A screening test is not enough to define the prognostic role of cognitive impairment after hip fracture: a short-term prospective study

Background Cognitive impairment assessed by easy-to-administer tests successfully predicts function after hip fracture, whereas the prognostic role of additional cognitive evaluations is largely unknown. Aims To investigate the capability of further assessments to discriminate cognitive impairment with prognostic relevance in hip-fracture women defined cognitively intact or mildly impaired on the Short Portable Mental Status Questionnaire (SPMSQ). Methods We prospectively investigated women with subacute hip fracture admitted to our rehabilitation facility. The women who made <= 4 errors on the SPMSQ were further assessed by 4 tests: Montreal Cognitive Assessment, Rey Auditory Verbal Learning Test (immediate and delayed recall) and Frontal Assessment Battery. Activities of daily living (ADL) were measured by the Barthel index. Successful rehabilitation was defined with a Barthel index score >= 85. Results Data from 127 women were available. Each of the 4 cognitive tests assessed at admission significantly predicted the Barthel index scores measured at discharge. The predictive role persisted after multiple adjustments. For a change in cognitive scores corresponding to the difference between 25 degrees and 75 degrees percentiles in their distribution in the sample, the adjusted odds ratio to achieve successful rehabilitation roughly ranged from 2 to 4, depending on which cognitive test was adopted. Discussion The women with subacute hip fracture defined cognitively intact or mildly impaired on the SPMSQ could have cognitive impairment revealed by further examination, with prognostic disadvantages in ADL. Conclusions Assessing cognition by the SPMSQ seems not enough to exclude the presence of cognitive impairment with relevant prognostic disadvantage in hip-fracture women.