Numerous instruments have been developed to evaluate frailty in hospitalized patients, including several versions of the Risk Analysis Index (RAI) among surgical patients and Canadian Study of Health and Aging Clinical Frailty Scale (CSHA-CFS) among medical patients.

Two versions of the RAI, the clinical (RAI-C) and administrative (RAI-A), were developed and validated in a 2017 study to evaluate frailty among surgical patients.¹ Variables included in the RAI-C were selected from the Minimum Data Set Mortality Risk Index-Revised using stepwise logistic regression to identify those that most reliably predicted 6-month mortality. The RAI-A was comprised of variables in the Veterans Affairs and American College of Surgeons National Surgical Quality Improvement Project (VASQIP and ACS-NSQIP) that most closely corresponded to variables in the RAI-C.

The RAI-C, scored on a scale of 0 to 81, was validated using data from 6856 patients at outpatient Veterans Health Administration surgery clinics in Nebraska and Western Iowa between July 2011 and September 2015. Information for 2785 patients with available VASQIP and ACS-NSQIP data was used to validate the RAI-A. Primary outcomes were 30-day, 180-day, and 365-day mortality. Secondary outcomes included postoperative complications. Instrument performance was assessed using C statistics, sensitivity, and specificity.

Among the 2785 patients with data available to calculate both the RAI-C and RAI-A, the C statistics for predicting mortality ranged from 0.744 to 0.824 for the RAI-C and from 0.797 to 0.901 with the RAI-A. Using a cutoff of 21 points utilized in prior studies, the RAI-C had a sensitivity of 50% and specificity of 82%, compared to a sensitivity of 25% and specificity of 97% for the RAI-A. Study caveats included potential bias from incomplete morbidity and mortality data.

The CSHA-CFS was developed and validated as part of a 5-year prospective cohort study conducted among 2305 community dwelling adults ≥65 years of age without dementia.² CSHA-CFS is a 7-point scale of increasing frailty that was derived from the authors’ earlier theoretical model of fitness and frailty (Table 28.1). Study physicians examined and assigned a CSHA-CFS score for each study participant. Primary outcomes were death and need for institutional care.

Overall, higher CSHA-CFS scores were associated with the primary outcomes (HR 1.30 for death, 95% CI 1.27-1.33; HR 1.46 for institutional care, 95% CI 1.39-1.53; P-value not reported). Incremental 1-category change along the CSHA-CFS scale was associated with mortality (21%, 95% CI 13%-31%; P-value not reported) and need for institutional care (24%, 95% CI 9%-41%; P-value not reported). Study caveats include a disproportionately large number of individuals in the study cohort with cognitive impairment and living in institutional facilities.

A 2013 consensus statement³ from six international professional societies recommend screening for frailty using a validated tool (without preference among available tools) in all adults >70 years of age.