<scp>eFrailty</scp> : Making frailty assessment accessible to clinicians and researchers

“Which frailty assessment tool should I use?” This is the most frequently asked question from clinicians and researchers who are aware of the concept of frailty but are unfamiliar with the dozens of definitions and tools available for assessment. Even when one chooses a specific tool, locating it on the web and following instructions for scoring and interpretation can be cumbersome at the point of care. As geriatricians, researchers, and educators who measure, study, and teach frailty, we felt the need to fill this gap by creating a “one-stop shop” for commonly used frailty assessment tools. We are excited to introduce the product of our efforts over the past two years—eFrailty.org (Figure 1). “Our main goal in creating eFrailty is to make frailty assessment accessible to clinicians and researchers, within and outside geriatrics and aging research.” To achieve this goal, we aimed to (1) summarize the features of commonly used, validated frailty assessment tools; (2) create web-based calculators to allow real-time scoring and interpretation at the point of care; and (3) help users choose the most appropriate tool for their clinical situation. First, we curated 15 commonly used and validated frailty assessment tools (Table 1) through team consensus, informed by the literature (original research articles reporting each tool)1 and our clinical experience. Since the deficit accumulation frailty index (FI) has been operationalized in various ways due to adaptations made depending on the clinical and research datasets used, we chose the version based on items routinely collected in a comprehensive geriatric assessment (CGA).2 For each assessment tool, we indicate the average time to administer each assessment tool, unique features, number of test items, and included domains. The site also features a crosswalk to allow comparison of scores across nine common frailty assessment tools.3 Second, we created web-based interactive calculators for scoring and interpreting each frailty tool within the website. For example, for the Fried physical phenotype, we applied sex- and height-specific cutpoints for slow gait, sex- and body mass index-specific cutpoints for weak grip strength, and kilocalorie cutpoints from selected items of the Minnesota Leisure Time Activity Questionnaire to determine low physical activity.4, 5 These cutpoints are automatically incorporated within the calculator such that, without a calculator, manually identifying these cutpoints and computing scores in clinical practice can be otherwise challenging and time-insensitive. Third, we developed an algorithm to guide the choice of frailty assessment in different clinical settings (Figure 2). We identified three clinical scenarios: screening or brief risk assessment (“does this patient have frailty or an elevated risk for adverse health outcomes?”), comprehensive assessment and care planning (“what interventions and supportive services does this patient need to reverse frailty or prevent adverse health outcomes?”), and risk assessment before stressful treatment (“what is this patient's risk for experiencing adverse outcomes after surgery or chemotherapy?”). For screening in primary care or acute care settings, a simple tool that can be completed within 3–5 min based on self-report with or without a brief physical performance test (e.g., chair stands) is appropriate. For comprehensive assessment and care planning, tools based on a CGA or similar multi-domain assessment, which often require cognitive tests and physical performance tests, are needed to identify modifiable causes of frailty and tailor interventions to prevent frailty progression and optimize health. For risk stratification before stressful treatment, such as surgery or chemotherapy, it is preferable to choose a tool validated in the patient population undergoing the treatment of interest. In each scenario, we recommend one main tool and several alternative tools which users can opt to choose from. Users should be aware that the assessment tools included on eFrailty are not intended to be a comprehensive list of validated frailty assessments. We did not include frailty scores that are solely calculated from existing healthcare data, such as claims-based FIs and electronic FIs.8 Because the calculators were created by our team based on the information reported in the published articles, any revisions or updates to the tools after the publication of the original articles might not be reflected. The frailty tool selection algorithm (Figure 2) for different clinical scenarios is based on the opinions of the authors, informed by our collective clinical experience, rather than research evidence. Therefore, the content of eFrailty should be used as a guide but not as a replacement for professional judgment. In conclusion, eFrailty.org is designed to address the pressing need of having frailty assessment tools readily accessible at point of care. This resource will empower clinicians to discuss frailty with their patients, personalize clinical management, and facilitate informed decision-making about treatment based on physiologic vulnerability, rather than chronological age. It will also enable researchers to conduct valid assessments of frailty, generating clinically relevant evidence for older adults at risk for or living with frailty. Furthermore, eFrailty can be useful in training and educating medical students, trainees, and clinicians by providing a centralized resource on frailty assessments. We hope eFrailty represents a significant step forward in making standardized frailty assessments possible, an essential step toward frailty-guided and person-centered clinical care. Drafting of the manuscript: Dae Hyun Kim. Critical revision of the manuscript for important intellectual content: Dae Hyun Kim, Megan Cheslock, Stephanie M. Sison, Ariela R. Orkaby, and Andrea W. Schwartz. Administrative, technical, or material support: Dae Hyun Kim, Ariela R. Orkaby, and Andrea W. Schwartz. Study supervision: Dae Hyun Kim. The development of eFrailty was a collaborative team effort of Megan Cheslock, MD, Dae Hyun Kim, MD, ScD, Yuchen Liu, MS, Natalie Newmeyer, MS, RD, LDN, Ariela R. Orkaby, MD, MPH, Vaishnavi Raman, MD, Andrea W. Schwartz, MD, MPH, Stephanie M. Sison, MD, MBA, Saran Thanapluetiwong, MD, and Lily Zhong, BA. The authors are thankful to Alexander Smith, MD, MPH, MS, and Eric Widera, MD, who inspired us and shared their experiences of creating eprognosis.org. Dr. Kim receives grants from the National Institute on Aging for unrelated work and received personal fee from Alosa Health (ended on 12/31/2022) and VillageMD (ended on 12/13/2022). The views expressed in this manuscript are those of the authors and do not necessarily reflect the position or policy of the Hebrew SeniorLife, the Department of Veterans Affairs or the United States government. The funding sources had no role in the design, collection, analysis, or interpretation of the data or the decision to submit the manuscript for publication. The eFrailty website was created with the funding support and resources from the Hinda and Arthur Marcus Institute for Aging Research at Hebrew SeniorLife. In addition, some of this material is the result of work supported with resources and the use of facilities at the Veterans Affairs Boston and Bedford Healthcare Centers and the New England Geriatric Research Education and Clinical Center, as well as a Clinical Innovation award from the VA Office of Geriatrics and Extended Care.