P=0.007). The adjusted odds ratio to achieve successful rehabilitation for a seven-point change in MoCA score was 1.98 (CI 95% from 1.02 to 3.83; P=0.042).

Conclusions: Contrary to our hypothesis, MoCA but not RAVLT and FAB retained the prognostic role when the scores from the three tests were evaluated together as potential predictors of functional ability in activities of daily living.

Clinical rehabilitation impact: In the presence of a normal (or mildly altered) score on the SPMSQ in subacute hip fracture, MoCA scores improve prediction of activities of daily living and should be routinely performed.