Disclosure and handling conflicts of interest in the clinical practice guidelines programme in Colombia




Short oral session 3: Tools for guideline development


Wednesday 13 September 2017 - 14:00 to 15:30


All authors in correct order:

Florez ID1, Acosta-Reyes JL2, Sierra JM3, Plata JA3
1 McMaster University & University of Antioquia, Canada & Colombia
2 Universidad del Norte, Colombia
3 University of Antioquia, Colombia
Presenting author and contact person

Presenting author:

Ivan D. Florez

Contact person:

Abstract text
Background: Conflicts of interest (CoI) during clinical practice guidelines (CPG) development may bias the recommendations and threaten the trustworthiness of CPGs. The Colombian Ministry of Health (CMH) started a CPG programme in 2011.

Objective: To describe the process of CoI disclosure and handling in the Colombian CPG programme.

Methods: This was a cross-sectional study. We included all the CPG funded by the CMH. We extracted the information from the guideline document appendices. We described: number of panelists, disclosures, CoI characteristics and handling. A CoI Classification, a modified version of the National Institute for Health and Care Excellence (NICE) classification was used as a framework to carry out the analysis. Descriptive statistics were calculated

Results: In total 50 CPGs were included. On average, each guideline panel had 24.1 members (SD=10.1), 12 (SD=6.74) clinical experts, 6.7(SD=2.6) methodologists. 8.1 members/per CPG (38.5%) had any CoI, while 4.61 clinicians/per CPG (50.3%) had financial CoI. None of the experts were excluded from a CPG, and only one expert was excluded from one question/section of the CPG because of their CoI in 13 CPG of 37 CPG that reported CoI. The most frequent financial CoI were: receiving support for educational purposes (32% of CPG), being a speaker (26% of CPG), consultancies (10%) and owning stock in industry (8%). In total, 30 CPG (60%) and 12 CPG (24%) had more than 50 and 75% of members with any CoI, respectively. In 4 CPGs (8%) CoI were discussed by an independent group/committee, it was discussed by all members in 15 CPGs (30%) and by a subgroup in 17 cases (34%). In 24% of CPG, CoI disclosure and handling were not adequately reported.

Conclusions: CoI are common among members of Colombian CPG panels. In most panels the majority of members had financial CoI. However, less than half of cases with financial CoI excluded members from questions/sections. In general, handling was not appropriately performed or reported. Further research and guidance in how to disclose and handle CoI is urgently needed to reduce the impact and increase the trustworthiness of recommendations.