Disciplinary differences in conflict of interest policy communication, attitudes, and knowledge.

Author:Canary, Heather E.


Financial conflicts of interest (COI) in research exist when an investigator potentially benefits financially from the research in which that individual is involved. One common conflict occurs when an investigator establishes a financial relationship with an industry partner through receipt of industry-sponsored research, gifts, or remuneration from consultation, speaking engagements, etc. Several studies suggest that industry partnerships can cause conscious or subconscious bias on study design, data collection and analysis, and reporting of results, no matter the nature of the relationship (Bekelman, Li, & Gross, 2003; Berger, 2015; Bes-Rastrollo, Schulze, Ruiz-Canela, & Martinez-Gonzalez, 2013; Lexchin, Bero, Djulbegovic, & Clark, 2003; Sah & Fugh-Berman, 2013). Another common conflict of interest occurs when an investigator generates intellectual property through research. Generating intellectual property alone does not cause a financial conflict of interest, but a conflict does arise if an existing industry partner or a university-funded start-up company commercializes the intellectual property. Although remarkably little research exists on conflicts of interest involving startup companies, financial and commitment conflicts are an obvious concern, especially if the researcher also holds an equity interest in the company (Smith, 2011).

The volume of industry-physician relationships and number of start-up companies appears to be increasing. According to Ornstein, Weber, and Nguyen (2013), pharmaceutical and medical device companies made payments to physicians of $4 billion from 2009-2013. That figure was nearly matched ($3.53 billion) between August 2013 and December 2014 (Groeger, Ornstein, Tigas, & Jones, 2015). (1) Similarly, the number of university start-up companies created per year increased from 330 in 2003 to 647 in 2012 (Valdivia, 2013). (2) These data together demonstrate that as industry-physician relationships and university investments in start-ups and licensing of intellectual property increase, so do the number of potential conflicts of interest related to research.

The increasing likelihood of conflicts of interest and lack of transparency of physician's industry relationships have received the attention of government regulators, leading to policy changes. In 2008, Senator Chuck Grassley (R-IA) called for policy revisions concerning industry payments to individual physicians. This led to the Physicians Payments Sunshine Act (2010) and the Department of Health and Human Services adopting new regulations to Public Health Services funding, which included changes to the National Institute for Health (NIH) conflict of interest policy in 2011. Some of the new NIH regulations set standards on investigators' industry relationships and intellectual property. Because the NIH is the largest research funding agency outside of the Department of Defense (White House Office of Science and Technology Policy, 2014), changes to its COI policy have had a major impact on COI policies and procedures of the nation's universities and colleges that seek NIH grant funding. The new NIH COI policy holds universities accountable for ensuring that researchers comply with federal regulations. As a result, many research universities made significant changes to their COI policies and procedures in 2012 based on NIH standards.

Implementation of a new university COI policy (COIP) is challenging. The policy must be robust enough to account for an array of possible conflicts of interest, including the aforementioned examples, as well as accommodating a diverse population of researchers. Universities with closely associated medical schools face an extraordinary challenge if the university desires a unified and comprehensive COIP. Clinical researchers at medical schools may have more industry-related conflicts based on the large amounts of money pharmaceutical and medical device companies give to physicians. Furthermore, attitudes, objectives, motivations, and philosophies may differ between researchers in different fields. A comprehensive COIP must be communicated effectively to such a diverse audience. This may mean different communication strategies customized to unique audiences, or one strategy that transcends the differences between researchers across all fields with conflicts of interest (which may or may not be field-related). Clearly, much remains to be understood about how COIPs are perceived, interpreted, and communicated, and with what consequences. Prior research indicates that three key concerns for COIPs are communication, knowledge, and attitudes, which we discuss below.


Policy making, implementation, and compliance are communicative practices. Policy practice is not linear, but rather implies a circuit of communicative behaviors that might include attending to potential unintended consequences of policy provisions, addressing ambiguities in the policy text, and managing paradoxes inherent to the academic research context (LeGreco, 2012). Canary, Blevins, and Ghorbani (2015) found that the bulk of organizational communication research on policy communication represents an information transfer perspective to policy communication among practitioners. Lee and Garvin (2003) suggest that practitioners ought to move away from such notions of information transfer and toward methods of informational exchange, including reciprocity and valuing open, two-way communication channels.

Policy communication requires a high degree of openness, clearly explaining every phase of the process, and being open about expectations by developing relationships based on mutual trust (Janse & Konijnendijk, 2007). However, conflicting pressures from multiple parties may lead to reservations about the COI process, and result in secrecy rather than transparency. Frankel (1996) warns that secrecy will persist if researchers continue to view it as a necessary strategy for maintaining industry ties. Furthermore, concerns for proprietary gain can subvert traditional processes of openness and sharing among scientists. Industry partnerships and agreements place restrictions on researchers that in many cases lead to unwillingness or inability to share information, including data, findings, and methods. Marchington, Rubery, and Grimshaw (2011) advise that when implementers are faced with difficult undertakings across organizational boundaries, they should be keenly aware of issues of alignment, integration, and consistency. That is, when a governing body has not aligned meaning and goals with members, and members are poorly integrated into systems of knowledge sharing (or if the system does not support integration), problematic inconsistencies may ensue. To address this, Beachy, Berger, & Olson (2014) concluded from their Institute of Medicine roundtable on COI management that goals of communicating to the public about COI are to promote innovation, meet client needs, and increase and maintain public trust.

Clearly, policy communication is not a simple issue, nor is it a process that has been clearly defined in the COIP context. Accordingly, this study investigates multiple aspects of policy communication across disciplines to shed light on such practices as well as on how communication relates to policy knowledge and attitudes.


If researchers are to be expected to comply with policy, they should be informed as to what a conflict is and what will be required if they are deemed to have a potentially conflictual relationship. As researchers are informed, their knowledge of COI processes increases. Gabler (2010) notes, "Social learning [is] the source of policy integration. Learning implies actors' improved understanding of alternative ideas, reflected in changes to frames that underlie identities, interests, policies, and institutions" (p. 83). Lipton, Boyd, & Bero (2004) provide several useful pieces of advice. Implementers should know which gaps in knowledge exist to understand how to properly assess what information needs to be conveyed. Lipton et al. (2004) found that while researchers report a desire for self-regulation and personal integrity, they approach policy only to the extent that it applies to them. They likely underappreciate their responsibility to understand the scope and implications of conflicts of interest.

For the most part, researchers understand why COIPs exist, and understand appropriate procedures to ensure conflicts are managed (Lipton et al., 2004). However, many are not aware of the actual impact offinancial incentives in research, and faculty researchers generally lack understanding of specific details regarding COIPs (Glaser & Bero, 2005; Lipton et al., 2004). This is may be due to COIPs lacking detail regarding what constitutes COI and how it should be managed, or simply not being readily accessible to organizational members (Ancker & Flanagin, 2007). When the policy is available, those who are affected by the policy may develop the requisite knowledge for using it effectively, increasing compliance rates (Zelisko, Baumann, Gamble, Laporte, & Deber, 2014). Gabler (2010) asserts that researchers become strongly integrated with policy mandates when stakeholder goals are congruent and the institution provides opportunities for complex and reciprocal learning. In a recent study, Sacco, Bruton, Hanjal, & Lustgraaf (2015) found that participants who had taken college-level ethics courses demonstrated more sensitivity to the importance of disclosing financial COIs than those without such education. Although participants in that study were students rather than researchers, results indicate the influence of knowledge development opportunities on responses to research ethics issues. Overall, prior research indicates that researchers may have cursory knowledge of COIPs but details that could influence compliance are of ten lacking. Studies have more of ten focused on attitudes...

To continue reading