Research as Development

Front matter

RESPARDH As ~ © DEVELOPMENT

Biomedical Research, Ethics, and Collaboration in Sri Lanka®*

JALLA SARIOLA AND BOB oIMPoON =,

Contents

Acknowledgments

1. International Collaborative Research in Biomedicine: A Form of Development?

9. Research as Development: Unintended Consequences vii vi Contents Notes 175 Bibliography 179

Chapter 2

[Collaboration is] . . . human behavior that facilitates the sharing of meaning and completion of activities with respect to a mutually shared superordinate goal and which takes place in a particular social, or work, setting.

(IrvonEN AND SONNENWALD 2000, 79)

Science, like any other activity involving social collaboration, is subject to shifting fortunes.

The research collaborations that we focus on in this book all took place in Sri Lanka, an island republic of some 20 million people off the southern tip of India. This location forms the backdrop for our study of how biomedical research can become tangled with economic, infrastructural, epistemological, and scientific development. To understand the specificities of this entanglement by means of ethnographic enquiry in the present day, it is necessary to begin with some sense of how things came to be as they are. In this chapter we move through a series of layers en route to our own engagement with the field of clinical trials, collaboration, and bioethics. We provide concise histories of biomedicine, biomedical research, and the parallel development of medical ethics, research ethics, and subsequently bioethics in Sri Lanka. This historical overview sets the scene for our account of the clinical trials that we observed. In the second half of the chapter we describe how we as researchers entered into this context as both ethnographers and collaborators. In a context so infused with the idea of collaboration, we had to collaborate as well. We explore the implications of this simple methodological fact for ethnographic work in this context and more generally for researching expert systems and professional practice.

Biomedicine in Sri Lanka

The tradition of biomedicine is very firmly established in Sri Lanka. The Civil Medical Department of the British colonial administration was first established as a department separate from the one dealing with the occupying military forces in 1858. Although aimed specifically at the control of smallpox, the creation of a department to address the health needs of the local population, as distinct from the needs of the colonizers, marked an important step on the road to a national health service (Uragoda 1987, 81). Nonetheless, historians of colonial medicine have noted a circularity in how it developed. As Margaret Jones put it, “Western scientific medicine in its tropical form, along with Western sanitary methods, were imposed on populations whose health problems largely stemmed from the exploitative drives of the metropolitan power in the first place” (Jones 2000, 88; see also Arnold 2000; Hewa 2012 on hookworm eradication). Investment in medical research was thus necessary, in the first instance, to preserve a reasonably healthy workforce. The Colombo Medical School opened in 1870 and is the second oldest in Asia after the Calcutta Medical College in India which opened in 1848. The school developed a distinguished tradition of providing biomedical education to the local population. Teaching followed British curricula, and the degrees were verified from the United Kingdom. Any postgraduate medical training had to be obtained in the United Kingdom. The colonial backdrop is important for our efforts to understand how contemporary international collaborations work in Sri Lanka and elsewhere.

Indeed, it is evident that Britain, Europe, and the colonies have been enmeshed in a complex network of exchanges over a long period. Similar to what we postulate analytically about research, Hormoz Ebrahimnejad, writing on the encounter between traditional and modern medicine in Iran, has argued that rather than a one-way diffusion from the latter to the former, there was, in fact, an ongoing circularity. Ebrahimnejad goes on to claim that Western medicine itself became modernized in the colonies rather than being exported to them in an already “modern” form (Ebrahimnejad 2009). Then as now, there were flows in both directions. As Mark Harrison has suggested, it would be mistaken to think that the colonial countries were excluded from knowledge production; many doctors made their careers in the colonies, which provided sites for research, experimentation, and innovation (Harrison 2010; see also Johnson and Khalid 2012; Raj 2013). Knowledge transfer between Britain and the colonies was thus possible through the networks that connected colonial medical practitioners and administrators and their metropolitan counterparts. The transfer of resources also played its part. Soma Hewa (1995), for example, has documented the role of the American Rockefeller Foundation in developing public health campaigns such as the one aimed at hookworm eradication among tea estate laborers between 1915 and 1920, which became an important platform upon which local public health services were eventually built.

Throughout the nineteenth and early twentieth centuries and until independence in 1948, Sri Lanka, or Ceylon as it was previously known, was considered a model colony within the British Empire. Local elites were highly Westernized and politically moderate, and the medical sector in particular had seen generations of collaborative interaction with “foreigners.” Even today, the medical profession preserves a strong sense of identity modeled on the structures that took root during the British colonial period.

The Anglicization of medicine was, and continues to be, reinforced by the use of English as the medium of medical (and other professional) education. Indeed, the language issue remains a source of tension, as the majority of the population receive primary and secondary education in their mother tongue (Sinhala or Tamil) but then have to take their medical degrees in English. Furthermore, an entrenched Anglophone biomedicine has had the effect of marginalizing the indigenous systems of medicine. As Kamalika Pieris has noted, in the nineteenth century the discussions regarding the development of a medical school identified one of its aims as being to “send out well-educated young men to open up the dispensaries of the Island and to diffuse a knowledge of European medicine among the poorer classes of the community and [thereby] in time supersede the ignorant vedarala” (Pieris 2001, 17). The vedarala were native physicians practicing a variety of indigenous medical techniques, including Ayurveda, and antipathy toward them was well established among practitioners of biomedicine. Despite the efforts to reclaim and elevate indigenous traditions (Wickramasinghe 2014, 104), this antipathy has continued to manifest itself among some present-day doctors and researchers. For much of its history, the Sri Lankan medical profession has not only been highly Westernized, English-speaking, and of high status but also cut off from the beliefs and values about health and illness held by the majority of the population.

In more recent times, an inexorable shift toward deregulation and an open-market economy has shaped health care access. After the election in 1977 of J. R. Jayawardena, who led the government with an agenda of fundamental market reform and economic liberalization, the private sector has come to play an increasingly important role in health care. In 1978, for example, doctors were for the first time allowed to maintain private practices in addition to their government responsibilities, a development that spurred a spectacular growth in private provision, particularly in tertiary care. International, market-priced pharmaceuticals also became available (to those who could afford to pay for them). In an overview of the health care system, Dulitha N. Fernando argued that it is difficult to assess the exact contribution made by private health care providers as the sector is not regulated, but she estimated that approximately 50 percent of primary medical care in 2000 was provided by private sector general practitioners (Fernando 2001). The government sector has been further affected, particularly in primary care provision, as pressures from the World Bank and other international financial institutions have increasingly undermined the earlier patterns of welfare expenditure (Jayasinghe 2002, 6—7).! Furthermore, the import of equipment to private medical institutions was greatly accelerated after 1977 by the granting of duty-free concessions. The result of these and other developments has been an overall trend toward private medicine.

In contrast, the public sector, which to date has provided free health care at the point of need, has fallen further behind the private sector in some fields.’ Steven Russell’s investigation of health care in Sri Lanka provided further evidence for the widespread use of private medicine in Sri Lanka (2005). His study examined a wide range of income groups and concluded that government services, which continue to be provided free at the point of need, remain important. Despite the growth of private medicine, people do not choose their providers based on whether they have to pay for them or not. The interweaving of public and private sectors is now dense both in terms of users and providers. The turn to private medicine raises concerns that private provision might even displace the state as the key provider of health services altogether. The ability to respond to health care problems has been impeded by disjointed management of health matters, administration, and budgeting. For example, in writing about the impact of technology on health care, Abeykoon highlighted the “confusion and conflicts over roles, responsibilities and lines of accountability between central and provincial levels of the Ministry of Health” as a “major issue” that resulted from devolution after 1977 (Abeykoon 2003, 11). Saroj Jayasinghe and Nalaka Mendis also reported that the health care system was “fragmented” with poor processes and structure and “lacks data and reliable information” (2002, 100; also see Rannan- Eliya and Sikurajapathy 2009, 329-36). It is in an environment where health care costs rise inexorably and privatization seems to present itself at every turn that international medical collaborations are currently being assembled.

Yet, Sri Lanka is cited by many as a success story for having managed, despite low spending and gross domestic product, to run a free national health system (Bhutta, Nundy, and Abbasi 2004). Part of this health system was the Sri Lankan pharmaceutical policy regime set in place in the early 1970s (Cooray 2016). At the time, it was internationally groundbreaking. As one of these internationally trained scholars, the pharmacologist Senaka Bibile, who had studied in Sri Lanka and obtained his doctorate from the University of Edinburgh, devised a so-called Essential Medicines List for Sri Lanka. At the time, Sri Lanka depended on imports from the United States and Europe and suffered from prices that were too high for people to pay. The Bibile system consisted of a list of drugs that were considered crucial to the health of the population; these drugs were either bought in bulk by the local State Pharmaceutical Corporation or were manufactured locally and provided for free to patients in public hospitals. The list initially included only fifty-two pharmaceuticals, so drug availability was rather narrow, but these generics kept prices down (Lall 1977, 692). For example, with the institutionalizing of a state-regulated pharmaceutical sector, the prices for antibiotics dropped by 85 percent in 1972 even while they were rising elsewhere globally (Lall 1977). The World Health Organization (WHO) adopted the system, and a worldwide Essential Medicines List has been released every two years.

Successes in such areas as family planning, nutrition, and female literacy have helped bring about substantial increases in life expectancy. Rasika Jayasekara and Tim Schultz (2007) noted that Sri Lankan life expectancy increased from around forty-three years of age just before independence to seventy-three years in 2004. However, as people began to live longer, they in turn contributed to the increase of noncommunicable and degenerative diseases such as heart disease and diabetes (Ghaffar, Reddy, and Singhi 2004). Palitha Abeykoon (1998) estimated that in the twenty-five years between 1995 and 2020, the number of those aged sixty and older will have more than doubled in Sri Lanka. As such, the problems that affect the elderly—such as falls, depression, cognitive dysfunction, and severe visual impairment—are set to increase significantly (Weerasuriya and Jayasinghe 2005, cited in Jayasekara and Schultz 2007). Jayasekara and Schultz also suggested that social and demographic changes have resulted in a weakening of “traditional and extended family networks”; younger generations becoming more dispersed and parental generations becoming more isolated has put further pressure on the health care system (Jayasekara and Schultz 2007; Widger and Kabir, n.d.). Moreover, noncommunicable diseases are on the rise at a time when the country is still managing infectious diseases as well as those rooted in poverty such as maternal anemia, stunted growth, and childhood malnutrition (Abeykoon 2003). As such, the health care system is forced to “fight on two different fronts” (Gunawardene 1999), and these conditions weigh heavily on the resources available in Sri Lanka (Deshapriya and Welikala 2004; Jayasekara and Schultz 2007). As we will see in chapter 3, this emerging epidemiological profile is important for international research collaborations: the increasing convergence with populations in the West opens the possibility for new populations of human subjects to participate in externally sponsored trials, and it is into this setting that the research collaborations we describe in this book were introduced.

Finally, it is impossible to give an overview of contemporary health care provision in Sri Lanka without reference to the communal and political violence that has consumed the island in recent decades. Over twenty-five years of conflict between the Liberation Tigers of Tamil Eelam (LTTE) and the government of Sri Lanka has taken a major toll on health and public services (Siriwardhana and Wickramage 2014). The bulk of fieldwork for this study was conducted during the time leading up to the end of the war, a very tense, oppressive era. Ongoing stressful living conditions and fear of deadly attacks and more silent dangers such as landmines were the norm for much of the population (Reilley, Abeyasinghe, and Pakianathar 2002). The war and internal population displacement placed great strain on hospital beds and services with the increased number of people suffering from long-term mental and physical health problems and physical disabilities. These conditions have left Sri Lanka facing significant challenges for both delivering health care and conducting research. When it comes to improving health care and developing infrastructure, powerful humanitarian and economic imperatives are at work, to which international research collaborations might speak. In documenting the practices of such collaborations, we describe the novel strategies at work to effect transformative change on the wider health system in Sri Lanka. This points to not only material changes but also conceptual ones, in which research, evidence, and experimentation begin to figure as essential drivers of policy formulation and change. In the next section, we provide a brief overview of past biomedical research in Sri Lanka as a necessary prelude to our discussion of contemporary research collaborations.

Biomedical Research in Sri Lanka

When international collaborations are conducted in Sri Lanka, they do not occur in an empty epistemic space. As we have seen, layers of interaction and exchange already existed that had built up over many centuries. Within this complex mosaic, local scholars will also proudly draw attention to Sri Lanka’s own traditions of medical research. Medieval ayurvedic physicians are credited with having developed early surgical techniques; their attempts at rhinoplasty to repair facial disfigurement and techniques for repairing anal fistula have been documented (Aluwihare 1982). More recent times also have seen medical advances, particularly in the fields of surgery and management of tropical diseases. For example, Pieris referred to the mid-nineteenthcentury surgeon P. D. Antonisz being lauded for performing the first successful esophagotomy and ovariotomy (Pieris 2001, 129). Trials carried out by local doctors have included a trial to test a cure for hookworm, in which patients were randomly selected from local hospitals (Goodwin, Jayewardene, and Standen 1958). Pieris also referred to a report in the Journal of the Ceylon Branch of the British Medical Association describing a trial performed at the general hospital using morphine and hyoscine as anesthetics during labor (Pieris 2001, 132). Other notable interventions in the field of experimental trials included one of the first trials of the contraceptive pill in the 1950s, which led Darshi Thoradeniya to describe Sri Lanka as a “laboratory for pill trials” (forthcoming).

Despite this history of experimentation, scientific research using randomized controlled trials in Sri Lanka has been limited. In a global survey of science capacity calculated according to networks of collaboration, international and national publications, and availability of funds for research, Sri Lanka landed fifth from bottom in a list of seventy-five countries (Wagner 2008, 132). This result led Caroline Wagner to describe Sri Lanka as one of a number of countries that was scientifically “lagging” (Wagner 2008). Although, as Amit Prasad (2014) reminds, it is important to treat such surveys with caution, they also serve a particular rhetorical purpose in justifying capacity-building activities.

In numerous essays and talks published over the last two decades, the eminent Sri Lankan microbiologist Professor S. N. Arseculeratne has maintained a scathing critique of the state of scientific research in his native Sri Lanka and provided his own views of why more developments have not been made (Arseculeratne 1999, 2008). In Arseculeratne’s view, the problem stems not from any global economic or political disparities but from a widespread deficiency in the way that science is promoted and understood in Sri Lanka. He describes a kind of scientific malaise that runs through the major teaching and research institutes. He sees this as particularly evident in the way that medical education is delivered. Medical doctors are not expected to publish research to progress in their careers; for university researchers, publication expectations are low or nonexistent. One of Arseculeratne’s main criticisms is that medical education is largely based on factual recall and rarely encourages analytical, experimental, or creative approaches. In one of his talks, he spoke of medicine needing to be seen “not merely as a craft with utilitarian ends, but as an important component of science with an intellectual background that includes the history and philosophy of modern science” (Arseculeratne 2010).

According to Arseculeratne, an unfortunate corollary of this deeply engrained utilitarianism is that the curiosity that drives medical research and innovation is only weakly present in Sri Lanka. He sees a further inhibitor of young doctors’ curiosity in the authoritarian ethos of medical education and practice. For medical students, relationships are marked by strong vertical hierarchies based on the status, knowledge, charisma, and reputation of their teachers. The steep power gradients that separate junior medical staff from their superiors manifest in a good deal of fear, a concern to avoid offence, and a tendency to replicate rather than challenge received wisdom; to fall foul of a powerful senior is to risk long-lasting damage to one’s reputation and future prospects. Such conditions are hardly conducive to a free and open exchange of ideas that challenges orthodoxy and generates innovative approaches.

Given the problems faced by the health sector in Sri Lanka, it is hardly surprising that researcher malaise has been further compounded by starved research budgets and a significant drain of expertise out of the country. These factors go some considerable way to explaining why Sri Lanka has been pronounced scientifically lagging.

In contrast to the view of the rote-learning medics practicing in an enfeebled research environment, Sri Lankan medical students are also pointed toward a venerable tradition in which local doctors working in resource-poor settings overcome tremendous odds by using their ingenuity and skills of improvisation. As Pieris put it in her history of the Sri Lankan medical profession, “the dedicated government doctor could be described as possessed of a perpetual pioneering spirit in that the government service invariably held shortcomings which the doctors had to somehow overcome if they were to deliver a satisfactory service” (Pieris 2001, 139). In contemporary settings, doctors still have to be able to perform with limited diagnostic testing facilities and make decisions drawing on basic clinical expertise, judgment, and experience. The heroic image of the doctor that medical students are taught to emulate is that of the healer, who brings relief from suffering in the service of others and without thought of material gain—a benevolent, paternalistic role that has particularly powerful resonances within Sri Lankan culture beyond medicine.

The reason for this brief excursion into Sri Lankan biomedical traditions is that it provides some sense of what was there before and what some would see as an impediment to change. As we will discuss in chapter 6, the extant research culture and efforts to transform it through international collaboration result in a reshaping of the subjectivities that underpin many of the roles and relationships that feature in the practice of biomedicine. In particular we focus on the relationship between doctor and patient, and the ways in which new modes of acting and being are informed by the relationship between researcher and research participant. This transformation, however, is only partially wrought through practice. Another significant source of such transformation is found in the systems of research governance that the new modes of experimental research bring with them. In the next section we provide, by way of backdrop, a brief history of the rise of research ethics and regulation in Sri Lanka.

Collaboration in Context 35 Research Governance in Sri Lanka

In developing world settings, biomedicine and biomedical research not only have become increasingly imbricated with each other but are both now shaped by the demands of research governance and regulation. This is particularly the case where human participants are involved. The shift from a physiciancum-scientist virtue-based ethics to one in which there is externalized evidence of ethical intent captured through audit and contract is in many respects readable from the rise of ethics reviews by committee. In the most basic of terms, the approval of a formally constituted body of experts in the form of an ethics review committee (ERC) is intended to ensure that research is beneficial, scientifically valid, and above all safe for those who participate (Stark 2011). At the time that our research was carried out, the machinery for ethics review was being rapidly developed across South Asia, with ERCs following a broadly similar institutional and procedural format to those established in Europe and the United States (Douglas-Jones 2012).

The genealogy underpinning these developments is well established. The principles of ethics review, as laid out in the Declaration of Helsinki (World Medical Association 2013, paragraph 23), include the need for a clearly formulated research protocol as well as information about funding, sponsors, institutional affiliations, conflicts of interest, and the use of incentives for participants. When the research is to take place in a country different from that of the research sponsor, it is expected that there will be a local ethics review based on the Helsinki principles, or what is generally referred to as a “two-ended review.” The European Medicines Agency (EMA) has emphasized the need for local ethics committees even more strongly, insisting that a local ethics committee is a prerequisite for a clinical trial to take place at all (European Medicines Agency 2012; CIOMS 2016, 29-31). The Universal Declaration on Bioethics and Human Rights takes these requirements one stage further, toward research as (a catalyst for) development—the focus of our interest here. Article 19, which deals with the topic of ERCs, states the following:

Independent, multidisciplinary and pluralist ethics committees should be established, promoted and supported at the appropriate level in order to a) assess the relevant ethical, legal and scientific and social issues related to research projects involving human beings, b) provide advice on ethical problems in clinical settings, c) assess scientific and technological developments, formulate recommendations and contribute to the preparation of guidelines on issues within the scope of this Declaration and d) foster debate, education and public awareness of, and engagement in, bioethics. (UNESCO

Here the remit extends beyond the mere setting of standards and begins to move into the realms of capacity building and the creation of a climate in which the objectives of the Universal Declaration will be realized. The encouragement for economically developed countries to grow local capacity in research governance is typically expressed in terms of empowerment and the hope that various forms of ethical imperialism can be mitigated. The involvement of host-country ethics committees in research governance also brings into play their knowledge and experience of local conditions, which it is believed will facilitate the research and benefit its participants.

However, a number of problems are usually identified in the development of the two-ended ethics review model. As Ruth Macklin pointed out, the tieup between the ends often leaves much to be desired; even though it is now common to have host and sponsor engaged in ethics review, there are few link-ups between ERCs and little by way of significant and potentially productive exchange (Macklin 2004, 152). Concerns have been expressed about the competence of local bodies to review research in a way that is broadly equivalent to the sponsor country’s procedures. Questions also arise when host and sponsor country ethics committees appear to work with fundamentally different ethical standards and principles. Such discrepancies have been the case in a number of clinical trials where local committees, after full ethics review of a project, appeared to be prepared to accept standards much lower than would pass in economically developed countries. Paul Farmer goes further still, identifying these discrepancies as the tacit acceptance within international biomedical research of fundamental health inequalities between low- and high-income countries (Farmer 2004, 200; see also Geissler 2013). He suggested that structural differences are at times read as “cultural” differences and dismissed as acceptable local practices.

Notwithstanding these concerns, ERCs are now a ubiquitous feature of research governance. During the period covered by our research, many Sri Lankan doctors and academics were fervently engaged in capacity building in bioethics, specifically in relation to research ethics and governance

(Douglas-Jones 2012). The reasons were simple: without corresponding local review procedures, the experimental research activity taking place in Sri Lanka would not be recognized as part of the global episteme of ethically validated knowledge production. In other words, the opportunities for international collaboration would be severely limited, and with this would come the kind of epistemic exclusion discussed in the introduction.

Partly in response to the potential loss of opportunity, between 1980 and 2006 the number of ERCs in Sri Lanka rose steadily from two to fifteen. At the time of our research in 2010, the figure was twenty, with more planned. Ethics committees were established in the universities of Colombo (1981), Peradeniya (early 1980s), Galle (1984), Jaffna (1985), and Kelaniya (1995) as well as the Sri Lanka Medical Association national committee (1999) (Dissanayake, Mendis, and Lanerolle 2006). According to local commentators, in the 1990s training in medical ethics was badly in need of updating and development in response to scientific and technological advances in biomedical sciences and to the changing structure of health care delivery. Vajira Dissanayake, N. Mendis, and R. D. Lanerolle (2006) described the transition as being from a well-established medical ethics concerned essentially with the conduct of doctors in relation to their patients to one that was more closely aligned with the directives of a global bioethics.

In 2007, the Forum of Ethics Committees Sri Lanka drafted a guideline: “Ethics Review Committee Guidelines: A Guide for Developing Standard Operating Procedures for Committees That Review Biomedical Proposals.” This document was intended to unify the island’s diverse procedures in the context of ethics reviews. The direction of travel was at that time strongly influenced by a widely read article by Ezekiel Emanuel, David Wendler, and Christine Grady (2000) entitled “What Makes Clinical Research Ethical?” This article has been influential in providing ethics committees in the developing world with a checklist of sorts when it comes to dealing with outside collaborations. Although this national guideline now exists for Sri Lanka, many committees have in fact developed their own standard operating procedures, rendering the landscape of ethics review quite diverse.

Capacity-building initiatives around ethics review and research governance in Sri Lanka have drawn strong support from medical elites keen to engage with the opportunities that international research collaboration makes possible (see Douglas-Jones 2012). This support has resulted in the opening up of new institutional spaces in which novel forms of advocacy and activism are evident. Support for local capacity-building initiatives has come from organizations such as the Forum for Ethical Review Committees in Asia and the Western Pacific (FERCAP), the Strategic Initiative for Developing Capacity in Ethical Review (SIDCER), and the Global Forum on Bioethics (GFB), all of which signal the bioethical collaborations which necessarily underpin scientific collaboration. Affiliation to these organizations and the establishment of local branches (for example, FERC—Sri Lanka) has been an important route to harmonization and the dissemination of good practice. It is also a way for local researchers to connect with regional networks and to participate in transnational research assemblages. For example, the Colombo Medical Faculty committee has had access to training from the U.S. National Institutes of Health located in Bethesda, Maryland. A researcher from the faculty has completed a bioethics degree there, and bioethics lectures have been streamed to students. The Colombo Medical Faculty Ethics Committee is also accredited by the Strategic Initiative for Developing Capacity in Ethical Review (SIDCER), an initiative which emerged out of collaboration between the WHO and the Special Programme for Research and Training in Tropical Diseases (TDR). At the time of writing, no other ethics committee in Sri Lanka had this accreditation. The National Science Foundation, whose ethics committee authorizes applications that have national reach, has developed its provision through training provided by UNESCO. Finally, only eight out of the twenty ethics committees operating in the country at the time of this writing are accredited by the Sri Lankan Drug Regulatory Authority to give ethics clearance to clinical trials. These ERCs were in the University of Colombo, University of Kelaniya, University of Sri Jayewardenapura, University of Peradeniya, University of Ruhuna, University of Jaffna, the Sri Lanka Medical Association, and the Medical Research Institute. As we will see in chapter 8, awareness of the regulatory landscape is important when it comes to understanding the internal debates and positions that are thrown up when bioethics and research governance are addressed in the local context.

Arguably, however, when it comes to clinical trials regulations, most ERCs follow the latest version of the International Conference on Harmonization Good Clinical Practice Guidelines (ICH-GCP). These guidelines provide an authoritative summary of good practice for industry-funded clinical trials from the European Union, Japan, the United States, and, more recently, from