Effect of Electronic Behavioral Nudges on Low-Value Preoperative Testing
Catherine A. Sarkisian, Chad Wes Villaflores, Sitaram S. Vangala, Eric M. Cheng, Suzanne Bliven Shu, Noah J. Goldstein, Ira S. Hofer, Victor F. Duval, John D. Bartlett, James M. Moore, Antonio M. Pessegueiro, Samuel A. Skootsky, Cheryl L. Damberg, John N. MafiImportance
Robust randomized trials have demonstrated that routine testing before cataract surgery does not improve outcomes; however, despite this evidence and long-standing professional society recommendations discouraging routine testing, preoperative testing remains frequent in real-world practice, with most patients undergoing cataract surgery receiving at least 1 preoperative test.
Objective
To measure whether electronic health record (EHR)–embedded, behavioral science–informed nudges reduce preoperative testing.
Design, Setting, and Participants
This pragmatic randomized clinical trial was conducted at a large academic health system (UCLA Health) in Los Angeles, California. Physicians performing preoperative evaluations were randomized to 1 of 3 EHR nudge alerts or usual care. All encounters with patients undergoing a preoperative visit with a randomized physician within 30 days of cataract surgery were silently enrolled. Patients were enrolled from June 24, 2021, to April 11, 2023. Data were analyzed from June 2020 (1 year prior to first enrollment) to April 2024 (1 year after last enrollment).
Interventions
The EHR interruptive alerts cited professional society recommendations against routine testing; each emphasized a different potential patient harm, including lack of safety benefit, financial burden, or psychological stress.
Main Outcomes and Measures
The primary outcome was the change in the proportion of preoperative cataract surgery encounters with any signed testing order(s) between the 12-month period before the physicians’ first encounter (enrollment) and the 12-month postencounter period. Secondary outcomes included order cancellation rates among clinicians exposed to the intervention, separate comparisons of each nudge, and analyses comparing orders for specific categories of tests.
Results
During the enrollment period, 229 of 776 randomized physicians had 1 or more eligible encounter during which a patient was enrolled. Together, these clinicians had 1045 preoperative encounters during the intervention period with 967 unique patients. Mean (SD) patient age was 72.1 (9.1) years, and 595 encounters (56.9%) were with female patients. Mean (SD) signed preoperative testing rates decreased from 83% (29%) to 79% (31%) in the combined intervention arms vs 84% (30%) to 82% (31%) in the control arm. In mixed-effects models, exposure to any nudge was not associated with a reduction in overall testing (adjusted odds ratio [aOR], 0.68; 95% CI, 0.34-1.34). In prespecified secondary analyses, 16% of orders were cancelled following exposure to a nudge, and the psychological stress nudge was associated with decreases in rates of complete blood cell count (aOR, 0.37; 95% CI, 0.18-0.75; P = .01) and electrocardiogram orders (aOR, 0.42; 95% CI, 0.21-0.83; P = .01) vs control.
Conclusions and Relevance
This randomized clinical trial of EHR nudges did not reach statistical significance on the primary outcome of overall preoperative testing. Secondary outcomes suggest future studies might consider the role of a psychological stress nudge in combination with other behavioral, workflow, and/or payment changes.
Trial Registration
ClinicalTrials.gov Identifier: