Many patient organisations collaborate with drug companies, resulting in concerns about commercial agendas influencing patient advocacy. We contribute to an international body of knowledge on patient organisation-industry relations by considering payments reported in the industry’s centralised ‘collaboration database’ in Sweden. We also investigate possible commercial motives behind the funding by assessing its association with drug commercialisation.
Our primary data source were 1,337 payment reports from 2014–2018. After extraction and coding, we analysed the data descriptively, calculating the number, value and distribution of payments for various units of analysis, e.g. individual companies, diseases and payment goals. The association between drug commercialisation and patient organisation funding was assessed by, first, the concordance between leading companies marketing drugs in specific diseases and their funding of corresponding patient organisations and, second, the correlation between new drugs in broader condition areas and payments to corresponding patient organisations.
46 companies reported paying €6,449.224 (median €2,411; IQR €1,024–4,569) to 77 patient organisations, but ten companies provided 67% of the funding. Small payments dominated, many of which covered costs of events organised by patient organisations. An association existed between drug commercialisation and industry funding. Companies supported patient organisations in diseases linked to their drug portfolios, with the top 3 condition areas in terms of funding–cancer; endocrine, nutritional and metabolic disorders; and infectious and parasitic disorders–accounting for 63% of new drugs and 56% of the funding.
This study reveals close and widespread ties between patient organisations and drug companies. A relatively few number of companies dominated the funding landscape by supporting patient organisations in disease areas linked to their drug portfolios. This commercially motivated funding may contribute to inequalities in resource and influence between patient organisations. The association between drug commercialisation and industry funding is also worrying because of the therapeutic uncertainty of many new drugs. Our analysis benefited from the existence of a centralised database of payments–which should be adopted by other countries too–but databases should be downloadable in an analysable format to permit efficient and independent analysis.
Citation: Mulinari S, Vilhelmsson A, Rickard E, Ozieranski P (2020) Five years of pharmaceutical industry funding of patient organisations in Sweden: Cross-sectional study of companies, patient organisations and drugs. PLoS ONE 15(6): e0235021. https://doi.org/10.1371/journal.pone.0235021
Editor: Joel Lexchin, York University, CANADA
Received: May 5, 2020; Accepted: June 5, 2020; Published: June 24, 2020
Copyright: © 2020 Mulinari et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: The dataset used in this study is available here: http://doi.org/10.5281/zenodo.3886331.
Funding: This work was supported by grants from the Swedish Research Council for Health, Working Life and Welfare (www.forte.se) (FORTE #2016-00875 to SM (PI) and PO) and the Crafoord Foundation (www.crafoord.se) (to SM (PI) and PO and AV). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Patient organisations are increasingly involved with healthcare services and systems [1, 2] including in relation to pharmaceuticals [3, 4]. The increasing scope of activity results in patient organisations encountering many challenges in developing their capacity and expertise [1, 5]. Some of these challenges may be addressed by funding and other support provided by pharmaceutical companies [6, 7, 8, 9, 10, 11]. However, partnering with companies also raises concerns about conflicts of interest weakening patient organisations’ independence from commercial agendas [4, 11,12, 13, 14, 15, 16, 17, 18] that do not always coincide with the interest of patients and public health [19, 20]. For example, several industry-funded patient groups at the European Union level supported the industry’s efforts to relax the ban on direct-to-consumer advertising, but this agenda was strongly opposed by patient groups that did not accept industry funding .
One response to these concerns has been to enhance the transparency of financial relationships between the two sides . This is part of a global policy trend in which transparency, primarily understood as public disclosures, is applied to individuals’ and organisations’ ties to pharmaceutical companies [23, 24]. Perhaps the most comprehensive disclosure initiative pertaining to patient organisations has been introduced by the European Federation of Pharmaceutical Industries and Associations (EFPIA), representing pharmaceutical companies operating in Europe, via its Patient Organisation Code of Practice . Implemented through industry self-regulation since 2012, the Code requires pharmaceutical industry trade group members to disclose payments to patient organisations, which are typically published as annual reports on company websites .
As a result of this and similar transparency initiatives elsewhere in the world, there has been a spur in research on pharmaceutical company funding of patient organisations in the United Kingdom (UK) [4, 22, 26], Australia  and United States  using industry disclosure data. Industry disclosures have clear advantages over other data sources, such as surveys of patient organisations [14, 29] or their websites [30, 31], due to their greater coverage and standardisation . The emerging picture from this research is that many patient organisations receive industry funding but that relatively few receive a large share of it [26, 27]. Most of the funding goes to research and public engagement, including advocacy, campaigning and disease awareness, and lobbying . Furthermore, companies predominantly fund organisations in commercially high-profile areas, such as cancer and diabetes, indicating mercantile rather than philanthropically motivated support . However, only a few countries have been surveyed making generalisations problematic, especially since the pattern of patient organisations’ relations with industry may vary between countries .
Here, we contribute to building an internationally comparative body of knowledge by considering industry reports of payments to patient organisations in Sweden, using the same methodology as in the previous UK study . Sweden has a history of pharmaceutical industry self-regulation  and a large patient organisation community: roughly 1 in 20 Swedes are affiliated to a patient organisation . Another reason for selecting Sweden for this case study is that the Swedish pharmaceutical industry trade group (Läkemedelsindustriföreningen, LIF) has gone further than trade groups in other European countries that deem it sufficient to disclose payments on the sponsoring company’s webpage, and which in the UK was shown to make it near-impossible to establish the scope of industry involvement for any patient organisation [22, 26]. In contrast–and consistent with recommendations from the UK –since 2005 LIF has a centralised and searchable (albeit not downloadable) database where companies upload reports for every payment. The aims of this study were therefore to assess:
- the pattern of industry engagement with patient organisations by looking at available payment reports
- the association between drug commercialisation and industry funding of patient organisations
- the advantages of a centralised database for identifying and understanding the nature of industry funding of patient organisations
Data and coding
We extracted 1,412 payment reports from the Swedish pharmaceutical industry’s trade group’s “collaboration database” , covering a 5-year period, from January 2014 to December 2018. We selected a 5-year period to have a fairly large but still manageable sample, but also because industry rules allow reports to be deleted after three years which limited data availability. Data was collected in two phases: data from January 2014 to December 2016 were collected in November and December 2016, and data from January 2017 to December 2018 were collected in April and May 2019. Two authors (SM and AV) manually extracted the data into an Excel database, including the report title, project start and end dates, sponsoring company, recipient organisation, and project and payment descriptions, which are all contained in separate fields. SM extracted the data from January 2014 to December 2016 and AV the data from January 2017 to December 2018 in dialogue with SM. AV checked the data extracted by SM from June 2016 to December 2016 but found no discrepancies. Each entry was coded separately for payment category (e.g. sponsorship, grant, fees for service and consultancy) and goal (e.g. policy engagement, education and training) . Given our focus on the national level, we aggregated local and regional patient organisations into their parent national association. Each patient organisation was coded according to their condition and disease area using ICD-10 (International Classification of Diseases, 10th revision). Most payments were reported in Sweden’s currency, SEK. We expressed all payments in 2018 EUR using average annual exchange rates available from Sweden’s Central Bank. To adjust for inflation, we used a yearly consumer price index obtained from Statistics Sweden (2014 = 4.76%, 2015 = 4.81%, 2016 = 3.79%, 2017 = 1.95%, 2018 = 0).
Descriptive data analysis
To examine the pattern of the disclosed payments, we performed descriptive analyses calculating the number, value and distribution (median, interquartile range (IQR)) of payments. The descriptive analysis was specified beforehand to the following units of analysis validated in a previous UK study : all companies and patient organisations, individual companies and patient organisations, condition and disease areas, and payment categories and goals (S1 Table). In short, the payment category codes were initially devised based on the codes used by EFPIA to categorise pharmaceutical company payments to healthcare organisations (e.g. hospitals, universities, medical associations), specifically, “grants”, “contributions to costs of events”, “travel, accommodation and registration fees”, “fees for service and consultancy”, and “sponsorship”‘. These codes were then supplemented with an inductive approach for any emerging payment categories that were unique to the patient organisation payment descriptions, such as “support and help” [22, 26]. Separately, payment goal was coded based on close iterative reading of payment descriptions and aggregating similar descriptions under the same codes [22, 26]. When coding payment goals we looked for the main purpose of activities funded by drug companies. Furthermore, for each patient organisation, condition and disease area, we identified the supporting companies as well as the main donor’s share of the overall funding.
During the coding, we realised that a substantial amount of the funding–especially of policy engagement and advocacy–concentrated to the so-called Politician’s Week held every summer, known nationally also as the Almedalen Week, which we therefore included as another unit of analysis. The Politician’s Week can be described as a micro-cosmos of Swedish politics where politicians, ministerial advisors, governmental and special interest organisations, companies, think thanks, lobbyist, policy- and PR consultants and journalists, gather for debate, advocacy and lobbying, marketing, and social and political networking. Thus, Politician’s Week funding may provide special insight into joint lobbying efforts.
Association between drug commercialisation and industry funding
We began by assessing the concordance (Cohen's κ) between companies marketing drugs in specific ICD-10 disease areas and their funding of corresponding patient organisations. To make the analysis practical, we considered the top ten donors overall in the ten most funded disease areas. The online Swedish Medicines Compendium–FASS–has webpages for every pharmaceutical company which lists all products marketed in Sweden with weblinks to the Summary of Products Characteristics that contain information on approved indications. We used information on approved indications to determine whether or not companies marketed at least one drug in each of the ten selected diseases. For a drug to be considered marketed in a disease it had be indicated for the disease, e.g. drugs approved in HIV/AIDS patients for combating non-HIV infections, such as fungal infections, were not HIV drugs. Similarly, analgesics indicated for cancer patients were not cancer drugs.
In a separate analysis we calculated the correlation between the commercialisation of new drugs in ICD-10 condition areas and payments to patient organisations in those condition areas by all LIF companies over the study period. We performed the analysis at the level of broader conditions (e.g. neoplasms), rather than narrower diseases (e.g. breast or prostate cancers), because the former is less sensitive to the fact that our drug sample for this analysis only included drugs approved 2014–2018 and that drugs’ indications may broaden over time. Furthermore, an important share of the funding, especially to cancer patient organisations, could not be differentiated at the level of specific diseases. In a first step, we compiled data on all new prescription drugs (i.e. New Active Substances (NAS) but excluding vaccines) from LIF companies that were approved by the European Medicines Agency (EMA) between 2014–2018, including information on date of approval, drug company and approved indication, based on information on the EMA and FASS websites. Next, we coded the drugs according to ICD-10 categories. Finally, using Spearman’s rank-order correlation, we assessed the correlation between the number of new drugs approved in 2014–2018 for various ICD-10 conditions as well as the number of LIF companies marketing these drugs per ICD-10 condition area and, separately, the number and value of payments in ICD-10 condition areas. Spearman’s rank-order correlation, rather than Pearson correlation, was chosen because of the large number of cancer drugs and payments to cancer patient organisations. A p-value below 0.05 was considered significant. Prism 8.2.1 for Macintosh (GraphPad Software Inc.) was used all statistical analyses.
Overview of industry payments
Over the 5-year period, 2014–2018, 46 pharmaceutical companies reported 1,412 relationships with the local, regional or national branches of 77 patient organisations. However, 75 reports (5.3%) lacked information on the value of the payment, in violation of industry rules, and were subsequently excluded. This left us with 1,337 reports totalling €6,449,224, including 56 (4.2%) reporting zero-value payments (i.e. relationships without financial support). As shown in Table 1, there was an increase of annual payments between 2014–2016, reaching €1,583,581 (n = 298) in 2016, followed by a drop in 2017 and in 2018 –to €1,091,558 (n = 257). The funding landscape was dominated by small payments with roughly one quarter being smaller than €1,000 (26.0%; n = 348), and three quarters (77.9%; n = 1041) below €5,000 (median €2,411; IQR €1,024–4,569). However, the ten largest payments constituted roughly 15% of the total value (€942,623).
Donors and recipients
There was a high concentration of payments among companies. The top ten donors provided 67.9% (€4,379,604) of the funding (Table 2; S2 Table for full table). Pfizer was the major donor with 14.8% (€954,234) of the reported funding whereas AbbVie had the greatest number of payments (10.9%; n = 146). Pfizer also made the largest single payment, worth €353,179, to the Swedish Breast Cancer Association in 2016 to support the creation of a website for information dissemination and advocacy (S3 Table). Pfizer also made the second largest payment to the same patient organisation in 2018 worth €97,497 supporting the same project. Besides Pfizer, AbbVie was the only other company that had more than one payment in the top ten.
As shown in Table 3, there was also a high degree of concentration of payments among patient organisations. Of the 77 patient organisations that received funding, the top ten recipients amassed 62.4% (€4,026,410), and the top recipient, the Swedish Breast Cancer Association, alone received 10.3% (€663,774), from just five companies. The Network Against Cancer–an umbrella coalition of cancer patient and advocacy organisations and networks–had the greatest number of company funders (n = 16) but was only the fourth most funded organisation.
Category and goal of funding
The top funding priority was supporting patient organisations’ engagement with outside audiences (Table 4). Thus, the most common payment category was “contribution to costs of events” organised by the recipient patient organisation or third parties, for example an external meeting or lecture, attracting 35.7% (€2,300,750). Other main payment categories were “support and help” (20.5%; €1,325,363), “partnership arrangements” (15.8%; €1,021,968) and “sponsorships” (12.9%; €831,430).
Furthermore–looking at the goal of the funding–public involvement, including “communication”, “advocacy, campaigning and disease awareness” and “policy engagement” together attracted the lion’s share of the funding (61.3%: €3,953,597). In contrast, “support for patients” attracted just 5.5% (€356,801) and “research-related activities” 0.9% (€59,544).
There were some payments for which the category or the goal of funding (or both) could not be determined due to lack of information, totalling €181,610 (2.8%) and €59,228 (0.9%), respectively.
Conditions and diseases
Table 5 reports on broad condition areas and narrower disease areas by amount of funding received. Of the 17 ICD-10 condition areas, the top five amassed 69.6% (€4,485,252), with neoplasms alone attracting 37.5% (€2,419,186). This was followed by infectious and parasitic (9.1%; €590,026), endocrine, nutritional and metabolic (9.1%; €587,605), musculoskeletal (6.9%; €446,718) and digestive system (6.8%; €441,717) diseases.
However, within the broader condition areas, we observed a clear hierarchy among diseases. Thus, within neoplasms, cancers of breast, blood (lymphoid, hematopoietic and related tissues) and prostate amassed the bulk of funding (60.6%) whereas, for example, cancer of the female genital organs (1.9%) and skin (0.1%) received comparably little funding. This pattern was even more evident for some other condition areas. For example, virtually all funding for infectious and parasitic diseases went to HIV and hepatitis C. Overall, the top ten diseases in terms of funding accumulated 58.6% (€3,777,683). By far, most funding went to breast cancer with 10.3% (€663,774), followed by HIV (7.3%; €469,074) and blood cancers (7.2%; €463,378).
As shown in Table 5, companies tended to invest selectively and heavily in certain disease areas, which had a major effect on the overall pattern of funding. Pfizer, for example, provided most of the funding of breast cancer advocacy (74.1%; €492,074). Similarly, Astellas provided 43.9% (€148,674) of the funding of prostate cancer advocacy, whereas GlaxoSmithKline dominated HIV (76.9%; €360,702) and AbbVie Psoriasis (47.2%; €103,626) and hepatitis C (95.4%; €96,526) funding. At the same time, certain disease areas had many funders, such as cancers in general (n = 16) and blood cancers (n = 10) and diabetes (n = 9), resulting in overall substantial sums.
The politician's week: Joint industry-patient organisation lobbying
A particular aspect of the industry funding landscape in Sweden was the substantial and recurrent funding of patient organisations during the yearly Politician’s Week, or Almedalen Week. Payment reports shows that the Politician’s Week is an important arena where patient organisations, together with one or more companies, seek to frame policy debates on certain diseases and treatments, especially cancers. Between 2014 and 2018, 8.7% (n = 116) of all payment reports pertained to the Politician’s Week, with a total value of €704,763 (10.9%). Nineteen patient organisations (24.7%) received funding from 22 companies (46.8%). The companies mostly contributed to “events” organised by patient organisations (51.0%; €359,458), followed by “partnership arrangements” between patient organisations and industry (23.2%; €163,433). The goals of the Politician’s Week payments were mainly “advocacy and disease awareness” attracting 46.6% (€327,938), followed by “policy engagement” (22.2%; €156,643) and “communication” (20.0%; €140,776). About half of the funding during the Politician’s Week concerned neoplasms (51.1%; €360,186), mostly cancer in general (all through The Network Against Cancer) (56.8%; €204,694). Payments combining “policy engagement” and neoplasms attracted 21.3% (€150,455) of all funding during the Politician’s Week, representing as much as 73.6% of the total funding for “policy engagement” between 2014–2018.
Notably, 68.1% (n = 79) of all reported Politician’s Week payments pertained to activities co-funded by more than one company (52.8%; €372,154), often targeting policy- and decision-makers through “advocacy” (29.2%; €108,563) or “policy engagement” (27.6%; €102,882). Co-funded activities could involve as many as twelve companies, for example, aiming to lift the issue of regional inequalities in the use of cancer drugs onto the political agenda prior to the September 2018 parliamentary elections; or, as in 2015, a seminar on the “patient perspective” which, in addition to cancer patient organisation representatives, included speakers such as the Minister for Health and other high-level politicians, and the Chairpersons of the Swedish Medical Association and the Swedish Association of Health Professionals.
Commercial motives for funding
Table 6 shows the concordance between commercialisation of drugs and the funding of patient organisations in the ten most funded disease areas (including cancer in general) for the top ten donors overall. If a company marketed at least one drug in a disease area, there was an 83% chance that it supported a patient organisation in the disease area (κ = 0.78, 95% confidence interval: 0.66–0.90). Companies only supported patient organisations in disease areas linked to their drug portfolio. AbbVie’s support of Hemophilia patient advocacy appears at first to be an exception to this rule; however, hundreds of hemophilics in Sweden were infected by hepatitis C in the 1980s due to the use of virus-contaminated blood in transfusions, and AbbVie markets hepatitis C drugs.
To further investigate the possible link between commercial motives and funding, we assessed the correlation between the number of new drugs in 2014–2018 marketed by LIF companies in different condition areas and payments to patient organisations in those condition areas. Overall, 139 new drugs were marketed by LIF companies (n = 42). As many as one third (n = 46) were cancer drugs, followed by drugs for treating endocrine, nutritional and metabolic disorders (n = 23; 16.5%), and infectious and parasitic disorders (n = 19; 13.7%). Together, these three condition areas accounted for 56% of the funding over the study period (Table 5). In contrast, there were only three new drugs for treating mental, behavioural and neurodevelopmental disorders, and only one for treating ophthalmologic conditions. Across condition areas, there was a very strong, positive monotonic correlation between the number of new drugs marketed by LIF companies and the number of payments (Rs = 0.85, n = 16, p = 0.00003) and the value of those payments (Rs = 0.78, n = 16, p = 0.0004). There was also a very strong, positive monotonic correlation between the number of LIF companies marketing new drugs in different condition areas and the number of payments (Rs = 0.84, n = 16, p = 0.00005) and the value of those payments (Rs = 0.77, n = 16, p = 0.0005) in those condition areas. Results were similar for the subset of LIF companies (n = 30) that marketed new drugs (n = 118; 85%) and reported payments over the study period (€5,868,363; 90.0%): Rs = 0.84–0.75, n = 16, p<0.0009, for all analyses.
This study’s finding of close relations between patient organisations and pharmaceutical companies is consistent with what has been reported from other countries, including the UK [4, 26], United States [13, 14, 17, 28, 35], Australia [11, 27], Canada , Italy  and Finland . Between 2014 and 2018, 77 Swedish patient organisations were reported to receive support from 46 companies, involving 1,337 payments with a cumulative value of roughly €6.4m. In addition, there were 75 (5.3%) payments reported which lacked information on the value of the payment, and which were therefore excluded from the analysis. The support was dominated by small payments, many of which were provided to cover costs of events organised by patient organisations, also consistent with findings from other countries [26, 27]. However, our data shows how even small payments may result in substantial financial support from industry since companies may co-fund projects, or one company may repeatedly fund the same or similar projects.
To contextualise our findings, and to enable broader generalisations, we tabulate selected features of industry payments in Sweden (this study) and the UK (reported in previous study using similar methodology)  (Table 7). Taking population size into account, the total value of payments was about 1.5 times greater in the UK, and the number of payments 1.9 times greater. There are two likely contributors to this difference. First, there were more reporting companies and recipient patient organisations in the UK. This may reflect country differences in, for example, drug commercialisation strategies (e.g., leading global role of UK drug market), government administration (e.g., healthcare in the UK is devolved to England, Scotland, Wales and Northern Ireland, possibly potentiating UK country-specific patient organisations) and availability of other funding (e.g., government funding seems more prominent in Sweden, possibly reducing the need for commercial sources of funding ). Second, large-size payments were more common in the UK, and they were often of a greater value, as evidenced by the larger median and upper quartile values. Notably, many large-size payments in the UK were linked to research-related activities which accounted for 25% of the value of payments but only 3% of the number of payments. In Sweden, we found almost no funding of research-related activities. In contrast, most funding (61%) went to public involvement activities, encompassing communication, advocacy, campaigning, disease awareness and policy engagement, which typically involves smaller-size payments. This finding suggests marked country-differences in the role of patient organisations in the development of new medicines.
Nevertheless, some striking similarities existed in funding patterns in Sweden and the UK. The hierarchy of funding across condition areas was very similar, with the same top 3 condition areas receiving virtually identical shares of funding in the two countries, including 38% and 36% for neoplasms in Sweden and the UK, respectively. This finding suggests that strong commercial motives pattern companies’ funding decisions, including an impetus to support cancer patient advocacy following the many cancer drug launches [35, 37]. However, our study further strengthens this argument. First, we show how companies tend to invest selectively and heavily in diseases linked to their drug portfolios, which shaped the overall pattern of funding across conditions and diseases. Second, we show a very strong correlation between, on the one hand, the number of new drugs and number of companies marketing new drugs in different condition areas and, on the other hand, the payments to patient organisations in different condition areas.
Another striking resemblance between Sweden and the UK was the concentration of funding across donors. In Sweden, only ten companies provided 68% of reported funding; in the UK this was 69%. The picture of a limited number of companies dominating relations is strengthened by findings from the UK showing high concentration of payments at the donor level to healthcare professionals (50% for the top 10 companies)  and organisations, such as hospitals and primary care centres, (59% for the top 10 companies) .
The broad picture, then, is of relatively few companies dominating relations by channelling their funding to patient organisations (and most likely other actors) in disease areas linked to their drug portfolios. As a consequence, much funding has been awarded to organisations for (some) cancer patients, such as those suffering from breast, blood or prostate cancers, but also diabetes and HIV. In stark contrast, limited funding goes to organisations for patients facing other major public health challenges, such as mental illness, but for which there have been few or no recent drug launches. Indeed, the top ten areas in terms of disease burden in Sweden in 2017 were ischemic heart disease (8.5% of total Disability-Adjusted Life Years), low back pain (6.1%), stroke (4.2%), headache disorders (3.9%), diabetes (3.7%), Alzheimer’s disease (3.6%), COPD (3.6%), falls (3.3%), depressive disorders (3.1%) and lung cancer (2.8%) [40, 41]. Yet, diseases of the circulatory system only amassed 4.6% of the value of payments compared to cancer’s 37.5%, despite ischemic heart disease being the leading cause of death and disability in Sweden (although decreasing since 2007), whilst mental, behavioural and neurodevelopmental disorders only received 0.7% of all funding, and depressive disorders were not represented at all.
A concern with this commercially patterned funding is how it may create inequalities between patient organisations representing different diseases not only in terms of differential resources but also differential access to medical and political expertise and the media . This risk is exemplified by the extensive joint campaigning and lobbying around certain diseases, especially cancers, during Sweden’s annual Politician’s Week. At this political, social and media event certain patient organisations, with support from companies, gain access to politicians and other policy- and decision-makers as well as the media. We found that 10.9% of all funding, and 14.9% of all cancer patient group funding, concentrated on the Politician’s Week, and a substantial portion of this went to policy engagement. The Politician’s Week payments also underscores the importance of taking into account the industry-level agenda for understanding funding patterns. Thus, as many as twelve companies could fund a single event targeting high-level policy and decision-makers.
In addition, we believe that the close link between funding and commercialisation are concerning in light of the uncertainty around the therapeutic value of many new drugs [42, 43, 44, 45, 46]. The problem of therapeutic uncertainty was exemplified by the EMA’s recent withdrawal of the soft tissue cancer drug Lartruvo–which the EMA had approved in 2016 on the basis of preliminary data–after the company failed to demonstrate the drug’s benefits in a post-authorisation confirmatory trial . More generally, a recent study found that more than half of randomised controlled trials for cancer drugs in Europe approved in 2014–2016 had flaws likely to exaggerate treatment benefits . Only one-quarter measured survival as a key outcome, and fewer than half reported on patients’ quality of life. Ten of these new cancer drugs–Cometriq, Mekinist, Zydelig, Vargatef, Imbruvica, Kyprolis, Darzalex, Ibrance, Opdivo and Empliciti–were marketed by companies that supported a corresponding Swedish patient organisation and/or the Network Against Cancer. The support included several key cases of policy engagement. We cannot make judgements regarding the impact of industry funding on any particular patient organisation or activity. However, there is arguably a danger that patient organisations–who are unlikely to have access to all trial documentation for novel drugs, nor the capacity to undertake independent review of this material–may become direct or indirect champions of drugs that are putting health systems under great financial strain but whose therapeutic benefits remain uncertain.
Prior research has debated the advantages and shortcomings of industry self-regulation and government regulation for ensuring public disclosure of payments to patient organisations. For example, Kang et al  suggested implementing a legal disclosure requirement in the United States modelled on the Physician Payments Sunshine Act, rather than self-regulation, to ensure centralised payment reporting by companies because government regulation is better at compelling universal compliance. However, the Swedish example shows that a centralised database can in principle be achieved with self-regulation. The Swedish example also shows that national drug industry trade groups can, should they chose to do so, go beyond the rules set out by the industry at the European level that only requires of companies to disclose shorter descriptions of payments to patient organisations on their websites on an annual basis . In April 2020, the Swedish pharmaceutical industry trade group updated its Code of Practice with some further innovations that could also be adopted by other countries . Most importantly, companies in Sweden should now report payments to individual ‘expert’ patients and caregivers in the database in addition to the payments to patient organisations. Moreover, companies are no longer allowed to pay for patient organisations’ costs for travel, accommodation and food at meetings and conferences, unless the patient organisation representatives are acting as consultants for the company. This latter rule extends the ban that already existed on paying for health professionals’ costs for travel, accommodation and food at meetings and conferences .
It is tempting to speculate that the industry’s more stringent rules and disclosure requirements in Sweden compared to, for example, the UK [22, 26] relate to Sweden’s historically more transparent approach to government and business regulation, including in the pharmaceutical area . However, we note that for the disclosure of payments to healthcare professionals and organisations it is the industry in the UK, but not in Sweden, that established a centralised database of payments, and that several European countries have opted for government regulation to guarantee public disclosure of payments to healthcare professionals . This underscores the point that Sweden’s self-regulatory system also has significant room for improvement . Most pertinent to this study, disclosure databases need to be downloadable in an analysable format (e.g. CSV) to permit efficient and independent analysis. In addition, there is no reason why payment reports should be deleted after three years as is permitted under the current rules. Much can be also done to improve the standardisation and consistency of reporting, for example by linking payment descriptions to pre-established payment categories and goals; ensuring that costs are reported in the same way by all companies; and that information is always complete and represent actual costs (see below); and requiring that companies specify–in a standardised way–the disease or diseases in relation to which the payment is made. Finally, the fact that more than 5% of reports lacked clear information on the value of the payment demonstrates the need for improving compliance and quality control of the database.
Limitations and strengths
One key limitation of our study is that we had to rely on the information provided by companies with no possibility of independently verifying the data. Furthermore, whilst most drug companies active in Sweden are trade group members or subscribe to the Code, and are therefore expected to report in the database, there are exceptions, e.g. Vertex and Gedeon Richter; therefore, our study likely underestimates the volume of payments. A further limitation is that many companies appear to enter information in the database after a formal arrangement between parties has been reached, but before the payment has been made, which means they report expected or maximum costs rather than the actual cost. This is a major limitation because we cannot know if there are differences between expected or maximum costs and the real costs. Another limitation is that each data entry was extracted and coded by one author, but our spot checks by a second author did not identify any discrepancies. Finally, we had to exclude 75 reports (5.3%) that lacked information on the value of the payment. Because of these limitations the values we present should be interpreted with some caution.
A further limitation pertains to our analysis of the association between drug commercialisation and industry funding across conditions and disease areas. Most importantly, we cannot exclude that some of the observed differences are caused by differences in patient organisations’ willingness to accept commercial funding. Related to this, we cannot know what effect, if any, funding had on patient organisations. Future studies should therefore increasingly investigate the perspectives and motives of patient organisations and their representatives .
That said, the industry payment reports in Sweden have clear advantages compared to previously used data sources. Most importantly, Sweden’s centralised and searchable database of payments means greater transparency and greater certainty that all disclosures are included in the analysis. Furthermore, our impression is that the Swedish database contains more detailed descriptions of the payments compared to the reports published on company websites. This seems to translate into less ambiguity regarding the purpose of the funding. Indeed, compared to the UK, likely as a result of more detailed information, much fewer payments were coded as unclear with respect to the goal of funding (0.9% vs. 7.6%). The more detailed information also allowed us to characterise the extensive patient organisation-industry joint campaigning during a national lobbying event that merits follow up in future research.
Finally, a key strength of our study is the use of the same methodology as in the previous UK study  which allowed us to identify similarities and differences in the pattern of funding between the countries. Future research should collect and analyse comparable data from a larger sample of countries to improve the generalisation of results.
This Swedish study reveals close and widespread ties between patient organisations and drug companies. The broad picture is of a relatively limited number of companies dominating relations by supporting patient organisations in disease areas linked to their drug portfolios. The Swedish disclosure initiative’s online database has clear advantages for identifying and understanding the nature of industry funding, but the lack of downloadability and standardisation made the analysis burdensome. It is possible that the close ties between patient organisations and companies reflect coinciding interests. However, because funding is commercially motivated it will create inequalities between patient organisations in different disease areas even if there are coinciding interests between donors and recipients. Furthermore, the strong association between drug commercialisation and industry funding is arguably problematic in light of the uncertainty surrounding the therapeutic value of many new and typically expensive drugs.
S1 Table. Coding manual for payment categories and goals.
S2 Table. Drug industry funders of Swedish patient organisations (2014–18).
- 1. Löfgren H, de Leeuw E, Leahy M. 2011. Democratizing health: Consumer groups in the policy process: Edward Elgar Publishing.
- 2. Wehling P, Viehöver W, Koenen S. 2014. The public shaping of medical research: patient associations, health movements and biomedicine: Routledge.
- 3. Menon D, Stafinski T. Role of patient and public participation in health technology assessment and coverage decisions. Expert Rev Pharmacoeconomics Outcomes Res. 2011;11(1):75–89.
- 4. Mandeville KL, Barker R, Packham A, Sowerby C, Yarrow K, Patrick H. Financial interests of patient organisations contributing to technology assessment at England’s National Institute for Health and Care Excellence: policy review. BMJ. 2019;364:k5300. pmid:30651227
- 5. Baggott R, Forster R. Health consumer and patients' organizations in Europe: towards a comparative analysis. Health Expect. 2008;11(1):85–94. pmid:18275405
- 6. Kent A. Should patient groups accept money from drug companies? Yes. BMJ. 2007;334(7600):934. pmid:17478845
- 7. Taylor J, Denegri S. Industry links with patient organisations. BMJ. 2017;356:j1251 pmid:28292970
- 8. Haerry D, Landgraf C, Warner K, Hunter A, Klingmann I, May M, et al. EUPATI and patients in medicines research and development: guidance for patient involvement in regulatory processes. Front Med. 2018;5:230.
- 9. di Priolo SL, Fehervary A, Riggins P, Redmond K. Assessing stakeholder opinion on relations between cancer patient groups and pharmaceutical companies in Europe. Patient. 2012;5(2):127–39. pmid:22299759
- 10. Pearcy J. ABPI guidance on drug companies and patient organisations. BMJ. 2019;364:l740. pmid:30777855
- 11. Parker L, Fabbri A, Grundy Q, Mintzes B, Bero L. “Asset exchange”-interactions between patient groups and pharmaceutical industry: Australian qualitative study. BMJ. 2019;367:l6694. pmid:31831471
- 12. Lexchin J. Association between commercial funding of Canadian patient groups and their views about funding of medicines: An observational study. PLOS ONE. 2019;14(2):e0212399. pmid:30768629
- 13. McCoy MS, Carniol M, Chockley K, Urwin JW, Emanuel EJ, Schmidt H. Conflicts of interest for patient-advocacy organizations. NEJM. 2017;376(9):880–5. pmid:28249131
- 14. Rose SL, Highland J, Karafa MT, Joffe S. Patient advocacy organizations, industry funding, and conflicts of interest. JAMA Internal Med. 2017;177(3):344–50.
- 15. Batt S. Health Advocacy, Inc.: How Pharmaceutical Funding Changed the Breast Cancer Movement: UBC Press; 2017.
- 16. Lau E, Fabbri A, Mintzes B. How do health consumer organisations in Australia manage pharmaceutical industry sponsorship? A cross-sectional study. Austral Health Rev. 2018;43(4): 474–80.
- 17. Li DG, Singer S, Mostaghimi A. Prevalence and disclosure of potential conflicts of interest in dermatology patient advocacy prganizations. JAMA Dermatology. 2019;155(4):460–4. pmid:30698625
- 18. Perehudoff SK, Alves TL. The patient and consumer voice and pharmaceutical industry sponsorship. Amsterdam: HAI Europe. 2011.
- 19. Davis C, Abraham J. Unhealthy pharmaceutical regulation: Innovation, politics and promissory science: Basingstoke: Palgrave Macmillan; 2013.
- 20. Mulinari S. Unhealthy marketing of pharmaceutical products: An international public health concern. J Pub Health Policy. 2016;37(2):149–59.
- 21. Mulinari S. Regulating drug information in Europe: a pyrrhic victory for pharmaceutical industry critics? Sociol Health Illn. 2013;35(5):761–777. pmid:23094890
- 22. Rickard E, Ozieranski P, Mulinari S. Evaluating the transparency of pharmaceutical company disclosure of payments to patient organisations in the UK. Health Policy. 2019;123(12):1244–50. pmid:31455562
- 23. Fabbri A, la Santos A, Mezinska S, Mulinari S, Mintzes B. Sunshine policies and murky shadows in Europe: disclosure of pharmaceutical industry payments to health professionals in nine European countries. Int J Health Policy Manag. 2018;7(6):504–9. pmid:29935127
- 24. Grundy Q, Habibi R, Shnier A, Mayes C, Lipworth W. Decoding disclosure: Comparing conflict of interest policy among the United States, France, and Australia. Health Policy. 2018;122(5):509–18. pmid:29605526
- 25. EFPIA. EFPIA Code of Practice on relationships between the pharmaceutical industry and Patient Organisations. 2011. https://www.efpia.eu/media/24310/3c_efpia-code-of-practice-on-relationships-pharmapluspt-orgs.pdf Accessed 30 Dec 2019.
- 26. Ozieranski P, Rickard E, Mulinari S. Exposing drug industry funding of UK patient organisations. BMJ. 2019;365:l1806. pmid:31122928
- 27. Fabbri A, Swandari S, Lau E, Vitry A, Mintzes B. Pharmaceutical Industry Funding of Health Consumer Groups in Australia: A Cross-sectional Analysis. Inter J Health Serv. 2019;49(2):273–93.
- 28. Kang SY, Bai G, Karas L, Anderson GF. Pharmaceutical Industry Support of US Patient Advocacy Organizations: An International Context. Am J Pub Health. 2019;109(4):559–61.
- 29. Hemminki E, Toiviainen HK, Vuorenkoski L. Co-operation between patient organisations and the drug industry in Finland. Soc Sci Med. 2010;70(8):1171–5. pmid:20163903
- 30. Ball DE, Tisocki K, Herxheimer A. Advertising and disclosure of funding on patient organisation websites: a cross-sectional survey. BMC Public Health. 2006;6(1):201.
- 31. Rothman SM, Raveis VH, Friedman A, Rothman DJ. Health advocacy organizations and the pharmaceutical industry: an analysis of disclosure practices. Am J Pub Health. 2011;101(4):602–9.
- 32. Zetterqvist AV, Merlo J, Mulinari S. Complaints, complainants, and rulings regarding drug promotion in the United Kingdom and Sweden 2004–2012: A quantitative and qualitative study of pharmaceutical Industry self-regulation. PLoS Med. 2015;12(2):e1001785. pmid:25689460
- 33. Myndigheten för Vårdanalys. Sjukt engagerad—en kartläggning av patient-och funktionshinderrörelsen (Rapport 2015: 4). Stockholm; 2015. https://www.vardanalys.se/rapporter/sjukt-engagerad/. Accessed 30 Dec 2019.
- 34. LIF. Samarbetsdatabaser. https://www.lif.se/etik/samarbetsdatabaser/ Accessed 30 Dec 2019.
- 35. Abola MV, Prasad V. Industry funding of cancer patient advocacy organizations. Mayo Clin Proc. 2016;91(11):1668–70. pmid:27814841
- 36. Colombo C, Mosconi P, Villani W, Garattini S. Patient organizations’ funding from pharmaceutical companies: is disclosure clear, complete and accessible to the public? An Italian survey. PLoS One. 2012;7(5):e34974. pmid:22590498
- 37. Hughes D, Williams-Jones B. Coalition Priorité Cancer and the pharmaceutical industry in Quebec: conflicts of interest in the reimbursement of expensive cancer drugs? Healthc Policy. 2013;9(1):52–64. pmid:23968674
- 38. Mulinari S, Ozieranski P. Disclosure of payments by pharmaceutical companies to healthcare professionals in the UK: analysis of the Association of the British Pharmaceutical Industry’s Disclosure UK database, 2015 and 2016 cohorts. BMJ Open. 2018;8(10):e023094. pmid:30344175
- 39. Ozieranski P, Csanadi M, Rickard E, Tchilingirian J, Mulinari S. Analysis of pharmaceutical industry payments to UK healthcare organizations in 2015. JAMA Netw Open. 2019;2(6):e196253–e. pmid:31225896
- 40. Institute for Health Metrics and Evaluation (IHME). Sweden profile. 2018. http://www.healthdata.org/sweden. Accessed 30 Dec 2019.
- 41. Institute for Health Metrics and Evaluation (IHME). Global Health Data Exchange. GBD 2017 data. https://vizhub.healthdata.org/gbd-compare/ Accessed 30 Dec 2019.
- 42. Tibau A, Molto C, Borrell M, Del Paggio JC, Barnadas A, Booth CM, et al. Magnitude of clinical benefit of cancer drugs approved by the US Food and Drug Administration based on single-arm trials. JAMA Oncol. 2018;4(11):1610–1. pmid:30267037
- 43. Davis C, Naci H, Gurpinar E, Poplavska E, Pinto A, Aggarwal A. Availability of evidence of benefits on overall survival and quality of life of cancer drugs approved by European Medicines Agency: retrospective cohort study of drug approvals 2009–13. BMJ. 2017;359:j4530. pmid:28978555
- 44. Prasad V, Kim C, Burotto M, Vandross A. The strength of association between surrogate end points and survival in oncology: a systematic review of trial-level meta-analyses. JAMA Internal Med. 2015;175(8):1389–98.
- 45. Ferrario A, Kanavos P. Dealing with uncertainty and high prices of new medicines: a comparative analysis of the use of managed entry agreements in Belgium, England, the Netherlands and Sweden. Soc Sci Med. 2015;124:39–47. pmid:25461860
- 46. Hoekman J, Boon W. Changing standards for drug approval: A longitudinal analysis of conditional marketing authorisation in the European Union. Soc Sci Med. 2019;222:76–83. pmid:30605802
- 47. Pontes C, Zara C, Torrent-Farnell J, Obach M, Nadal C, Vella-Bonanno P, et al. Time to review authorisation and funding for new cancer medicines in Europe? Inferences from the case of Olaratumab. Appl Health Econ Health Policy. 2019. pmid:31696433
- 48. Naci H, Davis C, Savović J, Higgins JPT, Sterne JAC, Gyawali B, et al. Design characteristics, risk of bias, and reporting of randomised controlled trials supporting approvals of cancer drugs by European Medicines Agency, 2014–16: cross sectional analysis. BMJ. 2019;366:l5221. pmid:31533922
- 49. LIF. 2020. Ethical Rules for the Pharmaceutcial Industry in Sweden. https://www.lif.se/globalassets/etik/dokument/ler-english-version-2020-0501-jd-24apr-pdf.pdf. Accessed 2 June 2020.
- 50. Abraham J, Lewis G. 2000. Regulating medicines in Europe: competition, expertise and public health. Routledge