Going Boldly Where No One Has Gone Before? How Confidentiality Risk Aversion is Killing Research on Sensitive Topics
- 189 Downloads
Bernhard and Young (Journal of Academic Ethics, 7, 175-191, 2009) allege that a “myth of confidentiality” plagues research in North America because of the absence of statute-based legal protections and the requirements of some REBs to limit confidentiality “to the extent permitted by law.” In this commentary we describe statute-based protections for research confidentiality available in the United States, clarify the legal situation regarding research confidentiality in Canada, and explain that REBs that “require” confidentiality to be limited by law are imposing a doctrine that is not required by the TCPS and may violate researchers’ academic freedom. The paper laments how excessive REB risk aversion and inaction by the granting agency Presidents has created a situation where some REBs are encouraging researchers to download research risks to research participants and forcing researchers to choose between exposing themselves to the prospect of going to jail to protect confidentiality, watering down their research objectives, or conducting “vanilla” research rather than engaging in controversial and/or sensitive areas of study. The paper urges the granting agency Presidents to seek legislative change to protect research participants who provide information that could cause them harm if their identity were to be revealed.
KeywordsConfidentiality Statute-based protections Wigmore criteria REB review Confidentiality certificates Ethics and law
In their important article “Gaining institutional permission: Researching precarious legal status in Canada”1 Bernhard and Young (2009) offer a case study of their efforts to design and implement an interview-based field study of the experiences of persons residing in Canada illegally. Bernhard and Young begin by pointing out that little is known about this largely invisible segment of the population—which is estimated to be between 40,000 and 600,000 people—and offer many persuasive arguments about the social value of doing such research. But then, instead of proceeding with the discussion of methods and results one would expect, the narrative transforms into a cautionary tale about what can happen when a risk-averse REB insists that researchers limit confidentiality to account for the possibility of court-ordered disclosure of research information, or of immigration authorities attempting to secure from them the identities of illegal residents. Instead of following the REB’s suggestion to limit confidentiality in a way that would download these risks to research participants or “changing the focus to participants who were at some point living with precarious status but had regularized their situations” (p.185) Bernhard and Young abandoned their longitudinal research design in favour of a cross-sectional approach so that the identities of research participants would not need to be recorded. Instead of the 1,700 surveys and 250 interviews hoped for in the originally proposed project, they ended up with just 18 persons participating in a pilot project (p. 186). Because of the change in their research design, they were unable to pursue many of their original research objectives. Reflecting on this dismal outcome, Bernhard and Young argue that researchers need to be made aware of the “myth of confidentiality” created by the threat of subpoenas and other lawful attempts to obtain confidential research information and the lack of any rock-solid legal basis on which to guarantee it.
Bernhard and Young’s sense of ethical responsibility to their research participants is laudable: they did what they thought was necessary to avoid harming any of their research participants. However their REB appears not to have apprised Bernhard and Young of the full range of alternatives that the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (Canadian Institutes of Health Research et al. 1998/2005; hereafter “the TCPS”) recognizes are available to researchers for approaching research confidentiality. And while Bernhard and Young’s legal analysis goes beyond the scant information in the TCPS by identifying some important legal possibilities that researchers should consider, the situation may not be quite as bleak as they suggest.
Bernhard and Young’s experience reflects the contemporary ethos of risk aversion that is undermining the contemporary university’s ability to fulfill its social research mandate, although one does not usually see it so clearly manifest as this. Indeed, one wonders how big the “dark figure” of forsaken research is, as researchers in a publish-or-perish culture avoid REBs entirely and pursue methods that are exempt from ethics review, or conduct “vanilla” research because they are unwilling to spend the time it takes to negotiate with REBs in order to be able to conduct research on sensitive topics. In the process, REBs may have unwittingly become a state-sponsored instrument for sanitizing potentially controversial research on sensitive topics.
More generally, Bernhard and Young’s experience draws attention to the absence of leadership being shown by universities and the Presidents of Canada’s three granting councils—the putative stewards of Canada’s system of protecting research participants from harm—in addressing the “myth of confidentiality.” As a complement to Bernhard and Young’s provocative narrative, we elaborate these arguments below.
Statutory Law Regarding Research Confidentiality
Bernhard and Young’s portrait of Canadian law is correct on two counts: (a) there are no statute-based protections for research participants in Canada that would allow academic researchers to offer a legally-backed guarantee of “complete confidentiality” to research participants; and (b) in lieu of a research shield law, an academic researcher’s best legal defence would be to claim privilege by invoking the Wigmore criteria, which the Supreme Court of Canada has recognized as the appropriate test to adjudicate claims of privilege where no statute-based privilege exists (Slavutych v. Baker et al. 1976).
Research on HIV, AIDS, and other STDs;
Studies that collect information on sexual attitudes, preferences, or practices;
Studies on the use of alcohol, drugs, or other addictive products;
Studies that collect information on illegal conduct;
Studies that gather information that if released could be damaging to a participant’s financial standing, employability, or reputation within the community;
Research involving information that might lead to social stigmatization or discrimination if it were disclosed;
Research on participants’ psychological well being or mental health;
Genetic studies, including those that collect and store biological samples for future use;
Research on behavioral interventions and epidemiologic studies.2
refuse to disclose identifying information on research participants in any civil, criminal, administrative, legislative, or other proceeding, whether at the federal, state, or local level. Certificates of Confidentiality may be granted for studies collecting information that, if disclosed, could have adverse consequences for subjects or damage their financial standing, employability, insurability, or reputation. By protecting researchers and institutions from being compelled to disclose information that would identify research subjects, Certificates of Confidentiality help achieve the research objectives and promote participation in studies by assuring confidentiality and privacy to participants.3
Further, had their research been conducted in the U.S. and funded by the Office of Justice Programs (OJP), 42 U.S.C. §3,789g provides that information collected using OJP funds are not admissible as evidence “in any action, suit, or other judicial, legislative, or administrative proceedings.” OJP-funded research is subject to 28 CFR Part 22 (§22.23), which requires funding applicants to certify that they will not divulge confidential research information pertaining to any identifiable private person (Lowman and Palys 2001; Palys and Lowman 2002).
The Common Law Regarding Research Confidentiality
It seems that if we followed the ERB5 suggestion to indicate that we could assure confidentiality ‘to the extent permitted by law’ we would fail the Wigmore. If we said that we absolutely guaranteed confidentiality and would be willing to go to jail to uphold it, we might meet the Wigmore criteria; however this defence might or might not hold up in court. In other words, even if we risked going to jail for both our study participants and for the greater public good of conducting the research, we might promise confidentiality but not be able to deliver it (p. 185).
The first requirement for privilege is that the communications at issue have originated in a confidence that they will not be disclosed. The Master held that this condition was not met because both the appellant and Dr. Parfitt had concerns that notwithstanding their desire for confidentiality, the records might someday be ordered disclosed in the course of litigation. With respect, I do not agree. The communications were made in confidence. The appellant stipulated that they should remain confidential and Dr. Parfitt agreed that she would do everything possible to keep them confidential. The possibility that a court might order them disclosed at some future date over their objections does not change the fact that the communications were made in confidence (para. 24)
The problem with researchers limiting confidentiality a priori “to the extent permitted by law” is not that it will destroy the researcher’s ability to assert privilege by invoking Wigmore6—a researcher should do everything possible to protect confidential research information and assert privilege if necessary—but the information conveyed to prospective participants for the purpose of obtaining informed consent is itself problematic, because it conveys no information about what it means to protect confidentiality “to the extent permitted by law.” What exactly does the REB envision as “the extent permitted by law”? Is the subpoena the limit? Is there an obligation to challenge the subpoena? Will the REB or university take the case to the Supreme Court if necessary? For a participant to provide informed consent they presumably would need answers to all these questions. Most research participants would probably not have the knowledge necessary to pose such questions in the first place.
Bernhard and Young suggest that lawful threats to research confidentiality in Canada are “far from hypothetical,” in which case one might ask, when has a lawful threat to confidentiality in Canada ever resulted in the loss of research confidentiality? Indeed, how many lawful threats to confidentiality have there ever been in Canada? And while there is a much more extensive record of lawful threats to confidentiality in the U.S., how many have resulted in the disclosing of confidential research information?
Our encounter with these issues prompted us to further investigate the issue of confidentiality in academic research. How had other researchers coped with these limitations that we had encountered? Would the legal system seek to apprehend representatives of the academic system engaging in research that could be clearly proven to be for the public good? To our dismay, we learned that these issues were far from hypothetical.
The fact was that researchers in a number of areas had been subpoenaed to produce their field notes in order to identify participants. For example, in 1971 a researcher was subpoenaed for his study about whether participants had collected welfare while receiving income maintenance in the US. Another of the cases that gave us pause was that of Samuel Popkin, a Harvard professor who was jailed for 8 days in 1972 for refusing to reveal the identities of the people he interviewed for a secret war study (Lowman and Palys 2001). In 1973 the US Federal Bureau of Investigation threatened to subpoena the research records of the Kinsey Institute’s work on human sexuality. We discovered that in some cases, upon failure to produce field notes, researchers had been jailed for several months at a time (Caroll and Knerr 1975; Lowman and Palys 2001). (p.183)
While researchers have been the subject of many subpoenas in the United States, particularly in the context of civil trials and grand jury proceedings, being subpoenaed is one thing; being ordered to violate research confidences after challenging the subpoena is entirely another. The judicial track record in relation to research participants is very encouraging. The published record of U.S. cases indicates that U.S. judges have shown consistent respect for the research enterprise and the research participant’s role in it (Lowman and Palys 2001; Palys and Lowman 2000, 2002). While at least two researchers in the United States have indeed been jailed because of their refusal to disclose confidential research information (Samuel Popkin in 1972; Rik Scarce in 1994),7 the proceedings both involved grand juries. Canada does not utilize a grand jury system of pre-trial disclosure, using the “show-cause” hearing instead.
Overall, then, the situation in Canada is very different from the U.S. To our knowledge, only one researcher in Canada has ever been subpoenaed—Russel Ogden—who has received three (in 1994, 2003 and 2004). The first was initiated by the Vancouver Coroner in relation to Ogden’s Criminology MA thesis at Simon Fraser University; the Coroner sought the names of two of Ogden’s research participants who might provide useful information in the death of an unknown female that was reported in Ogden’s thesis and a local newspaper. Ogden refused, was threatened with a charge of contempt of court, but invoked the Wigmore criteria to claim a public interest privilege. The Coroner agreed that the public interest in protecting research confidentiality was greater than the Coroner’s interest in the data, and there ended his attempt to force Ogden to testify. The two other subpoenas came years later when prosecutors in a trial of a woman charged with counselling suicide subpoenaed Ogden in the hope that his research on assisted suicide might provide information about that woman’s activities. Each time Ogden received a subpoena he challenged it. Both subpoenas were withdrawn, thus negating the need to argue the case for privilege. He suspects they were withdrawn because it would have been obvious to the Court that the subpoenas constituted a “fishing expedition” for evidence (Ogden 2010). The Crown would have had to demonstrate that Ogden was likely to have material evidence for the subpoena to be granted, and not merely demonstrate a suspicion that he might have such evidence. Announcing his intention to challenge each of these subpoenas was enough to ward off the threat, which suggests that we may be our own worst enemies if we are cowed into thinking that any threat a researcher might receive has substance.
In fact, the legal issues we or our participants were at risk of encountering became evident to the ERB immediately upon receipt of our application form. This form includes a section in which the researcher estimates the level of risk involved in participating in the project. In our applications, due to the focus on precarious legal status, we indicated that the participants faced ‘more than minimal risk’ by talking to us. (p.182)
How can risk be classified as “more than minimal” when it has never materialized in what must be hundreds of thousands of research projects conducted in Canada—in which case the prima facie risk is zero? The answer to this question depends on how one applies the TCPS definition of “minimal risk” research, i.e., that in which “the probability and magnitude of possible harms implied by participation in the research [is] no greater than those encountered in those aspects of [the participant’s] everyday life that relate to the research” (p. 1.5). Put another way, in the research context the “reasonably foreseeable harms” required in TCPS 2.4(c) for informed consent are difficult to identify. If one argues that the mere fact of participating in research creates risks greater than those encountered in everyday life, then all research creates greater than a minimal risk. Alternatively, one could argue that risk occurs only when it involves identifiable participants divulging “sensitive information,”8 in which case all research involving collection of such information that is not anonymous from the outset constitutes greater than minimal risk. Conversely, if one argues that participating in Bernhard and Young’s research creates no more risk than anyone who is unlawfully residing in Canada would confront everyday—which is why they are loathe to seek medical help, social services, family violence counselling, and all the other things that Bernhard and Young see as the price of residing unlawfully in Canada—then participation in their research would involve no more risk to participants than they would encounter every day. One problem created by the TCPS definition is that an REB can interpret minimal risk in any one of these ways, and has the power to impose it. But is this a reason to abandon or fundamentally alter a research project?
[G]iven the local context at the time, we felt we had reason to be wary. There had been publicized immigration raids in malls in the city and a non-status woman was detained and deported on Ryerson’s campus during an international women’s day event. Given this intensified focus on non-status issues, the uncertainty around confidentiality made us unwilling to take that risk (Bernhard 2010).9
Just to confirm whether we were truly at risk of encountering these legal issues, we decided to contact the Canadian Border Services Agency (CBSA) asking for a letter undertaking to respect the confidentiality that we would be promising our participants. We argued that although it was clear that the proposed research was for the public good, in order to conduct the study, we had to be able to assure participants that their responses were confidential and that their participation in this study would in no way trigger enforcement to investigate or deport them. The response that we received was that the agency was mandated to remove inadmissible persons from Canada as soon as is practical. Therefore, it would be contrary to the CBSA’s legislative obligations to entertain the request not to pursue enforcement action of the study participants (p.185)
When sensitive data are gathered from identifiable sources, one of the ways researchers protect participants is by anonymizing any notes and/or transcripts as soon as possible. This makes any risk fleeting because the information on file no longer has an identifiable source. By the time any third party knows that the information exists and seeks to challenge confidentiality, researchers are further aided by the passage of time that makes their recollections of who exactly said what and when more fuzzy, and thus of questionable evidentiary value. By contacting CBSA, the researchers drew attention to themselves and their participants before the research began, which might be rather like waving a red flag at a bull.
Ethics and Law in the TCPS
[L]egal and ethical approaches to issues may lead to different conclusions. The law tends to compel obedience to behavioural norms. Ethics aim to promote high standards of behaviour through an awareness of values, which may develop with practice and which may have to accommodate choice and liability to err. Furthermore, though ethical approaches cannot preempt the application of the law, they may well affect its future development or deal with situations beyond the scope of the law.
In response to a query about how to interpret this passage, the granting council Presidents have recognized that, because “legal and ethical approaches to issues may lead to different conclusions,” research ethics may, on very rare occasions, supersede law. Consequently, should a lawful challenge to research confidentiality ultimately result in a court ordering a researcher to divulge confidential information, it is for the researcher to decide as a matter of conscience whether to comply.10 This interpretation of the TCPS is consistent with several disciplinary ethics codes, including the Canadian Psychological Association Code of Ethics (section IV.1711) and the Academy of Criminal Justice Sciences Code of Ethics (Para 1912).
Researchers shall maintain their promise of confidentiality to research participants within the extent permitted by law and/or ethical principles. This may involve resisting requests for access, such as opposing court applications seeking disclosure. Researchers’ conduct in such situations should be assessed on a case-by-case basis and guided by consultation with colleagues, any relevant professional body, the REB, and/or legal counsel. (lines 1749–1755).
It is thus clearly recognized in both the former and proposed policy statement that law and ethics may lead to different conclusions. If there is a conflict, researchers must choose whether they will fulfill their ethical obligation or comply with a lawful order to disclose confidential research information. By absolutely subjugating ethics to law and imposing a “Law of the Land” solution in the final instant, Bernhard and Young’s REB may well have imposed a doctrine that violated their academic freedom to conduct research, depending on how academic freedom is defined in the collective bargaining agreement that applies to them. Indeed, we successfully grieved Simon Fraser University for infringing our academic freedom to conduct research when it tried to impose precisely this confidentiality protocol on criminologists in the late 1990s (Lowman and Palys 2000). Consequently Bernhard and Young (2009) are not quite right when they suggest that the wording imposed on them—confidentiality “to the extent permitted by law”—“is the wording around confidentiality that all Canadian university ethics boards recommend for informed consent documents” (p.185). Since our grievance, the SFU REB has given researchers a choice between the Law of the Land and the ethics-first “strict confidentiality” approach to potential conflicts between ethics and law. More recently, it has stopped making this distinction, and does not require confidentiality to be limited to the extent permissible by law.
The “Myth of Confidentiality” Revisited
On the basis of their analysis of law, Bernhard and Young conclude that research confidentiality in Canada is a myth: even if confidentiality is essential—as was clearly agreed by everyone concerned with their proposed research—threats to confidentiality are real, protections are nil. There is no choice but to promise confidentiality “to the extent permitted by law”—whatever that means—and the law is no friend. Although Bernhard and Young argue convincingly that more protections for research participants are needed, is the current situation really this bleak?
Four parties are involved in the protection of research confidentiality in Canada, each of whom bears some responsibility for the protection of research participants: (1) researchers; (2) research ethics boards; (3) university administrations; and (4) the federal granting agency Presidents and their advisors, the Interagency Advisory Panel on Research Ethics (PRE) and Secretariat on Research Ethics (SRE). We consider each of their responsibilities in turn.
While the TCPS seems to assume that researchers will take ethical shortcuts and require extensive monitoring in order to rein in these propensities, in practice researchers—particularly field-based researchers in the more qualitative tradition of face-to-face and ongoing interaction with socially marginalized and vulnerable people—have been the primary line of defence against threats to research confidentiality and have frequently gone the distance to protect their research participants. We have not found a documented example in the history of the social sciences in North America of a research participant being harmed because of a university researcher disclosing confidential information. However, there are many examples of considerable courage in the face of lawful threats to confidentiality. Russel Ogden continued with the protection of the confidentiality of his research participants and won his case, even after he was effectively abandoned by his university (Lowman and Palys 2000). Alfred Kinsey would have no part of the FBI’s efforts to get access to his data base on sexual histories. Samuel Popkin of Harvard University (in 1972) and Rik Scarce of Washington State University (in 1994) withstood the pressure of grand juries—the “prosecutor’s dream”—and even went to jail to protect their research participants. A special issue of Law and Contemporary Problems entitled Court-Ordered Disclosure of Academic Research: A clash of Values of Science and Law (Cecil and Wetherington 1996) gives example after example of academic researchers—some with and some in the absence of support from their universities—doing whatever it took to cooperate with courts while simultaneously defending their academic freedom to conduct research and ensuring their research participants were protected, as their ethical mandate requires.
Are researchers losing their resolve? Bernhard and Young feel the options available to them in the absence of statute-based protection involve “misleading study participants [because of what they call the “myth of confidentiality”] or risking our own legal apprehension” (p.181). However, as they recognize, researchers do have a vehicle for claiming privilege—the Wigmore criteria—and thus far the record of those either explicitly invoking the Wigmore criteria or fulfilling the four criteria even when not explicitly invoking them13 is unblemished (Palys and Lowman 2002). If researchers design their research to anticipate the criteria, they should be able to ensure that they meet the first three. The wild card lies in criterion 4, where the courts balance the damage to be done to the research enterprise by violating confidentiality against the court’s need for evidence in the case at hand. Although researchers can and should ask themselves ahead of time who might be interested in the information they gather and who would also possess the resources and motivation to challenge research confidences, one can never know for sure what the specific context will be until one gets there.
Our negotiations of the ethics review process suggested that if put to the test (e.g., through a Wigmore defence if our records were subpoenaed), the confidentiality and security of people living with precarious immigration status in Canada would fail to be protected. (p.189)
Does experience support this pessimism? The record of lawful challenges to research confidentiality suggests that researchers should have more faith in the courts than this. In the one case in Canada to be heard—the Ogden case in Coroner’s Court—the Coroner placed a high value on the research enterprise. He understood not only that the ability of researchers to gather information for the public good requires participants to have faith that the information they supply will not subsequently be used to cause them harm, but also that the information itself would not exist for anyone’s benefit—neither researchers, society, nor the courts—were it not for volunteer participants’ faith in the integrity of the research enterprise. In a formal legal opinion, Jackson and MacCrimmon (1999) suggested that the Ogden decision likely would have withstood judicial review, and suggest the most likely situation in which the courts would give the legal need for evidence greater weight than the research participant’s right to confidentiality would be if a serious crime was involved and researchers were the only ones who held information that could keep an innocent person from being imprisoned. In other words, it would take a constitutionally protected right—in this case a defendant’s right to a fair trial—to overcome a research participant’s constitutionally protected right to privacy.14 Although we cannot find a documented instance of a researcher anywhere in the world facing such a dilemma, if we were to confront an “innocence at stake” situation, we would probably not need a court order to take an ethically appropriate course of action that balanced all the interests involved.
The attitude of courts to research is similar in the United States where many researchers have been subpoenaed. The importance judges at all levels attach to the social value of the research enterprise, and the respect shown for research participant privacy, is cause to keep fighting rather than throwing in the towel (Lowman and Palys 2000).
Whenever they begin a sensitive project where violation of confidentiality would harm participants, researchers would do well to engage in some soul-searching about the relative value of research in relation to other social values. If the research enterprise is important and confidentiality integral to gathering valid data on controversial and/or sensitive topics, then researchers should vigorously defend their research and the rights and interests of their research participants. Indeed, given the current apparent lack of interest on the part of the granting agency Presidents to lobby for the enactment of research shield laws in Canada similar to the confidentiality certificate system in the U.S., or some other type of legislated protection, asserting privilege through the common law is currently the only way that researchers can protect research participants and ever hope to have a class privilege recognized.
Research Ethics Boards
In some ways REBs like researchers are victims of university administrator and granting agency inaction. The REB’s role involves identifying prospective risks and ensuring that researchers have dealt with them. The current version of the TCPS identifies the prospect of subpoena as a general risk, but offers no guidance about how to respond to or minimize that risk. The lack of any clear statute-based protection for identifiable research data may give REBs legitimate cause for pause. Nevertheless, Bernhard and Young’s (2009) allegory suggests several issues that REBs ought to consider.
The first is the need to be realistic about risks to confidentiality. Bernhard and Young’s research on persons residing in Canada illegally is an excellent example of the kind of research where specific threats to research-participant confidentiality could reasonably be anticipated, i.e. it is a controversial issue, the research poses prospectively huge costs to participants if their identities were disclosed to authorities, the research was conducted at a time when there was heightened Canada Border Services Agency arrest activity, and the Border Agency constituted a specific third party with the resources and motivation to make a formal challenge to research confidentiality. But how would CBSA even find out the research was being conducted? And did the REB consider what circumstances might actually precipitate such a challenge? Would it be a deportation hearing; a subpoena compelling a court appearance with the prospect of the researcher being asked to provide information about a research participant; a show-cause hearing? If the one Canadian and many US cases involving researcher-participant privilege (see Lowman and Palys 2001) are any indication, the courts would be highly likely to value the research enterprise much more than a fishing expedition for evidence or to provide information to further an investigation (Jackson and MacCrimmon 1999). Indeed, it would be exactly such a case that could help establish a researcher-participant privilege in common law.
In those situations where a specific threat can be envisioned and the legal means and resources to mount a challenge are available, and it is plausible the third party will find out about the research before all data are anonymized, REBs should ensure that the research is designed with the Wigmore criteria in mind to provide the best evidence possible to address any potential challenge. For the most part, this is quite easily done. While the TCPS does not describe the common law defence of confidential communications, general advice about how to design research to enhance the possibility of a successful Wigmore defence is available in the literature (Palys and Lowman 2000, 2002). Where there is a reasonably foreseeable threat by a specific third party—such as a coroner or specific border services agency—formal legal advice should be sought to examine the specific case at hand. However, in order not to infringe their academic freedom to conduct research, REB’s should insist that the researcher decides whether to follow an ethics-first perspective—in which case they should respect a researcher’s proposal to offer strict confidentiality (see footnote 11)—or promise confidentiality to the fullest extent permitted by law.
Being realistic about risks to confidentiality means that REBs should not force researchers to assure confidentiality “to the extent permitted by law” in situations where there is no tangible specific risk to participants. “In-the-realm-of-possibility” threats that have never actually materialized can hardly be said to constitute “reasonably foreseeable risks.” In studies where data are of no conceivable interest to anyone in a criminal case or in the context of litigation, making pledges “to the extent permitted by law” cheapens the research and may undermine the duty of confidentiality, a core ethical responsibility. In such cases REBs should be satisfied with an undertaking of strict confidentiality and expect the researcher to uphold it. In those situations where researchers and/or REBs decide to make a pledge of confidentiality “to the extent permitted by law,” researchers ought to explain to research participants what they mean by that phrase, so that truly informed consent can be achieved. To do otherwise is to invite a caveat emptor ethic in which researchers and universities can treat the mere receipt of a subpoena as the boundary of their obligation, when that is by no means the extent of confidentiality permitted by law. Indeed, if we value research, universities should resist subpoenas at every turn, which brings us to the role of university administrations.
Researchers no doubt would be more likely to make an unlimited pledge of confidentiality if they could feel confident that their universities would support them. However, as Bernhard and Young note, such has not always been the case: “Most surprising to us were the cases where universities had not supported their researchers, allowing them to be sued and even to go to jail in spite of conducting the research as part of the terms of their employment” (p.183). Such inconsistency has indeed been the hallmark in the United States. While institutions such as Harvard and MIT have supported their researchers (e.g., see In Re: Cusumano and Yoffie1997), others such as the University of California at Berkeley (Leo 1995), Washington State University (Scarce 1994), the State University of New York at Albany (Brajuha and Hallowell 1986) and the University of Georgia (Fischer 1996) have not.
The situation in Canada began poorly when SFU effectively abandoned Russel Ogden when he became the first researcher ever to be subpoenaed and asked to divulge confidential information to a Canadian court (Lowman and Palys 2000). Although he was required to do research for his MA thesis, and had followed all university policies, including ethics review, to the letter, Ogden transformed overnight from SFU’s media darling to its pariah. Matters went from bad to worse when the university’s then-Vice-President of Research/Dean of Graduate Studies/Chair of the University Research Ethics Review Committee15 sought to ensure the university would never again have to consider fighting a subpoena. He introduced a change to the ethics policy that required researchers to guarantee confidentiality to the extent permitted by law (Clayman 1997; Lowman and Palys 2000). Because “law” to him meant statute, not common law, a subpoena would have marked its limit.
The vague statements of personal support as expressed by the president of the University, Dr. Stubbs, and the dean of Graduate Studies, Dr. Clayman, sound hollow and timid when compared with the opportunity they had as leaders of the University, to promote the demonstrated value of academic freedom and academic privilege as evidenced in this case. To set aside this opportunity because of fear that if they were to financially support Ogden by paying his legal fees in this context, some people might misapprehend that they were in favour of euthanasia, demonstrates a surprising lack of courage.
Where a graduate student, who is a candidate for a degree, undertakes in good faith research approved by the University under faculty supervision, and where that individual’s academic freedom is challenged or compromised by an external body, the University has an obligation to provide legal advice, representation and/or indemnification to him/her in defending against those actions unless the matter relates to allegations of plagiarism or libel. (Driver 2007)
We know of no other university that has explicitly stated that it will support its graduate students in this way,16 and do not know how many university collective bargaining agreements contain this kind of indemnification for faculty.
Challenges to academic freedom can come both from within and without the university. A university can guarantee to protect academic freedom against actions inside the institution that are within its legal and moral jurisdiction. It can, of course, give no such guarantee about threats to academic freedom that come from outside the university. But a university has the obligation to try to protect this freedom from such external threats and challenges. If universities do not take on this obligation to protect such a basic institutional right, who will?
The extent of PRE’s encouragement in the second draft version of the TCPS (circulated in 2009) is that “Institutions should support their researchers in maintaining promises of confidentiality” (Interagency Advisory Panel on Research Ethics 2009a). This is welcome admonition—assuming it appears in the next incarnation of the TCPS—but when are PRE and the granting councils going to walk this kind of talk?
The Presidents of Canada’s Granting Agencies
The TCPS (CIHR et al. 1998/2005) and terms of reference available online at the Interagency Advisory Panel on Research Ethics (PRE) web site17 outline the governance structure for research ethics in Canada that has been imposed on all research institutions that seek funding from the three federal granting agencies (Interagency Advisory Panel on Research Ethics 2009b). The Presidents of the granting agencies are the official stewards of the TCPS, while PRE advises the Presidents on policy. The Secretariat on Research Ethics (SRE), a government body, provides administrative support. The Presidents possess a power that none of the rest of us enjoy, which is the ability to lobby directly to their respective Ministers on matters of importance to the research enterprise.18
Although the Presidents of the granting agencies have formally recognized researchers’ academic freedom to make decisions of conscience when and if ethics and law part company,19 PRE and SRE have done nothing to protect research-participant confidentiality from the lawful threats that give rise to the “confidentiality to the extent permitted by law” ethos that Bernhard and Young realize gives rise to the “myth of confidentiality.” While the pressures they received from their REB to dilute their research and their decision to abandon their original research design represents a tangible and lamentable example of the price of this myth, their description of its effect in the publish-or-perish culture of contemporary academia is much more troubling. Faced with potentially lengthy delays and possibly insurmountable hurdles when proposing to conduct research on sensitive or controversial topics, who knows how many academics undertake “vanilla research” or avoid controversial areas entirely so that they do not have to compromise their prospects for tenure and promotion. The obvious solution to the mere fact of potential conflicts between ethics and law is to create a legal situation where law and ethics do not have to lead to different conclusions. There would seem to be three ways forward.
The first, Pollyanna’s route, is to argue that because ethics and law have always coincided in Canada, there is no “problem” to address. In support of this view, there is only one instance in the entire history of Canada’s research enterprise where a researcher was actually subpoenaed, brought to court, and threatened with contempt. In that instance the researcher asserted researcher-participant privilege, and won. If there really is no “problem”—we have a quiet social consensus that the research enterprise is a vital social institution, that confidentiality is integral to its proper and diverse functioning, and that researchers should be allowed to go about their business with their research participants undisturbed—then the Presidents of the granting agencies should ensure that the revised TCPS makes no reference to threats that have never materialized, and say no more than that researchers have a duty of confidentiality and should go on promising “complete confidentiality” or “strict confidentiality” just as they did in the years before Russel Ogden was subpoenaed, the TCPS was created, and the controversy over the “limits of confidentiality” began.
Alternatively, if the threat of conflict between law and ethics is as real as the advocates of limited confidentiality believe it to be (Jones and Interagency Panel on Research Ethics 2007), there are two other ways to proceed. One is to create law that supports research ethics. That is exactly what the federal government has done for itself to protect Statistics Canada, the government’s research agency, i.e., it has provided a statute-based privilege through the Statistics Act.20 In the United States this sort of privilege has been extended beyond the Census Bureau to other branches of government, and to non-governmental researchers, including university-based researchers, who do research on “sensitive topics,” via certificates of confidentiality.
The rock-solid protections that certificates of confidentiality provide are not news. They have been discussed in the literature and brought to the attention of PRE, SRE and the granting agency Presidents (Palys and Lowman 2000, 2003, 2006). SSHWC also encouraged PRE/SRE to advise the granting agency Presidents to lobby for the creation of a Canadian equivalent of certificates of confidentiality (Social Sciences and Humanities Research Ethics Special Working Committee 2004, 2008), as has the Canadian Association of University Teachers (CAUT). Whether the specific certificate of confidentiality model is appropriate for Canada is another matter. While they have existed in the United States for 40 years and have never been breached, the additional bureaucracy that has been created to administer them gives considerable discretion to government agencies to control research agendas and may add yet another impediment to the free flow of inquiry that should characterize the academic research enterprise. Another way to protect research confidentiality would be to create the equivalent of a class privilege for research participants under the Canada Evidence Act. Alternatively, universities and the granting agencies might attempt to collaborate with Canada’s Information and Privacy Commissioner to lobby for statutory recognition of the right to research privacy in order to place the onus on any party seeking confidential research information to justify why a confidentiality agreement should be set aside.
The other way to proceed is to maintain the status quo, with the common law Wigmore test being the main way to assert research-participant privilege. Researchers have reacted to the status quo in three ways.
The first is to follow an “ethics-first” perspective. Researchers who adopt this perspective are not prepared to throw research confidentiality into the trash bin of history, and thus feel obliged to pledge confidentiality without any a priori limit—save for mandatory reporting laws—knowing that there is a theoretical possibility that, given the current state of Canadian law, such a pledge could land them in jail. They make their research as impervious as possible to external challenge by designing it with the Wigmore test in mind. They weigh the value of confidentiality against any third party interests that seem plausible sources of a lawful challenge, and have faith that courts will continue to find in favour of protecting research-participant confidentiality. According to this perspective, it would be unethical for research participants to pay the price for any mistakes that researchers might make, in which case ethics-first researchers are prepared to go the distance to ensure that research confidentiality is not violated. The main issue for them is why the Interagency Advisory Panel on Research Ethics and granting agency Presidents would leave Canadian researchers in this position instead of lobbying strenuously for a research shield law or kindred legislation that would protect the research participants who are the foundation of the research enterprise. Perhaps Bernhard and Young’s cautionary tale will encourage them to rethink their position.
We reject the balancing component of the privilege implemented by [the Court of Appeals] and a small number of States. Making the promise of confidentiality contingent upon a trial judge’s later evaluation of the relative importance of the patient’s interest in privacy and the evidentiary need for disclosure would eviscerate the effectiveness of the privilege. As we explained in Upjohn, if the purpose of the privilege is to be served, the participants in the confidential conversation “must be able to predict with some degree of certainty whether particular discussions will be protected. An uncertain privilege, or one which purports to be certain but results in widely varying applications by the courts, is little better than no privilege at all.” 449 U. S., at 393.
As Bernhard and Young have shown, Canadians pay a huge price for this uncertainty in the research realm. It means that some researchers are reluctant to engage in research on sensitive topics and longitudinal analyses because they do not want to expose research participants to harm. They avoid asking important questions and opt for safer cross-sectional designs in order to avoid the huge waste of time that struggling with risk-aversive REBs entails. We do not know the size of the “dark figure” of watered-down research, or the extent to which sensitive and controversial topics are being avoided, or if the granting agency Presidents consider such a loss significant enough of a problem to resolve it by lobbying for some kind of research shield law.
The third possibility—caveat emptor (un)ethics—is the one that is the most troublesome because it is research participants who are made to pay the price. Ironically, this is the perspective that PRE has encouraged, and which the agency Presidents through the TCPS may yet embrace. This position involves accepting perceived limits to confidentiality, and then, in the guise of “informed consent” and maintaining the “autonomy of participants,” telling research participants that confidentiality will be protected “to the extent permitted by law,” and allowing them to make the choice as to whether to participate. To ethics-first researchers, this approach abrogates the researcher’s responsibility to protect research participants. Instead it downloads the responsibility to research participants to know the law in order to protect themselves from researchers. Is it really “ethical” to warn research participants that we reserve the right to abandon confidentiality in the face of legal pressure, especially when we do not inform them what this pressure might entail, or how researchers and universities would react to it? Letting research participants who share their lives with us swing in the wind because we warned them we might “have” to be witnesses for the prosecution is a form of caveat emptor befitting the stereotypical used-car salesman. Research participants who share the intimate details of their lives for the greater public good deserve far better.
The potential for conflicts between ethics and law in relation to research confidences was recognized in the United States in the late 1960s and 1970s when the FBI, grand juries, and some civil litigants sought to use legal pressure to obtain confidential research information. In response to these interventions federal authorities created certificates of confidentiality administered through the National Institutes of Health for research in which confidentiality is essential for the gathering of valid information. They do not preclude the possibility of disclosure if researchers feel ethically obliged to report certain kinds of information, such as threats of harm to third parties, but researchers are required to make prospective participants aware of any such self-imposed limit before the research begins.
In Canada, we face a different set of circumstances. Only one researcher has ever been subpoenaed, brought to court, and threatened with a charge of contempt if he did not supply the names of his research participants. The researcher asserted a claim for a public interest privilege and won. Accordingly, the ‘threat’ that exists to research confidentiality in Canada is theoretical, a creation of government bureaucrats and university administrators who paralyze research to ward off phantoms that are the product of their own risk aversion mentality. There is no documented case of a Canadian court ordering disclosure of confidential research information, and not a single case of a researcher in Canada being thrown in jail for refusing to disclose such information. This has not stopped some REBs and some researchers from treating the threat as overwhelming. Bernhard and Young’s allegory of trying to secure REB approval for a study of persons with precarious legal status living in Canada reveals the cost to the research enterprise that risk aversion entails. Had humans always lived by this philosophy, they never would have sailed out of sight of land. So much for going boldly where no one has gone before.
The remote possibility of a conflict between ethics and law has created the untenable situation where researchers could end up in jail for violating a court order, research topics are being avoided because of the difficulty of getting them approved by excessively risk-aversive REBs who do not fully understand the ethical and legal options and issues, and an unknown number of REBs are encouraging a form of caveat emptor dressed up as ethics to resolve the potential conflict. None of these outcomes is desirable, which is why the next edition of the TCPS should either: (a) Clarify that, aside from mandatory reporting laws that can be accounted for in advance, there is no general threat to research confidentiality in Canada, in which case researchers can pledge strict confidentiality without REB hindrance. In cases where tangible threats to research confidentiality can be foreseen, the TCPS should encourage researchers to design their research with the Wigmore test in mind. Or (b) the granting agency Presidents should explain to Canadian researchers why they have done nothing about this untenable state of affairs and have chosen instead to merely re-paint the woodwork while the research enterprise they are supposed to foster quietly rots from within.
This includes people whose legal status is in-process, undocumented or unauthorized, many of whom entered Canada as refugee claimants, through a temporary resident visa, or through a family sponsorship arrangement.
The four requirements that must be met when asserting a case-by-case public interest privilege are: “(1) The communications must originate in a confidence that they will not be disclosed; (2) This element of confidentiality must be essential to the full and satisfactory maintenance of the relation between the parties; (3) The relation must be one which in the opinion of the community ought to be sedulously fostered; and (4) The injury that would inure to the relation by the disclosure of the communications must be greater than the benefit thereby gained for the correct disposal of litigation” (Wigmore 1905: 3185; italics in original).
Bernhard and Young refer to their ERB or “Ethics Review Board.” We use the more common (in Canada) REB or “Research Ethics Board.”
REBs and researchers who would limit their promise of confidentiality nonetheless should seek legal advice on how best to word such limitations. In one US case—Atlantic Sugar v The United States (1980)—where the researchers promised confidentiality “unless required by law,” the trial judge simply said that he “required” the information, precisely the exception to confidentiality that was anticipated in the informed consent statement.
Neither of these researchers held a certificate of confidentiality.
The U.S. certificates of confidentiality website defines as “sensitive” any information that “could have adverse consequences for subjects or damage their financial standing, employability, insurability, or reputation.”
For media accounts of these events, see, for example, http://www.kersplebedeb.com/nzinga/index.html, http://toronto.nooneisillegal.org/taxonomy/term/25 and http://toronto.nooneisillegal.org/node/390
Psychologists should “Familiarize themselves with the laws and regulations of the societies in which they work, especially those that are related to their activities as psychologists, and abide by them. If those laws or regulations seriously conflict with the ethical principles contained herein, psychologists would do whatever they could to uphold the ethical principles. If upholding the ethical principles could result in serious personal consequences (e.g., jail or physical harm), decision for final action would be considered a matter of personal conscience.”
“Confidential information provided by research participants should be treated as such by members of the Academy, even when this information enjoys no legal protection or privilege and legal force is applied.”
In the United States some researchers have based their defence on the First Amendment right to freedom of expression and freedom of the press.
Although “privacy” as such is not mentioned in the Charter, the Supreme Court of Canada has taken section 7 regarding "life, liberty and security of the person" and section 8, which protects the citizen from “unreasonable search or seizure,” to mean that privacy is a fundamental right of all citizens in a democratic society (Palys and Lowman 2000: 45).
These events occurred before the existence of the TCPS, at which time the SFU ethics policy did not preclude university administrators from Chairing the Research Ethics Committee. The VP-Research resigned the Chair in recognition of his administrative conflict of interest when the TCPS was released.
Graduate students are over-represented among the subpoenaed but under-represented in terms of being supported by their university administrations. These include Russel Ogden of Simon Fraser University (Lowman & Palys 2000) who did research regarding assisted suicide in Vancouver’s HIV/AIDS community; Richard Scarce (1994) of Washington State University, who researched animal rights activists; Richard Leo (1995) of the University of California at Berkeley, who did field-based research regarding police interrogations; and Mario Brajuha (Brajuha and Hallowell 1986) of the State University of New York at Albany for his research on the sociology of the American restaurant.
The President of the Canadian Institutes of Health Research (CIHR) reports to Parliament through the Minister of Health; the Presidents of the Social Sciences and Humanities Research Council (SSHRC) and the National Science and Engineering Research Council (NSERC) report to Parliament through the Minister of Industry.
See http://www.statcan.gc.ca/about-apercu/act-loi-eng.htm, and particularly Section 18, entitled, “Information is privileged,” which states in part, “Except for the purposes of a prosecution under this Act, any return made to Statistics Canada pursuant to this Act and any copy of the return in the possession of the respondent is privileged and shall not be used as evidence in any proceedings whatever.”
- Atlantic Sugar, Ltd., v. U.S., 85 Cust. Ct. 128 (1980).Google Scholar
- Bernhard, J. K. (2010). Personal communication, 22 March.Google Scholar
- Blomley, N., & Davis, S. (1998). Russel Ogden Decision Review. Report prepared for the President of Simon Fraser University. Online at http://www.sfu.ca/~palys/ogden.htm
- Brajuha, M., & Hallowell, L. (1986). Legal intrusion and the politics of fieldwork: the impact of the Brajuha case. Urban Life, 14(4), 454–478.Google Scholar
- Canadian Institutes of Health Research, Natural Sciences and Engineering Research Council of Canada, Social Sciences and Humanities Research Council of Canada (1998) [with 2000, 2002 and 2005 amendments]. Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Online at http://pre.ethics.gc.ca/eng/policy-politique/tcps-eptc/
- Caroll, J., & Knerr, C. (1975). Confidentiality and social science research sources and data: The Popkin case. Political Science Quarterly, 6, 268–280.Google Scholar
- Cecil, J. S., & Wetherington, G. T. (Eds.) (1996). Court-ordered disclosure of academic research: a clash of values of science and law. Law and Contemporary Problems (Special issue, Vol.59).Google Scholar
- Clayman, B. (1997). The law of the land. Simon Fraser News, 10(5), 30 October. Online at http://www.sfu.ca/sfunews/sfnews/1997/Oct30/opinion2.html
- Driver, J. C. (2007). Memorandum from dean of graduate studies [of Simon Fraser University] to faculty deans, chairs & directors, graduate program chairs and senate graduate studies committee entitled, “Protection of Academic Freedom – Graduate Students, 5 March.”Google Scholar
- In Re: M. A. Cusumano & D. B. Yoffie [United States of America v Microsoft Corporation]. (1997). US Court of Appeals for the First Circuit, No.98-2133. Online at http://lw.bna.com/lw/19990105/2133.htm
- Interagency Advisory Panel on Research Ethics (2009a). Draft 2nd Edition of the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans (TCPS) (Revised). Online at http://pre.ethics.gc.ca/pdf/eng/Revised%20Draft%202nd%20Ed%20PDFs/Revised%20Draft%202nd%20Edition%20TCPS_EN.pdf
- Interagency Advisory Panel on Research Ethics (2009b). Governance structure for the Tri-Council Policy Statement: Ethical Conduct for Research Involving Humans. Online at http://pre.ethics.gc.ca/eng/panel-group/about-apropos/reference/
- Jackson, M. J., & MacCrimmon, M. (1999). Research confidentiality and academic privilege: a legal opinion. Report commissioned by the SFU Task Force on Research Ethics. Online at http://www.sfu.ca/~palys/JackMacOpinion.pdf
- Jaffee v. Redmond (1996). 518 U.S. 1Google Scholar
- Jones, D. J., and Interagency Panel on Research Ethics (2007). Interface of Law and Ethics in Canadian Research Ethics Standards: An Advisory Opinion on Confidentiality, Its limits & Duties to Others. McGill Health Law Publication, 1 1: 101-105 <http://mhlp.mcgill.ca/pdfs/vol1/jones-pre.pdf> Accessed 14 March 2010.
- M. (A.) v. Ryan (1997) 1 S.C.R. 157Google Scholar
- Palys, T. S., & Lowman, J. (2000). Ethical and legal strategies for protecting confidential research information. Canadian Journal of Law and Society, 15(1), 39–80.Google Scholar
- Palys, T., & Lowman, J. (2003). Déjà Vu: Ogden Subpoenaed Again. [An Open Letter dated 28 January, to: Marc Renaud, President, Social Sciences and Humanities Research Council, Tom Brzustowski, President, National Science and Engineering Research Council, Alan Bernstein, President, Canadian Institutes for Health Research, appearing on the National Council for Ethics in Human Research (NCEHR) listserv.] Online at http://www.sfu.ca/~palys/NCEHR-DejaVu-OpenLetterToAgencyPresidents.pdf
- Slavutych v. Baker et al. (1976) 1S.C.R. 254.Google Scholar
- Social Sciences and Humanities Research Ethics Special Working Committee (2004). Giving Voice to the Spectrum: Report of the Social Sciences and Humanities Research Ethics Special Working Committee. Report prepared for the federal Interagency Advisory Panel on Research Ethics. Online at http://www.sfu.ca/~palys/SSHWC-GivingVoice-2004.pdf
- Social Sciences and Humanities Research Ethics Special Working Committee. (2008). SSHWC Recommendations Regarding Privacy and Confidentiality. Report prepared for the federal Interagency Advisory Panel on Research Ethics. Online at http://www.pre.ethics.gc.ca/policy-politique/initiatives/docs/SSHWC_PC_Policy_Recommendations_-_January_2008_-_EN.pdf
- Wigmore, J. H. (1905). A treatise on the system of evidence in trials at common law, including the statutes and judicial decisions of all jurisdictions of the United States, England, and Canada. Boston: Little, Brown and Company.Google Scholar