Abstract
In the context of both chronic pain and opioid crises, this large-system quality improvement project sought to increase use of evidence-based multimodal pain management strategies. Primary care providers (PCPs) in internal medicine and family medicine identified as above-median prescribers of 30-day opioid supplies were selected for intervention. PCPs received individualized email letters showing their opioid prescribing patterns relative to peers and urging them to view an internal pain/opioid educational video and related system guidelines. The median number of patients receiving 30-day opioid supplies from our target PCPs decreased over a 24-month period. For cohort patients identified at baseline and remaining in treatment over time, those receiving opioid prescriptions decreased, and those receiving nonopioid prescriptions increased. Percentages of PCPs prescribing nonopioids for cohort patients increased over the first year and nonpharmacologic referrals increased in range. Our evidence suggests that PCPs who are higher opioid prescribers will change their practices voluntarily when given feedback about their opioid prescribing patterns relative to their peers, as well as education regarding evidence-based pain management and opioid prescribing.
Keywords: Pain management, multimodal treatment, evidence-based medicine, chronic pain, clinical practice patterns, instructional films and video, medical education
Target audience: All physicians
Learning objectives: After completing the article, the learner should be able to
1. Assess chronic pain clinical practice in relation to current evidence-based guidelines.
2.Integrate additional chronic pain education into his or her own practice.
Faculty credentials/disclosure: Dr. Embry is the E. Rhodes and Leona B. Carpenter Foundation Chair in Family Medicine and co-chair of the Baylor Scott & White Pain Management and Opioid Prescribing Ambulatory Taskforce. Dr. Reis is senior vice president and chief medical officer of the Temple Regional Clinics, medical director of family medicine, and co-chair of the Baylor Scott & White Pain Management and Opioid Prescribing Ambulatory Taskforce. Dr. Couchman is senior vice president and chief medical officer, clinical operations, and executive sponsor of the Baylor Scott & White Pain Management and Opioid Prescribing Ambulatory Taskforce. Dr. Ledbetter is senior vice president and chief medical officer of the Dallas–Fort Worth Central Region and vice chairman of the Baylor Scott & White Quality Alliance Board of Managers. Mr. Zolfaghari is a biostatistician for the Center for Applied Health Research, Baylor Scott & White Health. The authors and planner have no conflicts of interest to disclose.
Accreditation: The A. Webb Roberts Center for Continuing Medical Education of Baylor Scott & White Health is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.
Designation: The A. Webb Roberts Center for Continuing Medical Education of Baylor Scott & White Health designates this journal CME activity for a maximum of 1.0 AMA PRA Category 1 Credit™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.
Ethics: This course has been designated for 1 credit of education in medical ethics and/or professional responsibility.
ABIM MOC: Successful completion of this CME activity, which includes participation in the evaluation component, enables the participant to earn up to 1.0 Medical Knowledge points in the American Board of Medicine’s (ABIM) Maintenance of Certification (MOC) program. The CME activity provider will submit participant completion information to ACCME for the purpose of granting ABIM MOC credit.
Process: To complete this CME activity, read the entire article and then go to https://ce.bswhealth.com/Proceedings2020. You will register for the course, pay any relevant fee, take the quiz, complete the evaluation, and claim your CME credit. For more information about CME credit, email ce@bswhealth.org.
Expiration date: October 1, 2022.
The opioid crisis in the United States has been described as a “perfect storm” of converging factors that resulted in a steep increase in opioid prescribing in the US throughout the 2000s, followed by opioid misuse, diversion, addiction, illegal opioid use, and related overdose deaths. 1 Multiple agencies, regulatory bodies, state governments, and health care systems began to try to curb opioid prescribing and encourage addiction identification and treatment. 2 , 3 When the Centers for Disease Control and Prevention (CDC) published guidelines for management of chronic pain in 2016, opioid prescribing declined. 4 , 5 However, there were unintentional side effects that included forced tapering, patient dismissal, and hard limits on opioids; the opioid pendulum had swung too far, resulting in inadequate management of pain. 6
Today, the CDC, along with other entities such as The Joint Commission and the US Department of Health and Human Services, continues to call for individualized care and multimodal pain management, including nonpharmacologic and nonopioid modalities. 7–9 Barriers to use of modalities other than opioids remain, however, including insufficient provider and/or patient buy-in, out-of-pocket patient costs, geographical and other accessibility issues, lack of provider expertise and confidence, and inadequate pain/opioid medical education. 10–12
A number of health care system pain and opioid initiatives have been published, the majority focusing on reductions in opioid use and increased opioid safety. 13 , 14 A published Veterans Health Administration project is an exception, demonstrating decreased opioid prescribing as well as increased use of multimodal strategies in primary care, where most chronic pain is managed. 15 For the present large-system initiative, our aim was to show decreased use of chronic opioid therapy and increased use of nonopioid medications and nonpharmacologic interventions for chronic pain, consistent with CDC and other guidelines, in primary care clinics.
To mitigate risks associated with longer-duration opioid prescribing, 16 as well as to improve pain management practices, we chose to focus on primary care providers (PCPs) who were higher prescribers of 30-day opioid supplies. Data showed that while about half of our PCPs prescribed such supplies to ≤10 patients over a 3-month data collection period, some providers prescribed 30-day supplies to well over 100 patients. While differences in patient panels might account for some variability, this wide range of prescribing patterns suggested a need for provider education aimed at evidence-based pain management. Our premise was that, following intervention, higher prescribers of opioids would voluntarily alter their practices to align more closely with current recommendations and guidelines. 17
METHODS
This initiative took place in a large health care system. We focused on two groups of PCPs, each made up of internal medicine and family medicine providers, from two geographically and administratively disparate regions of our system. Data for the two groups were analyzed separately, to assess for any differing facilitators or barriers to improved pain practices. PCP opioid prescribing data were extracted from system electronic health records (EHR), excluding those PCPs chiefly practicing pediatrics, oncology, palliative care, or urgent care. All above-median prescribers of 30-day opioid supplies were selected for intervention, as indicated by the number of unique patients who were prescribed such supplies during a 3-month period. Thirty-day prescriptions were targeted to increase the likelihood that data represented prescriptions for chronic pain rather than for acute pain. 18 Final data analyses included all above-median providers who remained system PCPs at the end of 24 months from the start of the project (117 in group 1, 133 in group 2).
Intervention
On the first day after baseline data collection, each PCP received an email letter, with group 1 letters coming from their chief medical officer and group 2 letters from the chairman of their board of directors. The letter included a bar graph showing each PCP’s own prescribing compared to his or her peer group, based on number of unique patients for whom they prescribed 30-day opioid supplies over a 3-month period (Figure 1). The letter directed PCPs to an internal educational video, as well as to system guidelines. Importantly, the PCPs were not instructed to cease or reduce opioid prescribing. Additionally, although viewing the video was strongly encouraged, it was not required.
Figure 1.
Sample audit and feedback showing an individual primary care provider who prescribed 30-day opioid supplies to 61 to 70 patients during a 3-month period.
System pain experts representing interventional pain management, addictions psychiatry, physical medicine and rehabilitation, family medicine, and pain psychology produced the educational video, Acute and Chronic Pain Management and Opioid Prescribing for Outpatients (see link at end of article). Topics included acute pain, pain neurophysiology, chronic pain, nonopioid treatments, behavioral health interventions, opioids, opioid and benzodiazepine tapering, interventional pain management, addiction, interdisciplinary pain programs, opioids risk assessment, and difficult conversations. Emphasis was on multidisciplinary, multimodal treatment planning, based on pain etiology. Providers were able to earn 2.25 AMA PRA Category 1 credits TM for video completion. Those not completing the video within several weeks of the email letter received follow-up email letters, again urging them to view the video and guidelines; 99% of PCPs in group 1 and 89% in group 2 completed the video within 9 months of receipt of their first email letter.
Multimodal measures
For a 24-month period (3 months baseline and 21 months postintervention), we captured data from the EHR for 30-day opioid prescribing by our PCPs, as well as for nonopioid medication prescriptions and referrals known to be helpful for pain. For nonopioid medications, we identified those in the EHR database that were discussed in our own video and guidelines and/or in CDC guidelines. 4 Categories included acetaminophen, corticosteroids, gabapentin, nonsteroidal anti-inflammatory drugs, pregabalin, serotonin-norepinephrine reuptake inhibitors, tricyclic antidepressants, topical lidocaine, and topical anti-inflammatories. For nonpharmacologic modalities, we identified referral categories that could be appropriate for patients with pain, per our own guidelines and those of the CDC, including behavioral health, case management, chemical dependency, integrative medicine, massage therapy, occupational therapy, interventional pain management clinic, physical medicine and rehabilitation, physical therapy, psychiatry, psychology, sleep medicine, and social work. We expected decreases in opioid prescribing soon after the intervention. In addition, we expected to see increases in nonopioid prescribing and an increased range of pain-related referrals over time, given the video’s emphasis on multimodal, multidisciplinary strategies.
Initial analyses
To assess opioid prescribing, we compared the number of unique patients, per study provider, receiving 30-day opioid prescriptions during 3-month baseline data collection to the number of unique patients, per provider, receiving such prescriptions during the two 3-month periods ending at 1 year (Time 1) and at 2 years (Time 2). Wilcoxon signed-rank tests were performed for analysis of this data. The same process was used to examine changes in nonopioid prescriptions and referrals.
Interrupted time series (ITS) analyses were performed for the period beginning the first day of baseline data collection and ending 24 months later to assess for an intervention effect on opioid prescribing. The ITS model used quasi-Poisson distribution regression with three time-based covariates estimating the preintervention slope, the change in level at the intervention point, and the change in slope from preintervention to postintervention. The preintervention slope quantified the outcome trend before the intervention. The level change estimated the change in level attributed to the intervention immediately after the intervention, and accounting for the preintervention trend. The change in slope quantified the difference between the preintervention and postintervention slopes. 19 An autoregressive correlation structure was applied for numbers of unique patients receiving 30-day opioid prescriptions over time, adjusting for the fact that patients per month are correlated for a single provider.
Secondary (ad hoc) analyses
Given the absence of conclusive evidence for increased use of nonopioid medications and nonpharmacologic strategies in initial analyses, we attempted to eliminate patients who may have been prescribed 30-day supplies of opioids for acute pain during baseline. Data regarding any patient who did not receive 30-day supplies of opioids by a study PCP throughout baseline were excluded from further analyses, with the remaining patients (2380 in group 1, 1746 in group 2) being followed as cohort groups for 21 months postintervention. Patient exclusions did not change number of PCPs in either group.
For changes in 30-day opioid prescribing for these cohort groups, Wilcoxon signed-rank tests compared numbers of patients receiving opioids during baseline to numbers of patients remaining in treatment and receiving opioids at Time 1. “Remaining in treatment” was defined as receiving a 30-day opioid prescription, a nonopioid prescription, and/or any pain referral during Time 1. The same process was used to compare baseline to Time 2, and Time 1 to Time 2 prescribing.
To assess nonopioid prescribing by group 1 and group 2 PCPs, we performed chi-square analyses to examine percentages of cohort patients receiving nonopioid prescriptions during baseline, at Time 1, and at Time 2. The same process was used to examine referrals.
To examine PCP practices regarding specific nonopioid medications, we calculated cumulative percentages of study PCPs in each group who wrote at least one nonopioid prescription from a specific category, for at least one cohort patient during baseline, at Time 1, and at Time 2. The same process was used to calculate cumulative percentages of providers who ordered specific referrals over time.
RESULTS
Initial findings
Note that these results are based on data for unique patients, including both new and established patients, over a 24-month period. The median number of unique patients receiving 30-day opioid supplies from group 1 PCPs decreased by Time 1 compared to baseline (median 8, interquartile range [IQR] 13; P < 0.0001). Median numbers also decreased from baseline to Time 2 (median 12, IQR 15; P < 0.0001), as well as from Time 1 to Time 2 (median 4, IQR 8; P < 0.0001). These analyses were repeated for group 2, showing similar statistically significant results. Results of analyses of change in number of patients receiving nonopioid medications or referrals were not statistically significant for either group.
Using ITS analysis, we were able to attribute lower 30-day opioid prescribing by group 2 providers to the intervention (email letter with prescribing data and access information for video and guidelines). An incident rate ratio of 0.97 (confidence interval 0.95–0.99, P = 0.02) was observed, indicating a slope decrease of 3% per month attributable to the initial intervention. ITS analysis failed to reach statistical significance for group 1, likely because opioid prescribing declined during baseline (Figure 2).
Figure 2.
Average number of unique patients, per primary care physician, receiving 30-day opioid prescriptions over the duration of the study.
Secondary (post hoc) findings
Figure 3 shows the percentages of patient cohorts receiving any 30-day opioid prescription, any nonopioid prescription, and/or any pain-related referral order by target PCPs over eight 3-month periods. At Time 1, PCPs were no longer prescribing 30-day opioid supplies to 18% of cohort patients remaining in treatment, but these patients were now receiving nonopioid prescriptions and/or pain-related referrals. By Time 2, 42% of cohort patients in group 1, and 37% in group 2, were no longer receiving opioids, nonopioids, or referrals from study PCPs.
Figure 3.
Percentage of cohort patients receiving 30-day opioid prescriptions, nonopioid prescriptions, and/or pain-related referrals, by calendar quarter.
Consistent with initial analyses, we found a decrease in median number of group 1 cohort patients still in treatment at Time 1 and receiving 30-day opioid supplies, compared to those receiving opioids at baseline (median 8, IQR 9; P < 0.0001). A decrease was also found between baseline and Time 2 (median 9, IQR 10.5; P < 0.0001), and between Time 1 and Time 2 (median 2, IQR 3; P < 0.0001). Analyses for group 2 yielded similar statistically significant results.
Results of nonopioid data analysis showed that group 1 PCPs prescribed nonopioids to 32.7% of cohort patients during baseline. By Time 1, they were prescribing nonopioids to 43.5% of patients remaining in treatment (unadjusted chi-square: P < 0.0001). Compared to a baseline of 32.7%, by Time 2, group 1 PCPs had prescribed at least one nonopioid to 41.7% of patients remaining in treatment (unadjusted chi-square: P < 0.0001). For group 2, cohort percentages at baseline and Time 1 were 37.0% vs 47.0%, respectively (unadjusted chi-square: P < 0.0001), and percentages at baseline and Time 2 were 37.0% and 46.0%, respectively (unadjusted chi-square: P < 0.0001). Neither group showed an increase in number of patients receiving nonopioids from Time 1 to Time 2.
There was an increase in pain-related referrals between baseline and Time 1 by group 1 PCPs, with 9.5% of cohort patients being referred during baseline, and 12.8% receiving referrals by Time 2 (unadjusted chi-square: P = 0.001). In group 2, changes in numbers of referrals for cohort patients were not statistically significant over time.
Examination of specific nonopioid medications prescribed by study PCPs for patient cohorts showed that at baseline, higher percentages of PCPs in both groups prescribed corticosteroids, gabapentin, and nonsteroidal antiinflammatory drugs. Group 1 providers were more likely than group 2 PCPs to have prescribed antidepressants to cohort patients at baseline. The percentages of providers prescribing topical medications were low over the entire span of the study for both groups. Prescribing of acetaminophen also was low over time, but any over-the-counter recommendations would not have been captured in EHR data. Specific nonopioid prescribing patterns for each group were similar at baseline, Time 1, and Time 2, but cumulative percentages increased (Table 1).
Table 1.
Cumulative percentages of primary care physicians prescribing nonopioid medications or making referrals during three calendar quarters (24-month span)
| Variable | Group 1 |
Group 2 |
||||
|---|---|---|---|---|---|---|
| Baseline | Time 1 | Time 2 | Baseline | Time 1 | Time 2 | |
| Nonopioid medication | ||||||
| Acetaminophen | 12.2% | 15.7% | 19.1% | 9.2% | 10.8% | 13.1% |
| Corticosteroid | 54.8% | 65.2% | 68.7% | 49.2% | 62.3% | 64.6% |
| Gabapentin | 80.9% | 82.6% | 82.6% | 60.8% | 65.4% | 68.5% |
| NSAID | 67.8% | 70.4% | 72.2% | 64.6% | 70.8% | 73.1% |
| Pregabalin | 38.3% | 40.0% | 40.0% | 25.4% | 29.2% | 30.0% |
| SNRI | 51.3% | 56.5% | 60.9% | 40.0% | 46.2% | 50.0% |
| Topical anti-inflammatory | 18.3% | 27.0% | 31.3% | 10.0% | 18.5% | 20.0% |
| Topical lidocaine | 16.5% | 22.6% | 26.1% | 10.8% | 13.8% | 16.9% |
| Tricyclic antidepressant | 35.7% | 38.3% | 40.0% | 24.6% | 29.2% | 33.1% |
| Referral | ||||||
| Chemical dependency | 0.0% | 0.9% | 1.8% | 0.0% | 0.0% | 0.9% |
| Case management | 0.9% | 7.1% | 15.2% | 0.9% | 10.4% | 27.8% |
| Integrative medicine | 0.0% | 1.8% | 2.7% | 0.0% | 0.0% | 0.0% |
| Massage therapy | 0.0% | 0.0% | 0.9% | 0.0% | 0.0% | 0.9% |
| Occupational therapy | 6.3% | 21.4% | 28.6% | 0.9% | 5.2% | 6.1% |
| Pain management clinic | 44.6% | 82.1% | 86.6% | 10.4% | 37.4% | 73.9% |
| PMR | 6.3% | 24.1% | 27.7% | 2.6% | 19.1% | 35.7% |
| Psychology/behavioral health | 17.0% | 44.6% | 48.2% | 0.9% | 12.2% | 15.7% |
| Psychiatry | 12.5% | 27.7% | 36.6% | 6.1% | 12.2% | 20.0% |
| Physical therapy | 39.3% | 72.3% | 81.3% | 7.8% | 27.8% | 53.0% |
| Sleep medicine | 9.8% | 28.6% | 33.9% | 4.3% | 17.4% | 28.7% |
| Social work | 12.5% | 25.9% | 33.9% | 0.0% | 0.0% | 0.9% |
NSAID indicates nonsteroidal antiinflammatory drug; PMR, physical medicine and rehabilitation; SNRI, serotonin-norepinephrine reuptake inhibitor.
Data regarding specific referrals ordered for cohort patients suggested that at baseline, the highest percentages among group 1 PCPs were for referrals to occupational therapy, interventional pain management, psychology/behavioral health, psychiatry, physical therapy, and social work. For group 2 at baseline, the highest percentage among providers was for interventional pain management referrals; percentages were low for all other services. Over the 24-month data collection, provider percentages in both groups increased for referrals to case management, interventional pain management, physical medicine and rehabilitation, psychology/behavioral health, psychiatry, physical therapy, and sleep medicine. Group 1 percentages also increased for occupational therapy and social work. Neither PCP group was likely to refer to chemical dependency, integrative medicine, or massage therapy at any time (Table 1).
DISCUSSION
Over 24 months of data collection, we showed declines in the numbers of patients (both new and established) receiving 30-day opioid prescriptions from higher-prescribing PCPs. For one of the two PCP groups, we were able to attribute this change to the intervention through ITS. The other group showed a similar trend, although not statistically significant. Fewer members of the patient cohorts, identified in baseline data as most likely to have chronic pain, received 30-day opioid prescriptions from our PCPs over time, and more of them received nonopioid prescriptions. One PCP group showed an increase in cohort patients receiving nonpharmacologic referrals.
The two PCP groups had fairly similar patterns for specific nonopioid medications, but group 1 providers appeared to highly favor gabapentin and were more likely to prescribe antidepressant medications. Neither PCP group showed high use of topical pain medications or acetaminophen, although there may have been over-the-counter recommendations for the latter that would not have been captured in the EHR data.
Examination of specific referrals in each PCP group showed a lower likelihood of referrals by group 2 at baseline. However, both groups showed increases in referrals for cohort patients over time, as well as a widening of the range of referrals. Referral patterns were fairly consistent with availability/unavailability of services across the system. For example, chemical dependency treatment is very limited throughout the system, and internal psychology and psychiatry services are especially sparse near most of the group 2 clinics. Massage therapy and integrative medicine services are available only on a limited basis across the system. A closer look at the data may determine how accessibility to services, or referrals to specific services, impacted opioid prescribing.
As with other quality improvement projects, there are limitations associated with this one, as well as some design improvements that likely would have produced more conclusive data. Whether this intervention would be effective in promoting evidence-based chronic pain management by non-PCP providers, or in other systems, is unknown. Given the absence of a comparison group, we cannot assert that all findings were due to our interventions. We did not consider short-duration or intermittent opioid prescribing by target PCPs. We do not know for certain that nonopioid medications were prescribed for pain; for example, seasonal allergies could account for high corticosteroid prescribing. Similarly, some referrals may have been ordered for conditions other than pain. We did not follow-up on patients in the cohort groups who were no longer receiving 30-day opioids, nonopioid medications, or referrals at the end of 2 years; whether they left the PCP or system, learned to self-manage, no longer had pain, or were dismissed by their PCP is unknown.
In conclusion, this intervention was distinct from those employed in most currently published studies, in that we did not set external limits on opioid prescribing and we focused on PCPs who were higher prescribers of opioids. Our assumption that providers who were shown evidence of their higher opioid prescribing would be motivated to learn and use other pain management strategies was supported, consistent with previous evidence of the utility of audits and feedback. 20 While we cannot assert that the educational video was more effective due to the information being presented by internal rather than external experts, other studies have shown this approach may be effective in changing provider practices. 21 Our two PCP groups showed fairly similar changes in practice over time, suggesting these outcomes may be replicable. Referral patterns reinforced the need for more uniform availability of certain pain-related services across geographical regions of our system. Finally, our study likely would have produced more conclusive data by requiring PCPs to complete the video within a specified time, as dates of completion ranged from 1 week to >21 months after intervention.
Based on our positive outcomes, we continue to monitor PCPs, urging those who prescribe 30-day opioid supplies to higher numbers of patients to review the video and guidelines for treatment planning. We offer PCP consultation and support as needed, and our video is now assigned to new providers and trainees during the onboarding process.
We invite health care systems and providers to view or use our video or guidelines at https://blog.bswhealth.md/pain-management-opioid-prescribing-guidelines/. A short “trailer” for the video also is available at https://www.youtube.com/watch?v=GHI-d3LGiXI&feature=youtu.be. Feedback, questions, or collaboration are welcomed.
ACKNOWLEDGMENTS
The authors acknowledge the following video presenters: James Albers, MD, Christopher Burnett, MD, Timothy Clark, PhD, Emily Garmon, MD, Rodney Lange, MD, Glenn Ledbetter, MD, Jason Sapp, DO, and Amber Whittenburg, MD, with Cinamon Romers, PhD, and Jae Ross, PsyD, acting as patients. The authors also acknowledge the contributions of project manager Layne Stone in addition to those of Robert Probe, MD, and the Baylor Scott & White Clinical Leadership Council.
References
- 1. Jones MR, Viswanath O, Peck J, Kaye AD, Gill JS, Simopoulos TT.. A brief history of the opioid epidemic and strategies for pain medicine. Pain Ther. 2018;7(1):13–21. doi: 10.1007/s40122-018-0097-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Soelberg CD, Brown RE, Du Vivier D, Meyer JE, Ramachandran BK.. The US opioid crisis: current federal and state legal issues. Anesth Analg. 2017;125(5):1675-1681. doi: 10.1213/ANE.0000000000002403. [DOI] [PubMed] [Google Scholar]
- 3. Beaudoin FL, Banerjee GN, Mello MJ.. State-level and system-level opioid prescribing policies: The impact on provider practices and overdose deaths, a systematic review. J Opioid Manage. 2016;12(2):109–118. doi: 10.5055/jom.2016.0322. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Dowell D, Haegerich TM, Chou R.. CDC guideline for prescribing opioids for chronic pain—United States, 2016. JAMA. 2016;315(15):1624–1645. doi: 10.1001/jama.2016.1464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Bohnert ASB, Guy GP, Losby JL.. Opioid prescribing in the United States before and after the Centers for Disease Control and Prevention’s 2016 opioid guideline. Ann Intern Med. 2018;169(6):367–375. doi: 10.7326/M18-1243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kroenke K, Alford DP, Argoff C, et al. Challenges with implementing the Centers for Disease Control and Prevention opioid guideline: a consensus panel report. Pain Med. 2019;20(4):724–735. doi: 10.1093/pm/pny307. [DOI] [PubMed] [Google Scholar]
- 7. Dowell D, Haegerich T, Chou R.. No shortcuts to safer opioid prescribing. N Engl J Med. 2019;380(24):2285–2287. doi: 10.1056/NEJMp1904190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. The Joint Commission . Joint Commission enhances pain assessment and management requirements for accredited hospitals. Off Newsl Jt Comm. 2017;37(7):2–4. [Google Scholar]
- 9. US Department of Health and Human Services. Pain Management Best Practices Inter-Agency Task Force Report: Updates, Gaps, Inconsistencies, and Recommendations. Washington, DC: HHS; 2019. https://www.hhs.gov/ash/advisory-committees/pain/reports/index.html [Google Scholar]
- 10. Giannitrapani KF, Ahluwalia SC, McCaa M, Pisciotta M, Dobscha S, Lorenz KA.. Barriers to using nonpharmacologic approaches and reducing opioid use in primary care. Pain Med. 2018;19(7):1357–1364. doi: 10.1093/pm/pnx220. [DOI] [PubMed] [Google Scholar]
- 11. Desveaux L, Saragosa M, Kithulegoda N, Ivers NM.. Family physician perceptions of their role in managing the opioid crisis. Ann Fam Med. 2019;17(4):345–351. doi: 10.1370/afm.2413. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Lippe PM, Brock C, David J, Crossno R, Gitlow S.. The First National Pain Medicine Summit—final summary report. Pain Med. 2010;11(10):1447–1468. doi: 10.1111/j.1526-4637.2010.00961.x. [DOI] [PubMed] [Google Scholar]
- 13. Meisenberg BR, Grover J, Campbell C, Korpon D.. Assessment of opioid prescribing practices before and after implementation of a health system intervention to reduce opioid overprescribing. JAMA Netw Open. 2018;1(5):e182908. doi: 10.1001/jamanetworkopen.2018.2908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Boyle KL, Cary C, Dizitzer Y, Novack V, Jagminas L, Smulowitz PB.. Reduction of opioid prescribing through the sharing of individual physician opioid prescribing practices. Am J Emerg Med. 2019;37(1):118–122. doi: 10.1016/j.ajem.2018.09.052. [DOI] [PubMed] [Google Scholar]
- 15. Edmond SN, Moore BA, Dorflinger LM, et al. Project STEP: implementing the Veterans Health Administration’s Stepped Care Model of Pain Management. Pain Med. 2018;19(suppl_1):S30–S37. doi: 10.1093/pm/pny094. [DOI] [PubMed] [Google Scholar]
- 16. Paulozzi LJ, Zhang K, Jones CM, Mack KA.. Risk of adverse health outcomes with increasing duration and regularity of opioid therapy. J Am Board Fam Med. 2014;27(3):329–338. doi: 10.3122/jabfm.2014.03.130290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Owen GT, Bruel BM, Schade CM, Eckmann MS, Hustak EC, Engle MP.. Evidence-based pain medicine for primary care physicians. Proc (Bayl Univ Med Cent). 2018;31(1):37–47. doi: 10.1080/08998280.2017.1400290. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Mundkur ML, Franklin JM, Abdia Y, et al. Days’ supply of initial opioid analgesic prescriptions and additional fills for acute pain conditions treated in the primary care setting—United States, 2014. MMWR Morb Mortal Wkly Rep. 2019;68(6):140–143. doi: 10.15585/mmwr.mm6806a3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Kontopantelis E, Doran T, Springate DA, Buchan I, Reeves D.. Regression based quasi-experimental approach when randomisation is not an option: interrupted time series analysis. BMJ. 2015;350(5):h2750. doi: 10.1136/bmj.h2750. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Chauhan BF, Jeyaraman MM, Mann AS, et al. Behavior change interventions and policies influencing primary healthcare professionals’ practice—an overview of reviews. Implement Sci. 2017;12(1):3. doi: 10.1186/s13012-016-0538-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Flodgren G, O’Brien MA, Parmelli E, et al. Local opinion leaders: effects on professional practice and health care outcomes. Cochrane Database Syst Rev. 2019;6(8):CD000125. doi: 10.1002/14651858.CD000125.pub4 [DOI] [PMC free article] [PubMed] [Google Scholar]