Changes in Opioid Marketing Practices After Release of the CDC Guidelines

ABSTRACT

Objectives: After the release of the CDC guidelines in March 2016, the rate of opioid prescriptions decreased. How or whether pharmaceutical companies changed their opioid marketing practices post release of the CDC guidelines is unknown. Our objectives were to (1) evaluate whether the release of the guidelines was associated with changes in total monthly marketing spending received per physician, monthly marketing encounter frequency per physician, and spending per encounter during opioid marketing; and (2) evaluate whether such changes in marketing differed between specialist physicians and primary care physicians (PCPs) and between urban and rural primary care service areas (PCSAs).

Study Design: Retrospective observational cross-sectional study using opioid marketing spending data from the CMS Open Payments database between August 2013 and December 2017.

Methods: Single-group and multiple-group interrupted time series analyses were used to evaluate differences in the immediate changes in level and trend over time in opioid marketing practices post release of the CDC guidelines.

Results: Post release of the CDC guidelines, the monthly number of marketing encounters per physician and total monthly amount received per physician decreased. However, the amount spent at each marketing encounter increased. The release of the CDC guidelines was associated with an immediate increase in level of opioid marketing spending per encounter by $0.59 (95% CI, $0.51-$0.68; P < .001) and an over-time increase in rate of spending per encounter of $0.04 per month (95% CI, $0.03-$0.05; P < .001). These changes differed between specialists and PCPs and between urban and rural PCSAs.

Conclusions: It is important to continue ongoing education for physicians on changes in pharmaceutical opioid marketing practices.

Am J Manag Care. 2022;28(10):507-513. https://doi.org/10.37765/ajmc.2022.89248

Takeaway Points

• It is unknown how the CDC guidelines on opioid prescribing have been associated with pharmaceutical opioid marketing practices to physicians; the current study sheds light on this.
• The CDC guidelines on opioid prescribing were associated with a decrease in total monthly marketing spending received per physician, a decrease in monthly frequency of opioid marketing encounters per physician, and an increase in spending per physician encounter during opioid marketing.
• Pharmaceutical marketing changes after the release of the CDC guidelines on opioid prescribing differed between specialist and primary care providers and also between rural and urban primary care service areas.
• Physicians who received higher spending per encounter also had higher encounter frequencies.

Death from drug overdose remains an important public health crisis in the United States and is the leading cause of death among those younger than 50 years.¹ Two-thirds of the 63,632 drug overdose deaths in 2016 involved an opioid,² and 40% of the 63,632 deaths involved prescription opioids specifically.³ Prescription opioid overdose deaths increased from 3442 in 1999 to 17,029 in 2017.⁴ After the onset of the COVID-19 pandemic in the United States in March 2020, there has been a further spike in drug overdose deaths; 107,270 drug overdose deaths were reported in 2021, with 80,816 of the deaths attributed to all opioids and 13,503 specifically to prescription opioids.⁵

About 20% of US adults reported experiencing chronic pain in 2016,⁶ and 20% of noncancer pain is treated with opioids.⁷ Although the prevalence of pain reported by Americans has not changed since the 1990s, opioid prescriptions for pain have increased.^8,9 In an effort to reduce the burden of prescription opioid overdose, the CDC in March 2016 issued new guidelines on opioid prescription for chronic pain by primary care physicians (PCPs).¹⁰ The guidelines recommend initiating opioid treatment with immediate-release opioids at doses less than 50 morphine milligram equivalents (MME) per day. One of the guidelines also reads: “Nonpharmacologic therapy and nonopioid pharmacologic therapy are preferred for chronic pain. Physicians should only consider adding opioid therapy if expected benefits for both pain and function are anticipated to outweigh risks to the patient.” The guidelines have in fact shown to be associated with a reduction in opioid prescription rate,¹¹ but whether the pharmaceutical companies implemented different opioid marketing strategies to counter the reduction in prescriptions after the release of the CDC guidelines is unknown. Pharmaceutical marketing to physicians has been shown to be associated with increases in physician drug prescriptions and formulary addition requests by physicians for the promoted drugs.^12-20 Pharmaceutical sales representatives meet with physicians to talk about their drugs and encourage prescriptions. They engage in promotional activities with physicians that may involve gifts, food and beverages, dinners, sponsorships for conferences, expense-paid travel and lodging, honoraria, consulting fees, compensation for serving as faculty or speaker, and the like.^21,22 About 48% of physicians accepted industry-associated payments in 2015.²³ In an article that discussed pharmaceutical marketing tactics, one of the authors, a former pharmaceutical sales representative, wrote, “It’s my job to figure out what a physician’s price is. For some it’s dinner at the finest restaurants…” and “During training, I was told, when you’re out to dinner with a doctor, ‘The physician is eating with a friend. You are eating with a client.’ ”²⁴ Indeed, about 95% of nonresearch opioid marketing encounters involve food and beverages.²⁵ Simple acts such as providing food enable marketing messages to be more positively received.²⁶ When people receive gifts they feel indebted and have a tendency to return the favor,²⁶ and in these cases of pharmaceutical opioid marketing, the return may be more opioid prescriptions.

In 2010, the Sunshine Act was passed into law through the Affordable Care Act and it required that medical product manufacturers report payments made to physicians to CMS, which publishes the data annually in a publicly searchable Open Payments database.²⁷ Payments for such things as meals, gifts, and speaking fees started being reported in the Open Payments database in August 2013. The American Medical Association recommends that any gifts accepted by physicians should primarily be of benefit to patients and should not be of significant value.²⁸ Although the majority of internal medicine program directors did not find pharmaceutical support desirable, more than half (56%) of them reported accepting support from the pharmaceutical industry.²⁹

Between 2013 and 2015, nonresearch opioid-related marketing to physicians exceeded $46 million.²⁵ In the Medicare Part D population, it has been shown that physicians who receive opioid-related payments prescribe more opioids than those who do not.^30-33 Another study showed the association between increases in county-level pharmaceutical marketing of opioids and higher overdose mortality.³⁴ However, to our knowledge, no study has evaluated whether the CDC guidelines have been linked with changes in pharmaceutical opioid marketing spending per encounter, changes in frequency of marketing encounters with physicians, or changes in total marketing amount spent per physician. Therefore, we evaluated whether the dollar value of food and beverages spent per physician encounter during opioid marketing, monthly number of encounters per physician, and total monthly amount received per physician changed post release of the CDC guidelines. We focused on food and beverage gifts because they account for about 95% of the opioid marketing encounters.²⁵

Although the CDC guidelines on opioid prescribing were focused on PCPs, clinicians often adopt evidence-based recommendations from outside of their own areas of practice,^35,36 and some specialist providers also adopted the CDC guidelines.^36,37 Hence, we examined whether associated changes in opioid marketing post release of the CDC guidelines differed between PCPs and specialists. To our knowledge, these questions have not been previously examined.

Certain differences between rural and urban areas with regard to opioids have been documented, including rates of opioid prescriptions being higher in rural than urban areas,^38,39 which may in turn influence pharmaceutical opioid marketing practices. However, whether postguideline changes in pharmaceutical marketing practices differed between urban and rural primary care service areas (PCSAs) is unknown. Hence, we also evaluated whether associated changes in postguideline opioid marketing differed between urban and rural PCSAs.

Overall, our study evaluates whether, post release of the CDC guidelines, the dollar value of food and beverage gifts per physician encounter for opioid marketing, monthly number of encounters per physician, and total monthly amount received per physician changed. Also, we evaluated whether these changes in marketing practices post release of the CDC guidelines differed between specialist physicians and PCPs, as well as between urban and rural PCSAs.

METHODS

Study Population

We used the CMS Open Payments database,²² a database that collects information on financial relationships between physicians and drug or device companies, which are required by the federal government to publicly report payments to provide transparency in the health care system. We extracted all opioid marketing spending on food and beverages for physicians between August 2013 and December 2017.

Outcomes

The outcome variables were the amount spent on each opioid marketing encounter with food and beverages, monthly number of encounters per physician, and total marketing amount spent on each physician per month. All yearly monetary values were converted to 2016 US$ equivalents using the 2016 consumer price index⁴⁰ to address inflation during the study period.

Measures

Physicians receiving marketing spending were classified into PCPs and specialist physicians. PCSAs are standardized systems of geographical units that measure access to primary care resources, utilization, supply, and associated outcomes.⁴¹ Physicians’ practice locations were assigned to PCSAs using their zip codes according to the Dartmouth Atlas.⁴² A single PCSA is made up of several Census tracts. We calculated the population-weighted proportion of Census tracts classified with a US Department of Agriculture rural-urban commuting area code of 3 or lower.⁴³ If the weighted proportion was greater than or equal to 0.75, that PCSA was considered to be urban; if less than 0.75, the PCSA was considered to be rural. Other cutoff values for population-weighted proportion of PCSAs different from 0.75 were explored and had no effect on the rural/urban classification or results.

Statistical Analysis

We used single-group interrupted time series analysis (ITSA)^44,45 with a linear probability model clustered on physicians to examine changes in opioid marketing pre- and post release of the CDC guidelines. We clustered SEs on physicians to account for repeated observation of physicians. The ITSA model allows for the evaluation of population-level interventions without case and control groups and is useful for teasing out immediate change in outcome measures, as well as changes in trajectory (slope) over time, following an intervention.⁴⁴ We chose a linear probability model over a logit model to produce readily interpretable estimates⁴⁶ and show how much opioid marketing practices changed post release of the CDC guidelines.

Changes in amount spent per opioid marketing encounter, monthly number of encounters per physician, and total monthly marketing amount received per physician were analyzed as a one-time change in level (intercept) at the time of exposure to the CDC guidelines and as change in trend over time after release of the CDC guidelines. The single-group ITSA was modeled in the form of equation 1. We used a multiple-group ITSA to evaluate whether changes in opioid marketing practices after release of the CDC guidelines are different when marketing to PCPs compared with specialist physicians, as well as when marketing in urban PCSAs compared with rural PCSAs. The multiple-group ITSA was modeled in the form of equation 2 using interaction terms.

y_ijt = β₀ + β₁*Monthst + β₂*CDC_guidelinet + β₃*MonthsSinceCDC_guidelinet + β₄*X_j + εt

y_ijt = β₀ + β₅*Monthst + β₆*Months#(group2−group1)t + β₇*CDC_guidelinet + β₈*CDC_guideline#(group2−group1)t + β₉*MonthsSinceCDC_guidelinet + β₁₀*MonthsSinceCDC_guideline#(group2−group1)t + β₁₁*X_j + εt

where y_ijt is the outcome variable for physician i from pharmaceutical company j at month t, Months is a linear time trend in months that starts at the beginning of our sample period, CDC_guideline is a binary indicator variable that equals 0 prior to the release of the CDC guidelines, and 1 after release of the CDC guidelines, MonthsSinceCDC_guideline is a linear time trend that equals 0 prior to the release of the CDC guidelines and starts counting up each month afterward, (group2 – group1) is a binary indicator variable that is either equal to group 1 or group 2 (eg, specialist/PCP for physician types or rural/urban for PCSAs), # is an interaction term, X_j is a vector of pharmaceutical marketing company fixed effects entered as a categorical variable, and εt is the random error at observation t.

β₀ is the intercept (starting level of the outcome), β₁ is the slope or trajectory of the outcome until the introduction of the CDC guidelines, β₂ represents the one-time change in level of outcome immediately at the time of implementation of the CDC guidelines (immediate treatment effect intercept change), β₃ represents the difference in the post–CDC guidelines trend/slope and pre–CDC guidelines trend/slope of outcome variable (treatment effect over time after the CDC guidelines), β₄ is a vector of the outcome variable by each pharmaceutical company, β₅ is the slope or trajectory of the outcome until the introduction of the CDC guidelines in the reference group, β₆ is the difference in slope or trajectory of outcome until the introduction of the CDC guidelines between groups compared, β₇ represents the one-time change in level of outcome immediately at the time of implementation of the CDC guidelines in the reference group, β₈ represents the difference in immediate one-time change in level of outcome at the time of implementation of the CDC guidelines in March 2016 between groups compared, β₉ represents the difference in the post–CDC guidelines trend/slope and pre–CDC guidelines trend/slope of outcome variable in the reference group, β₁₀ represents the difference between the 2 groups’ compared change in the post–CDC guidelines and pre–CDC guidelines trend/slope of the outcome variable (difference in treatment effect over time post CDC guidelines), β₁₁ is a vector of the outcome variable by each pharmaceutical company, and ε is the error term.

We repeated the analysis using 1-month and 2-month washout periods and our findings remained unchanged (with similar directions and magnitudes of changes in intercept and slope post release of the CDC guidelines). A washout period is a time period that is excluded from the analysis with the aim of allowing for some time for implementation of the intervention. For example, with the release of the CDC guidelines in March 2016, a 1-month washout leaves out the month of March 2016 from the analysis, and a 2-month washout leaves out the months March and April 2016 from the analysis.

Systemic differences in dollar amount spent on food and beverages during marketing across pharmaceutical companies were accounted for by including an indicator variable for each company as controls in all models.

All analyses were performed using DbVisualizer version 10.0.15 (DbVis Software AB) and Stata 14 (StataCorp LLC).

RESULTS

During the study period, 94.8% of all opioid marketing encounter payments involved food and beverages and a total of 86,101 unique physicians received opioid marketing spending with food and beverages; 35.5% of the physicians were PCPs. A total of 684,343 opioid marketing encounters occurred during the study period. Of the total encounters, 89.9% occurred in urban PCSAs compared with 10.1% in rural PCSAs (eAppendix Table 1 [eAppendix available at ajmc.com]). Seven pharmaceutical companies accounted for 91% of all opioid marketing encounters. The mean (SD) amount spent on meals and beverages on each encounter was $16.6 ($25.7), the mean (SD) monthly number of encounters per physician was 1.5 (3.9), and the mean (SD) total monthly spending per physician per month was $25.3 ($75.5).

The Table shows the association of total monthly opioid marketing spending per physician, monthly number of encounters per physician, and spending per opioid marketing encounter with the release of the CDC guidelines. Prior to the release of the CDC guidelines, the total monthly opioid marketing spending and monthly number of encounters per physician were increasing at a rate of $0.32 (95% CI, $0.14-$0.49; P < .001) and 0.03 (95% CI, 0.02-0.04; P < .001), respectively. Post release of the CDC guidelines, the change in total monthly opioid marketing spending and monthly number of encounters per physician decreased over time by $1.07 per month (95% CI, –$2.09 to $0.05; P = .040) and 0.08 encounters (95% CI, –0.14 to –0.02; P = .008) per month, respectively. The direction of the change remained the same using 1-month and 2-month washout periods post release of the CDC guidelines (eAppendix Table 1 and eAppendix Table 2). In contrast, before the release of the CDC guidelines, the amount spent per opioid marketing encounter on food and beverage was declining at a rate of $0.03 per month (95% CI, –$0.03 to –$0.02; P < .001). Post release of the CDC guidelines, there was an immediate increase in level of spending per encounter by $0.58 (95% CI, $0.49-$0.67; P < .001) and the change in amount of spending per encounter increased at a rate of $0.04 per month (95% CI, $0.03-$0.05; P < .001). The direction of the change remained the same using 1-month and 2-month washout periods post release of the CDC guidelines, even though the magnitudes of increase in spending were slightly higher (eAppendix Tables 1 and 2).

Also, physicians who received higher marketing spending per encounter also received more encounters (Figure 1). For example, in January 2014, the top 1% of physicians who received the highest payment per encounter accounted for 3.5% of all opioid marketing encounters. The top 5% accounted for 10.4% of all opioid marketing encounters (Figure 1).

We also found variations in the changes in opioid marketing practices post release of the CDC guidelines by physician specialty and geographic locations. The decline in total monthly opioid marketing spending and number of encounters per physician post release of the CDC guidelines was greater among specialist physicians than PCPs and in urban PCSAs than rural PCSAs (Figure 2 and Figure 3). Post release of the CDC guidelines, the immediate increase in level of spending per encounter was greater among specialist physicians than PCPs by $0.32 (95% CI, $0.16-$0.48; P < .001); however, the rate of increase in amount of spending per encounter was lower among specialist physicians compared with PCPs by $0.02 per month (95% CI, –$0.03 to –$0.01; P = .005) (Figure 4). Likewise, post release of the CDC guidelines, the immediate increase in level of spending per encounter was greater in urban than rural areas by $0.26 (95% CI, $0.05-$0.46; P = .014); however, the rate of increase was lower in urban PCSAs compared with rural PCSAs by $0.02 per month (95% CI, –$0.04 to –$0.01; P = .039) (Figure 4).

DISCUSSION

In this study that examined 684,343 non–research-related opioid marketing encounters involving food and beverages between August 2013 and December 2017, we found that post release of the CDC guidelines pharmaceutical companies reduced the total monthly amount spent on opioid marketing per physician and monthly number of opioid marketing encounters per physician. However, the amount spent on food and beverage per encounter increased. We also noted a shift in trend in prescription opioid marketing practices; before the release of the CDC guidelines, the amount spent per encounter on opioid marketing was declining, but post release of the guidelines, it started to increase.

In addition, we found that physicians who received higher marketing spending per encounter accounted for a disproportionately larger share of all encounters. This finding may be consistent with more focused marketing targeting of high-volume prescribers. For example, in February 2021, a charge was brought against a large consulting company, alleging that it advised “opioid manufacturers to target prescribers who write the most prescriptions, for the most patients, and thereby make the most money.”⁴⁷

We also observed that the rate of change in spending per encounter post release of the CDC guidelines compared with before the release of the guidelines was significantly higher among PCPs than specialist physicians. The CDC guidelines targeted PCPs, and PCPs prescribe about half of all prescription opioids,^48,49 which may explain why they were more heavily targeted with a higher rate of opioid marketing spending per encounter post release of the CDC guidelines.

The rate of increase in marketing spending per encounter post release of the CDC guidelines was also significantly higher in rural than in urban PCSAs. Perhaps this is because rural areas have a higher rate of opioid prescriptions,^38,39 potentially making them more attractive for marketing spending per encounter.

Several studies’ findings have shown that pharmaceutical opioid marketing targeted to physicians is associated with increases in opioid prescriptions by physicians receiving these payments.^30-33 It is important to continue ongoing education for physicians to increase their awareness of changes in pharmaceutical opioid marketing practices. Study findings have shown and suggested that educational interventions for physicians, as well as legislation that limits the value of gifts received by physicians, may be beneficial in addressing the potential influence of pharmaceutical opioid marketing on physician prescribing.^16,50,51

Limitations

Our study focused on food and beverage encounters and did not evaluate other forms of pharmaceutical opioid marketing, such as spending on education, consulting fees, honoraria, and grants. Also, our analysis evaluates response to the CDC opioid guidelines at the national level and may not have accounted for some state-level policy changes on opioid prescribing during the study period. For example, in 2016, Massachusetts limited first-time opioid prescriptions to 7 days, and in 2017, Utah recommended that alternatives to opioid treatment should be tried before initiating opioid treatment.^52,53

CONCLUSIONS

After the release of the CDC guidelines on opioid prescribing, total monthly amount spent per physician and monthly frequency of marketing encounters per physician decreased and instead the value of food and beverage gift items during each encounter increased. It is important to continue to educate physicians to maintain awareness of pharmaceutical opioid marketing practices.

Author Affiliations: Department of Health Services Research (ATT) and Division of Health Pharmacy Coverage Change and Management (RW, TB), School of Public Health, University of Minnesota Twin Cities, Minneapolis, MN; Department of Finance, University of Minnesota Carlson School of Management (PK-M), Minneapolis, MN; National Bureau of Economic Research (PK-M), Minneapolis, MN; Department of Health Services Research, Mayo Clinic (MMJ), Rochester, MN.

Source of Funding: This research was funded by Agency for Healthcare Research and Quality (AHRQ) (R01 HS025164) (principal investigator: Karaca-Mandic).

Author Disclosures: Dr Karaca-Mandic has provided consulting services to Sempre Health, holds an executive position and equity in XanthosHealth, serves as advisor to Koya Medical, all for unrelated work, and served as the principal investigator on the AHRQ grant that partially funded this work. The remaining authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.

Authorship Information: Concept and design (ATT, PK-M, RW, MMJ, TB); acquisition of data (PK-M, TB); analysis and interpretation of data (ATT, PK-M, MMJ, TB); drafting of the manuscript (ATT, RW, MMJ, TB); critical revision of the manuscript for important intellectual content (ATT, PK-M, RW, MMJ, TB); statistical analysis (ATT); obtaining funding (PK-M, TB); administrative, technical, or logistic support (PK-M, RW, MMJ, TB); and supervision (PK-M, TB).

Address Correspondence to: Adeniyi T. Togun, MD, PhD, MPH, MS, Department of Health Services Research, School of Public Health, University of Minnesota Twin Cities, 420 Delaware St SE, MMC 729, Minneapolis, MN 55455. Email: togun001@umn.edu.

REFERENCES

1. Underlying cause of death, 1999-2017 request. CDC. Accessed September 3, 2022. https://wonder.cdc.gov/controller/saved/D76/D15F907

2. Scholl L, Seth P, Kariisa M, Wilson N, Baldwin G. Drug and opioid-involved overdose deaths — United States, 2013-2017. MMWR Morb Mortal Wkly Rep. 2019;67(51-52):1419-1427. doi:10.15585/mmwr.mm6751521e1

3. What is the U.S. opioid epidemic? HHS. Updated October 27, 2021. Accessed September 3, 2022. https://www.hhs.gov/opioids/about-the-epidemic/index.html

4. Overdose death rates. National Institute on Drug Abuse. Updated January 20, 2022. Accessed September 3, 2022. https://www.drugabuse.gov/drug-topics/trends-statistics/overdose-death-rates

5. Provisional drug overdose death counts. CDC. Updated August 18, 2022. Accessed September 3, 2022. https://www.cdc.gov/nchs/nvss/vsrr/drug-overdose-data.htm

6. Dahlhamer J, Lucas J, Zelaya C, et al. Prevalence of chronic pain and high-impact chronic pain among adults — United States, 2016. MMWR Morb Mortal Wkly Rep. 2018;67(36):1001-1006. doi:10.15585/mmwr.mm6736a2

7. Daubresse M, Chang HY, Yu Y, et al. Ambulatory diagnosis and treatment of nonmalignant pain in the United States, 2000-2010. Med Care. 2013;51(10):870-878. doi:10.1097/MLR.0b013e3182a95d86

8. National Academies of Sciences, Engineering, and Medicine; Health and Medicine Division; Board on Health Sciences Policy; Committee on Pain Management and Regulatory Strategies to Address Prescription Opioid Abuse; Phillips JK, Ford MA, Bonnie RJ, eds. Trends in opioid use, harms, and treatment. In: Pain Management and the Opioid Epidemic: Balancing Societal and Individual Benefits and Risks of Prescription Opioid Use. The National Academies Press; 2017. Accessed September 3, 2022. https://www.ncbi.nlm.nih.gov/books/NBK458661/

9. Manchikanti L, Atluri S, Hansen H, et al. Opioids in chronic noncancer pain: have we reached a boiling point yet? Pain Physician. 2014;17(1):E1-10.

10. Dowell D, Haegerich TM, Chou R. CDC guideline for prescribing opioids for chronic pain — United States, 2016. MMWR Recomm Rep. 2016;65(1):1-49. doi:10.15585/mmwr.rr6501e1

11. Bohnert ASB, Guy GP Jr, 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

12. Lieb K, Scheurich A. Contact between doctors and the pharmaceutical industry, their perceptions, and the effects on prescribing habits. PloS One. 2014;9(10):e110130. doi:10.1371/journal.pone.0110130

13. Spurling GK, Mansfield PR, Montgomery BD, et al. Information from pharmaceutical companies and the quality, quantity, and cost of physicians’ prescribing: a systematic review. PLoS Med. 2010;7(10):e1000352. doi:10.1371/journal.pmed.1000352

14. Caudill TS, Johnson MS, Rich EC, McKinney WP. Physicians, pharmaceutical sales representatives, and the cost of prescribing. Arch Fam Med. 1996;5(4):201-206. doi:10.1001/archfami.5.4.201

15. Zipkin DA, Steinman MA. Interactions between pharmaceutical representatives and doctors in training. a thematic review. J Gen Intern Med. 2005;20(8):777-786. doi:10.1111/j.1525-1497.2005.0134.x

16. Fickweiler F, Fickweiler W, Urbach E. Interactions between physicians and the pharmaceutical industry generally and sales representatives specifically and their association with physicians’ attitudes and prescribing habits: a systematic review. BMJ Open. 2017;7(9):e016408. doi:10.1136/bmjopen-2017-016408

17. Chren MM, Landefeld CS. Physicians’ behavior and their interactions with drug companies. a controlled study of physicians who requested additions to a hospital drug formulary. JAMA. 1994;271(9):684-689. doi:10.1001/jama.1994.03510330062035

18. Yeh JS, Franklin JM, Avorn J, Landon J, Kesselheim AS. Association of industry payments to physicians with the prescribing of brand-name statins in Massachusetts. JAMA Intern Med. 2016;176(6):763-768. doi:10.1001/jamainternmed.2016.1709

19. DeJong C, Aguilar T, Tseng CW, Lin GA, Boscardin WJ, Dudley RA. Pharmaceutical industry-sponsored meals and physician prescribing patterns for Medicare beneficiaries. JAMA Intern Med. 2016;176(8):1114-1122. doi:10.1001/jamainternmed.2016.2765

20. Fleischman W, Agrawal S, King M, et al. Association between payments from manufacturers of pharmaceuticals to physicians and regional prescribing: cross sectional ecological study. BMJ. 2016;354:i4189. doi:10.1136/bmj.i4189

21. Weiss J. Medical marketing in the United States: a prescription for reform. George Washington Law Rev. 2010;79:260.

22. Open Payments: about. CMS. Accessed November 8, 2020. https://openpaymentsdata.cms.gov/about

23. Tringale KR, Marshall D, Mackey TK, Connor M, Murphy JD, Hattangadi-Gluth JA. Types and distribution of payments from industry to physicians in 2015. JAMA. 2017;317(17):1774-1784. doi:10.1001/jama.2017.3091

24. Fugh-Berman A, Ahari S. Following the script: how drug reps make friends and influence doctors. PLoS Med. 2007;4(4):e150. doi:10.1371/journal.pmed.0040150

25. Hadland SE, Krieger MS, Marshall BDL. Industry payments to physicians for opioid products, 2013-2015. Am J Public Health. 2017;107(9):1493-1495. doi:10.2105/AJPH.2017.303982

26. Wall LL, Brown D. The high cost of free lunch. Obstet Gynecol. 2007;110(1):169-173. doi:10.1097/01.AOG.0000268800.46677.14

27. Richardson E. The Physician Payments Sunshine Act. Health Affairs. October 2, 2014. Accessed March 9, 2021. https://www.healthaffairs.org/do/10.1377/hpb20141002.272302/full/

28. AMA Council on Ethical and Judicial Affairs. AMA Code of Medical Ethics’ opinions on physicians’ relationships with drug companies and duty to assist in containing drug costs. AMA J Ethics. 2014;16(4):261-264. doi:10.1001/virtualmentor.2014.16.4.coet2-1404

29. Loertscher L, Halvorsen A, Beasley B, Holmboe E, Kolars J, McDonald FS. Pharmaceutical industry support and residency education: a survey of internal medicine program directors. Arch Intern Med. 2010;170(4):356-362. doi:10.1001/archinternmed.2009.524

30. Hadland SE, Cerdá M, Li Y, Krieger MS, Marshall BDL. Association of pharmaceutical industry marketing of opioid products to physicians with subsequent opioid prescribing. JAMA Intern Med. 2018;178(6):861-863. doi:10.1001/jamainternmed.2018.1999

31. Carey C, Lieber EMJ, Miller S. Drug firms’ payments and physicians’ prescribing behavior in Medicare Part D. National Bureau of Economic Research working paper No. 26751. February 2020. Updated March 2021. Accessed September 3, 2022. https://www.nber.org/papers/w26751

32. Beilfuss S, Linde S. Pharmaceutical opioid marketing and physician prescribing behavior. Health Econ. 2021;30(12):3159-3185. doi:10.1002/hec.4424

33. Nguyen TD, Bradford WD, Simon KI. Pharmaceutical payments to physicians may increase prescribing for opioids. Addiction. 2019;114(6):1051-1059. doi:10.1111/add.14509

34. Hadland SE, Rivera-Aguirre A, Marshall BDL, Cerdá M. Association of pharmaceutical industry marketing of opioid products with mortality from opioid-related overdoses. JAMA Netw Open. 2019;2(1):e186007. doi:10.1001/jamanetworkopen.2018.6007

35. Sanson-Fisher RW. Diffusion of innovation theory for clinical change. Med J Aust. 2004;180(S6):S55-S56. doi:10.5694/j.1326-5377.2004.tb05947.x

36. Togun AT, Karaca-Mandic P, Wurtz R, Jeffery M, Beebe T. Association of 3 CDC opioid prescription guidelines for chronic pain and 2 payer pharmacy coverage changes on opioid initiation practices. J Manag Care Spec Pharm. 2021;27(10):1352-1364. doi:10.18553/jmcp.2021.27.10.1352

37. Togun AT, Mandic PK, Wurtz R, Jeffery MM, Beebe T. Association of opioid fills with Centers for Disease Control and Prevention opioid guidelines and payer coverage policies: physician, insurance and geographic factors. Int J Clin Pharm. 2022;44(2):428-438. doi:10.1007/s11096-021-01360-w

38. García MC, Heilig CM, Lee SH. Opioid prescribing rates in nonmetropolitan and metropolitan counties among primary care providers using an electronic health record system — United States, 2014-2017. MMWR Morb Mortal Wkly Rep. 2019;68(2):25-30. doi:10.15585/mmwr.mm6802a1

39. Lister JJ, Ellis JD, Yoon M. Opioid prescribing and opioid-overdose deaths in Michigan: urban-rural comparisons and changes across 2013-2017. Addict Behav Rep. 2019;11:100234. doi:10.1016/j.abrep.2019.100234

40. Consumer Price Index. US Bureau of Labor Statistics. Accessed November 8, 2020. https://www.bls.gov/cpi/

41. Goodman DC, Mick SS, Bott D, et al. Primary care service areas: a new tool for the evaluation of primary care services. Health Serv Res. 2003;38(1 Pt 1):287-309. doi:10.1111/1475-6773.00116

42. Wennberg JE, Cooper MM. The Dartmouth Atlas of Health Care. The Center for the Evaluative Clinical Sciences, Dartmouth Medical School, American Hospital Publishing; 1996.

43. Documentation: 2010 rural-urban commuting area (RUCA) codes. US Department of Agriculture. Updated August 17, 2020. Accessed October 29, 2020. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/documentation/

44. Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27(4):299-309. doi:10.1046/j.1365-2710.2002.00430.x

45. Penfold RB, Zhang F. Use of interrupted time series analysis in evaluating health care quality improvements. Acad Pediatr. 2013;13(suppl 6):S38-S44. doi:10.1016/j.acap.2013.08.002

46. Hellevik O. Linear versus logistic regression when the dependent variable is a dichotomy. Qual Quant. 2009;43(1):59-74. doi:10.1007/s11135-007-9077-3

47. Commonwealth of Massachusetts v. McKinsey & Company Inc. United States. Mass.gov. Accessed July 8, 2021. https://www.mass.gov/doc/massachusetts-mckinsey-complaint/download

48. Why guidelines for primary care providers? CDC. Accessed September 3, 2022. https://www.cdc.gov/drugoverdose/pdf/guideline_infographic-a.pdf

49. Gifford AL, Quach C. Prescription of opioid medications in VA outpatient care. US Department of Veterans Affairs. Accessed October 29, 2020. https://www.hsrd.research.va.gov/meetings/2001/HSRD2001AMab126.htm

50. Randall ML, Rosenbaum JR, Rohrbaugh RM, Rosenheck RA. Attitudes and behaviors of psychiatry residents toward pharmaceutical representatives before and after an educational intervention. Acad Psychiatry. 2005;29(1):33-39. doi:10.1176/appi.ap.29.1.33

51. Grande D. Limiting the influence of pharmaceutical industry gifts on physicians: self-regulation or government intervention? J Gen Intern Med. 2010;25(1):79-83. doi:10.1007/s11606-009-1016-7

52. Update for prescribers: new law regarding opioids. Mass.gov. March 15, 2016. Accessed September 4, 2022. https://www.mass.gov/news/update-for-prescribers-new-law-regarding-opioids

53. Prescribing policies: states confront opioid overdose epidemic. National Conference of State Legislatures. June 30, 2019. Accessed September 4, 2022. https://www.ncsl.org/research/health/prescribing-policies-states-confront-opioid-overdose-epidemic.aspx