Background and aims: Retractions are published in journals in order to correct, ensure the accuracy and integrity of literature and avoid of misconducts. Hence, retraction is an important mechanism in scientific accreditation especially in health subject. In the recent years, the numbers of retracted scientific articles have been increased but compared with the growth of total publications, it has allocated very small portion. In this study, the retracted articles indexed in PubMed have been assessed.

Methods: In this descriptive study, all retracted articles were searched in PubMed with limitation to publication types but without date limitation considered. Then retrieved data were analyzed.

Results: The results of this survey showed a total of 9211 retractions published during 1959 to 2016. According to these findings, the least retraction was published in 1959, 1966 and 1973 with one retracted article each year and the most retractions were 871 retractions in 2013. In addition, of 1781 journals published retractions, THE JOURNAL OF BIOLOGICAL CHEMISTRY (298 (3.24%)), PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (181 (1.97%)), SCIENCE (NEW YORK, N.Y.) (153 (1.66%)) and NATURE (129 (1.4%)) allocated the most rate of retractions. Furthermore, the assessing of indexing cituation of these findings showed 87.08% were in Medline, 11.26% in PubMed-not-Medline and 0.79% in In-Data-Review. Moreover, the most rate of publication types of retractions were related to “JOURNAL ARTICLE”, “SEARCH SUPPORT, NON-U.S. GOV'T”,” RESEARCH SUPPORT, N.I.H., EXTRAMURAL”, “COMPARATIVE STUDY “and” RANDOMIZED CONTROLLED TRIAL”, also the highest rate of Mesh subject headings in retracted articles were belong to “CELLS, CULTURED”, “OLECULAR SEQUENCE DATA”, “TREATMENT OUTCOME”, “CELL LINE, TUMOR”. Finally, the majority of retracted articles were English (98.5%), then French (0.39%), German (0.33%) and Chinese (0.31%)

Conclusion: Retractions publish in two types of self-reported by author or editors discernments. Also two main reasons for retraction are “misconducts” and “unintentional mistakes” of authors that the second reason has the largest numbers of retractions. Therefore, notifying of retracted articles and raising awareness of readers and authors are the best mechanisms for controlling of negative effects of retracted articles. However, publishing of retraction in journals reflects how much importance they are considering to the accuracy of published information.