3
Conflicts of Interest, Bias, and Ethics

Conflict of interest has been the subject of discussion and concern in many areas of science. With the increased reliance on funding from sources outside of government, there has been renewed interest in the debate. Avoiding conflict of interest in order to achieve sound, unbiased science is in the vested interest of the scientific community as well as the general public. However, creating a system in which scientific decisions are made in an ethical manner while free of conflict and individual bias is a challenge. The workshop therefore explored how to identify, manage, and, in some situations, eliminate the conflict.

GENERAL OBSERVATIONS REGARDING CONFLICTS OF INTEREST

Thomas H. Murray, Ph.D., President, The Hastings Center


Conflicts of interest are ubiquitous. In and of themselves, conflicts of interest are not necessarily a failing, but rather a description of a set of circumstances that can take many forms and be manifested in many settings. They are usually based on situations in which there is reliance on the judgment of an outside party with some very specific professional expertise. Exercise of that judgment should promote the interest of a party, who may be an individual (e.g., the patient in a doctor-patient relationship) or, on a broader level, an institution. The judgment that the professional owes the client, patient, or institution can include specific recommendations, but often it includes just interpretation of information for the receiver (the party to whom loyalty is owed), who does not have the necessary expertise to understand the information without assistance.

Often there are questions as to who is the primary party to whom loyalty is owed. For scientists on a panel of the Food and Drug Administration, for example, it isn’t immediately clear to whom they owe their primary loyalty; this is more complex than the doctor-patient example. Even at the largest relationship



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 35
3 Conflicts of Interest, Bias, and Ethics Conflict of interest has been the subject of discussion and concern in many areas of science. With the increased reliance on funding from sources outside of government, there has been renewed interest in the debate. Avoiding conflict of interest in order to achieve sound, unbiased science is in the vested interest of the scientific community as well as the general public. However, creating a system in which scientific decisions are made in an ethical manner while free of conflict and individual bias is a challenge. The workshop therefore explored how to identify, manage, and, in some situations, eliminate the conflict. GENERAL OBSERVATIONS REGARDING CONFLICTS OF INTEREST thomas H. Murray, ph.D., president, the Hastings Center Conflicts of interest are ubiquitous. In and of themselves, conflicts of inter- est are not necessarily a failing, but rather a description of a set of circumstances that can take many forms and be manifested in many settings. They are usually based on situations in which there is reliance on the judgment of an outside party with some very specific professional expertise. Exercise of that judgment should promote the interest of a party, who may be an individual (e.g., the patient in a doctor-patient relationship) or, on a broader level, an institution. The judgment that the professional owes the client, patient, or institution can include specific recommendations, but often it includes just interpretation of information for the receiver (the party to whom loyalty is owed), who does not have the necessary expertise to understand the information without assistance. Often there are questions as to who is the primary party to whom loyalty is owed. For scientists on a panel of the Food and Drug Administration, for example, it isn’t immediately clear to whom they owe their primary loyalty; this is more complex than the doctor-patient example. Even at the largest relationship 

OCR for page 35
 EnVIRonMEntaL HEaLtH SCIEnCES DECISIon MaKInG level—a scientist conducting research—the primary loyalty isn’t always clear. Is it to the scientific community, the public, or policy makers? While this is not an easy question to answer, there is a shared understanding that scientists have profound obligations and therefore need to avoid being improperly influenced by extraneous considerations—most visibly, financial ones. Types of Conflict A number of factors can make one type of conflict more significant than others: • Intensity of the conflict • Expectations and transparency • Conflicts of commitment • Power and status • Financial aspects The intensity of the conflict—the profound “push or pull” on a person pro- viding the advice—is important. For example, a large sum of money can be a more worrisome source of conflict than a small token. However, there are certain types of conflict that people are known to have and society is willing to accept if these conflicts are both expected and transparent. One such example is the doctor- patient relationship, in which the patient knows a doctor is being paid to give advice and provide an opinion. Because this relationship is open and transparent, it causes less concern about the intensity of the conflict. In other situations, in which a conflict that is expected is not disclosed, this “hidden” nature may take on a greater level of importance and may be seen as an attempt to keep an issue from the public. While society tends to focus on financial conflicts of interest because they are the most visible and measurable, other types of conflicts can be quite power- ful. Conflicts of commitment—being employed by one organization yet having other interests competing for time and attention away from the primary point of employment—have been cited as a concern at the National Institutes of Health and other government agencies. The question was how committed an intramural investigator was to his government job when he was making $150,000 a year above his salary from consulting and other commitments. Another form of con- flict, at times overlooked, involves status and power as prime motivators. Most of the time and effort put into creating rules and guidelines focuses on a larger set of concerns about the influence of power and money on public policy decisions. It is widely agreed that these decisions should be based on an unbiased evaluation of the weight of scientific evidence. Examples range from decisions about the regulation of drugs to the regulation of workplace or environmental toxins.

OCR for page 35
 ConfLICtS of IntERESt, BIaS, anD EtHICS In conclusion, to say that someone has a conflict of interest is not a moral criticism, but rather a description of a set of circumstances. That person has a primary interest that he or she needs to fulfill, although other interests may push or pull the person in different directions. A moral failure would be if the person neglected their primary interest and allowed these other interests to rule. Values of Science and Impact on Policy The values that govern scientific inquiry are not necessarily in line with the values that govern policy making, because they have different purposes. There is an inherent conservatism in the way science is taught and practiced. Scien- tists are taught not to go beyond the data. This strict emphasis on the data can often be used to inform policy decisions. However, this conservatism can also be used by those with an interest in denying that a certain relationship exits. Policy should be informed by the best science possible, yet in some situations, consensus cannot be reached. The absence of consensus does not mean there is insufficient evidence to warrant caution and regulation. For example, opponents of regulation called into question just how much data was needed to define the exact risks of asbestos exposure. In the rule- making process surrounding asbestos, both sides agreed that the science was not definitive in answering what exposure levels should be permissible, which led to differing views to solve the problem. Opponents of strict regulation, the asbestos industry and its allies, argued that in the absence of conclusive scientific evidence, exposures at the level in question were not harmful and the limit should be set significantly higher. Those concerned with the health of workers argued from the identical evidence that the value of protecting health, when the threat was plausible if not definitive, meant that permissible exposure levels should be set very low. This was an argument over what could be called “allocating the burden of uncertainty.” Such arguments are typical when public policy confronts scientific data over causal relationships, whether they involve environmental or occupational hazards or global warming. The misuse of scientific information in making, interpreting, and enforcing public policy is illustrative of the concern regarding the overall issue of conflict of interest. For example, uncertainty about the data can become magnified and misrepresented. This is true even though the nature of the scientific endeavor may not lead to full agreement and acceptance by all parties. Another example of the broad concern is the impact of money and power on the generation of scientific information and, in some cases, the selective generation of scientific informa- tion. An example is the tobacco industry–sponsored research on the relationship between viruses and cancer, presumably undertaken to deflect attention away from cigarettes as a cause of cancer.

OCR for page 35
8 EnVIRonMEntaL HEaLtH SCIEnCES DECISIon MaKInG Managing Conflict to Create the Best System Several aspects are key in detecting, managing, and addressing all types of conflict of interest, including those financial in nature: • Clarity • Simplicity • Fairness • Predictability When there is clarity about what constitutes a conflict because criteria and definitions have been set around types of relationships, there is greater acceptance of and adherence to the policies set forth by an organization. Confusion begets indifference and inattention. Next, a well-working system simplifies the issues that must be considered. For example, when clinical investigators are work- ing under two types of federal regulations, they may be confused about which regulations to follow. In order to avoid such situations, it is best to pare down differences and strive for one set of overarching rules and regulations. Otherwise, researchers will make judgments to follow one system and not the other. Further- more, any system needs to apply principles fairly to all parties involved in order to promote faith in the system and lessen the chances of its being undermined. Finally, the system should be made predictable by constructing a framework to determine the positive and negative repercussions of certain actions. Essentially, someone who follows the criteria and guidelines set forth is ultimately provided with protection against accusations of conflict of interest. This course also reflects the understanding that if these guidelines are not followed, known consequences will ensue. Most people in the scientific community consider disclosure to be the fun- damental element in addressing conflict of interest. However, it is crucial for disclosure to be done correctly and set up in a way that allows for the right type of information to be disclosed. While not a panacea, a disclosure system should be created to ask relevant questions while incorporating insights into typical human behavior. Such a system would not rely on individual judgment or be built around reporting every single relationship, but rather designed to elicit significant rela- tionships that may require further scrutiny. This would raise the signal to noise ratio and thus make it possible to identify potentially significant conflicts. MANAGING CONFLICTS OF INTEREST: THE INTERNATIONAL AGENCY FOR RESEARCH ON CANCER Vincent Cogliano, ph.D., Head of the Monographs programme, International agency for Research on Cancer The International Agency for Research on Cancer’s (IARC’s) mission “is the identification of causes of cancer, so that preventive measures may be adopted

OCR for page 35
 ConfLICtS of IntERESt, BIaS, anD EtHICS against them” (http://www.iarc.fr/en/About-IARC). The agency’s work has four main objectives: monitoring global cancer occurrence, identifying the causes of cancer, elucidating mechanisms of carcinogenesis, and developing scientific strategies for cancer control. These objectives are achieved through a number of programs such as the IARC Monographs. The objective of the program “is to prepare, with the help of international Working Groups of experts, and to publish in the form of Monographs, critical reviews and evaluations of evidence on the carcinogenicity of a wide range of human exposures. The Monographs represent the first step in carcinogen risk assessment, which involves examination of all relevant information in order to assess the strength of the available evidence that an agent could alter the age-specific incidence of cancer in humans” (IARC, 2000). These monographs are used worldwide as scientific support in decision making. For this reason, IARC has created a system to restore confidence in the scientific process in the development of the monographs by identifying and avoid- ing relationships that introduce conflict of interest. Identifying Conflict of Interest The many challenges in identifying conflicting interests include variations in individual interpretations of what is considered to be conflict. These can include not recognizing the potential for conflict or assuming impartiality because the monetary outcome is the same no matter the decision. Having criteria to deter- mine the nature of conflict is therefore necessary in any scientific organiza- tion. One such set of criteria used by IARC is the World Health Organization’s (WHO’s) Declaration of Interests, which outlines three main categories: • Employment and consulting. A review of whether in the previous four years the expert was employed by an interested party or consulted on a matter before a court or government agency. • Research support. An account of support for the expert’s own research and that of his or her unit, including supplies and equipment. • Financial interests. Stock, other securities, business interests, and patents or other intellectual property. To identify these interests, IARC asks specific, objective questions. A ques- tion such as “Have you served as an expert witness in a court case involving the interested party?” leaves no room for ambiguity. These questions were also designed to highlight activities that are suggestive of an ongoing relationship with an interested party and not just a one-time offering of scientific information on a particular issue. As an extra cautionary step, IARC has also instituted a policy to verify the absence of conflicting interest by conducting a review of recent papers for acknowledgment of research support as well as simple Internet searches. As a final step to ensure proper identification of conflict, there is a requirement for experts to update their declarations of interests at the start of each meeting. Not

OCR for page 35
0 EnVIRonMEntaL HEaLtH SCIEnCES DECISIon MaKInG only does this allow for clarification, but it also provides an opportunity to iden- tify newly acquired or solicited conflicting interests. Addressing Conflicts of Interest Different organizations have different techniques for addressing conflicts of interest. Some organizations ignore the issue altogether. Other organizations are addressing conflict by opting for a simple disclosure policy. A final group of organizations has chosen to build on a disclosure policy by limiting the number of experts who have conflicts. In other words, the experts with conflicts are assumed to be diluted by the experts who have no conflicts. A better approach is to actively balance experts who have a conflicting interest with experts with an opposing interest. IARC, on the other hand, strives to avoid conflicts of interest altogether. A 2003 Lancet editorial criticized the IARC conflict identification process by saying “it only needs the perception, let alone the reality, of financial conflict and commercial pressures to destroy the credibility of important organizations such as IARC and its parent, WHO” (Baines, 2003). This criticism was taken very seriously and led to changes in how IARC addresses conflict. Depending on the type of conflict identified, a threshold for concern was developed with an accompanying period of relevance (see Table 3-1). One such example would be employment by an interested party, where a higher threshold for avoiding conflict of interest was established. Such activities as sponsored travel and consulting on a particular process or new product would also have a corresponding threshold and period of relevance. To further manage the conflicts of interest process, IaRC Monographs for the past 2 years has provided for independent neutral-party verification of conflicting interests of its experts through the Lancet oncology. For example, after IARC identifies its experts and screens them for their conflicting interests, the conflicts of interest form used by the Lancet oncology is distributed to the experts at the start of their meetings. Six to eight weeks after a meeting, a summary is published in this journal, along with the editor’s account of any conflicts of interests of those at the meeting, which provides another layer in addressing conflicts and making them transparent. The Question of Best Versus Impartial Experts With the reduction in federal funding and the reality that researchers turn to private sources of funding, another conflict emerges: the controversy surrounding the issue of “best experts” or “impartial experts.” In other words, what does an agency do when an expert with relevant knowledge and experience also has a real or apparent conflict of interest? This issue raises a dilemma involving two valid, yet different, ideals. On one hand, the selection of experts with real or apparent

OCR for page 35
 ConfLICtS of IntERESt, BIaS, anD EtHICS TABLE 3-1 IARC Guidelines for Addressing Conflict of Interest Type of Conflicting Interest Threshold for Concern Period of Relevance EMPLOYMENT by an All 1 year with no collaborations interested party CONSULTING on All 3 years matters before a court or government agency CONSULTING on new 2% of professional time or 1 year products or process compensation changes SPONSORED TRAVEL 2% of professional time or 1 year or sponsored compensation presentations at scientific meetings RESEARCH SUPPORT All 1 year after last publication for the expert’s own research RESEARCH SUPPORT 5% of research budget 1 year after last publication for the expert’s research unit STOCK and other $10,000 Current interests only financial instruments PATENTS and other All Current interests only intellectual property SOURCE: Cogliano, unpublished. conflicts of interest can erode confidence in the integrity and impartiality of the results. On the other hand, the omission of prominent experts can create the perception of reduced scientific quality. IARC has strived to achieve both ideals through a category of meeting participant known as the “invited specialist.” Invited specialists are experts with critical knowledge and experience who are recused from certain activities because of a conflicting interest. They are available at IARC meetings to contribute their unique knowledge and experi- ence but are not serving in key decision-making or influencing positions. This approach ensures that IARC meetings include the best qualified experts, but the meeting positions are developed by experts with no conflicting interest. The use of the invited specialist was also reviewed by an advisory group, which recom- mended continued use of these individuals in a limited capacity. Freedom from Interference Maintaining conflict-free and unbiased input throughout the scientific decision-making process is the ultimate goal. To do this requires keeping the

OCR for page 35
 EnVIRonMEntaL HEaLtH SCIEnCES DECISIon MaKInG entire process, in this case the IaRC Monographs process, free from interference. To maintain a transparent process, IARC, over the past few years, has made it a policy to publish the list of participants two months before each meeting. While this does foster a more open environment, many were concerned about interfer- ence with members prior to the meeting. Therefore, along with the names, a statement is published discouraging outside parties from lobbying and contacting meeting participants. Reminders are also provided for the panel members to not accept certain invitations in order to safeguard the integrity of the process. By taking effective measures to identify and avoid conflicts of interest, it is possible to do good science while promoting public confidence in the impartiality and integrity of the results. SESSION DISCUSSION: CONFLICTS OF INTEREST IN THE CURRENT RESEARCH CLIMATE During the discussion, participants noted that managing conflicts of inter- est should be embraced by all parties. Although easily achieved in theory, in practice there are many points that remain problematic and that can contribute to additional forms of conflict in the scientific process. These issues are not only financial in nature, such as public–private partnerships, but also can include cultural differences in science, asymmetry in the decision-making process, and individual scientific bias. While creating a process in which the original meeting on a particular issue is free from conflict, balance is needed throughout the entire process, especially in regulatory aspects, asserted John Balbus. Equal scrutiny should be applied not only to the data reviewed at a public regulatory meeting, but also to the review of the public comments submitted in support of, or against, a particular regula- tory issue. Cogliano agreed that the interests of any person or party submitting a public comment should be clearly disclosed, yet this should not weigh down the decision-making process. Drawing from his work at IARC, he mentioned that the agency has implemented a process in which the public can submit comments within a certain time period and of a certain length, thus allowing for review even if there are limited resources to do so. Some participants noted that scientists have an intellectual bias and a per- sonal financial interest in continuing funding of their research in an area. While there was not a definitive answer to how to balance this bias, Murray asserted that once a particular issue is no longer relevant from a research perspective, then research monies should not be going toward that area. The scientists remain employed by applying their skills to other scientific research. There was recognition that as federal dollars for research become tighter, privatization of research might be necessary in the future. Some participants questioned how this potential bias could be managed as the scientific questions put forth for study are developed by an interested party. This led to the overall

OCR for page 35
 ConfLICtS of IntERESt, BIaS, anD EtHICS question of whether science can have public-private partnerships and not have society dismiss the science because of bias. Murray responded that the same prin- ciples apply to this as to any other level of support: a transparent system in place and a firewall to separate the science from the funding. Murray and Cogliano stated that while the pioneers in this area may experience significant backlash, it is necessary to fund research. They favor a system with multiple levels to protect confidence in the scientific endeavor. In closing, Portier raised the issue of cultural differences in science and how these differences may lead to a level of conflict because of the many differ- ent countries involved in research. Cogliano agreed it is difficult to understand the public–private structure in some countries and suggested that the best way to address the potential for conflicts is for researchers around the world to continually refine the definition of conflicting interest and disclose whatever is pertinent.

OCR for page 35