Background

The World Health Organization (WHO), like many other organisations around the world, has recognised the need to use more rigorous processes to ensure that health care recommendations are informed by the best available research evidence. This is the fourth of a series of 16 reviews that have been prepared as background for advice from the WHO Advisory Committee on Health Research to WHO on how to achieve this.

A conflict of interest exists when an individual's secondary interests (e.g. personal financial) interfere with or influence judgments regarding the individual's primary interests (e.g. patient welfare, education, research integrity). There is evidence demonstrating the association of financial ties with a breakdown in research integrity. Recent studies and reviews have found that industry funding for research is associated with favourable outcomes for the sponsor [1–5]. Financial ties of investigators with their sponsors (stock ownership, consulting income, etc.) are also associated with favourable research outcomes for the sponsor [5]. This scholarly evidence has been accentuated by lay media stories documenting how financial conflicts of interest have led to biased and even dangerous research (e.g., [6, 7]). Biased research may be intentional or unintentional [8] and may result from damaged objectivity at multiple stages in the research process, including conceptualization of the question, design or conduct of the research, interpretation of the results, and publication (or not) of the research [9, 10]. Regardless of its source, the bias associated with financial and other conflicts of interest may damage both the public's and other researcher's trust in science [11]. The type of conflict most likely to affect the public's trust is a financial conflict where the scientist tends to gain financially from a particular research outcome [11–16], although other competing interests, such as professional advancement, are important. Conflict of interest policies are designed to protect the integrity of research and decision-making processes through disclosure and transparency.

The following report relies heavily on published research related to conflicts of interest in the context of U.S. academic research and U.S. and U.K. biomedical journals because there is little empirical research from other areas.

In this paper we address the following questions:

  • What is the best way to obtain complete and accurate disclosures on financial ties and other competing interests?

  • How to determine when a disclosed financial tie or other competing interest constitutes a conflict of interest?

  • When a conflict of interest is identified, how should the conflict be managed?

  • How could conflict of interest policies be enforced?

Related questions about group composition, consultation and group processes are addressed in another paper in this series [17, 18].

What WHO is doing now

Expert Advisory Panel members are currently required to disclose "all circumstances that could give rise to a potential conflict of interest as a result of their membership on an expert committee." [19]

According to the WHO Declaration of Interests for WHO Experts, a conflict of interest occurs when "the expert or his/her partner (a spouse or other person with whom s/he has a similar close personal relationship), or the administrative unit with which the expert has an employment relationships, has a financial or other interest that could unduly influence the expert's position with respect to the subject matter being considered." An apparent conflict of interest exists when the existence of an interest could result in the expert's objectivity being questioned by others, and a "potential conflict of interest exists with an interest which any reasonable person could be uncertain whether or not should be reported" [20].

The Declaration identifies 5 types of financial and other interests that must be disclosed by all experts, including proprietary interests and patents, shares or bonds in a related commercial entity, employment or consultancies, paid work or research, and grants or fellowships from a commercial entity that has an interest in the subject-matter or work of the committee. [20]

We are not aware of specific WHO documents providing guidance on how to avoid or manage conflicts of interest, and we know of no processes required to ensure that the committees discuss potential conflicts of interest on a case-by-case basis and handle them appropriately. There may be some variability in how departments collect and manage the disclosed information.

In October 2005, the WHO Office of Legal Counsel recommended a set of proposed revisions to the existing conflict of interest procedures that are similar to the recommendations in our report. These revisions would clarify the definition of a conflict of interest, include recommendations for avoiding situations that might result in conflicts of interest, and expand the relationships and affiliations that must be disclosed. The draft guidelines also recommend that a determination be made as to whether the expert's declared interest is insignificant, clearly significant, or potentially significant (para. 26). Suggestions for making this determination include weighing the nature and extent of the interest, the context of the work, and the importance of the expert's contribution (para. 29). The draft Guidelines also suggest three options for managing a conflict: 1) continue with public disclosure of the interest; 2) limit the expert's involvement; or 3) exclude the expert from the meeting or work altogether (para. 30)[21]

The draft guidelines also include a requirement that WHO experts disclose ties to the tobacco industry. This recommendation is in response to a 2000 commissioned report investigating the influence of the tobacco industry on WHO's global tobacco control policies. That report recommended that WHO formally vet prospective experts, consultants, and advisers for possible conflicts of interest related to the tobacco industry and that staff should be barred from having links with the tobacco industry [22]. In 2003, WHO's hypertension guidelines were revised in response to criticism about possible conflicts of interest among expert members [23].

What other organizations are doing

Many organizations recognize the importance of protecting against actual and potential conflicts of interest and require special employees, advisory committee members, and participants to disclose their financial ties to the organization. For example, the US Food and Drug Administration (FDA), the Cochrane Collaboration, the UK National Institute for Health and Clinical Excellence (NICE), and the US National Academies of Science all require advisory committee members and other special participants to disclose financial relationships, including research sponsorship, equity ownership, consulting fees, honoraria, related to the work or topic of the committee. These organizations use a structured disclosure form to solicit information; they employ different standards for determining conflicts of interest and for managing them (see below).

Methods

The methods used to prepare this review are described in the introduction to this series [24] Briefly, the key questions addressed in this paper were vetted amongst the authors and the ACHR Subcommittee on the Use of Research Evidence (SURE). We searched PubMed and the Cochrane Methodology Register [25] for existing systematic reviews and relevant methodological research that address these questions. We did not conduct systematic reviews ourselves. The answers to the questions are our conclusions based on the available evidence, consideration of what WHO and other organisations are doing, and logical arguments.

For this review, we searched PubMed for original qualitative and quantitative research using the terms "conflicts of interest" and "disclosure" and the Cochrane Methodology Register using "conflict of interest". We searched the reference lists of all relevant publications, consulted references from the Council of Medical Editors meeting on disclosure (Sept 2004) and selectively searched for the published policies of several organizations, including the Cochrane Collaboration, NICE, FDA, and National Academies of Science [26–30].

Findings

Our database searches did not yield any systematic reviews of conflict of interest or financial disclosure policies. We found several systematic reviews of literature examining the association between commercial sponsorship and outcomes favorable to the sponsor and the financial ties of investigators and favorable outcomes. We also found a number of empirical studies of particular aspects of industry involvement in science and medicine, case studies and commentaries.

When a conflict of interest is identified, how should the conflict be managed?

The only empirical studies of management decisions in conflicts of interest detail a number of management strategies that are commonly used by university conflict of interest committees [35, 40]. These possible management strategies include: disclosure of the financial tie(s) in publications and public presentations; reducing equity holdings to below 5%; altering consulting agreements to ensure separation between consulting and research work; eliminating the financial tie; appointing oversight committees to review the scientific process and resulting research; and recusal. Government and professional society guidelines recommend that institutions "manage" the financial conflicts of interest of their researchers. Disclosure of financial ties in all publications and presentations is the most frequently used management strategy [40–43].

Scientific journals are also encouraging disclosure as a way of dealing with financial conflicts of interest [44], however, the adequacy of disclosures in scientific articles has been questioned [32, 45]. Even when financial sponsorship is disclosed, few studies describe the role of the sponsor [46]. A study of the relationships between authors of clinical practice guidelines and the pharmaceutical industry found considerable interaction between guideline authors and the pharmaceutical industry [47]; another study found that clinical practice guidelines published in journals almost never published conflict of interest statements along with the guidelines [48].

The FDA in 2002 issued draft guidance amending their disclosure regulations related to Advisory Committee members. The draft guidance now requires that Advisory Committee members granted waivers of their conflicts of interest will have the nature and magnitude of their conflicts of interest disclosed and read into the public record at the start of the committee hearings. [28]. NICE, which is currently reviewing its policies on disclosure and conflicts of interests, recommends that "members should declare all interests at the beginning of all appraisals" and that those declarations of interests be kept in files available for public scrutiny or are recorded in the minutes of the meeting [27]. The Cochrane Collaboration publishes the declarations of interests of its Steering Group members [26].

Although disclosure of financial ties is becoming more accepted within the scientific and policy communities, there are widely varying opinions about the adequacy of disclosure as a management strategy for financial conflicts of interest. Some critics of disclosure feel that it is unnecessary and can taint the reputation of "good" researchers [49, 50]. Others believe that "the key to avoiding conflict of interest is public disclosure" [51]. Studies that disclose industry sponsorship have a systematic bias towards outcomes that favor the sponsor [3, 5, 52, 53], so, therefore, disclosure does not eliminate bias. Although disclosure does not eliminate the association of research funding with outcomes favorable to the sponsor, many argue that it can minimize perceived conflicts of interest.

Additional research is necessary to be able to evaluate different methods for defining conflicts of interest and to determine their relative impact on the decision-making capabilities of the organization.

How should conflict of interest policies be enforced?

There is no empirical evidence evaluating the enforcement of conflict of interest policies. Most organizations and academic institutions convene a standing or ad hoc committee to review financial interest disclosures and, where deemed necessary, recommend management strategies. The US FDA vets all financial disclosure statements through a multi-stage process, beginning with initial review, followed by consultation with the individual and an FDA official, review by the FDA Ethics staff, and final approval by the appointing official. The FDA operates under federal regulations and thus has the power to enforce its decisions [28, 29]. The Cochrane Collaboration directs unclear cases of financial disclosure for reviews to a "Funding Arbiter" who convenes a standing panel of four to give guidance [26].

Further work

There is currently a lack of empirical evidence regarding the most effective ways to determine the existence of conflicts of interest, manage conflicts of interest, and enforce conflict of interest policies. Additional research is necessary to evaluate different methods for defining conflicts of interest and to determine their relative impact on the decision-making capabilities of the organization. WHO's proposed draft recommendations (October 7, 2005) represent a more rigourous evaluation of conflict of interest because it requires more complete disclosure, clearer standards for evaluating conflicts of interest, and explicit management strategies.