Positive association between conflicts of interest and reporting of positive results in randomized clinical trials in dentistry

Abstract

Background. The relationship between industry funding and study results has been explored widely in medicine but not in dentistry. The authors aimed to assess the relationship between conflicts of interest (COIs) and study results. Methods. The authors assessed all randomized clinical trials (RCTs) published between July 2010 and June 2012 in the 10 dental journals with the highest impact factors in dentistry. The authors used three definitions of COI and explored their associations with positive study results. Results. Depending on the definition of COI, the odds ratio for reporting positive results varied between 2.40 (95 percent confidence interval [CI], 1.16-5.13) and 9.19 (95 percent CI, 1.71-170.64). The authors found no association between positive study results and journal of publication or area of practice. Conclusions. RCTs in which authors have some type of COI are more likely to have results that support the intervention being assessed. Practical Implications. When reviewing the results of RCTs, clinicians need to be aware of the association between reporting positive study results and the type of COI disclosure and be even more careful when critically appraising and applying their results.

Background. The relationship between industry funding and study results has been explored widely in medicine but not in dentistry. The authors aimed to assess the relationship between conflicts of interest (COIs) and study results. Methods. The authors assessed all randomized clinical trials (RCTs) published between July 2010 and June 2012 in the 10 dental journals with the highest impact factors in dentistry. The authors used three definitions of COI and explored their associations with positive study results. Results. Depending on the definition of COI, the odds ratio for reporting positive results varied between 2.40 (95 percent confidence interval [CI], 1.16-5.13) and 9.19 (95 percent CI, 1.71-170.64). The authors found no association between positive study results and journal of publication or area of practice. Conclusions. RCTs in which authors have some type of COI are more likely to have results that support the intervention being assessed. Practical Implications. When reviewing the results of RCTs, clinicians need to be aware of the association between reporting positive study results and the type of COI disclosure and be even more careful when critically appraising and applying their results.