Accountability in global health systems: insights from a network analysis of Purdue Pharmaceuticals

Abstract

Beginning in 1996, Purdue Pharmaceuticals (Purdue) knowingly mislabeled and mass marketed OxyContin (oxycodone), an opioid painkiller, catalyzing the opioid crisis which has been responsible for more than 600 000 deaths in and beyond North America. This case is an extreme example of how transnational pharmaceutical companies prioritize shareholder profits over public wellbeing. As such, the field of global health faces the critical challenge of better understanding how transnational pharmaceutical companies, like Purdue, can be held to account for the harms they cause. Within the framework of Actor-Network Theory, a sociomaterial approach to analyzing complex networks, this case study uses key informant interviews (n = 18) to examine how accountability is taken up in and by global health systems in response to the harms caused by Purdue. Findings highlight the multiple co-existing versions of accountability enacted within global health systems organized as three separate but interrelated networks: social accountability, political accountability, and legal accountability. Though often interconnected, these diverse networks mobilized distinct tools, resources, and strategies, such as news articles, scholarly literature, and policy guidelines, to construct and stabilize enactments of accountability. Through this in-depth examination of the complex interactions involved in global health and pharmaceutical systems, this study offers a nuanced understanding of the diverse actors mobilized and the unique strengths leveraged within and by accountability networks. Further, in examining these networks’ differences, interconnectedness, and peculiarities, we broaden the scope of how accountability is defined, conceptualized, and operationalized in global health systems.

Introduction

In 1996, Purdue Pharmaceuticals (Purdue) launched OxyContin, an opioid pain killer generically known as oxycodone, with the largest marketing strategy in pharmaceutical history [1, 2]. Up until this launch, opioids were prescribed largely to palliative patients or those with cancer due to their efficacy in pain relief but high risks of dependence and overdose [3, 4]. However, looking to expand this market, Purdue executives decided to promote OxyContin as a cure for all untreated pain first in the United States (US) and Canada, and then internationally through the international arm of their company Mundipharma. Cornerstone to Purdue’s marketing plan were two claims: 1) that OxyContin lasted longer than any other prescription opioid on the market, and 2) that OxyContin was the first ever non-addictive opioid painkiller [1]. It is now well-established that not only were these claims strongly misleading, but importantly, Purdue executives and its owners, the Sackler family, knew these claims were false prior to the launch of the drug in 1996 [4–6].

In the early 2000s, prescription rates of OxyContin were rising dramatically along with dependence, overdose deaths, and company profits [3, 7]. At this time, Purdue also began receiving hundreds of patient lawsuits from families and communities of people struggling with OxyContin-related dependence and overdose deaths. However, through procedural delays and technicalities, Purdue’s extensive legal department effectively prevented any of these lawsuits from reaching court. As addiction rates and evidence against the company mounted, the West Virginia Attorney General was finally able to bring Purdue to court, and in 2007 three top executives were fined US$635 million in criminal and civil penalties for knowingly misbranding OxyContin as only mildly addictive [5, 8]. Though this fine is seemingly significant, Purdue’s sales of OxyContin continued to climb even after the ruling and the company profited over US$30 billion between 1996–2016 solely from the sale of OxyContin [1].

Moving into 2010, prescription rates in North America began to drop due to the rising awareness of the risks associated with OxyContin. In response, the Sackler family, while continuing their marketing within Canada and the US, expanded to international markets. To do so, they employed the same deceptive set of marketing tactics to bring OxyContin to countries including China, Brazil, Chile, and Peru through Mundipharma [9]. To further promote opioids on a global scale, Purdue aimed to influence the World Health Organization’s (WHO) international guidelines to favour prescription opioids as the primary mechanism for the treatment of pain [10]. To accomplish this, Purdue founded, led, and infiltrated pain advocacy groups, including the American Pain Society and the International Association for the Study of Pain, to advocate in their interest. Purdue did so through providing funding, compensating group leaders, or positioning Purdue executives in leadership roles within these groups [10], US Senate Homeland Security & Governmental Affairs Committee Ranking Member’s Office]. As a result, the groups consulted on the WHO guidelines, leading to two international guidelines containing Purdue’s misleading marketing claims on the safety of prescription opioids [10–12].

By 2016, Purdue had expanded beyond its operations in the US, Canada, and Europe to more than 122 “developing markets,” in Asia and South America through Mundipharma; seeing their revenue from these markets increase by more than 800% [13]. At present, Purdue has filed for bankruptcy in the US and remains in a long legal battle negotiating the resolution of thousands of lawsuits in Canada and the US levied against them by individuals, hospitals, Indigenous communities, and local and state governments [14, 15]. Despite this ongoing litigation, and the hundreds of thousands of opioid-related deaths in and beyond the US [16, 17], Mundipharma continues to market opioids and other products in international markets with impunity.

Meanwhile, the opioid crisis continues to progress globally, exacerbated by social, economic, and political forces such as a lack of affordable and accessible housing, an unpredictable illicit drug supply, and the criminalization of drug use. This evolution of the opioid crisis includes a shift from a prescription opioid-driven crisis, as catalyzed by Purdue and other actors such as other opioid manufacturers, pharmaceutical distributers, medical professionals, and legislators, to an illicit opioid crisis [18]. This shift towards illicit opioids occurred as prescription opioids became more strictly regulated in the 2010s and was characterized by the increased use of heroin, and later by the proliferation of illicit synthetic opioids such as fentanyl. The term “opioid crisis” is employed in this paper to denote the ever-evolving spectrum of harms, including dependence and overdose death, associated with Purdue, OxyContin, and the illicit crisis as it exists today on a global scale.

Due to the substantial number of opioid-related deaths, the opioid crisis and Purdue’s actions have been examined in scholarly literature, organizational reports, media, film, and television, well-establishing Purdue as the root of both the prescription and illicit opioid crises in North America for example [19, 20, 21, 4]. However, at present, there is limited literature examining Purdue within the greater context of global health systems, meaning the health systems consisting of “transnational actors that have a primary intent to improve health and the polylateral arrangements for governance, finance, and delivery within which these actors operate” [22]. Within this definition, “polylateral” refers to interactions involving multiple actors, including non-state actors and individuals.

This gap in the literature is significant since Purdue, the Sackler family, OxyContin, and the opioid crisis continue to have deleterious impacts on the health of populations across borders. As such, the global health community faces the challenge of better understanding how transnational pharmaceutical companies, like Purdue, are held to account by global health systems.

Accountability and global health systems

The concept of accountability has been taken up in diverse ways across scholarly disciplines including business management, domestic and supranational law, international relations, public policy, and global health [23–25]. Because of its extended analytical domains, as described by Newell and Belour [26], “accountability has become a malleable and often nebulous concept, with connotations that change with the context and agenda” (p. 2). To find focus considering this complexity, this paper concentrates on accountability within the interconnected fields of global governance, international relations, and global health.

To begin, a lack of accountability is often cast in technocratic terms as a “problem” for which one actor or set of actors must hold another to account for their actions and decisions. Such framing, describing a “principal-agent” conceptualization, is common because it describes the accountability that may exist between citizens and elected public officials [27]. However, this conceptualization fails to explain how accountability functions in settings where the roles of principals and agents are unclear, interdependent, unstable, and laden with unequal power dynamics [28]. In the multifaceted field of global health, including its diverse actors, institutions, laws, policies, and ever-changing dynamics, this binary conceptualization of accountability is challenged to consider a large network of actors with varying degrees of power and influence [29, 30].

Addressing these shortcomings [28], propose a “second generation” perspective on accountability that moves away from a logic of compliance towards a logic of participation [31]. This second-generation perspective has given rise to a number of accountability frameworks that consider pluralistic power structures, the engagement of multiple actors, and accountability as grounded in shared values, such as justice and respect [32, 33]. Within this framing, accountability on the global stage is ambiguous because “it is a socially embedded, politicized, pluralistic, and value-heavy construction,” meaning it is shaped by its social, economic, cultural, and political contexts [28, p. 3]. Accountability does not operate in isolation, but rather is a fundamentally relational concept employed by diverse actors as a means of social control [30]. This paper therefore draws on Actor-Network Theory (ANT) to build on Bruen et al.’s approach to accountability in the context of global health as a “complex problem, not necessarily reducible to one set of actors holding another to account, but rather as an ongoing process with built-in tensions that continually shape relations between a diverse set of actors” [29, p. 12]. ANT expands this approach to conceptualize and explore accountability as an “actor” in the global response to Purdue, providing a dynamic framework for understanding the complex, interconnected relationships that shape accountability in global health systems. ANT frames accountability not as a static or singular concept but as something constructed and maintained by a network of diverse actors, both human and non-human, who interact within these systems. Approaching accountability in this manner facilitates this paper’s engagement with the complex networks of actors, influences, and relationships that shape accountability in global health systems.

Aim

This paper provides a case study of Purdue and OxyContin to examine accountability in global health systems in response to the harms caused by transnational pharmaceutical companies. Guided by ANT, this study explores the multiple enactments of accountability within global health systems, focusing on how influence is exerted and maintained within these networks to create enduring accountability.

Theoretical frame

This study draws specifically on ANT’s interrelated concepts of multiplicity, enrollment, and inscription devices to trace how accountability comes to be constructed, mobilized, and transformed within global health networks. To begin, De Laet and Mol’s [34] concept of multiplicity is employed to understand the multiple and interacting enactments (networks) of accountability [35]. The concept of multiplicity recognizes that accountability does not exist in global health networks as one stable form but rather is dynamic, taken up, and enacted differently by different actors, each representing a co-existing multiple reality. For example, accountability as enacted by legal systems may differ from accountability as sought from health advocacy groups, or international health organizations. The lens expands upon the limited dimensionality of existing conceptualizations of accountability to explore how accountability’s multiple realities differ from and interact with one another, and why some enactments are privileged over others.

The concept of enrollment in ANT describes the processes through which actors, both human and nonhuman, are enlisted into a network to support the network’s goals or objectives [36]. Non-human actors, as per ANT, can be understood as anything that effects influence on another actor, for example, policies, laws, institutions, and publications, are all actors relevant to this case study. The final ANT concept central to this study is inscription devices, which refers to tools, instruments, or mechanisms, for example organizational reports or scholarly publications, that translate actions or events into a material form [37]. Inscription devices play a key role in stabilizing and maintaining networks as they allow a phenomenon to be recorded and circulated, making them central to the production of knowledge. Within global health systems, inscription devices transform complex realities into concrete records that can be used to assess, enforce, and communicate accountability across various actor-networks. Applying these concepts of ANT to this research enables a more nuanced understanding of how accountability is constructed in response to complex transnational issues like the opioid crisis, identifying pathways for durability and influence within these networks.

Methods

Study design

Data presented here are derived from a larger qualitative study examining accountability within global health systems in response to the harms caused by Purdue. This paper explores the multiple enactments of accountability with a specific interest in the actors and inscription devices mobilized in each enactment. Data was generated through key informant interviews with transnational actors, defined as individuals that operate in a way that transcends national political borders [22].

Ethics statement

Ethics approval for this study was received from the University of Toronto Research and Ethics Board on July 13^th, 2022 (protocol number 0004367). Informant consent was obtained in written form.

Data generation

Between February-October 2023, 18 key informant interviews were conducted with experts relevant to this case. This purposive sample was recruited [38, 39] guided by the criteria listed below. Purposive sampling was combined with conceptually driven sequential sampling [40] which allowed selection criteria to evolve as actors-networks were identified. Recruitment continued until thematic saturation was reached, resulting in 18 interviews being conducted. Key informant inclusion criteria included: Transnational actor (as defined by Hoffman and Cole) with exposure to the topic of Purdue and OxyContin (personal or professional) AND content expertise in global health and pharmaceutical systems; Adults over the age of 18; English-speaking; and able and willing to sign informed consent. In alignment with ethical requirements, interview transcripts were stored on a password encrypted device, accessible only to the research team.

Informants included activist group representatives (n = 3), healthcare providers (n = 5), attorneys (n = 2), pharmaceutical industry representatives (n = 2), academics (n = 16), international organization representatives (n = 3) and regulators (n = 2). Many informants held multiple roles, most with academic appointments at an educational institution. The majority of informants were located within the US (n = 8) and Canada (n = 5) due to the concentration of the opioid crisis within these countries. Other informants were located within the United Kingdom (n = 2), Australia (n = 2), and India (n = 1). Informants provided written consent to participate in this study, and monetary compensation was not provided.

From a sociomaterial approach such as ANT, interviewing is designed to facilitate learning about the performative entanglement of both humans and non-humans within complex networks. As relevant to this study, the case of Purdue was the focal point for the interviews, referenced to elicit insight into the multiple enactments of accountability within our global health systems. Semi-structured interviews were conducted with pre-determined questions but with the flexibility to add other questions or prompts as deemed appropriate by the interviewer. The interview guide was developed by author AB with input from JCK, LF, and APB. Based on the research aims, as well as previous literature, the interview guide aimed to capture three domains: 1) actors and relationships mobilized to respond to the harms caused by Purdue; 2) accountability systems; and 3) strengths, weakness, and opportunities within identified global accountability systems. All interviews were conducted virtually between February-October 2023 by author AB due to the geographic dispersion of informants. Interviews lasted 30–45 minutes, using Zoom meeting software, and were audio- and video-recorded.

Analysis

Data analysis was conducted by author AB and centred narrative description and the need to remain as close to the data as possible [41]. This means that data remained the centre of analysis, rather than analysis being guided by a predetermined framework. Data analysis brought actors, networks, and relations implicated in accountability networks into focus specifically, how they were constructed, deconstructed, and/or reconstructed, to form, strengthen, and maintain networks through an inductive coding process [42]. Each instance where accountability was discussed was extracted and coded in Microsoft Excel for 1) actors (human and non-human), 2) relationships (including processes of enrollment), 3) inscription devices, 4) implications, and 5) demonstrative excerpts.

Findings

The data generated from key informant interviews with policy makers, academics, and industry representatives, reveals a multifaceted landscape of accountability. The multiple co-existing and interacting versions of accountability as enacted by various networks within global health systems are presented below, organized as three separate but interrelated sociomaterial networks: social accountability, political accountability, and legal accountability. Though these networks are overlapping and interrelated, the sections below are characterized according to the type of accountability enacted. Specifically, social accountability refers to the actions and consequences that generated awareness, mobilized public opinion, and drove social pressure against Purdue. Political accountability refers to the responsibility of public officials, politicians, institutions to be answerable to the public, stakeholders, and oversight bodies for their actions and decisions, and in the case of Purdue, was enacted as congressional and homeland security investigations. And finally, legal accountability refers to the legal obligations and consequences faced by Purdue, the Sackler family, and other implicated parties within their networks. Within each of these sections, results detail actors and inscription devices enrolled as well as the avenues through which influence is exerted to enact and stabilize each version of accountability.

Social accountability

The most discussed form of accountability for the harms caused by Purdue was public awareness and social consequences. Though multiple definitions exist, in this paper social accountability encompasses the collective processes through which citizens, communities, and civil society organizations hold others, such as public officials, services providers, and institutions, accountable for their actions and decisions [43–45]. Central actors in such networks included victims’ families, activist groups, the general public, and public officials. Key informants described various inscription devices that played a role in these networks, including news articles and scientific literature. These devices contributed to the stabilization and durability of social enactments of accountability by translating knowledge within and between networks.

It is important to recognize that temporal ordering may obscure the complexity and simultaneity of influences within the network, therefore the results are organized to prioritize narrative clarity beginning with the actions of activist groups, particularly Physicians for Responsible Opioid Prescribing (PROP), given their prominence in discussions of accountability. PROP’s grassroots efforts were identified as one of the first to raise awareness of Purdue’s actions and the risks and harms associated with the over-prescription of OxyContin. One key informant described the sudden shift in approach to prescribing opioids as a motivator for PROP’s enrollment:

“Well, there were a few physicians who were kind of shocked by this big change that happened where opioids were once viewed as, you know, to be used only for a few days, only for people with clear pathology for either cancer or some type of trauma, like a car accident or something, and not to be used long term, nor were they effective long term. And that group was kind of like a really true grassroots effort of several concerned physicians, Physicians for Responsible Opioid Prescribing, PROP.” (Academic; US)

To address their concerns with this shift in prescribing, PROP drew on their existing relationships with health regulators such as the FDA, as well as on academic literature, to resist and “push back” against Purdue:

“[PROP] would basically function as activists to push back against these advances of opioids to the FDA complaining about the trials that were being used, reminding these organizations of the data and what we know, what we don’t know, and what we need to know.” (Academic; US)

PROP leveraged their technical knowledge and status as health professionals to increase awareness of the issue of overprescription and apply pressure on public officials and regulatory bodies. Similarly, key informants described the importance of academic articles (inscription devices) that contained the knowledge generated from previous studies outlining the risks associated with prescription opioids. PROP leveraged the data inscribed in such literature as a reference point in their communication with regulators to strengthen and stabilize their arguments, providing a foundation for action.

To extend the reach of their advocacy and push back against the advances of opioid manufacturers, PROP and other activist groups also enrolled the media and public officials in their networks:

“You know, there were definitely professional groups, like PROP, that were very opposed to the use of OxyContin and were calling out the over marketing and over prescribing of OxyContin pretty early. And they were also talking to public officials and the media.” (Regulator and academic; US)

The media was a second central actor in social accountability networks, t referring to networks of journalists, editors, newspapers, magazines, and more recently books, television series, and podcasts, and film. The media was enrolled in social accountability networks because of the work of activist groups. Put plainly, a key informant stated, “a purpose of the media to hold powerful people to account” (Pain advocacy group representative and academic; UK).

Key informants also discussed how the media, meaning investigative journalists and their news outlets, played a central role in not only expanding public awareness of Purdue’s actions but also conducting investigations. One key informant described the role such investigations played in exposing harmful actions by Purdue, such as their creation and cooptation of pain advocacy groups to exert influence on public policy:

“So, you know there’s a couple of investigative outlets that have reported on [pain advocacy groups], like Pro Publica, maybe Mother Jones, Milwaukee Journal Sentinel, and John Fauber, they have done really good work on this front. Work that would lead ultimately to investigations by congressional bodies” (Activist group representative and academic; US)

As described here, the media played a crucial role in social accountability networks, as it exposed Purdue’s harmful actions and embedded this knowledge in a wide range of inscription devices (e.g. news and magazine articles) that increased the reach of this information and stabilized it within material representations. In other words, investigative journalists made Purdue’s actions visible to the public through publishing articles in prominent news outlets including the New York Times, the Los Angeles Times, and the Globe and Mail. The implications of the media’s awareness-raising efforts are multifold and interconnected with the implications of the political and legal enactments of accountability discussed below. However, within social accountability networks, informants outlined how widespread media coverage increased international awareness of the scale and severity of OxyContin and opioid-relate harms:

“I mean the term opioid epidemic is now probably a household term. I don’t think it’s just a professional term and I think everyone has seen the news stories and seen the graphs of rates of prescribing in the States and rates of opioid-related deaths” (Healthcare provider and pain advocacy group representative, India)

This quote demonstrates the importance of inscription devices in shaping narratives due to the ripple effect they can have when shared among networks. In addition to awareness-raising as social accountability, one informant described that the increase in awareness led to other social consequences, including social ostracization: “The social ostracism was one of the worst punishments. You know the Sackler name coming off all the buildings. That’s a pretty serious penalty in that” (Regulator and academic; US). Here, accountability is enacted as a set of social consequences. The Sackler family, owners of Purdue, faced social ostracism for the harms their company caused with OxyContin with many prominent cultural and educational institutions, such as the Guggenheim Museum and Tufts University, removing the Sackler name from their buildings, and refusing to receive further donations from the family.

Political accountability

The second version of accountability, termed “political accountability,” denotes the responsibility of public officials, politicians, institutions to be answerable to the public, stakeholders, and oversight bodies for their actions and decisions [46, 47]. In this analysis, political accountability is enacted largely in the US as a network of congressional and homeland security investigations involving diverse actors. Such actors enrolled in this network include 1) the media, whose initial investigations laid the ground work for the public officials; 2) pain advocacy groups, who used their position as non-profit patient or expert organizations to influence policy at national and international levels; 3) Purdue and other opioid manufacturers who influenced pain advocacy groups through funding, leadership, or other avenues; and 4) the public officials and their offices who conducted the investigations. The term “public officials” is employed to describe individuals elected by the public at local, state, and national levels. Key informants spoke about, as illustrated in the previous section, the influence of both media and activist groups in engaging public officials in accountability efforts. To do this, media and activist groups drew on existing relationships, or created new relationships with public officials to urge them to investigate, expose, and enforce accountability measures for Purdue and the ever-expanding opioid crisis. Key informants spoke specifically about the importance of the congressional investigations in exposing the ties between pain advocacy groups and Purdue as a form of accountability in itself:

“There were two Senate Finance Committee investigations and a Homeland Security investigation in which reports were published outlining the relationships between industry and pain advocacy groups. Yeah, maybe that counts on some level as accountability.” (Activist group representative and academic; US)

Such descriptions of political accountability networks also included key inscription devices, such as industry documents and congressional reports. For example, public officials examined Purdue’s 1996 Budget Plan that exposed their marketing plans to influence international guidelines through their relationships with pain experts and pain advocacy groups [10]. This Budget Plan provided material evidence, representing Purdue’s intentions, that could be shared with other actors. Further, the findings of such investigations were embedded, and stabilized in public reports [10, 48]. Such reports were then used as the foundation for news and magazine articles [49, 50] and further congressional investigations [48]. These examples illuminate how political accountability is enacted and maintained by an array of interconnected human and non-human actors.

The implications of this political accountability network, as described by informants, were multifold and intertwined with the efforts of social accountability networks outlined above. To begin, one informant described the congressional investigations and hearings themselves as a measure of accountability:

“There are challenges in having to testify in front of an angry Congress. So, I think that there have been a range of different consequences being written about [Purdue] that are quite negative. So, this kind of public, I don’t want to say shaming, but this public outcry has played a role in the accountability for their, for the harms caused.” (International organization representative and academic; Australia)

As illustrated by this quote, having to testify in front of US Congress was a measure of accountability for Purdue executives and Purdue-funded groups. The reach and impact of this political accountability is furthered through its inscription in media outputs that reached the public and led to further outcry, or social accountability, for the actions of Purdue and the harms they caused. This increasing public awareness due to the action of public officials and the media was described by another informant as a “rippling effect” causing the public to reflect on the issue:

“I think that the recognition of the opioid crisis kind of rippled through all the people who were involved at one level or another. And so, I think when people realized that there was a major issue that so many more people were dying, that we were such an outlier globally, I think that led to some introspection, and I’m sure, like the congressional hearings and journalism contributed to that significantly.” (Healthcare provider and academic; US)

The increased awareness in the public had further implications including the cessation of some Purdue-funded pain advocacy groups [51]. Social accountability efforts increased awareness of Purdue’s role in the crisis, catalyzing political accountability mechanisms. In tandem, these forms of accountability exposed conflicts of interest between public officials, educational institutions, professional and advocacy groups, and opioid manufacturers. An informant described this form of accountability, and its efficacy in undermining the credibility of Purdue-funded organizations, attributing its impacts largely to the increased media coverage:

“[Media] exposure was helpful because some of these groups went under. You’re a front group getting money from industry. Once you’re exposed as a front group, you’re not very effective anymore. So, industry stops giving these groups money and then they disappear. I suppose to that extent there, if you call that accountability, there’s been some accountability” (Academic; Canada).

This description demonstrates the cumulative influence that the media, activist groups, and public officials had on pain advocacy groups by exposing their ties to Purdue and other opioid manufacturers. As explained here, industry disrupted the funding for many of these groups, limiting the sustainability of the groups themselves.

Legal accountability

Legal accountability can be understood as the responsibility or obligation of actors (individuals, organizations, or institutions) to adhere to laws and regulations, with consequences imposed through judicial processes when violations occur [52, 53]. Legal forms of accountability encompass a range of mechanisms with varied implications. Civil proceedings are common with respect to harms caused by the pharmaceutical industry and often include fines, and settlements. Criminal liability, in contrast, is less commonly pursued with respect to the pharmaceutical industry (with exceptions such as the sentencing of John Kapoor from Insys Therapeutics [54] and can carry different consequences such as the possibility of incarceration. Legal accountability in this case was described as made up by heterogeneous actors working together to create, enforce, and interpret legal obligations and consequences as related to Purdue and the broader pharmaceutical industry. These networks were composed of various human and non-human actors including legal professionals, regulatory bodies, courts, laws, and various documentation, all of which interact to enact legal forms of accountability. Importantly, though this study examined accountability in global health systems, legal accountability was discussed by key informants largely in reference to domestic laws and regulations; a testament to the lack of enforceable accountability mechanisms in international legal systems governing the pharmaceutical industry.

Differing from social and political networks, which mobilized largely in response to public awareness raising efforts, legal accountability networks mobilized once criminal action was identified by the West Virginia Attorney General’s office. However, this distinction is not absolute, since public awareness campaigns also entangled with legal networks by drawing attention to Purdue and OxyContin. While legal networks could not mobilize fully until laws were broken, their actions were not entirely independent of political and social pressures, highlighting the complex interplay between these networks in shaping accountability measures. For example, when asked about accountability for the harms caused by Purdue, a key informant described some of the actors involved in these legal networks:

“I would say also what has certainly been very instrumental is the role of the legal community. So, class action lawsuits, lawyers, from the plaintiff side, legal departments of health authorities, which have led prosecutions against opioid producers for the harms and the fraudulent billing or the fraud in the context of billing within the healthcare system. So, in the US, for example, there is a massive lawsuit trying to recuperate some of the additional health expenses that health insurance at the federal or state level have incurred as a result of the opioid crisis.” (Attorney; Canada)

This key informant provides insight into the multiple actors within legal accountability networks. In this variation, accountability is enacted as a set of legal actions including prosecutions and lawsuits. In these networks, inscription devices, once again, played important roles. Legal documents, regulations, electronic records, and organizational communications inscribed actions, decisions, and events into formal records that were referenced, analyzed, and acted upon to enact accountability. These inscriptions helped stabilize legal networks by providing material evidence that could be leveraged to hold Purdue to account.

The above quotation also illuminates how legal networks aimed to enact accountability within the broader pharmaceutical industry by including other opioid manufacturers in their proceedings. This expansion of accountability beyond Purdue is crucial since Purdue’s actions of fraud, product mislabelling, and mass-marketing, are not unique to their company, but rather widespread, and other opioid manufacturers have contributed substantially to opioid-related harms in and beyond the US. This point was reiterated by another informant who described how such harmful practices are so widespread within the industry that it has become commonplace for companies to be fined for their actions:

“[Purdue] did get fined in the $650 million for this. The inappropriate education and things and people say, well, that’s disgusting. But again, I can point to every pharmaceutical company in the United States who’s actually been fined because of this style of marketing” (Pain advocacy group representative, healthcare provider, and academic; UK)

As illuminated by this key informant, an important aspect of legal accountability networks is their seemingly limited ability to prevent and address harmful marketing practices within the pharmaceutical industry. Such limitations included the ineffectiveness of financial penalties, the shielding of company executives responsible for decision-making, and jurisdictional boundaries. This quote also highlights the tension in accountability mechanisms targeting high-profile cases such as Purdue, which can provide a clear focal point for mobilization, and the need for comprehensive accountability frameworks capable of encompassing all relevant actors. Multiple key informants emphasized that accountability as enacted by legal networks had various limitations in addressing the harms caused by Purdue. For example, one informant described the limited impact of the 2007 charges against Purdue executives:

“One would have thought that the 2007 litigation could have had been an accountability moment for [Purdue], but it, in retrospect, was really more of a speed bump and they continued with a lot of the same approaches, you know their market peaked in 2009, 2010, but the litigation was in 2005, 2006, and concluded in 2007. So, I mean, they really were still selling a truly enormous amount of OxyContin at that time. And then the other companies too, I think you can’t really just look at [Purdue] alone. But other companies realized that this was a pretty lucrative field even after that initial round of litigation. So, whatever happened in that initial round of litigation wasn’t enough” (Regulator and academic; US)

The key informant here describes the legal accountability measures in 2007 as a “speed bump,” meaning its enactment failed to generate the desired effect of Purdue taking responsibility for and stopping its actions to avoid perpetuating further harm. This informant also speaks to the necessity of expanding accountability to other pharmaceutical companies implicated in the opioid crisis and interestingly correlates the increased production and marketing of opioids by other pharmaceutical companies with the failure of legal networks to meaningfully hold Purdue to account. In keeping with this frame, another key informant describes the attitude of pharmaceutical companies towards fines incurred through legal networks as an accepted expense: “If I’m honest, sometimes the fines that are doled out for bad behavior are, well, if I could be blunt, they come into the cost factoring of a marketing program of a drug” (Attorney; Canada).

Fines as a legal accountability measure are described by informants as an acknowledged necessary expense for pharmaceutical companies looking to maximize profits rather than acting as a deterrent against harmful practices. This means that legal accountability measures in the form of lawsuits and fines (such as the 2007 fine against Purdue executives) often fail to meaningfully address and prevent harmful practices within the pharmaceutical industry.

A second limitation of legal accountability networks described by key informants was their inability to reach the individual executives within pharmaceutical companies who are responsible for decision making. Though there are examples of individuals within such companies being prosecuted, notably Purdue executives were fined in 2007, this was a rare case, and what is more, Purdue’s owners, the Sackler family, were shielded from penalties during these proceedings. One key informant described how it is atypical for executives to be held accountable for their decisions within legal systems:

“Without individuals being held accountable, it’s hard to change corporate behavior. Take, you know, Purdue Pharma or other opioid manufacturers that have been found criminally guilty, or the corporation pleads guilty to a crime. Both in 2007 and more recently than the 2007 case, there was a slap on the wrist for some executives, but aside from that, generally you have a corporation that pleads guilty. You have a corporation that pays a fine, and yet none of the individuals, no board members, are held accountable. Executives that made the decisions that led to their company committing crimes are not held accountable and at the end of the day, the corporations are able to pay, and the fine is nothing compared to the revenue that they brought in.” (Activist group representative and academic; US)

Legal networks seemingly struggle to penetrate company networks to reach individual actors responsible for decision-making. These actors are protected by companies that are able to pay the fines incurred for their crimes. These fines, as noted again by this key informant, are covered by the profits accumulated through the sale of their products.

Finally, since this study set out to examine the issue of Purdue and OxyContin in global health networks, key informants discussed how domestic legal networks were unable to prosecute companies and individuals acting beyond their jurisdictions. Legal systems differ greatly between countries, reflecting the heterogeneity of global legal frameworks and the lack of cross-border collaboration. This diversity in legal structures can create barriers to coordinated action, further complicating accountability efforts on a global scale. One informant located outside the US described the limitations of their domestic legal systems in comparison to the US’s:

“In our legal system, accountability has been hampered by the limits of the litigation context in the sense that we don’t have the huge damages that you have in the US and it’s harder to launch class action lawsuits for that reason as well. So, the fact that if the damages are less significant, it’s harder to mount, it’s less likely that there will be a successful launch of a class action lawsuit or individual lawsuits against producers of opioids and the States has been more successful. State authorities have also been able to negotiate settlements with large opioid producers for the harms that their marketing practices have caused including litigation against pharmacy chains, producers of opioids distributors, and so on.” (Attorney and academic; Canada)

Furthermore, the US example demonstrates how state authorities can act as powerful agents within the network, exerting influence through settlements with opioid producers and pharmacy chains. In countries with fewer options for such actions, accountability networks are less robust. The structure of the US network, with high damage awards, established legal practices for class actions and state authority involvement. As described, these factors may allow for more durable forms of accountability compared to countries with restrictive legal frameworks and financial limitations. Legal accountability, in the form of class action lawsuits are a form of accountability that have held actors involved in the opioid crisis, including opioid manufacturers, prescribers, and pharmacies, to account. These enactments of accountability through these lawsuits and settlements do not counteract, or negate the limitations of legal networks discussed above, but are rather another version of accountability as enacted by legal networks. It is a version of accountability that is deemed meaningful and impactful by this key informant, because even though, according to other key informants, Purdue itself has yet to be meaningfully held to account, some networks have been able to recuperate damages for the role actors have played in perpetrating opioid-related harms.

At this point, the complex dimensions of accountability as categorized within social, political, and legal networks have been discussed. Such networks are both separate but closely connected, all playing a role in enacting accountability for the harms caused by Purdue and OxyContin. While the consequences of each version of accountability may be limited or seemingly unimposing, when woven together, the implications are stable and wide-reaching. As summarized by one informant:

“Well, I think a lot of people in the country and the world know who was involved and I think that the journalism has played a really important role. The congressional hearings and the litigation have all served to really tell a story, you know, and that itself is a measure of accountability.” (International organization representative, healthcare provider, and academic; US)

To be sure, the accountability that has been enacted is not sufficiently able to address the entire range of harms caused by Purdue. Rather, participants highlighted how specific accountability measures were enacted, including the actors, actions, and inscription devices enrolled for the enactment to strengthen accountability processes. By emphasizing these components, key informants demonstrated the mechanisms that made certain accountability efforts possible and effective, while implicitly raising questions about the gaps, limitations, and areas where further accountability is needed.

Discussion

This study examined the complexity of accountability enacted in response to the harms caused by Purdue and OxyContin. Through eighteen key informant interviews with experts in regulatory bodies, industry, academia, healthcare, and law, this research unpacks accountability processes in global health contexts. By categorizing these enactments into social, political, and legal accountability, this study offers a nuanced understanding of the diverse actors mobilized and the unique tools, resources, and strengths leveraged within and by these networks. Further, in examining the differences, interconnectedness, and peculiarities of these networks, we broaden the scope of how accountability is defined, conceptualized, and operationalized in global health systems.

Findings build on existing literature establishing Purdue’s malfeasances [2, 55, 56], the regulatory failures that exacerbated the opioid crisis [57, 58], and the means through which accountability mechanisms have failed to hold implicated actors responsible for the harms caused [59–62]. Of note, this study aligns and reinforces findings that existing accountability measures, specifically through legal means, are limited in their effectiveness in addressing malfeasance and corruption in the pharmaceutical industry [63–65]. However, this study extends the existing body of literature through interrogating the multiple ways that accountability has successfully been enacted to deepen understandings of accountability, and the elements needed for accountability to be realized.

Importantly, this study leveraged De Laet and Mol’s concept of multiplicity to understand that these versions of accountability represent multiple co-existing realities enacted by different (but interrelated) networks [34]. Through exploring how multiple versions of social and political accountability are not only co-existing but often intertwined and mutually reinforcing, this study advances the understanding of how different forms of accountability are enacted and consolidated within complex networks and can at times be added together and integrated with other forms of accountability. For example, the removal of the Sackler name from cultural institutions was described as a form of social accountability enacted in response to the harms caused by Purdue. This form of accountability was closely related to the public awareness raising that was described as a form of accountability in itself as enacted by media and activist groups. Both forms are enacted as separate versions of accountability, but when added together, the collective impact is strengthened. Further, key informants often described accountability as enacted by social and political networks in unison. Political accountability in the forms of congressional investigations and hearings, were motivated by social accountability measures, largely public outcry and increased media attention. What is more, the impacts of the political accountability measures were amplified due to media coverage of the investigations and hearings. When these forms of accountability are added together, they formed a complex, but composite whole. Such findings reiterate existing conversations emphasizing the importance of cross-border and cross-disciplinary coordination and collaboration in the pursuit of accountability for harms caused by the pharmaceutical sector [66–68].

Conversely, findings demonstrate that certain forms of accountability can also remain distinct, distributed across different spaces and times, rather than converging to find singularity. In analyzing how different networks (social, political, and legal) can operate in parallel yet remain disconnected, this study highlights the ways in which some accountability processes can be segmented and mutually exclusive, offering insight into the occasionally disjointed nature of networked realities in global health. For example, as related to legal forms of accountability, such as the 2007 litigation and resulting fines, we can understand that this version of accountability was enacted in West Virginia, more than ten years prior to the congressional investigations and hearings. The congressional actions, representing a distinct political version of accountability, were enacted in a separate locality, Washington DC. Both versions were described by key informants as forms of accountability enacted in response to the harms caused by Purdue, but their enactments were acknowledged separately, never being layered together with one another. Here, mutual exclusivity between accountability processes refers to a fragmentation, or lack of coordination across networks, that results in distinct enactments of accountability existing in parallel with one another. In parsing out these unique realities of accountability we can better understand the dynamic spectrum of accountability enacted in response to ever-shifting social, economic, cultural, and political structures.

Finally, this research demonstrates how diverse networks, though interconnected, mobilize distinct tools, resources, and strategies to enact accountability. By mapping the specific ways in which legal, social, and political networks navigate their unique capacities and limitations, this study offers a more nuanced perspective on the multifaceted nature of accountability in global health. For example, in social accountability networks, media drew on their investigative and information distribution systems to raise public awareness and apply pressure on public officials. In contrast, legal accountability networks drew on their authority to enforce established laws, to bring Purdue to court and fine executives for their misconduct.

In examining the different networks, actors, resources, and inscription devices involved in the different versions of accountability, this analysis provides a deeper understanding of how actors mobilize resources and engage allies both within and across networks to influence accountability outcomes. We highlight the complexity of accountability in global health, revealing how varied networks overlap, diverge, and sometimes conflict in approaches. Further, we illustrate how the mobilization of narratives and resources by corporate actors, like Purdue, shapes both the direct and indirect responses of global health systems, offering insights into the mechanisms through which accountability can be strengthened or undermined in transnational public health contexts.

Limitations

The results and analysis presented in this paper have several limitations. To begin, results were generated from the interpretation of a small sample of key informants, meaning the results presented are limited to the actors and relations identified by informants. Though a range of experience was captured, including industry, regulators, attorneys, and activist groups, the findings may not be representative of all possible perspectives. Further, though this study aimed to examine the case of Purdue in the context of global health systems, most key informants and therefore the data collected centred an American perspective due to the concentration of the opioid crisis and Purdue’s operations in the US. However, it is important to note that due to the significant concentration of power and resources within US pharmaceutical companies, the accountability mechanisms governing these companies have global implications. Further, this analysis highlights the accountability that can be enacted through more informal processes (i.e. social networks) that can more easily transcend national borders. Such processes include “naming and shaming” tactics, as well as investigative journalism efforts such as the financial mapping of Mundipharma [69]. Finally, it is important to note the ever-evolving nature of the case of Purdue, the Sackler family, and OxyContin. Since proceedings surrounding this case were ongoing at the time of data generation and analysis, there are likely evolutions and updates to the case that post-date this analysis.

Conclusion

This paper argues that there is multiplicity, complexity, and heterogeneity of accountability as enacted for the harms caused by Purdue. Through the lens of ANT, this paper provides a novel perspective on the issue of Purdue, OxyContin, and the complex interactions involved in enacting social, political, and legal versions of accountability. In demonstrating that accountability is enacted through multiple overlapping networks, we can better understand that global health systems are formed by diverse and often fragmented processes. This disconnection can hinder coordinated efforts to address prominent issues such as corporate malfeasance and public health crises. Further, ANT and its concepts allow us to expand tightly bound definitions of accountability and challenge the traditional boundaries between sectors (e.g. medical/legal) by demonstrating the interconnectedness and flows of influence within and across sectors as well as borders.

Learning from one of the most devastating cases of pharmaceutical harms to date, this study contributes vital insight into the complexity that is accountability in global health systems. As the opioid crisis in and beyond North America continues to grow and evolve, it is essential to identify key levers, weaknesses, and opportunities of global health accountability systems so that we may strengthen regulatory frameworks, oversight mechanisms, and enforcement practices to better respond to and prevent future harms caused by transnational pharmaceutical companies and their allies. As such, we emphasize the need to prioritize stronger integrated accountability mechanisms that leverage the resources and strengths across sectors to safeguard public health and better hold transnational pharmaceutical companies, such as Purdue, accountable for the harms they cause.

Author contributions

A.B. acquired and analyzed manuscript data, conceptualized the manuscript, and wrote the main manuscript text. All authors contributed to the revision of the manuscript. All authors reviewed the manuscript.

Funding

This project was funded by a Canadian Graduate Scholarship (Doctoral) from the Canadian Government’s Social Sciences and Humanities Research Council awarded to author AB.

Data availability

The raw interview data that support the findings of this study are not openly available to protect the confidentiality of key informants.

Declarations

Competing interests

The authors declare no competing interests.