Half of systematic reviews about pain registered in PROSPERO were not published and the majority had inaccurate status.

OBJECTIVES

The aim of this study was to analyze the publication path of non-Cochrane systematic reviews (SR) in the field of anesthesiology and pain registered in the PROSPERO database.

STUDY DESIGN AND SETTING

We analyzed characteristics of SRs registered in PROSPERO from its inception to May 2017 and their publication status in August 2018. We surveyed corresponding authors of unpublished SRs about accuracy of PROSPERO status and reasons for nonpublication.

RESULTS

After screening 1,408 records from PROSPERO database, we found that the majority had "ongoing" (76.3%) and "completed not published" (9.2%) status. Survey of authors showed that most of the records had not been updated (82.4%, 526/638); SR had already been published in 75.2% (396/526), and work on SR had been discontinued in 7.8% (41/526) of cases. In total, based on PROSPERO status, survey of authors, and database searches, 53.6% (742/1,384) of SRs had been published within a period of 1.3 years or more following their registration. Main reasons for discontinuing work on SR were publication of an SR with similar or same topic by another author team and rejection of SR manuscript.

CONCLUSION

Only 16.3% of PROSPERO records had accurate status, and 46.4% of SRs were still unpublished. Further steps to ensure accuracy of PROSPERO status are needed, along with developing strategies for improvement of SR production process.