Background: The executive committee of the American College of Chest Physicians (ACCP) 9th edition of the Antithrombotic guidelines (AT9) developed a strategy to limit the potential impact of the conflict of interest (COI) on recommendations. This policy allowed conflicted panelist input into preparing, summarizing, and interpreting the evidence, but excluded them from the deliberations that ultimately determined the direction and strength of recommendations on which they had conflicts. In February 2011, a final conference to resolve controversial issues was held in Atlanta. Participants received repeated and forceful instructions not to vote on issues for which they had a primary intellectual or financial conflict of interest.

Objectives: To explore the compliance of participants at the final conference of the AT9 guideline with the policy for managing conflicts of interest

Methods: For each one of the controversial recommendations on which voting occurred during the final AT9 conference, we compared participants’ voting behaviour (the number of votes cast) with voting eligibility (the number of expected votes from unconflicted panel members in attendance, based in declarations of COI and MEDLINE records). We complemented this information with self-reported voting behaviour on the basis of a survey of the panellists who attended the meeting.

Results: We have collected from the ACCP the number of votes cast for each controversial recommendation and surveyed the attendants regarding their voting behaviour. The survey achieved 100% response rate. We are currently establishing the voting eligibility from the COI declarations and MEDLINE records. The full results will be available at the time of the colloquium.

Conclusions: Exploring the compliance with the new COI policy will help understand its feasibility and if there are refinements that should be considered for future guidelines.