Presurgical Frailty Assessment Linked to Improved Survival at 1 Year

Implementing a frailty screening initiative before elective surgery was associated with more referrals of frail patients for additional evaluation and improved survival at 1 year, a quality improvement study showed.

After controlling for confounding factors, there was a 22% reduction in the odds of mortality at 180 days post-surgery after an Epic Best Practice Alert (BPA) was implemented (OR 0.78, 95% CI 0.67-0.91) and an 18% reduction at 1 year (OR 0.82, 95% CI 0.72-0.92, P<0.001 for both), reported Daniel E. Hall, MD, MDiv, MHSc, of the University of Pittsburgh Medical Center (UPMC) Presbyterian, and co-authors.

An interrupted time series analysis showed that 1-year mortality was increasing at a rate of 0.12% per month before BPA implementation, but began decreasing by 0.04% per month after implementation. For patients who triggered the BPA, the estimated 1-year mortality rate changed by -4.2% (95% CI -6.0% to -2.4%), the authors noted in JAMA Surgery.

“These observations demonstrate that the interventions were both effective and sustainable, 2 key characteristics in determining the utility of quality improvement efforts within hospital systems,” Hall and team wrote.

He and his colleagues also found that, after BPA implementation, the proportion of frail patients referred to a primary care physician (9.8% vs 24.6%) or a presurgical care clinic (1.3% vs 11.4%) increased significantly (both P<0.001).

A previous study by the same researchers showed that implementing a widespread screening tool for frailty, known as the Risk Analysis Index (RAI), was feasible in large hospital systems and reliably estimated postoperative outcomes.

“Despite this documented success with respect to implementation of routine frailty screening, there remains a critical knowledge gap regarding whether efforts to screen and mitigate frailty-associated risk translate into improved outcomes at the level of the screened population,” Hall and colleagues wrote.

In an accompanying commentary, Deborah M. Stein, MD, MPH, of the University of Maryland School of Medicine in Baltimore, and colleagues acknowledged that while frailty assessment is an important tool to help manage the care of older patients, “we wonder if this system-level intervention is replicable outside of the University of Pittsburgh Medical Center. The infrastructure required to achieve sustained success includes a system-wide buy-in of multiple stakeholders, staff education, adaptation of screening tools into the workflow, and surgeon compliance, which may not be universally feasible.”

They also raised questions about frail patients being less likely to be offered elective surgery, and the long-term outcomes of those who were not offered surgery. “Although the nonrandomized controlled trial design aimed to account for confounders and the interrupted time series analysis aimed to control for secular trends in postoperative mortality, selection bias remains a concern,” they noted.

For this study, surgeons at all UPMC clinics in nine specialties were incentivized to complete RAI assessments starting in July 2016 for all patients considering elective surgery. Beginning in July 2017, additional BPAs alerted clinicians to patients with an RAI of 42 or greater.

A total of 50,463 patients with 1 year of postsurgical follow-up were included — 22,722 before implementation and 27,741 after. Mean age was 56.7, 57.6% were women, and 87.8% were white. Patient characteristics were similar between the time periods.

Of the patients, 1,367 had an RAI of 42 or greater, which triggered the frailty BPA, and surgeons acknowledged the alert in all cases.

Because the study was not randomized, Hall and team noted that they could not draw causal conclusions and there may be unobserved confounders. They also didn’t account for specific components of the services provided by primary care physicians and presurgical care clinics.

Also of note, they were unable to quantify whether their intervention influenced decisions to choose nonoperative care for frail patients.