Retracted publications in autism research are mostly concerned with ethical misconduct.

Abstract

As the prevalence of autism appears to increase, more research to guide effective diagnosis and intervention practices is needed. Findings disseminated through peer-reviewed publications are critical, but the number of retractions continues to rise. An understanding of retracted publications is imperative to ensure the body of evidence is corrected and current. The objectives of this analysis were to summarize key characteristics of retracted publications in autism research, examine the length of time between publication and retraction, and assess the extent journals are adhering to publishing ethical guidelines for reporting retracted articles. We searched five databases through 2021 (PubMed, EMBASE, Scopus, Web of Science, and Retraction Watch). A total of 25 retracted articles were included in the analysis. Ethical misconduct accounted for the majority of retractions rather than scientific error. The shortest time to retraction was 2 months and the longest length was 144 months. The time lag between publication and retraction since 2018 has improved considerably. Nineteen of the articles had retraction notices (76%), whereas six articles did not have a notice (24%). These findings summarize errors of previous retractions and illuminate opportunities for researchers, journal publishers and librarians to learn from retracted publications.