Menu
Menu
Dozens of frailty instruments have been developed and described.12 For the purposes of this module we will focus on five instruments that have been most commonly used and studied in the perioperative setting and that align with consensus on the multidimensional nature of frailty. While frailty is often thought of and assessed as a syndrome that is strictly present or absent based on a specific cut-off, in reality frailty is best thought of as a continuum where higher scores on a frailty instrument mean more severe frailty, which indicates greater risk of adverse outcome. Score ranges and cut-offs will be discussed below for each instrument.
Time: 5-20 minutes to administer; significantly longer to complete compared to the CFS
Resources: Hand-held dynamometer (for grip strength); space and timer for 15-foot walk; activity questionnaire
Other: Rated as being moderately difficult to use in preoperative clinic and less easy to use than the CFS. Adaptations (eg, FRAIL Scale20) exist to decrease measurement burden (replaces formal measurements with patient questions)
The Fried Phenotype2 is the most studied frailty instrument in surgical patients and uses five variables to define frailty. These include two questions asked directly to the patient (presence of unintentional weight loss (>10 lbs in the last year) and feeling easily exhausted). Three measurements are also required. Gait slowness is measured using a 15-foot timed walking test (slowness is defined as the lowest 20%, normalized for sex and height), grip strength using a hand-held dynamometer (weakness is defined as lowest 20% by sex and body mass index) and low activity levels using a standardized questionnaire (defined as <383 kilocals/week for men or <270 kilocals/week for women). These variables are used to construct a score ranging from 0-5; frailty is present with a score >3, while individuals with a score 1-2 are considered pre-frail.
Time: 10-13 minutes to complete
Resources: Does not require extra space or instruments
Other: Tends to tolerate some missing data points. Can be constructed using electronic health record data in settings with adequate access to multidimensional healthcare data.15 16
The frailty index13 is widely used in perioperative patients and non-surgical patients as well. A frailty index is constructed by assessing at least 30 variables that cover multiple frailty-related domains (physical performance, nutritional status, mental health, cognition and others).14 Importantly, the specific variables used are not as important as having an adequate number that represent multiple domains.13 Variables are considered as deficits and are typically measured as present (1 point assigned) or absent (0 points assigned). Deficits can be assessed via direct patient interview, questionnaires or (where adequate data are available) electronic health data.15 16 The number of points assigned is divided by the number of deficits measured to provide a score ranging from 0 (no frailty) to 1 (completely frail); however, frailty index scores exceeding 0.66 are rarely seen. A typical cut off point to define someone as frail would be >0.21. Categorizations are often applied (<0.10 not frail, 0.10-0.21 pre-frail, 0.22-0.44 frail and >0.45 severely frail).
Time: Add less than 1 minute to a clinical assessment, significantly faster than Fried Phenotype.
Resources: Does not require extra space or instruments
Other: Rated as very easy to use and logistically appropriate in a preoperative anesthesiology clinic. Feasibility data most strongly support use of the CFS compared to other instruments. Can be assessed via chart review or proxy history.21
The Clinical Frailty Scale (CFS)1 is a clinically-oriented frailty instrument based on clinical assessment and judgement combined with visual cues and brief vignettes. Vignettes prompt assessment of activity levels, disease symptoms, medical problems, activities of daily living and cognition. Although designed to be simple and feasible for clinical use, the CFS explains 80% of the information contained in the more complex FI. Scores range from 1 (very fit) to 8 (very severely frail). The ninth category denotes individuals nearing end of life who are not otherwise showing signs of frailty (e.g., otherwise well patient with end-stage cancer) and is rarely applicable in preoperative assessment of older people. Typically, the CFS cut off for assigning frailty is a score >4 or >5. Various categorizations have been applied, including 1-3 (no frailty), 4-5 (vulnerable or mildly frail), >6 (moderate to severely frail).
Time: 1-5 minutes to complete
Resources: Paper, pen and assessor needed for clock draw test, space and chair for timed up and go
Other: An adaptation, called the reported EFS, replaces clock draw and walk tests with patient or proxy queries.22
The Edmonton Frail Scale (EFS) is a 10-domain frailty assessment that assess cognition, general health, functional independence, social support, medications, nutrition, mood, continence and lower limb function.17 Nine questions are directly answered by patient or proxies. Cognition requires completion of a clock-draw test, while lower limb function is assessed using a timed up and go test. Answers to questions are graded on a 0-1 or 0-2 point scale, with the overall score ranging from 0 (no frailty) to 17 (very frail). A typical cut off point to assign frailty is a score of >8, while a three-level categorization has been used (0-3, no frailty; 4-7 pre-frailty; >8 frailty present).
Time: 2 minutes or less, longer in people with greater frailty23
Resources: Does not require extra space or instruments
Other: Clinical and database versions exist.18 Has greater weighting when cancer diagnoses present than other instruments.
The Risk Analysis Index (RAI) is a multidomain frailty instrument that assigns points for the presence and/or combination of specific demographic, comorbid, oncologic and disability states.18 The tool is an adaptation of the Minimum Data Set Mortality Risk Index, which used frailty-related variables to develop a smaller set of indicators that can be obtained from patients (or their surrogate). A clinical tool (RAI-C) and electronic database tool (RAI-A) have been developed, and in each higher scores suggest greater frailty. The RAI questionnaire includes 14 questions assessing 11 variables and 2 statistical interactions with scores ranging from 0 to 81. Patients are typically categorized as robust (RAI≤20), normal (21-29), frail (30-39), and very frail (≥40). Use of a recent recalibration of RAI domains may improve performance.19
The accuracy of frailty instruments can be considered from two perspectives. Since there is no single gold-standard definition of frailty, we can’t compare each instrument to a common reference to determine how accurately each diagnoses frailty. This means that frailty instruments are assessed to ensure that they are 1) consistent with the agreed-upon multidimensional aspects of frailty, and 2) predictive of adverse outcomes relevant to older adults. This part of the module will primarily focus on point 2 (that is, prediction of outcomes), specific to the frailty instruments identified above.
Mortality: Recent systematic reviews show that each of the frailty instruments discussed above are associated with a higher risk of death after surgery, and that none is markedly more accurate that the others.12 The CFS was most strongly associated with mortality (odds ratio (OR) 4.89, 95%CI 1.83-13.05) across 6 studies, followed by the FP (OR 3.95, 95%CI 2.00-7.81) with data pooled across 10 studies.
All studied instruments were moderately effective at discriminating between high and low risk of death patients (area under the receiver operating curve values ~0.7) and added extra discriminative information when assessed in addition to traditional risk factors like age, ASA score and others.12 The RAI has consistently demonstrated higher discrimination for mortality (area under the receiver operating curve values often>0.8) than other instruments, although head-to-head comparisons are lacking.
Complications: All described frailty instruments also predict a higher risk of complications. The EFS had the strongest association (OR 2.92, 95%CI 1.52-3.46), followed by the FP (OR 2.47, 95%CI 2.00-3.04), the FI (OR 2.29, 95% CI 1.52-5.65) and CFS (OR 1.68, 95%CI 0.95-2.95). The FI may be the most effective frailty tool for discriminating between older people at high or low risk of experiencing a complication.12
Need for non-home discharge: The CFS was most strongly associated with not being discharged home after surgery (OR 6.31, 95%CI 4.01-9.36) followed by the FP (OR 5.18, 95% CI 3.34-8.03) and FI (OR 2.29, 95% CI 1.52-3.46). The CFS may provide more useful information related to providing older people with accurate estimates of how likely non-home discharge is compared to the FP and FI.12
Delirium: The FP (OR 3.79, 95% CI 1.75-8.22) and EFS (OR 2.11, 95% CI 1.06-4.21) both predict the risk of delirium after surgery. More data is needed for the CFS, RAI and FI.12
Because there is no single ‘best frailty’ instrument, your choice of tool for your practice will need to consider the outcomes you are trying to improve, as well as the setting you work in and resources available. Below are comments on feasibility and practical tips for each frailty tool assembled through a systematic search of the available literature.
Step 3: View the Case Studies ›
Step 4: View Key Papers and References ›
Step 5: See the Infographic Guides ›
‹ Back to Step 1: Frailty Introduction ›
You do not have to make a payment at this time, but you can reach out to your practice/group leader to find out when they plan to pay your dues. Email or call the ASA Support team for assistance: (630) 912-2552 or info@asahq.org.
