The use of “trend” statements to describe statistically nonsignificant results in the oncology literature
A partir de la littérature biomédicale publiée dans des revues de cancérologie à facteur d'impact élevé, cette étude évalue l'utilisation d'énoncés indiquant une tendance significative pour décrire des résultats statistiquement non significatifs
The use and interpretation of P values in the biomedical literature is problematic. The importance of a study is often inappropriately defined by the P value. This problem is highlighted by the use of “trend” to refer to statistically nonsignificant results. There is no definition of a trend toward statistical significance and, therefore, describing “almost significant” results as a trend introduces substantial subjectivity and the opportunity for biased reporting language that could mislead a reader (eg, assuming P < .10). To deemphasize P values, some journals prohibit the use of statements about a trend toward significance.1 Instead, presentation and discussion of observed differences and their uncertainty (eg, CIs) are encouraged. The degree of overreliance on P values, and how this overreliance results in unclear reporting practices, is not characterized in the oncology literature, to our knowledge. We examined recent original research articles in oncology journals with high impact factors to evaluate the use of statements about a trend toward significance to describe statistically nonsignificant results.
JAMA Oncology , résumé, 2017