Frailty and cognitive status evaluation can better predict mortality in older adults?

Objectives: to evaluate the improvement in one-year mortality prediction after adding a 2-min cognitive screening to a simple 1-min frailty detection instrument. Secondary outcomes were new activities of daily living (ADL) disability and falls.

Design: Prospective cohort study.

Setting: A geriatric day-hospital for intermediate care.

Participants: A total of 701 older adults with an acute or decompensated disease (79.5 (8.3) years, 64% female).

Measurements: A rapid and simple frailty evaluation was performed using the FRAIL questionnaire. The presence of cognitive impairment was defined by previous diagnosis of dementia or a score of five or less on an education-corrected 10-point cognitive screening tool.

Results: Frail participants with normal (hazard risk [HR] 4.0, 95% confidence interval [CI], 1.73-9.25) and impaired cognition had a higher risk of death (HR 4.38, 95% CI, 1.95-9.87) than robust participants. The presence of cognitive impairment increased the risk of death in prefrail (HR 3.60, 95% CI, 1.55-8.34) and robust participants (HR 3.49, 95% CI, 1.22-9.96). Cognitive impairment was associated with an increased risk of incident ADL disability in all frailty categories. The presence of cognitive impairment was associated with a significantly higher risk of fall in robust seniors. The predictive accuracy of the FRAIL scale was lower than expected (between 0.58 and 0.69), and a small improvement was observed after adding the cognitive screening (between 0.61 and 0.72).

Conclusion: Despite of significant results in predicting relevant clinical events, the present combination of the FRAIL and 10-CS scales may not be ideal in clinical practice.