Evaluation of a commonly used tool: Does the 5-item frailty index predict phenotypic frailty?

Background: The simplified frailty index (sFI) is a commonly used instrument to estimate postoperative risk, but its correlation with phenotypic frailty has been questioned. This study evaluates the relationship between sFI and phenotypic frailty, as measured by the Sinai Abbreviated Geriatric Evaluation (SAGE).Methods: Charts were retrospectively reviewed from patients >= 75 years old who underwent surgery between 2012-2022. The sFI score was calculated by adding 1 point for hypertension, COPD, congestive heart failure, functional dependence, and diabetes (score 0-5). SAGE was calculated by adding 1 point for normal gait speed, normal Mini-Cog (c), and independent activities of daily living (ADL) (0-3). Spearman rank correlation was used to test the relationship between sFI and SAGE. SAGE components were used as binary-dependent outcomes in covariate-adjusted logistic regression modeling to evaluate associations with sFI scores while adjusting for potential confounders.Results: 334 patients were assessed, with a mean age of 84.0. SAGE and sFI scores were significantly associated, with a modest inverse relationship (r=-0.24, p<0.0001). Each 1-point increase in sFI score was associated with increased odds of ADL deficit (OR 2.3, 95%CI [1.5-3.8], p<0.0001) and abnormal gait speed (OR 1.9, 95%CI 1.2-3.0, p<0.01). The sFI score was not associated with deficits in the Mini-Cog (OR 1.5, 95%CI [0.96-2.3], p=0.07).Conclusion: Higher sFI was significantly associated with increased phenotypic frailty, particularly with the loss of physical condition and function but not associated with cognitive deficit. Therefore, sFI may not be an appropriate tool to estimate postoperative complications related to cognition, such as delirium risk.

Automated summary

This study evaluated the relationship between the simplified frailty index (sFI) and phenotypic frailty. The results showed a significant but modest inverse correlation between sFI and phenotypic frailty, indicating that sFI may not be suitable for assessing postoperative complications related to cognition.