Background

The estimated prognos of a patient might influence the expected benefit/risk ratio of different interventions. The main purpose of this study was to investigate the Clinical Frailty Scale (CFS) score as an independent predictor of short-, mid- and long-term mortality in critically ill older adults (aged ≥ 70) admitted to the emergency department (ED). 

Methods

This is a retrospective, single-center, observational study, involving critically ill older adults, recruited consecutively in an ED. All patients were followed for 6.5–7.5 years. The effect of CFS score on mortality was adjusted for the following confounders: age, sex, Charlson’s Comorbidity Index, individual comorbidities and vital parameters. All patients (n = 402) were included in the short- and mid-term analyses, while patients discharged alive (n = 302) were included in the long-term analysis. Short-term mortality was analysed with logistic regression, mid- and long-term mortality with log rank test and Cox proportional hazard models. The CFS was treated as a continuous variable in the primary analyses, and as a categorical variable in completing analyses.

Results

There was a significant association between mortality at 30 days after ED admission and CFS score, adjusted OR (95% CI) 2.07 (1.64–2.62), p < 0.0001. There was a significant association between mortality at one year after ED admission and CFS score, adjusted HR (95% CI) 1.75 (1.53–2.01), p < 0.0001. There was a significant association between mortality 6.5–7.5 years after discharge and CFS score, adjusted HR (95% CI) 1.66 (1.46–1.89), p < 0.0001. Adjusted HRs are also reported for long-term mortality, when the CFS was treated as a categorical variable: CFS-score 5 versus 1–4: HR (95% CI) 1.98 (1.27–3.08); 6 versus 1–4: HR (95% CI) 3.60 (2.39–5.44); 7 versus 1–4: HR (95% CI) 3.95 (2.38–6.55); 8–9 versus 1–4: HR (95% CI) 20.08 (9.30–43.38). The completing analyses for short- and mid-term mortality indicated a similar risk-predictive value of the CFS. 

Conclusions

Clinical frailty scale score was independently associated with all-cause short-, mid- and long-term mortality. A nearly doubled risk of death was observed in frail patients. This information is clinically relevant, since individualised treatment and care planning for older adults should consider risk of death in different time perspectives.