E375: An Electronic Health Record-Based Nudge Intervention to Reduce Low-Value Axillary Surgery in Older Women with Breast Cancer: A Nonrandomized Controlled Trial

The Choosing Wisely recommendations advocate against routine use of SLNB for axillary staging in older women with early-stage, ER+/cN0 breast cancer. However, despite a multitude of data in recent years describing SLNB as a low-value surgery, high rates of SLNB use persist.

Methods:

We sought to evaluate whether an EHR-based nudge reminder targeting surgeons in the first outpatient visit with patients meeting Choosing Wisely criteria decreases rates of SLNB. This was a mixed-methods, hybrid type I effectiveness study within a single healthcare system comprised of a pre-study survey on acceptability, appropriateness, and feasibility of the intervention, a prospective, nonrandomized controlled study (NCT06006910) of the nudge intervention itself, and post-study semi-structured interviews with each of the participating surgeons. The study included a 12-month preintervention (control) period and a 12-month intervention period. Eligible surgeons were fellowship-trained faculty who saw new patients with breast cancer at 8 clinics across the region. Eligible patients were those that had a new consultative encounter with the breast surgical oncologist and had a tumor meeting the Choosing Wisely criteria for possible omission of SLNB (cT1-2, N0). The primary endpoint of the study was the change in the rate of SLNB usage after deployment of the nudge intervention.

Results:

The participating surgeons scored the designed intervention highly in a survey on acceptability, appropriateness, and feasibility. Similar numbers of patients were seen by the group of surgeons before (control period; n=194) and after (intervention period; n=193) nudge deployment. Patients seen in the control and intervention period had similar ages at diagnosis, degrees of frailty, and tumor characteristics (ER H-score, PR H-score, stage, grade, and Ki-67). Compared with the control period, unadjusted rates of SLNB decreased by 22.6 percentage points (45.8% SLNB rate in the pre-nudge period to 23.2% after; 95% CI -29.2 to -16.0) in the intervention period (Fig. 1). An adjusted interrupted time series model showed a statistically significant level decrease in the rate of SLNB following nudge deployment (p = 0.005). Semi-structured interviews to identify barriers and facilitators to nudge usage are ongoing.


Conclusions:

A 12-month EHR-based nudge intervention significantly decreased routine use of SLNB in older women who met Choosing Wisely criteria. Future work is aimed at expanding nudge use to other patient groups that may not benefit from axillary surgery.