Predicting mortality of psychogeriatric patients: a simple prognostic frailty risk score

Abstract

Background and aims: Frailty and mortality in psychogeriatric patients are hard to predict but important in counselling and therapeutic decision making. We have therefore developed a simple frailty risk score to predict mortality this population.

Study design: Prospective observational study including 401 community dwelling psychogeriatric patients (249 women; mean (SD) age 78.0 (6.5) years), who had been referred to a multidisciplinary diagnostic observation centre. We used Cox proportional hazards regression models to identify and select baseline characteristics for the development and validation of a risk score for the prediction of 3 year mortality.

Results: A total of 116 subjects died during follow-up (median follow-up duration of 26 months). Baseline characteristics associated with mortality were: age (hazard ratio (HR) 1.44, 95% confidence interval (CI)1.02 to 2.04), male sex (HR 2.93, 95% CI 1.89 to 4.59), living alone (HR 1.53, 95% CI 0.99 to 2.38), body mass index (BMI) <18.5 kg/m(2) (HR 4.09, 95% CI 2.06 to 8.14), cardiovascular disease (HR 1.42, 95% CI 0.94 to 2.15), elderly mobility score <20 (HR 1.92, 95% CI 1.24 to 2.98), number of medicines > or =2 (HR 2.28, 95% CI 1.21 to 4.31), and impaired motor (HR 1.47, 95% CI 0.93 to 2.32) and process skills (HR 1.92, 95% CI 1.12 to 2.98) in activities of daily living. These predictors were translated into an easy-to-use frailty risk score and patients were stratified into very good (<45 points), good (45-50) moderate (51-55), poor (56-61) and very poor (>61) prognosis groups. Three year mortality rates across these groups were 8.0%, 15.9%, 25.9%, 41.5%, and 68.8%, respectively (p<0.001). The area under the receiver operating characteristic curve (AUC) of the risk score was 0.78 (95% CI 0.73 to 0.82), indicating good discriminative performance.

Conclusions: We developed and validated a risk score for the prediction of 3 year mortality. This risk score can be used to stratify patients into different risk categories, thereby informing patient counselling and tailored diagnostic and therapeutic decisions in clinical practice.