Friedman Lee, Friedman Michael
University of Illinois at Chicago School of Public Health (L Friedman); and The Social Policy Research Institute, Skokie, Illinois (M Friedman).
J Occup Environ Med. 2016 Mar;58(3):238-47. doi: 10.1097/JOM.0000000000000671.
To date, there is no comprehensive analysis of the relationship between financial conflict of interest (COI) and a potential publication bias in environmental and occupational health studies.
We analyzed original research articles published in 2012 in 17 peer-reviewed journals. Multivariable ordinal logistic regression models were developed to evaluate the relationship between financial COI and the study outcome.
Of the 373 studies included in the analysis, 17.2% had a financial COI associated with organizations involved with the processing, use, or disposal of industrial and commercial products, and studies with this type of COI were more likely to report negative results (Adjusted Odds Ratio = 4.31), as were studies with any COI associated with the military (employment or funding; Adjusted Odds Ratio = 9.15).
Our findings show a clear relationship between direction of reported findings and specific types of financial COI.
迄今为止，尚未对环境与职业健康研究中经济利益冲突（COI）与潜在发表偏倚之间的关系进行全面分析。
我们分析了2012年发表在17种同行评审期刊上的原创研究文章。建立多变量有序逻辑回归模型来评估经济利益冲突与研究结果之间的关系。
在纳入分析的373项研究中，17.2%与涉及工业和商业产品加工、使用或处置的组织存在经济利益冲突，此类利益冲突的研究更有可能报告负面结果（调整优势比=4.31），与军事相关的任何利益冲突（就业或资助）的研究也是如此（调整优势比=9.15）。
我们的研究结果表明，报告结果的方向与特定类型的经济利益冲突之间存在明确的关系。
