Education as promotion

The resulting legal evidence stemming from United States litigation of brand name opioid manufacturers suggests that pharmaceutical industry promotion to health professionals was a key factor in seeding and exacerbating the epidemic of opioid-related overdose and mortality. Primarily, promotional and educational campaigns misrepresented and greatly downplayed the risk of addiction [7, 15]. Regulators have taken note: in Canada in June 2018, Health Canada called on opioid manufacturers and distributors to voluntarily stop all promotional activities and advertising to health professionals [16].

While there has been scrutiny of pharmaceutical sales practices, including large bonuses for representatives and coupons for free samples, opioid manufacturers’ involvement in medical and continuing education has also been flagged as another means by which companies promoted the over-prescribing of opioids [7, 8, 17]. Analysis of internal industry documents suggests that a key pharmaceutical industry marketing strategy is coordinated “education” campaigns comprised of a number of activities not typically recognized as ‘promotional’ including use of key opinions leaders, peer selling, and industry sponsorship of continuing education [18, 19]. Purdue Pharma’s promotion of a sustained-release oxycodone formulation included more than 40 national conferences attended by over 5000 health professionals, and sponsorship of more than 20,000 pain-related educational programs [7]. A case study of medical education at a research-intensive university found that an annual interprofessional pain curriculum, sponsored by pharmaceutical companies including opioid manufacturers, included pharmacotherapy lectures and free textbooks that emphasized the effectiveness of opioids for chronic non-cancer pain and minimized the risks of harm and addiction [15]. These lectures were delivered by physicians who received payments from opioid manufacturers, which was not always disclosed to students [15].

Despite being characterized as “educational,” industry-sponsored events are almost always associated with other payments and/or gifts: food and drink were provided at over 90% of pharmaceutical-industry sponsored educational events for health professionals in Australia between 2011 and 2015 [14]. Between 2013–2015, 1 in 12 United States physicians (and 1 in 5 primary care physicians) received a payment from an opioid manufacturer, which was most commonly in the form of food or beverages or speakers’ fees, suggesting that attendance at industry-sponsored drug dinners or other educational events is widespread [10].

Analyses of United States data suggest that this kind of sponsored education and related payments such as speakers’ fees, conference travel sponsorship, consultancies, and meals are associated with the increased and harmful prescribing of opioids. Researchers found that 7% of physicians who prescribed opioids under a national drug insurance program (Medicare Part D) received payments from opioid manufacturers; these physicians submitted 9.3% more opioid claims the following year compared with physicians who did not receive payments [11]. Analysing the impact of these payments at the county level per 1000 county population, researchers found that increases in the number of payments to physicians, the number of physicians receiving marketing, and the total monetary value of payments to physicians were associated with increased morality from opioid overdoses [12].

Opioid promotion to nurses

The focus of public, regulatory, legal, and academic scrutiny of opioid-related promotion has largely remained on promotion to physicians [10–12], and to some extent nurse prescribers [20]. In the United States, a nurse prescriber plead guilty to accepting $83,000 in kickbacks in exchange for prescribing an opioid medication, which she accepted in the form of speaking engagements at educational events that were often attended only by the company’s sales representatives or friends without prescribing authority [20]. However, the implications of opioid promotion to registered nurses without prescribing authority are poorly understood.

Nurses, which comprise the largest proportion of health professionals across health systems, frequently interact with industry representatives and also report receiving payments and gifts [21]. In Australia, 40% of pharmaceutical-industry sponsored educational events included nurses; in contrast, 21% of sponsored events included primary care physicians [22]. Through promotional campaigns, pharmaceutical companies seek not only to increase their market share, but also to expand their market through promoting awareness, screening, assessment of conditions for which the drug is indicated and expanding the indications for the drug, particularly to new settings and populations [23] Nurses play a key role in assessing pain, and administering opioid medications and particularly, medications prescribed ‘as needed,’ across multiple healthcare settings including acute care, primary care, and long term care. Nursing practice also attends to the multiple psychosocial factors that serve as determinants of pain and they play a large role in coordinating care, referrals, and the administration of alternative modalities for pain treatment [24, 25]. Thus, it is important to understand patterns in promotion targeted at nurses as well as physicians, particularly as nurses’ relationships with industry have received less policy attention than physicians’. In Australia, nurses’ relationships with industry are primarily governed by the industry Code of Conduct [26], and there is little guidance specifically for nurses from professional associations, the health system or the government. To date, there has been no investigation into the promotion of opioids specifically to nurses, thus we aimed to compare the nature of pain-related educational events sponsored by opioid manufacturers that were targeted at physicians and nurses.

Design

We conducted a cross-sectional content analysis of a public database of pharmaceutical industry-sponsored educational events for health professionals [14]. This study was exempt from ethical review according to the guidelines of the University of Toronto.

Setting

From October 2011 to September 2015, Medicines Australia required member companies to report the details of all industry-sponsored events for health professionals, defined as any registered health professional, and including registered nurses; companies posted PDF reports every 6 months on the Medicines Australia website. The database was created by Fabbri et al [14] who downloaded and compiled 301 separate PDFs from 42 member companies, converted these to Excel format, and cleaned the data for errors introduced in file conversion and to standardize the reporting. This dataset provides a unique case study to examine the nature of pharmaceutical industry-sponsorship of opioid- and pain-related events and particularly, those that include nurses.

In 2015, 12 opioid analgesics were available in Australia; 9 of these were also approved for subsidy by the Australian Government under the Pharmaceutical Benefits Scheme, including buprenorphine, codeine, fentanyl, hydromorphone, methadone, morphine, oxycodone, tapentadol and tramadol [27].

Sampling

Industry-sponsored educational events were included if they were sponsored by manufacturers of brand name prescription opioid analgesics approved by the Therapeutic Goods Administration (TGA) from 1996 to present, who are also member companies of Medicines Australia (Table 1). This time frame ensured that the product was likely on patent during the study time frame (Oct 2011 to Sept 2015). Thus, the inclusion criteria for events were sponsorship by a company:

• With Medicines Australia membership during the study period (Oct 2011 to Sept 2015)
• That manufactured a prescription opioid approved by the TGA from 1996 to present and whose primary indication was for analgesia

Exclusion criteria included:

• Sponsor not a member of Medicines Australia during the study period (e.g. generics manufacturers, opted out of membership)
• Product approved by the TGA prior to 1996 (e.g. codeine, codeine+paracetamol, pethidine, methadone, dextropropoxyphene)
• Product not on the market by September 2018
• Manufacturers of opioids whose primary indication is anaesthesia, substance use disorder treatment, or breakthrough cancer pain as these products have very low usage in comparison with other opioids [5]; in all cases, companies that manufactured opioids for these indications marketed only one product and sponsored few to no pain-related events
• Manufacturers of medicines for hospital-use only

Thus, we included all events sponsored by:

• Janssen-Cilag Pty Ltd: a subsidiary of Janssen Global, the pharmaceutical branch of Johnson & Johnson, manufactured pharmaceutical products in the areas of mental health, neurology, women health, haematology, gastroenterology, and pain management.
• Mundipharma Pty Ltd: founded in 1998, Mundipharma is a privately held company owned by the Sackler family (of Purdue Pharma) and part of the global network of Mundipharma companies, which marketed prescription drugs primarily in the area of pain management, as well as an asthma inhaler and oncology symptom management.
• bioCSL / Seqirus Australia Pty Ltd: a member of the CSL Group, a global specialty biotechnology company produced a range of vaccines, antivenoms, and pharmaceuticals related to pain management and allergies; the company is now known as Seqirus.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Prescription brand name opioid medicines approved by the TGA from 1996 to September 2018.

https://doi.org/10.1371/journal.pone.0248238.t001

Coding

We coded all events using an iteratively generated keyword search strategy that expanded the coding strategy established by Fabbri et al [14]. Fabbri et al [14] generated keywords based on theoretical variables of interest as outlined in the literature on industry-health professional interactions (e.g. clinical specialty, clinical setting, hospitality provided) and then iteratively expanded the set of keywords within each variable to account for the nature and range of events in the dataset (see S1 Table). We developed a coding scheme for pain-related content, which was first piloted by two independent coders on a random sample of pain-related events and refined among the team. Where information was missing in the “description of event” column, events were coded using keyword searches in the “venue” and “professional status” columns where content was unambiguous. We coded all Mundipharma events as “pain-related” unless the event explicitly stated otherwise, given the company’s limited product profile and the dominance of opioid medications within this portfolio. We further assumed that events were “pain-related” in the absence of details about the event content (e.g. 1 hour Journal club) given that Mundipharma is a member of the global network of Mundipharma companies (which are 100% owned by the Sackler family, the owners of Purdue Pharma) and Mundipharma’s aggressive promotion of opioid medications and particularly oxycodone internationally is well-documented [9]. We conducted additional keyword searches to exclude events as “not pain-related” based on Mundipharma’s non-opioid product offerings, which at the time of the study, included respiratory products for asthma, diabetes products, and oncology products related to lymphoma and anti-nausea agents (see S1 Table).

We coded all events for professional status (i.e. physician, nurse, etc) as ‘present’ or ‘absent’ using a keyword search in the “professional status” column. We then coded events into five mutually exclusive categories (see S1 Table):

1. Physician-only
2. At least one nurse in attendance
3. Nurses only
4. Other health professionals only
5. Other health professionals and physicians only

We elected not to analyse events for nurses with and without prescriptive authority separately, inferring that non-prescribing nurses are routinely included in pharmaceutical industry-sponsored events in Australia. In Australia, Nurse Practitioners (NPs) are a sub-category of registered nurses (RNs) who have prescriptive authority. In a previous analysis of transparency reports detailing payments from Australian pharmaceutical companies to individual health professionals, we found that companies did not always distinguish between RNs and NPs, referring to individuals as “nurses” [28]. However, nurses with an NP license likely account for a minority of nurses in receiving industry payments or attending sponsored events. Though NPs in Australia have a similar scope of practice to NPs in the United States (US), NPs in Australia are much fewer in number (1,319 in Australia vs 136,060 in US) and represent a smaller proportion of all RNs (representing about 0.005% of all Australian RNs vs about 5% of US RNs) [29]. From September 1st 2010, nurse practitioners were authorised to prescribe the opioid medications under study, though their prescribing was limited by scope of practice, state and territory prescribing rights, and contingent on the NP having a collaborative arrangement in place with a physician [30].

We coded the “venue” column to determine whether or not the event was held in a clinical setting and the “description” column to determine the type of event (e.g. journal club, grand rounds, conference etc) using the coding strategy established by Fabbri et al [14]. This coding strategy initially comprised types of events described in the literature on industry-health professional interactions and then iteratively added keywords to account for all event types in the dataset [14].

Analysis

To address the primary objective, the frequency and percentage of events were tallied overall and by company, and then broken down by whether the event was pain-related. Differences across companies in the latter were assessed using a chi square test or Fisher’s exact test when the expected count was less than 5. Among the pain-related events, the frequency and percentage of events were compared across professional groups and then further broken down by event location and format. Total costs of events and costs for food and beverages (in Australian dollars) were summed overall and by professional group. Median costs per event were calculated overall and by professional group and presented with an Inter-Quartile Range (IQR). Differences across groups in median costs were assessed using the Kruskal-Wallace test. The frequency and percentage of pain-related events was reported for each of the 12 content foci variables and compared across professional groups using a chi square test or Fisher’s exact test, as above. Statistical significance was established for omnibus tests at p < .05 and adjusted for post-hoc tests using the Bonferroni correction (i.e., p < .05/3 or .017). All analyses were undertaken in SAS (version 9.4).

Policy implications

Despite the routine attendance of nurses and other non-physician clinicians at industry-sponsored events, there is little policy guidance or regulation of these relationships. Non-prescribing registered nurses, pharmacists and other allied health professionals are covered by the Medicines Australia and European Federation of Pharmaceutical Industries and Associations (EPFIA) Codes of Conduct, both of which represent a form of industry self-regulation. While comprehensive in terms of the recipients covered, these self-regulatory approaches fall short of comprehensive disclosure of all forms of industry payments (i.e. frequent exclusion of food and beverages, drug samples) and apply only to members of these trade associations, who disclose on a voluntary basis [40, 41]. Legislation such as the United States Physicians Payments Sunshine Act has addressed these limitations in requiring comprehensive, standardized, and enforceable reporting. Though several categories of nurses with prescribing authority (e.g. nurse practitioners, clinical nurse specialists) will be covered by this legislation beginning in 2022 [42], policymakers should consider enacting transparency legislation that covers all registered health professionals. While transparency is necessary, it is insufficient to mitigate or prevent harms associated with industry relationships. Thus, professional associations, regulators and accrediting bodies for nurses, pharmacists and allied health professionals need to develop guidance related to ethical interaction with industry and management of conflicts of interest. This will likely be most effective in practice if developed and implemented in a way that is inclusive of all members of the healthcare team.

Regulators and industry trade associations have attempted to distinguish between companies’ “promotional” and “non-promotional” activities [26, 43], with health professional education and disease awareness campaigns typically included in the latter. Regulators should instead consider regulating promotional campaigns, comprised of sales, marketing, advertising, promotion, research and educational activities, as a whole to restrict the promotion of products that pose a high risk of harm. However, this also raises the question of who should be responsible for delivering education about pharmaceuticals to healthcare professionals and suggests pragmatic challenges as industry-independent educational interventions are in no way comparable in terms of resources, reach and duration to industry-sponsored educational events. Two Australian trials examined the effects of independent chronic pain management education programs among general practitioners and found no significant effect on prescribing habits or intent [44, 45]. However, a systematic review and meta-analysis of interventions to improve prescribing practices for chronic non-cancer pain found very low-certainty evidence that education could indirectly improve prescribing practices through increased uptake of treatment agreements, urine drug testing, and use of prescription drug monitoring programs [46]. Thus, the question is perhaps not how to deliver ‘counter’ education, which may not be effective on its own in mitigating the harms associated with industry-sponsored education, but how it might be incorporated into multi-faceted [46], multidisciplinary, and structural efforts to improve the quality use of medicines and reduce opioid-related harms.

Strengths and limitations

This was a cross-sectional, descriptive analysis of data that were publicly reported by pharmaceutical companies; thus, we were restricted by the format in which data were reported and could not verify their accuracy or completeness. Further, there was a high degree of variability in terms of the specificity and comprehensiveness of reporting with insufficient detail to determine the content focus of about 1/3 of sampled events. We iteratively coded data using keywords according to each variable of interest and due to the variability of the underlying reports, we may have missed some synonyms or categories. We may have overestimated the number of events that were “pain-related” and sponsored by Mundipharma in assuming that events were pain-related in the absence of details about the event content or attendee specialty. However, given that Mundipharma’s product portfolio is dominated by pain-related medications and the promotion of opioid medications by Mundipharma companies both overtly and covertly is well-documented, we believe this was a reasonable assumption [9]. The original reports provided the total number of attendees and listed the professional status of attendees (i.e. physician, registrar, nurse, pharmacist etc); however, the reports did not specify the number of attendees by professional status. Thus, we could only make comparisons based on whether certain professionals were present or absent. Despite these limitations, this analysis is the first to describe the content of educational events sponsored by opioid manufacturers and also the nature of opioid-related promotion to nurses.