Improving Surgical Outcomes Through Frailty Screening: An Overview of the Risk Analysis Index

Reports of Complications Related to Surgery or Invasive Procedures

More than 31,000 events involving complications related to a surgery or invasive procedure were reported to the Pennsylvania Patient Safety Reporting System (PA-PSRS) in the five-year period between January 1, 2020, and December 31, 2024^[1]. Of those events, more than 15,000 resulted in serious patient harm or death.

Patient Frailty and Preoperative Use of the Risk Analysis Index Tool

Patient frailty is a significant predictor of postoperative morbidity and mortality.^1–5 The collective body of research over recent years has made a strong argument for preoperative use of a Risk Analysis Index (RAI) tool to evaluate the patients for frailty, as a proxy for physiologic reserve.^3,5,6 The RAI tool can be completed by the patient/representative or staff and the scores are then tallied to estimate the degree of patient frailty. Higher RAI scores indicate that a patient would have greater frailty-associated risks. For ease of interpretation, the RAI scores are often grouped into the following four categories of patient status: robust, normal, frail, or very frail.

This tool is intended to be used with patients who are potential candidates for nonemergent or elective surgeries. Use of the RAI to identify patients who are frail or very frail may guide the surgeon and patient to consider preoperative rehabilitation (i.e., prehabilitation) or to not have surgery (i.e., opting for nonoperative treatments for symptom management, sometimes including palliative care).^3–5,7–10 This decision among frail patients could reduce the likelihood of a morbidity and premature mortality.^3,4,7–10 Overall, the use of RAI and related interventions have been associated with improved long-term outcomes for frail patients.^9,11

Background and Considerations for Use of the Risk Analysis Index

Validity and Predictive Ability of RAI

• The RAI tool has been validated for point-of-care use for clinicians to screen patients for frailty.^4,6,8,11,12 Readers should note that numerous other frailty indices exist, but some are not suitable for point-of-care use (e.g., other frailty indices use dynamometers and walking tracks, were designed to be used retrospectively, or impose excessive respondent burden).^4,6,13,14

• The RAI tool is effective for identifying approximately the 10% of patients that are at greatest risk for postoperative morbidity and mortality, due to their extent of frailty.^15,16

• The RAI has been validated to predict postoperative mortality¹³ across 30-,¹⁰ 90-,³ 180-,^8,10 and 365-day³ time frames.

• Studies demonstrated the predictive ability of the RAI across a range of clinical context,^4,8,9 including the following surgical specialties: oncology,¹⁶ spine,^3,5 cardiovascular,¹⁷ urology,¹⁸ neurology,^10,19 plastic,¹⁰ otolaryngology,¹⁰ orthopedic,¹⁰ gynecology,¹⁰ vascular,¹⁰ and thoracic.¹⁰

• Research indicates that the RAI can and should be used with patients of all ages¹⁷ and across the full range of surgical procedures, including procedures of low physiological stress (i.e., operative stress scores).^10,20,21

• In a study of both young and old patients, those with higher preoperative RAI scores (i.e., greater frailty) were associated with significantly more postoperative days away from home (e.g., rehabilitation, skilled nursing facilities) and a higher likelihood of losing independence after surgery.²²

• The RAI demonstrates greater predictive ability for both morbidity and mortality when compared with the five-factor modified frailty index^[2] (mFI-5)^5,24 and has similar predictive ability when compared to the other indices¹³; Fried frailty phenotype (FFP),²⁵ Edmonton Frail Scale (EFS),²⁶ and the Canadian Study of Health and Aging.²⁷

Versions and Access to RAI

• There are several similar versions of the RAI tool (e.g., RAI-A, RAI-C, RAI-ICD, and RAI-VQI) and the primary differences are the data sources used to estimate patient frailty.^4,12,28

• For most clinical applications, we suggest using the RAI-C version, based on its ease of use at the point-of-care.

• The RAI-C is available as a printable version and to users of Epic (through EpicShare),²⁹ Cerner, REDCap, Microsoft’s PowerApp, and Veteran Affairs’ Computerized Patient Record System (CPRS).^12,15,28

• To access the RAI-C and for information about the other RAI versions, see reference 28.

Implementation and Features of RAI-C

• For a user guide and supporting materials, see references 12 and 28.

• The RAI-C is intended to be used preoperatively within the existing workflow to estimate patient frailty.¹²

• The tool targets the following five domains of health: physical, functional, social, nutritional, and cognitive.^4,6

• The RAI-C consists of 14 questions^3,6,9,28 and is available in English, Spanish, Portuguese, and Chinese languages.²⁸

• The questions on the tool can be completed by staff, but are most often completed by the patient/representative, which will take them less than two minutes.¹²

• The score from a completed tool can be quickly tallied manually or with an online calculator (available at efrailty.hsl.harvard.edu/ToolRiskAnalysisIndex.html) by a range of staff, including medical assistants and nurses.^6,8,12,15 Staff have reported calculating the RAI-C score in a median of less than 40 seconds per patient.^8,13

• The cutoff scores used to classify frailty can be adjusted to fit the setting and available resources.⁴

• The RAI-C score, ranging from 0 to 81, will categorize the patient as being robust (≤29), normal (30–36), frail (37–44), or very frail (≥45).¹²

Treatment Decisions and Clinical Actions

• Patients identified through the RAI as being frail or very frail should then participate in a more thorough assessment that would be used to further inform patient and provider decisions.^6,15 The follow-up could target multiple areas of frailty through various tests, such as geriatric assessment, serological biomarkers, and functional performance (e.g., gait speed, grip strength).^6,15

• Patients and providers engage in shared decision-making based on the predicted trajectory of frailty-associated risks^4,6,9,11,22 and the patient’s health-related goals.¹⁵

• Depending on the degree of patient frailty and their goals, the patient may choose not to proceed with the surgery and instead opt for medical symptom management, sometimes including palliative care. Other patients may choose to prepare for surgery with preoperative rehabilitation (i.e., prehabilitation).^9,13 Use of preoperative rehabilitation may increase the odds of postoperative recovery at home, as opposed to being postoperatively placed in a rehabilitation facility.¹⁵

• Frailty-associated risks can be reduced through optimized care and prehabilitation,^3,7,9,11,15 which may include glycemic control, anemia management, adjustment of medication, multimodal anesthesia, nutritional supplementation, evaluation of home support, care coordination, respiratory muscular training, balance and strength training, cardiovascular exercise, and use of upper body ergometers.

Disclosure

Matthew A. Taylor has no conflicts of interest; Daniel E. Hall has an unpaid consulting relationship with FutureAssure, LLC, a medical device company seeking to facilitate point-of-care frailty assessment.

This article was previously distributed in an April 9, 2025, newsletter of the Patient Safety Authority, available at https://conta.cc/3GdjSNN.

About the Authors

Matthew A. Taylor (MattTaylor@pa.gov) is a research scientist on the Data Science & Research team at the Patient Safety Authority, where he conducts research, uses data to identify patient safety concerns and trends, and develops solutions to prevent recurrence.

Daniel E. Hall (Daniel.Hall2@va.gov) is a general surgeon and health services researcher based in Pittsburgh with national responsibility in the National Surgery Office for the Surgical Pause program. Informed by his clinical practice, his research and administrative efforts focus on improving the perioperative management and outcomes of high-risk, frail veterans considering surgical treatment.