Characteristics of retracted publications related to pain research: a systematic review

Abstract

Retraction is a mechanism for correcting the scientific record and alerts readers when a study contains unreliable or flawed data. Such data may arise from error or research misconduct. Studies examining the landscape of retracted publications provide insight into the extent of unreliable data and its effect on a medical discipline. We aimed to explore the extent and characteristics of retracted publications in pain research. We searched the EMBASE, PubMed, CINAHL, PsycINFO and Retraction Watch databases to 31 December 2022. We included retracted articles that (i) investigated mechanisms of painful conditions, (ii) tested treatments that aimed to reduce pain, or (iii) measured pain as an outcome. Descriptive statistics were used to summarise the included data. We included 389 pain articles published between 1993 and 2022 and retracted between 1996 and 2022. There was a significant upward trend in the number of retracted pain articles over time. Sixty-six percent of articles were retracted for reasons relating to misconduct. The median (interquartile range) time from article publication to retraction was 2 (0.7 to 4.3) years. The time to retraction differed by reason for retraction, with data problems, comprising data falsification, duplication, and plagiarism, resulting in the longest interval (3 (1.2 to 5.2) years). Further investigations of retracted pain articles, including exploration of their fate post-retraction, are necessary to determine the impact of unreliable data on pain research.