Part 4: Conflict of Interest Management Before, During, and After the 2010 International Consensus Conference on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations
The International Liaison Committee on Resuscitation (ILCOR, www.ILCOR.org) established a conflict of interest (COI) policy in 2004 to manage actual or potential conflicts of interest in an open and effective manner.1 This article describes the current ILCOR and American Heart Association (AHA) COI policies and their application throughout the 2010 evidence evaluation process. The purpose of the COI policies and procedures is to protect the integrity of the decision-making processes and ILCOR's Consensus on CPR and ECC Science and Treatment Recommendations, as well as the integrity of the 2010 AHA Guidelines for Cardiopulmonary Resuscitation and Emergency Cardiovascular Care. ILCOR and the AHA welcome readers' questions and feedback on this process.
The validity of the ILCOR evidence evaluation process depends on rigorous and objective expert review of published science. Therefore, it is essential that any potential professional conflict of interest be fully disclosed and managed transparently and effectively during the planning and conduct of the evidence evaluation process, especially when concerns are expressed or when issues arise. Many of the world's most qualified scientific experts have professional relationships that could pose a real or perceived conflict of interest. In order to benefit from their knowledge and expertise, it is necessary to manage the way these experts are involved in discussions of topics of potential conflict, and especially to minimize their influence over consensus statements or recommendations in such areas.
ILCOR COI procedures apply to all volunteers and staff working on any aspect of the evidence evaluation process. As host of the 2010 CPR Consensus Conference (C2010), the AHA required all participants to complete an AHA COI disclosure questionnaire and to comply with all current AHA COI policies, including new writing group authorship requirements noted below. Relationships that ended more than 12 months prior to COI submission were considered no longer relevant by current AHA policy.
Summary of COI Procedures
Because the 2010 Evidence Evaluation process has taken place over the past 5 years, COI management for C2010 has been an ongoing, dynamic process; it was modeled on the COI model used successfully in the 2005 CPR Consensus Conference.2–4 COI for the evidence evaluation process was overseen by co-chairs, who were appointed by the ILCOR Executive Committee in 2007. One of the identified co-chairs served that role for the C2005 process and provided continuity. Improvements in the COI process for C2010 (such as earlier identification of commercial relationships) were based on lessons learned from the C2005 process and were formally adopted by the ILCOR Steering Committee in March 2007 (see Table).
From 2005 onward, every participant joining the 2010 evidence evaluation process has completed and submitted an AHA COI disclosure form. Participants are required to update their disclosure form annually or whenever there is a substantive change from prior disclosure. AHA staff verified that a disclosure form was completed by each participant, C2010 attendee, and systematic review (worksheet) author, and they reviewed the disclosures.
ILCOR formed 6 task forces to review the evidence in key resuscitation areas: basic life support; advanced life support; pediatric life support; neonatal life support; acute coronary syndromes; and education, implementation, and teams. The ILCOR Executive Committee chose task force co-chairs to avoid any potential conflict of interest in the topics addressed by the task force. For 1 task force, both co-chairs had commercial conflicts in the same area relevant to the task force. ILCOR then created a separate “overlap” task force to handle the areas of conflict and appointed additional co-chairs who had no conflicts in the area of overlap.
Before assigning systematic review topics, ILCOR task force co-chairs reviewed the completed disclosure forms of proposed worksheet authors and avoided assigning worksheets to authors with potential conflicts. Although the COI process focused primarily on avoiding financial and commercial conflicts, worksheet topics were also assigned to minimize intellectual conflicts such as authorship of key studies relevant to the worksheet topic under review. As a second check, each evidence evaluation worksheet included a space for the author to disclose potential conflicts of interest relevant to that worksheet. Worksheets were not accepted unless the COI section was complete. The COI information submitted for each worksheet was reviewed by the task force co-chairs and by the evidence evaluation expert, and potential conflicts that had not been identified based on the pre–worksheet assignment COI disclosure were identified. When potential conflicts were discovered after initial assignment, the worksheet was reassigned to a nonconflicted author. In 1 case, a worksheet author was discovered to have a potential conflict of interest after completion of the worksheet first draft. A second worksheet author without any potential COI assumed responsibility for completing the worksheet, performed another literature search, and revised the consensus on science and the treatment recommendation sections.
If task force members or assigned worksheet authors received grants from industry within the past 12 months, the task force co-chairs attempted to reassign the author to avoid the conflict. In rare instances it was not feasible to replace a worksheet author who had industry grants. In these situations we investigated the conditions of the grant for details including investigator control of study design, control of data/analysis, lack of publication restrictions, and salary support. We initially planned not to scrutinize grants from nonprofit foundations, but in view of the heavy support some foundations receive from industry, we applied these rules to foundation funding as well.
Completed worksheets were electronically posted for public comment from October 2009. Representatives of industry and industry employees were not permitted to participate in the evidence evaluation process. However, to be able to benefit from the expertise of industry representatives who had valuable input to contribute, ILCOR did permit them to provide “public” comment to the posted worksheets. Those who commented were asked to declare potential COIs, but they were not required to identify themselves. Task Forces evaluated all posted comments on their merits, consistent with open peer review.
At all meetings for the 2010 evidence evaluation process, COI disclosure was required either orally or by simultaneous slide projection (described below), and monitors present in each room ensured compliance with COI procedures. Although potential COI was largely self-identified, all participants were encouraged to request clarification about affiliation or involvement during presentations and discussions. In several instances before and during the C2010 Conference, participants raised issues regarding potential conflicts of interest that were clarified by the COI co-chairs, resulting in modification of disclosures and sometimes a change in role. Anyone with a potential conflict of interest was permitted to participate in debate but not in decisions or votes regarding wording of consensus on science and treatment recommendations.
In the 5 years leading up to the conference, COI co-chairs fielded many questions and issues regarding the nature of diverse relationships and whether they posed a potential conflict of interest. The co-chairs maintained written documentation of the issues raised and their resolution, for transparency and to aid consistency. Many participants asked co-chairs for feedback about their personal disclosures. In at least 1 instance, a participant ended his relationship with commercial entities to eliminate potential conflict, in order to participate fully in the process.
Management of Potential COI During the C2010 Conference
All participants who attended the C2010 conference were required to submit a COI disclosure within 12 months of the conference; late registrants and those with outdated or missing disclosures were required to complete the COI disclosure when they arrived on-site. Throughout the conference, participants were afforded the opportunity to revise or update their COI disclosure, as appropriate. Five computers dedicated to COI use allowed rapid completion or updating of the online COI form; COI information was then immediately entered into the COI database.
All C2010 participants were asked to bring a laptop computer and were given complimentary wireless Internet access throughout the conference. All conference materials including COI disclosures were available electronically. Participants were directed to access the C2010 COI website listing each attendee's name and institution and the basic details of any declared professional relationship that could pose a potential conflict of interest, categorized by type of relationship (grant, consultant, speakers bureau, stock ownership, intellectual property, other; see Disclosures at the end of each part of the 2010 ILCOR Consensus for CPR and ECC Science With Treatment Recommendations). In addition, this information was available in a printed copy present at the moderator's table and the registration desk. As in C2005, each participant was assigned a participant number. Based on pilot testing at the 2009 Orlando ILCOR meeting, we assigned numbers in different ranges based on the participant's COI declaration, so all participants could easily determine whether the participant had declared commercial relationships. Participants were assigned numbers 1 to 399 if they declared no relationships and 400 to 599 if they declared relationships. Participants who had not completed COI disclosure in advance were assigned a temporary number (600–700), were required to complete the online COI form during on-site C2010 registration, and were required to make full oral disclosure prior to each comment until their COI slide was updated, usually within 4 hours. COI information for each participant was listed numerically on the COI website, and the website was updated daily with new or revised COI disclosure information.
As in C2005, continuous COI disclosure for all speakers (scheduled or unscheduled) was provided without interruption or delay in the proceedings throughout C2010. All participants who commented during the conference, whether they were a moderator, presenter, panelist, or conference attendee, were required to state their name and participant number prior to speaking. A slide listing the speaker's institution and COI disclosure information was projected on a designated screen for the duration of the speaker's presentation, question, response, or comment. This provided conference participants with immediate and continuous information about any relationships the speaker had that could pose a COI issue.
Questions from the audience, comments, and statements from all moderated sessions were audio recorded for future reference. Speakers' statement of participant number at the beginning of comments made the task of identifying recorded speakers easier and made it possible to assess the impact of potential conflicts of interest from the recordings.
A COI monitor was assigned to each session to ensure that policies were followed and to manage questions or issues arising during the session. The monitors' COI Attestation Forms were reviewed and retained as part of the AHA ECC COI documentation file. Conference participants were repeatedly reminded to raise COI issues with COI monitors, moderators, or COI co-chairs. COI co-chair mobile telephone numbers were printed at the bottom of all COI Attestation Forms. A confidential COI telephone “hotline” was announced and listed in the conference program to enable participants to report issues anonymously if they did not wish to make their comments in person. During the conference any new COI problems or questions that could not be resolved by the session moderators were referred to the COI co-chairs for rapid resolution. If a problem became apparent during a session, moderators were instructed to stop discussion if they could not resolve the problem immediately to enable the COI co-chairs time to resolve the issue. When the problem was resolved, the panel was then permitted to resume the earlier presentation and discussion. An Ad Hoc COI Committee composed of the 2010 Consensus Conference coordinator (William H. Montgomery), conference co-chairs (Robert W. Hickey and David Zideman), and COI co-chairs (John E. Billi and Michael Shuster) was available to deal with any issue the COI co-chairs deemed sufficiently challenging.
Potential COI Management Issues Arising During the C2010 Conference
COI co-chairs investigated and recommended resolution for a number of issues that arose during the conference. None of the COI issues required interruption of discussion or convening of the Ad Hoc COI Committee (each of these events did occur once in C2005).3,4 No anonymous calls were received on the COI hotline. Six participants voluntarily revised their COI disclosure forms once they observed the comprehensive level of disclosure of their peers or were reminded of relationships that might pose a potential conflict. In 1 instance, 2 participants provided information regarding a potentially conflicting, undisclosed relationship of another participant. A COI co-chair investigated each issue, and in both cases, the participant's disclosure form, online list, and slide were updated.
Minor difficulties with the planned COI procedures included difficulty experienced by the projectionist in hearing or understanding the speaker's number, the speaker forgetting to provide the number at the beginning of comments (this occurred less frequently as the conference progressed), occasional slow slide projection, the COI slide not yet being updated, or the speaker failing to highlight an intellectual or commercial relationship directly related to the issue under discussion. All of these minor difficulties occurred infrequently and were handled locally by the moderators or the COI monitor assigned to the session; no breaches of COI policy were observed or reported. One unintended benefit of the simultaneous projection of the COI slide was that the audience always knew who was speaking, something that can be difficult to discern in a large meeting room with floor microphones and a wide variety of native languages among participants.
Management of Potential COI Issues After the C2010 Conference
Writing groups were formed to write the 2010 Consensus on CPR and ECC Science and Treatment Recommendations. All writing group co-chairs and writing group members' COI forms were reviewed by AHA officers to ensure compliance with current AHA writing group COI policies. The chair of each writing group and more than 50% of the members of each writing group were required to be free of any relevant relationship with industry and of any significant intellectual bias or competing organizational relationship. Any chair of a writing group with any relevant COI was asked to step down as a chair.
Overall, the C2010 COI management process worked well. Nonetheless, ILCOR and the AHA continue to revise policies and procedures to enhance transparency and ensure scientific integrity throughout the evaluation of the evidence and development of consensus statements and treatment recommendations. Readers are welcome to provide feedback on any aspect of the ILCOR or AHA COI policies and implementation. Please contact the lead author.
The American Heart Association requests that this document be cited as follows: Billi JE, Shuster M, Bossaert L, de Caen AR, Deakin CD, Eigel B, Hazinski MF, Hickey RW, Jacobs I, Kleinman ME, Koster RW, Mancini ME, Montgomery WH, Morley PT, Morrison LJ, Munoz H, Nadkarni VM, Nolan JP, O'Connor RE, Perlman JM, Richmond S, Sayre MR, Soar J, Wyllie J, Zideman D; for the International Liaison Committee on Resuscitation and the American Heart Association. Part 4: conflict of interest management before, during, and after the 2010 International Consensus Conference on Cardiopulmonary Resuscitation and Emergency Cardiovascular Care Science With Treatment Recommendations. Circulation. 2010;122(suppl 2):S291–S297.
↵* Co-chairs and equal first co-authors.
© 2010 American Heart Association, Inc., European Resuscitation Council, and International Liaison Committee on Resuscitation.
1. ILCOR should be especially sensitive to potential COI issues among persons selected for a leadership role with oversight or responsibility for science review of a particular area or topic. These situations must be reviewed on a case-by-case basis. ILCOR may decide that the risk to the integrity of the process from the individual's relationship is not significant and that the individual still represents the best choice for ILCOR, taking into account the risks and benefits. If a person already playing a leadership role develops or is discovered to have a conflict that poses a significant risk to the integrity or credibility of the ILCOR process, another qualified individual without such a potential conflict should replace the first person. Such a substitution shall not imply any impropriety on the part of anyone but rather indicate a preventive step taken to avoid any perceived or real conflict from endangering the integrity of the process. A position of leadership can include the chair or vice chair of any committee, subcommittee, task force, working group, ad hoc group assigned to work on an issue, evidence panel, or evidence collection process. The fact that such perceived conflicts are usually without any improper intent does not protect the individual, ILCOR, or its work from the potential consequences of inadequate management of such a situation.
2. In addition to financial relationships, other bases of potential conflicts of interest must be considered, such as in-kind support, intellectual collaboration or intellectual investment in one's own ideas, or a long-term research agenda in which an investigator has invested substantial time. Although these situations will be considered on an ad hoc basis, financial relationships are more likely to adversely affect the credibility of ILCOR and the integrity of its process and products.
- © 2010 American Heart Association, Inc., European Resuscitation Council, and International Liaison Committee on Resuscitation.
ILCOR Conflict of Interest Policy and Procedures. Available at: http://ecccanadaheart.com/presenter.jhtml?identifier=3033464. Accessed September 13, 2010.
- Billi JE,
- Eigel B,
- Montgomery WH,
- Nadkarni VM,
- Hazinski MF
- Billi JE,
- Zideman DA,
- Eigel B,
- Nolan JP,
- Montgomery WH,
- Nadkarni VM
- Billi JE,
- Zideman DA,
- Eigel B,
- Nolan JP,
- Montgomery WH,
- Nadkarni VM