Developing an instrument to measure bias in CME.

Introduction: The pharmaceutical industry, by funding over 60% of programs in the United States and Canada, plays a major role in continuing medical education (CME), but there are concerns about bias in such CME programs. Bias is difficult to define, and currently no tool is available to measure it.

Methods: Representatives from industry and academia collaborated to develop a tool to illuminate and measure bias in CME. The tool involved the rating of 14 statements (1 = strongly disagree, 4 = strongly agree) and was used to evaluate 17 live CME events. Cronbach's alpha was used to assess the internal consistency of the scale.

Results: Cronbach's alpha for the total score was 0.82, indicating excellent internal consistency. Incomplete or biased data, data presented in an unbalanced manner, and experience not integrated with evidence-based medicine were found to correlate strongly with the total score. Use of trade names showed a low correlation with the total, and nondeclaration of conflict of interest correlated negatively with the total. These associations suggest that whereas sponsor companies may declare conflicts of interest, such a declaration may not ensure an unbiased presentation.

Discussion: The tool and the data from this study can be used to raise awareness about bias in CME. Policymakers can use this tool to ensure that CME providers meet the standards for education, and CME providers can use the tool for conducting random audits of events they have accredited.