Introduction

The overall objective of this work was to estimate misclassification of cognitive impairment (CI) by sex among hip fracture patients. The effects of sex and identified CI on hip fracture recovery outcomes, including trajectories of cognitive and physical function and mortality (all-cause and cognition-specific), were subsequently examined.

Methods

This study used secondary data from a cohort of hip fracture patients recruited from 8 Baltimore-area hospitals between 2006-2011 (N=339), with frequency matched enrollment of females (n=171) and males (n=168), tested within 22 days of hip fracture admission and at 2, 6, and 12 months. Death data were derived from the National Death Index 2006-2018 (n=260; females=113, females=147). Analyses differentiated source of CI identification (SCI, n=330) between clinical diagnosis/documentation and direct testing [Modified Mini Mental State Examination (3MS)] of cognition at baseline by patient sex. Analyses identified cognitive and functional recovery using group-based trajectory modeling (GBTM) and joint trajectory models; and estimated time to death for all-cause and cognition-related cause of death (CR-COD) by SCI and sex.

Results

Males had increased odds of CI identified by both hospital record and 3MS after adjusting for age, education, and comorbidities. There were 2-4 distinct groups of recovery for cognitive and functional recovery. In joint models of recovery, high levels of cognitive function were only seen in high physical functioning groups. Significantly more men died than women (147 vs 113, p<.0001) and died sooner from all-cause mortality (41 vs 54 months, p=0.001) but not CR-COD. Males and those with SCI Both were independently at increased risk for all-cause mortality but there was not a significant interaction. Those with SCI Both were 14 times more likely to die of CR-COD (p<.0001).

Conclusions

There is clinical underdiagnosis of active CI in males. Additional CI ascertainment can identify a sub-population of patients at excess risk for mortality and CR-COD. Cognitive testing after hip fracture should be instituted as standard practice to avoid sex bias in identification of CI. Cognition changes little over 12 months post fracture. Pre-fracture functional status informs pattern of physical function recovery. Sex and CI increase mortality risk, and to some extent cause-specific mortality.