Abstract
Background
Despite the introduction of the Centers for Disease Control and Prevention's opioid prescribing guidelines, studies indicate that a significant proportion of opioids prescribed at hospital discharge remains unused. Little is known if improved provider awareness of guidelines and metrics would facilitate rightsizing opioid prescriptions at hospital discharge. Our institution created opioid prescribing guidelines and a key metrics dashboard and subsequently disseminated these tools to our institution's leadership and prescribers. We aim to evaluate the effectiveness of these efforts in reducing hospital discharge opioid prescriptions, especially those for acute pain exceeding 100 morphine milligram equivalents (MME).
Methods
Following the development of practice-specific opioid prescribing guidelines in 2017, a key metrics dashboard was created in 2021 to display the percentage of hospital discharges with opioids prescribed and the percentage of discharges with opioids prescribed for acute pain exceeding 100 MME. These metrics were broken down into calendar years between 2018 and 2022, and by the 7 major practice regions across our institution spanning 5 U.S. States.
Results
From 2018 to 2022, all regions showed a decline in the percentage of hospital discharges with opioids prescribed (range 2.7%-9.4%). In the same period, 5 of 7 regions showed a decline in discharge opioid prescriptions exceeding 100 MME for acute pain (range 8.8%-23.2%). Two sites showed an increase of 2.4% and 2.7%.
Conclusion
Downward trends in hospital discharge opioid prescriptions were observed for most practice regions following the introduction of our institution's opioid prescribing guidelines and key metrics dashboard.
Keywords
Introduction
The number of overdose deaths in the United States has exceeded 100,000 annually since 2021. 1 Although national opioid dispensing rates have declined since 2012, 2 a significant proportion of opioids prescribed at hospital discharge for acute pain goes unused; this represents a potential source of diversion.3,4 Up to 5% of opioid naïve patients receiving opioids for acute pain end up using opioids chronically, 5 including those who received their initial opioid prescription after hospital discharge. 6
In 2017, our institution formed an Opioid Stewardship Program (OSP) which developed opioid prescribing and monitoring guidelines in line with recommendations from the Centers for Disease Control and Prevention (CDC), and later adjusted with revised guidelines in 2022.7–9 Our OSP was tasked with improving awareness of opioid prescribing metrics, including the number of index opioid prescriptions at hospital discharge for acute pain, and to implement quality improvement measures to rightsize opioid prescribing to reduce the number of prescriptions and the total quantity at discharge.
Method and Materials
Practice Gap Analysis
An internal audit of opioid prescribing practices in July 2021 looking at opioid prescribing patterns within our institution, including patients prescribed opioids at the time of hospital discharge for acute pain indications, revealed that reliable mechanisms to identify individual and/or practice prescribing outliers did not exist. Significant variations existed between regions and within the same specialty practice region location. Additionally, prescribers did not have easy access to morphine milligram equivalent (MME) conversions or opioid prescribing data (eg, number of prescriptions, MME per day, or total MME per prescription).
Data Collection and Metrics Display
Electronic health record (EHR) data starting June 1, 2018 and onwards on opioid prescriptions across all medical and surgical specialties and regions were reviewed. Since June 2018, all prescribers in our institution are required to select whether an opioid prescription is for an acute or chronic pain indication at the time of prescribing and the encounter setting is captured in our EHR. Opioid prescribing metrics were classified into indications for chronic versus acute pain (without relying on diagnostic or billing codes), and inpatient/hospital discharge versus outpatient prescriptions, and displayed on an Opioid Stewardship Key Metrics Dashboard in the EHR system in October 2021, which is available to all prescribers. Dashboard metrics are updated daily; users can drill down on prescribing metrics to the individual patient and prescription.
Practice Settings
Our healthcare organization consists of 3 main campuses in 3 urban locations (each with a population greater than 100,000): 1 in the Midwest, 1 in the Southwest, and 1 in the Southeast region of the United States. Our organization has a presence in more than 70 rural communities (each having a population of less than 100,000) in 3 Midwest states through a network of satellite hospitals and clinics. These rural communities are divided into 4 regions for administrative purposes. Over 1.3 million patients are served by our organization annually. Table 1 shows selected demographic information of patients served in each practice region.
Demographics and Discharge Opioid Prescribing Data in 2018 and 2022 Broken Down by 7 Major Practice Regions.
DC = discharge; Rx = prescription; MME = morphine milligram equivalent.
*For our study, an urban region is defined as having at least one town or city where the population is at least 100 000; a rural region does not have any cities or towns with a population of 100 000 or more.
Results not weighted.
Opioid Metric and Dashboard Dissemination
With the publication of our dashboard, OSP reached out to practice and quality leaders through written and electronic communications, as well as virtual meetings to bring awareness of this new functionality from October 2021 to March 2022. All medical and surgical specialties from all 7 regions were simultaneously engaged. Instructional materials on accessing and using the dashboard, including a “how-to” guide and a brief video, were disseminated. Links to our institution's opioid prescribing guidelines are included within the dashboard. All practices were requested to identify practice and educational gaps related to opioid prescribing.
Target Setting
Once the dashboard was created in 2021, we set a goal of a 5% reduction of hospital discharges with an opioid prescription for an acute pain indication with a total MME exceeding 100 by the end of 2022.
Data Analysis
The percentages of hospital discharges with opioid prescriptions from all regions were identified from 2018 and 2022. Separately, the percentages of discharges with an opioid prescription for acute pain exceeding 100 MME were identified. Data were broken down into 5 calendar years and 7 practice regions. Due to the descriptive nature of this study, no statistical analyses were conducted. This project was exempt from review by our Institution's Review Board.
Results
Percentage of Hospital Discharges With Opioid Prescriptions
In 2018, between 28.6% (3550/12,408 in Site 6) and 46.8% (18,649/39,849 in Site 3) of patients discharged from the hospital were prescribed opioids. The trend in percentage of opioid discharge prescriptions declined between 2018 and 2022. By 2022, all 7 practice regions have shown a decline, ranging from 2.7% in Site 6 to 9.4% in Site 5 (Table 1).
Percentage of Hospital Discharges With Opioid Prescriptions Exceeding 100 MME for Acute Pain
In 2018, between 35.4% (186/529 in Site 4) and 48.6% (238/490 in Site 6) of hospital discharges had an opioid prescription for acute pain exceeding 100 MME. By 2022, 5 of the 7 practice regions (Sites 3-7) showed a decline ranging from an 8.8% decline (Site 4) to 23.2% (Site 6). Sites 1 and 2 (both urban) showed an increase at 2.4% and 2.7%, respectively (Table 1).
The results displayed in Table 1 are not weighted.
Discussion
Opioid prescribing for acute pain had not been systematically studied until recently, resulting in a wide variability of prescribing practices often passed down from one provider's personal preference to their trainees. 3 Prior research demonstrates that the lack of a systematic way of prescribing opioids intended for as-needed use for acute indications has resulted in the majority of opioids prescribed going unused and not properly disposed of by the patient. 4 The goal of practice-specific guidelines is not to have a “one-size-fits-all” approach. Rather, guidelines provide an agreed-upon general approach that allows for prescribing flexibilities, as clinically appropriate. There is growing evidence that rightsizing acute opioid prescribing, including reducing the amounts of postoperative opioids prescribed, does not affect pain control or patient satisfaction.10–12
The implementation of our dashboard, together with our educational and awareness efforts, was well received and widely endorsed by practice leaders from all regions and across all specialties in our institution. We were humbled by anecdotal feedback from our leadership about the frequency with which they utilized the dashboard to monitor metrics and to identify outlying practices requiring additional education and remediation.
With the dashboard, the OSP identified a slight increase in prescriptions exceeding 100 MME for acute pain in sites 1 and 2 in 2022 (both urban locations). It is possible that differences in patient demographics (such as those listed in Table 1), urbanicity, medical complexities and local practice patterns could contribute to the differences observed. A quality improvement project is underway to evaluate causes and identify solutions.
Limitations and Future Directions
Our study was conducted in one institution with one set of institutional prescribing guidelines. Our practice locations also have higher percentages of individuals identifying as white compared to national statistics. 13 Therefore, our results may not be fully generalizable. However, our institution has a presence in multiple geographic regions in the United States serving urban and rural patients across diverse clinical settings. The declining prescribing trends observed in most regions are worth noting.
The declining trend in discharge opioid prescriptions began before the dashboard introduction. It is also unclear if these improvements are exclusively related to our opioid prescribing guidelines. The introduction of CDC's guidelines in 2016 together with increased regulatory pressure to reduce opioid prescribing, insurance limits, emphasis on multimodal pain management, and increased media attention on the opioid epidemic likely contributed to the declining trend. Future studies evaluating provider attitudes and reasons for changes in opioid prescribing could help clarify.
The COVID-19 pandemic, with its associated changes in practice patterns and access to care, may have also played a role in the trend in opioid prescribing observed in our study. Efforts to measure opioid prescribing changes and their effects on patient outcomes and satisfaction are currently being undertaken.
Our institution separately created a multidisciplinary Controlled Substance Advisory Group to review cases in real-time with outpatient providers on patients prescribed controlled substances, and then provide management recommendations and expedited referrals. 14 Work is underway to create a similar service for hospital providers on managing acute pain with an emphasis on multimodal pain management in the hospital and to provide recommendations for continued outpatient care. Such efforts could improve hospital providers’ confidence in rightsizing opioid prescribing without compromising pain control and functionality.
As our dashboard is still in its infancy, we intend to track the utilization of our dashboard in the coming years, including the number of users that access our dashboard and to which region and specialty they belong, and then to track practice patterns over time to determine if changes in prescribing correlate with dashboard utilization.
Conclusions
Our OSP focused on developing consensus opioid prescribing best practices and a dashboard to better inform practices and prescribers and to track progress. This has led to measurable improvements in hospital discharge opioid prescribing.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and publication of this article: Benjamin Lai receives financial support for research through the Mayo Clinic's Robert D. and Patricia E. Kern Center for the Science of Health Care Delivery.
Data Availability Statement
The datasets generated during and/or analyzed during the current study are available from the corresponding author upon reasonable request. Request may be made to Dr. Benjamin Lai at lai.benjamin@mayo.edu.