|
|
Background Little is known about the nature, extent, and consequences of financial relationships between industry and institutional review board (IRB) members in academic institutions. We surveyed IRB members about such relationships.
Methods We surveyed a random sample of 893 IRB members at 100 academic institutions (response rate, 67.2%). The questionnaire focused on the financial relationships that the members had with industry (e.g., employment, membership on boards, consulting, receipt of royalties, and paid speaking).
Results We found that 36% of IRB members had had at least one relationship with industry in the past year. Of the respondents, 85.5% said they never thought that the relationships that another IRB member had with industry affected his or her IRB-related decisions in an inappropriate way, 11.9% said they thought this occurred rarely, 2.4% thought it occurred sometimes, and 0.2% thought it occurred often. Seventy-eight respondents (15.1%) reported that at least one protocol came before their IRB during the previous year that was sponsored either by a company with which they had a relationship or by a competitor of that company, both of which could be considered conflicts of interest. Of these 78 members (62 voting members and 16 nonvoting members), 57.7% reported that they always disclosed the relationship to an IRB official, 7.7% said they sometimes did, 11.5% said they rarely did, and 23.1% said they never did. Of the 62 voting members who reported conflicts, 64.5% reported that they never voted on the protocol, 4.8% said they rarely did, 11.3% said they sometimes did, and 19.4% said they always did. Most respondents reported that the views of IRB members who had experience working with industry were beneficial in reviewing industry-sponsored protocols.
Conclusions Relationships between IRB members and industry are common, and members sometimes participate in decisions about protocols sponsored by companies with which they have a financial relationship. Current regulations and policies should be examined to be sure that there is an appropriate way to handle conflicts of interest stemming from relationships with industry.
The focus of concern has been on academic investigators and their relationships with industry.10,11 However, the relationships between academic institutions and industry are also being scrutinized.12 Of additional interest is the extent to which relationships with industry may affect members of institutional review boards (IRBs). Because IRBs are responsible for overseeing and protecting the safety and well-being of research participants, they should be free of undue influence by financial interests or by the appearance of such interests.12,13
Federal regulations anticipated the potential for conflicts of interest among IRB members and require that ". . . no IRB may have a member participate in the IRB's initial or continuing review of any project in which the member has a conflicting interest, except to provide information requested by the IRB."14 Close and remunerative associations between IRB members and industry raise questions about conflicting interests, yet little is known about the extent of these relationships or whether they influence the attitudes and behaviors of IRB members.15
Methods
Sample
We developed our sample in two steps. First, as in our previous research, we identified the 100 medical schools and the 15 independent hospitals that received the most funding from the National Institutes of Health (NIH) in 2003.16 Second, at each institution we sought lists of IRB members from the Office for Human Research Protections (OHRP) and from the institutions themselves. Of the 115 institutions, 15 were excluded because a list of IRB members was not received within a year (5 institutions), the list of names provided was more than 2 years out of date (8 institutions), or the list of names was for an IRB that did not review medical research (2 institutions).
This process resulted in a list of 3946 IRB members at 100 institutions. From this list we drew a simple random sample of 893 IRB members. The resulting sample was self-weighting and therefore reflected the underlying distribution of the population of IRB members across institutions.
Development of the Survey Instrument and Testing
The questionnaire was informed by four focus groups of IRB members. Seven cognitive interviews were used to test for uniformity in comprehension and for the comfort of the respondent with the response tasks.17 In addition, we mailed a pretest of the questionnaire to 52 IRB members. When completed, the final questionnaire was eight pages long and took approximately 20 minutes to complete (see the Supplementary Appendix, available with the full text of this article at www.nejm.org).
Administration of the Survey
The survey was administered by mail between April and June 2005. Subjects were sent the questionnaire, an introductory letter, a fact sheet, a confidentiality statement, a postage-paid return envelope, and a postcard. They were asked to return the survey (which contained no identifying information) separately from the postcard (which included the subject's name), to indicate that they had completed the questionnaire. This process simultaneously preserved the anonymity of the responses and identified those subjects who did not respond.
Approximately 2 to 3 weeks after the initial mailing, subjects who had not responded were contacted by telephone and encouraged to complete the questionnaire. A second set of survey materials was mailed to subjects who indicated that they would complete the questionnaire. Additional telephone contact was made with nonrespondents.
Relationships with Industry
We asked IRB members whether, in the past year, they had held any of the following positions with a company: officer or paid employee, member of a board of directors, paid consultant, member of a scientific advisory board, recipient of royalties (e.g., patent licenses or milestone payments), or member of a speakers bureau. Similarly, we asked whether, in the past year, they had received from an industry source any funding for university or hospital research, support for students or postdoctoral fellows, equity in exchange for professional services or intellectual property, or compensation for participation in meetings, conferences, or other activities.
To measure the potential positive consequences of the relationships between IRB members and industry, we asked members to estimate the benefits of the relationships with industry in terms of conveying the scientific superiority of the drug, device, or other product in comparison with others currently on the market, as well as providing an understanding of industry standards for data management, the effect of industry regulations on study design, the use by industry of multiphase or multisite trials, and industry requirements for secrecy. To measure the potential negative consequences of the relationships between members and industry, we asked members how often in the past year of IRB service they thought an IRB member's relationship with a company inappropriately affected his or her IRB-related decisions. We also asked how frequently they thought a protocol was presented in a biased way because of a member's relationship with industry, whether a member did not properly disclose a relationship with industry, and whether they believed that their IRB failed to take appropriate action regarding the relationship of an IRB member with industry.
Disclosure of Relationships with Industry
We asked members whether their IRB had a defined process by which they could disclose their financial and other relationships with industry to the IRB. We also asked them to indicate the actions of their IRB regarding such relationships, including whether "IRB members are specifically asked about or are expected to complete a form documenting their financial and other industry relationships when joining the IRB" or whether "financial and other relationships are openly discussed in IRB meetings or other IRB forums."
Conflicts of Interest
Conflicts of interest occur when primary interests may be superseded by a secondary interest.18 The primary interest of IRBs is to ensure the rights, safety, and welfare of human subjects.15 All other interests are secondary and may include financial gain, professional status, power, or recognition. These secondary interests, although not improper in and of themselves, can conflict with the primary interest. When the relationship of an IRB member with industry conflicts with the protection of human subjects, a conflict of interest exists. However, not all relationships with industry are conflicts of interest. Rather, conflicts of interest are a subgroup of all relationships with industry that may unduly influence or supersede the primary interest of protecting human subjects. In order to understand how conflicts of interest are identified, we asked, "Does your IRB have a written policy that defines when an IRB member's financial or other industry relationship is a conflict of interest?"
To estimate whether respondents had one specific form of conflict of interest with industry, we asked, "During your most recent year of IRB service, about how many protocols came before your IRB that were either sponsored by a company with which you personally had a relationship . . . or sponsored by a company that was in competition with a firm with which you had a personal relationship?" The possible responses were as follows: none, 1 or 2 protocols, 3 to 5 protocols, 6 to 10 protocols, and more than 10 protocols. Respondents who checked any category other than "none" were considered to have had a conflict of interest, even if their interactions with the company and the protocol under consideration were in different therapeutic areas (e.g., a consultation about antiinfective agents and a protocol for a study of antihypertensive agents). Respondents who had a conflict of interest were asked, "For the protocols involving companies with which you had relationships of these kinds, how often did you [do the following]: disclose the relationship to the IRB and an IRB official, leave the room when the protocol was under consideration, partially participate in the discussion by only responding to specific questions, fully participate in the general discussion, or vote on the protocol?"
Statistical Analysis
The data were analyzed with the use of the frequency and chi-square procedures in SPSS software. Because of the random-sample design, weighting after sampling was not necessary. Furthermore, because responses were anonymous, we could not weight the responses for differential nonresponse, nor were we able to perform analyses within individual IRBs or institutions.
Results
Characteristics of the Respondents
Of the 893 IRB members in the survey sample, 39 were ineligible because they either were deceased or were no longer members of the IRB at the institution in which they were sampled. Of the remaining 854 eligible subjects, 574 completed a questionnaire, for a response rate of 67.2%. The characteristics of the respondents are shown in Table 1.
|
Of the respondents, 22.6% had received funding for research from industry in the previous year; 17.4% had received compensation for participation in meetings and conferences; 14.5% had served as a consultant; 14.2% had served as a member of a speakers bureau; 10.0% had been members of a scientific advisory board; 2.0% had served as an officer, an executive, or a paid employee; 1.2% had received royalties or equity; and 1.1% had served on a board of directors (Table 2). Altogether, 36.2% of IRB members had had at least one of these types of relationship with industry.
|
Effects of Relationships with Industry
Most of the IRB members (85.5%) reported that in the past year they never thought that the relationship between another member and industry inappropriately affected his or her IRB-related decisions. However, 11.9% thought this happened rarely, 2.4% thought it happened sometimes, and 0.2% thought it happened often (Table 3). Similar rates were reported for questions about whether the respondent thought that a protocol had been presented in a biased manner because of a member's relationship with industry (sometimes, 2.7%; often, 0.4%) and about lack of proper disclosure by another member of his or her relationship with industry (sometimes, 1.8%; often, 0.2%).
|
Disclosure of Relationships with Industry
Of the respondents, 67.0% said that their IRB had a formal process for members to disclose their relationships with industry. The remaining 33.0% said no such process existed or they did not know of one. When asked what their IRB did regarding the disclosure of relationships with industry, 49.1% of the respondents reported that when joining the IRB, they were required to fill out a form documenting their relationships, and 45.6% said that the relationships were openly discussed by their IRB.
Conflicts of Interest
Almost half of all IRB members (45.8%) reported that their IRB had a formal written definition of what constituted a conflict of interest, 12.1% said no such definition existed, and 42.2% did not know. Table 4 provides the percentages of members who reported that at least one protocol came before their IRB for review in the previous year that was sponsored either by a company with which they had a relationship or by a company in competition with a firm with which they had a relationship. Of all respondents, 15.1% had this type of conflict of interest at least once in the previous year.
|
We performed an analysis of the 6.9% of IRB members who had a conflict of interest who either freely participated in the general discussion of a protocol or voted on a protocol from companies with which they had relationships. We found that these members were significantly more likely to be consultants to industry than were the following: IRB members who had relationships with industry but who did not engage in these activities (56.4% vs. 35.0%, P=0.01), members of speakers bureaus (66.7% vs. 32.7%, P<0.001), recipients of reimbursements for participation in conferences (74.4% vs. 41.8%, P<0.001), and members of a scientific advisory board (41.0% vs. 24.4%, P=0.03).
Discussion
These results raise several important issues regarding relationships between industry and IRB members at universities and independent hospitals. Clearly, such relationships occur frequently and multiple relationships exist, suggesting that IRBs need to attend to the disclosure and management of relationships between industry and their members in much the same way that during the past two decades, medical schools and universities have given attention to relationships between industry and faculty members. In addition, a small percentage of IRB members with conflicts of interest report having participated in discussions or voted on proposals, suggesting possible violations of federal regulations.
More than half the IRB members reported that their IRB did not have a formal process for disclosure of relationships with industry or that they did not know of one (which is the functional equivalent of not having a process). This may be viewed as problematic and suggests that current policies and practices regarding the disclosure of such relationships among IRB members should be examined.
The results were similar for the identification of conflicts of interest. Twelve percent of members said that their IRB had no formal definition of a conflict of interest, but more than 40% did not know whether such a definition existed for their IRB, suggesting that a large number of members are unfamiliar with their IRB's policies. These findings suggest a relative lack of clear policies and guidance, which if true, could lead to variation within and among IRBs and institutions regarding which relationships are considered conflicts of interest and which are not. Such variation may make it difficult for institutions and policymakers to measure the frequency with which conflicts of interest occur and what, if anything, is done about them.
Our survey suggests that the relationships of IRB members with industry have both positive and negative effects on IRBs. Although 15.6% of the respondents reported that they thought at least one protocol may have been presented in a biased way because of a member's relationships with industry, only 2.7% thought this happened "sometimes," and 0.4% thought it happened often. This finding suggests that the primary reviewer system, which IRBs depend on heavily, is functioning largely independently of the influence of the relationships between members and industry.
However, our data regarding the percentage of members who fully participate in discussions or vote on protocols with which they have a conflict of interest, even though these events are rare, suggest that additional scrutiny of IRB policy and practices related to the relationships between members and industry is warranted, since such behavior is in violation of federal regulations. Despite these concerns, it is reassuring that 92.9% of all respondents believe that their IRB always took appropriate action regarding a member's relationships with industry.
At the same time, many IRB members perceived beneficial effects of relationships with industry, including the conveyance of the scientific superiority of an investigational drug or medical device or product in comparison with others currently on the market, and improved understanding of industry standards for data management. The extent to which these benefits can be derived from sources outside the IRB is unknown, but it is likely that banning members from having industry relationships would have a deleterious effect on IRBs.
These perceived benefits may reflect the dependence of IRBs on investigators who have experience in clinical research and may also have relationships with industry. This causes a predicament — namely, that those who are most knowledgeable are also most likely to have relationships with industry that create conflicts of interest. Suggestions for balancing these perceived benefits and risks include encouraging experienced investigators who have no relationships with industry to serve on the IRB and consulting with experts who do not have industry relationships and who are not IRB members.
This research has several limitations. First, subjects may have been unwilling to admit to engaging in behaviors that might be viewed by others as undesirable, such as having relationships with industry or conflicts of interest. Thus, our data may underestimate the actual frequency of such behaviors. It is also possible that the conflicts we identified were de minimis and that our data overestimate their effect. For example, by our definition, a person who served as a consultant to industry once in the previous year and received a modest honorarium (e.g., $1,000) is treated similarly to a person who had a significant ongoing relationship with a commercial entity and who received substantially larger payments as a result of equity relationships.
Second, our results may not apply to IRBs at institutions that are not research intensive. Because most of the respondents in our study are members of IRBs that operate under institutional assurances of compliance to federal agencies and are subject to oversight and enforcement by the OHRP, it is possible that policies and practices are better developed at the institutions included in our sample than at many other institutions. Third, we were unable to determine how certain findings, such as reports of conflicts of interest, were distributed among the institutions we surveyed as a result of the procedures we used to protect the anonymity of the respondents.
Despite these limitations, our study provides both positive and negative findings with regard to relationships between industry and IRB members. Such relationships are not universally bad or good. Rather, they have risks and benefits. The goal from a public policy perspective is to encourage disclosure of these relationships and to identify conflicts of interest by means of clearly identified standards. When problematic relationships are discovered, IRBs must identify the steps that should be taken to eliminate or ameliorate the conflict. Failure to do so could call into question the ability of the IRB system to discharge its duty as the overseer of the safety and protection of human subjects in a fair and unbiased manner.
Supported by a grant (3R01CA095400-02-51) from the National Institutes of Health.
No potential conflict of interest relevant to this article was reported.
We thank Sanford Chodosh, Eric T. Juengst, and Dragana Bolcic-Jankovic.
Source Information
From the Institute for Health Policy, Massachusetts General Hospital (E.G.C., J.S.W., C.V., M.A., J.E.M., G.K.) and the Center for Survey Research, University of Massachusetts at Boston (B.R.C.) — both in Boston.
Address reprint requests to Eric G. Campbell at 50 Staniford St., 9th Fl., Boston, MA 02114, or at ecampbell{at}partners.org.
References
46.107(e) (2005). (Accessed November 13, 2006, at http://www.hhs.gov/ohrp/humansubjects/guidance/45cfr46.htm.)
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
