Combined use of the Clinical Frailty Scale and laboratory tests in acutely hospitalized older patients.

Aims: To evaluate the Clinical Frailty Scale (CFS) and a Frailty Index based on laboratory tests (FI-lab) in terms of what each assesses about frailty and to determine the appropriateness of combined use of these two frailty scales.

Methods: This was a prospective observational cohort study in an acute geriatric ward of a university hospital. The FI-lab is the proportion of laboratory parameters that yield abnormal results from a total of 23. The FI-lab and CFS were assessed at admission. Data on activities of daily living (ADL), cognition, geriatric syndromes, and comorbidities were also collected. Main outcomes were in-hospital mortality and 90-day mortality after admission.

Results: In total, 378 inpatients (mean age 85.2 ± 5.8 years, 59.3% female) were enrolled. ADL and cognition correlated strongly with the CFS (Spearman's |r|> 0.60) but weakly with the FI-lab (|r|< 0.30). Both the CFS and FI-lab correlated weakly with geriatric syndromes and comorbidities (|r|< 0.40). The correlation between the CFS and FI-lab was also weak (r = 0.28). The CFS and FI-lab were independently associated with in-hospital mortality and 90-day mortality after admission. The Akaike information criterion was lower for models using both the CFS and FI-lab than for models using either tool alone.

Conclusions: The CFS and FI-lab each reflected only some of the aspects of frailty in acutely hospitalized older patients. The model fit was better when the two frailty scales were used together to assess the mortality risk than when either was used alone.