The Usage of "Trend" Statements in Pediatric Orthopaedic Literature.

Abstract

In medical research, a P-value of <0.05 is generally considered to indicate statistical significance and serve as the foundation from which clinical recommendations are made. However, the P-value is not immune to subjectivity and misinterpretation, introducing variability into study results. This can subsequently negatively impact patient care when used to support clinical recommendations or practice guidelines. This study sought to identify the frequency with which "trend" statements are used to describe nonstatistically significant results within recent pediatric orthopaedic literature. An audit of all original research articles and meta-analyses published in 2024 across 4 major pediatric orthopaedic surgery journals was conducted to identify all instances of a trend statement (defined as a statement where a nonstatistically significant result was described as approaching or trending toward significance). Other data points collected included the location of the statement within the article, reporting of associated P-values, and the context of the statement. Editorials, commentaries, and literary reviews were excluded from consideration. A total of 21 of 451 (4.7%) articles evaluated in this study included at least one trend to describe a statistically nonsignificant value as approaching significance. Out of the 24 trend statements identified, half (n=12) explicitly acknowledged the lack of statistical significance. The average reported P-value was 0.0821, with most P-values falling between 0.05 and 0.10 (78.9%). Trend statements were most often found in the results section (66.7%) and used to describe a single statistical comparison. No articles provided a definition for the use of a trend statement. Pediatric orthopaedic literature demonstrates a low percentage of trend statements to describe statistically nonsignificant results. These statements, while infrequent, may mislead readers and negatively influence clinical decision-making. Efforts to minimize their usage and provide explicit clarifications regarding statistical significance should continue, with leading journals playing a crucial role in setting the standard for future scholarly works.