What if email reminders to surgeons could help them avoid overprescribing opioid pills after surgery and unwittingly contributing to the opioid epidemic? In a new study conducted by RAND, Sutter Health, and several universities, surgeons who received email “nudges” to align their prescriptions with recommended amounts prescribed significantly less — and the reductions persisted for at least a year after the emails stopped. If widely adopted, this intervention could keep millions of opioid pills from being misused. This research brief describes the study’s findings and offers considerations for clinical leaders, hospital system administrators, and other stakeholders who might wish to use this intervention as a model in their own settings.
Surgeons prescribe opioids to help patients control pain after surgery, but the number of pills prescribed often exceeds what patients need. Extra pills fuel the opioid epidemic by presenting opportunities for misuse and diversion. Patients may use more than they need and develop persistent opioid use, and the pills can end up circulating in the community. An estimated 8.6 million people misused prescription painkillers in the United States in 2023, and almost 40 percent got the medication from a friend or relative.[1] These problems continue despite widespread accounts of the role of overprescribing in the opioid epidemic.
Guidelines should help prescribers right-size prescription amounts, yet not all clinicians follow them. Some clinicians’ prescribing behavior deviates further from guidelines than others’; understanding why could help address the causes of overprescription.
The researchers in this study used data from Sutter Health’s electronic health record (EHR) system to examine the factors that contributed to variations in the amount of opioids prescribed, both within and between surgeons. The researchers then used a cluster randomized controlled trial that targeted surgeons who were overprescribing with an email intervention that used two types of social norms feedback to nudge, or encourage, surgeons to prescribe within guideline amounts. After the email nudges were stopped, researchers explored surgeons’ opinions about the intervention and observed whether changes in prescribing behavior persisted (Table 1).
| Study Period | Activities |
|---|---|
| Pre-intervention (October 2019-September 2021) | -Baseline prescribing behavior observed -Surgeons randomized into three groups: one control and two intervention groups |
| Intervention (October 2021-September 2022) | -Two types of email nudges sent to intervention groups; control group received no emails -Emails sent from an email distribution system and typically signed by the surgeon's department chair, chief medical executive, or chief of staff |
| Post-Intervention (October 2022-September 2023) | -Email nudges turned off -36 surgeons across the three study groups interviewed -Post-intervention prescribing behavior observed |
*NOTE: Surgeons did not know they were being studied.
Researchers first sought to understand why some surgeons prescribe so much more than others. They studied the health system’s data on opioid prescription amounts related to six common types of orthopedic, obstetric/gynecological, and general surgeries performed at 24 hospitals over two years and looked for characteristics of patients, prescribers, and hospitals associated with prescribing variation.
Statistical analyses revealed that some of the variation in prescribing was medically warranted and some was not. Medically warranted reasons for larger prescriptions included certain aspects of the patient’s medical history (having a high body mass index, a chronic pain condition, or a history of opioid use) and the complexity of the surgery (estimated by how long it took to perform). But other factors — such as the patient’s race, the type of health insurance the patient had, the time of day when the patient was discharged, and the type of clinician who wrote the prescription — also accounted for significant portions of the variation in prescribing, even though they are not related to the patient’s level of need for opioids. After controlling for the medically warranted factors, almost half of the variation was driven by differences in prescribing habits between prescribers (46 percent).
Overall, 59 percent of the more than 42,600 surgeries performed had an opioid prescription recorded at discharge, and more than half of the prescriptions exceeded guideline amounts, translating to more than 13,000 patients receiving 10.6 extra opioid pills each. These results clarified the need to address differences in surgeons’ prescribing habits at the individual level.
To curtail prescription of excess pills, researchers conducted a cluster randomized controlled trial with 640 surgeons across 19 hospitals to investigate the potential to nudge prescribing behavior closer to guideline amounts. Surgeons in the two intervention groups who overprescribed at least two patients in a month received one of two email nudges informing them that their prescription amounts exceeded either (1) what their peers prescribed or (2) what was recommended by clinical guidelines. About 35 to 40 percent of surgical discharge prescriptions exceeded guideline-recommended amounts at baseline.
Surgeons who did not receive email nudges continued to overprescribe at a rate similar to baseline (37 percent), but surgeons who received email nudges overprescribed significantly less often (25 to 28 percent of the time). The effects of the intervention strengthened in its last few months of the study period, with even more pronounced effects among surgeons who performed more procedures and those who prescribed above guidelines at baseline. These reductions translate to about 42,000 fewer pills of 5 mg oxycodone (320,000 morphine mg equivalents) for the 26,000 relevant procedures performed by surgeons in the two intervention arms during the study year. Importantly, these reductions did not lead to increased opioid refill requests or hospital visits.
“Surgeons who received email nudges overprescribed significantly less often.”
For the final year of the study, email nudges were turned off, and researchers asked 36 surgeons what they thought of the intervention. Generally, those interviewed were supportive of the intervention, but some raised concerns. Several themes emerged:
Researchers also continued to observe prescribing behavior. Surgeons in the control group prescribed above guidelines at about the same rate throughout the study and after the intervention ended. For those who received email nudges, the intervention effects continued through the post-intervention year, with reductions relative to the control group similar to those observed during the study period and no significant differences between either email type.
Targeted feedback to prescribers during this study’s intervention period reduced the flow of unused opioids from the Sutter Health system, stopping approximately 42,000 opioid pills from entering the community for potential misuse or diversion. With about 50 million inpatient surgeries performed per year, this intervention, if implemented nationwide, could prevent millions — or tens of millions — of pills from entering circulation.
Health systems interested in adapting this intervention to their own settings should consider some details:
Zachary Wagner, Allison Kirkegaard, Louis T. Mariano, Jason N. Doctor, Xiaowei Yan, Stephen D. Persell, Noah J. Goldstein, Craig R. Fox, Chad M. Brummett, Robert J. Romanelli, Kathryn Bouskill, Meghan Martinez, Kyle Zanocco, Daniella Meeker, Satish Mudiganti, Jennifer Waljee, and Katherine E. Watkins, “Effect of Peer Comparison or Guideline-Based Feedback vs No Feedback on Post-Surgery Opioid Prescriptions: A Cluster-Randomized Controlled Trial,” JAMA Health Forum, Vol. 5, No. 3, March 15, 2024.
Meghan Martinez, Allison Kirkegaard, Kathryn Bouskill, Xiaowei Sherry Yan, Zachary Wagner, and Katherine E. Watkins, “Surgeons’ Views of Peer Comparison and Guideline-Based Feedback on Postsurgery Opioid Prescriptions: A Qualitative Investigation,” BMJ Open Quality, Vol. 13, No. 2, April 5, 2024.
Kyle A. Zanocco, Zachary Wagner, Louis T. Mariano, Allison Kirkegaard, Xiaowei Yan, Craig R. Fox, Noah J. Goldstein, Chad M. Brummett, and Katherine E. Watkins, “Persistence of Social Norms Feedback on Postsurgery Opioid Prescribing Behavior: Secondary Analysis of a Randomized Clinical Trial,” JAMA Health Forum, Vol. 6, No. 1, January 31, 2025.
Kyle Zanocco, Robert J. Romanelli, Danielle Meeker, Louis T. Mariano, Rivfka Shenoy, Zachary Wagner, Allison Kirkegaard, Satish Mudiganti, Meghan Martinez, and Katherine E. Watkins, “Drivers of Variation in Opioid Prescribing following Common Surgical Procedures in a Large Multi-Hospital Healthcare System,” Journal of the American College of Surgeons, Vol. 239, No. 3, May 1, 2024.
Meghan C. Martinez, Kathryn Bouskill, Xiaowei Sherry Yan, Allison Kirkegaard, Jason N. Doctor, and Katherine E. Watkins, “A Qualitative Analysis on the Implementation of a Nudge Intervention to Reduce Post-Surgical Opioid Prescribing,” BMC Health Services Research, Vol. 25, 2025.
Zachary Wagner and Craig R. Fox, “Opinion: Surgeons Give Patients Too Many Opioids. A Few Simple Steps Could Curb Excess Prescribing,” Los Angeles Times, June 18, 2024.
