Error publication (published erratum) in neurosurgical journals worldwide using PubMed during the last 30 years

The issue of error of scientific publications has recently attracted the interest of medical researchers. However, there was no similar evaluation of errata in the field of neurosurgical literature. The aim of this study is to evaluate published errata in neurosurgical journals and to discuss the strategies that can be used in order to reduce errata frequency and to prevent their dissemination.

A literature search of error publication in 28 main neurosurgical journals was performed using PubMed (1990-2019). Extracted data included authors' name, chronology, country of origin, journal impact factor, subject area, research type, reason for published error, and source of responsibility.

A total of 441 published errata were identified and analyzed. Most studies were published within the last 6 years. The majority of publications had one single reason for the published erratum. The mean amount of time between the original publication date of the paper and the published erratum was 6.72 months. The most common reason given for published erratum was that of authorship, followed by text content, figures, and tables. The mean published error rate was 0.81% (2014-2019).

Unlike other specialties, errors are infrequently observed in neurosurgical journals and mostly without altering the interpretation of study findings. However, improvement is still needed. With the development of online journal publishers and scientific social media platforms, new strategies must be studied in order to track and correct errors better and faster. Also, authors and publishers have to work better together in order to produce high-quality scientific papers.