PLoS ONE (Jan 2013)

Conflicts of interest among authors of clinical practice guidelines for glycemic control in type 2 diabetes mellitus.

  • Susan L Norris,
  • Haley K Holmer,
  • Lauren A Ogden,
  • Brittany U Burda,
  • Rongwei Fu

DOI
https://doi.org/10.1371/journal.pone.0075284
Journal volume & issue
Vol. 8, no. 10
p. e75284

Abstract

Read online

BACKGROUND:Conflict of interest (COI) is an important potential source of bias in the development of clinical practice guidelines (CPGs). OBJECTIVES:To examine rates of disclosure of COI, including financial interests in companies that manufacture drugs that are recommended in CPGs on glycemic control in type 2 diabetes mellitus, and to explore the relationship between recommendations for specific drugs in a guideline and author COI. METHODS:We identified a cohort of relevant guidelines from the National Guideline Clearinghouse (NGC) and abstracted COI disclosures from all guideline authors for this observational, cross-sectional study. We determined which hypoglycemic drugs were recommended in each guideline, and explored the relationship between specific disclosures and whether a drug was recommended. RESULTS:Among 13 included guidelines, the percentage of authors with one or more financial disclosures varied from 0 to 94% (mean 44.2%), and was particularly high for two US-based guidelines (91% and 94%). Three guidelines disclosed no author financial COI. The percentage of authors with disclosures of financial interests in manufacturers of recommended drugs was also high (mean 30%). On average, 56% of manufacturers of patented drugs recommended in each guideline had one or more authors with a financial interest in their company. We did not find a significant relationship between financial interests and whether a drug was recommended in our sample; US-based guidelines were more likely to make recommendations for a specific drug compared to non-US based guidelines. DISCUSSION:Authors of this cohort of guidelines have financial interests directly related to the drugs that they are recommending. Although we did not find an association between author COI and drugs recommended in these guidelines and we cannot draw conclusions about the validity of the recommendations, the credibility of many of these guidelines is in doubt.