Pain management and sedation/original research
Opioid Prescription Reduction After Implementation of a Feedback Program in a National Emergency Department Group

https://doi.org/10.1016/j.annemergmed.2021.12.009Get rights and content

Study objective

Reducing excessive opioid prescribing in emergency departments (ED) may prevent opioid addiction. We evaluated the largest personalized feedback and peer comparison intervention to date on emergency clinician opioid prescription rates in a national emergency clinician group.

Methods

This interrupted time series analysis of a quality improvement intervention included data from adults discharged from 102 EDs in 17 states from January 1, 2019, to July 31, 2021. From June 16, 2020, to November 30, 2020, site-level ED directors received emails on local opioid prescription rates. From December 1, 2020, to July 31, 2021, all clinicians were granted electronic dashboard access, which showed prescription rates compared with peers, and national ED leaders sent emails to high-prescribing clinicians and engaged in one-on-one conversations. The primary outcome was opioid prescriptions per 100 discharges.

Results

The study included 5,328,288 ED discharges from 924 physicians and 472 advanced practice providers. Opioid prescription rates did not change meaningfully in the site-level director feedback period (mean difference = −0.3, 95% confidence interval [CI] −0.6 to −0.1). During the direct clinician feedback period, opioid prescription rates declined from 10.4 per 100 discharges to 8.4 per 100 discharges (mean difference = −2.0, 95% CI −2.4 to −1.5), a 19% relative reduction. Among prescribers in the highest initial quintile, opioid prescribing reduced by 35% among physicians and 41% among advanced practice providers in the direct feedback period.

Conclusion

We demonstrated a large, sustained reduction in opioid prescribing by emergency clinicians using direct, personalized feedback to clinicians and an electronic dashboard for peer comparison.

Introduction

In the past 30 years, US opioid prescribing rates have quadrupled and overdose deaths increased by 200% from 2000 to 2014.1,2 More recently, opioid overdoses have increased during the coronavirus disease 2019 (COVID-19) pandemic.3, 4, 5 Previous studies have suggested that physician prescribing behavior may cause or exacerbate addiction, driving opioid mortality.1,2 Although emergency clinicians (physicians, physician assistants, and nurse practitioners) typically prescribe small numbers of pills per prescription, they account for approximately one fifth of overall opioid prescriptions.6 Higher opioid prescribing by individual clinicians has been associated with long-term opioid use and addiction and may contribute to the opioid epidemic.1,7 Large variation has also been observed in emergency clinician prescribing, suggesting room for improvement by addressing high prescribing in outlier clinicians.8

Editor’s Capsule Summary

What is already known on this topic

Emergency care is one key opportunity to address prescriptive opioid use.

What question this study addressed

Does a feedback program using peer data on opioid prescribing fed directly to each clinician alter emergency department (ED) later opioid prescribing?

What this study adds to our knowledge

Using a time series approach from a large emergency services contract group and more than 5 million ED encounters, both physicians and advanced practice providers displayed drops in prescribing, especially in the highest pre-effort prescribers.

How this is relevant to clinical practice

Personal comparative information to ED prescribers of opioids can be a better tool than other generalized prescribing reduction efforts.

Several emergency department (ED) interventions to reduce opioid prescription variability, frequency, and quantity have been implemented. These include state prescription drug monitoring programs, guidelines on opioid prescribing, educational interventions, nudges within electronic health records, and feedback programs.9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24 These interventions have been successful to varying degrees in reducing opioid prescribing. Studies on feedback of individual prescribing practices with comparison relative to peers have been particularly effective.22, 23, 24, 25 One study found that 65% of attending physicians, residents, and advanced practice providers at 4 EDs underestimated their perceived opioid prescribing compared with peers.21 When shown their actual rates and peer group norms, opioid prescribing decreased more than controls. However, studies to date on peer feedback have been largely from academic settings and in small numbers of clinicians. One of the largest interventions was a multipronged strategy that included sharing dashboards allowing for clinician peer comparison of opioid prescribing rates, direct feedback to outliers from medical directors, and electronic medical record nudges at 14 EDs in a single system, which reduced opioid prescriptions by 7%.17 To our knowledge, no study has examined interventions aimed at reducing ED opioid prescriptions across a large number and variety of practice settings, health systems, clinicians, and clinician types on a national scale.

We evaluated the effect of an audit and feedback quality improvement program for emergency clinicians on opioid prescribing rates in a national emergency medicine (ED) group. We also examined clinician-, regional-, and condition-level factors associated with declines in opioid prescribing.

Section snippets

Study Design and Intervention

We conducted an evaluation of a 2-phase intervention to reduce opioid prescribing in a national ED practice organization that provides emergency physician and advanced practice provider staffing to a large sample of US EDs staffed by a single national ED group (Figure 1). The primary purpose was to reduce opioid prescribing, particularly among clinicians who were outlier prescribers. The definition of an outlier prescriber was determined by national clinical leadership team examination of the

Characteristics of Study Subjects

A total of 5,328,288 adult discharge visits treated by 1,396 clinicians (924 physicians and 472 advanced practice providers) in 102 EDs in 17 states were included. Patient and visit factors of the study population appear in Table 1. Clinician-level factors and characteristics of the 102 EDs appear in Table E1 and Table E2 (both available at http://www.annemergmed.com). The mean patient age among the included visits was 41.1 years (SD=22.4), with 56% of the patients being female. Most visits

Limitations

Several limitations should be considered. Regression to the mean may have contributed to the change in opioid prescription rates for visits treated by clinicians with high rates during the preintervention period. Without a control group, it is not possible to directly estimate what the reduction may have been in the absence of the intervention or to compare the trends seen in this study with that of clinicians not in this national emergency clinician group who did not receive the intervention.

Discussion

In this largest study to date of an ED opioid prescribing intervention, personalized feedback along with making opioid prescription rates visible to all clinicians and leaders through a dashboard led to immediate large reductions in opioid prescription rates, with the greatest decline among the highest quintile of prescribing. Several smaller studies have evaluated feedback programs at single institutions or in single systems.17,21, 22, 23, 24 This study builds on prior literature by expanding

References (37)

Cited by (2)

  • Sustaining the gains: Reducing unnecessary computed tomography scans in pediatric trauma patients

    2023, Journal of Pediatric Surgery
    Citation Excerpt :

    In addition, compliance was assessed more often and temporally closer to the child's presentation to our pediatric trauma center; therefore, direct provider feedback was given in a more timely fashion reinforcing the importance of adherence. Direct provider feedback is a useful quality improvement tool and has been shown to improve compliance with opioid [35] and venous thromboembolism prophylaxis [36] guidelines. A high degree of variability exists with regard to CT scanning injured children nationally and globally, despite existing consensus recommendations.

Readers: click on the link to go directly to a survey in which you can provide feedback to Annals on this particular article.

A podcast for this article is available at www.annemergmed.com.

Please see page 421 for the Editor’s Capsule Summary of this article.

Supervising editor: Donald M. Yealy, MD. Specific detailed information about possible conflict of interest for individual editors is available at https://www.annemergmed.com/editors.

Author contributions: All authors conceived the study and designed the trial. MSZ provided statistical advice on the study design and analyzed the data. JJO drafted the manuscript, and all authors contributed substantially to its revision. JJO takes responsibility for the paper as a whole.

Authorship: All authors attest to meeting the four ICMJE.org authorship criteria: (1) Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND (2) Drafting the work or revising it critically for important intellectual content; AND (3) Final approval of the version to be published; AND (4) Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding and support: By Annals policy, all authors are required to disclose any and all commercial, financial, and other relationships in any way related to the subject of this article as per ICMJE conflict of interest guidelines (see www.icmje.org). The authors have stated that no such relationships exist. The authors report that this article did not receive any outside funding or support.

View full text