Abstract
BACKGROUND: The Clinical Frailty Scale is an ordinal scale from 1 (very fit) to 9 (terminally ill) commonly used to assess frailty in older patients. It is simple for clinicians to apply and can help identify patients who may benefit from discussions around end-of-life care. We externally validated the Scale to assess its performance for predicting 90-day mortality in a cohort of admitted older patients who had screened positive for being at risk of nearing the end of life. METHODS: We used data from a randomised controlled trial assessing a tailored feedback loop for reducing non-beneficial care. Our study included patients aged 75 and above admitted between May 2020 and June 2021 from 3 Australian hospitals. We assessed whether the Scale could be used in a frail population to identify patients who were at risk of short-term mortality. Predictive performance was assessed using the c-statistic, smoothed calibration curves and decision curves. Models were tested for coefficient stability. RESULTS: Our dataset contained 4639 patients and 956 deaths within 90 days. The Clinical Frailty Scale had a c-statistic of 0.62 (95% CI 0.60 to 0.64) or 0.63 (95% CI 0.61 to 0.65) by adding age and transforming the Scale using a cubic spline. Risks were underestimated without a non-linear transformation as scores of 8 and 9 had a higher risk that diverged from a linear association. The net benefit of using the tool was greatest between 5 and 8 on the Scale. CONCLUSIONS: The utility of the Clinical Frailty Scale may be as a flag to encourage clinicians to become more comfortable with discussing ageing and death, rather than as a highly discriminating model to classify patients as high risk or low risk. Statistical uncertainty over mortality should not be a barrier to initiating end-of-life care discussions with frail older patients.