The reporting of Health Systems Data use in primary results publications of clinical trials: A systematic review

Background : Data collected within clinical trials often overlaps with routinely collected Health Systems Data (HSD). There is potential for HSD to reduce burdens for trials and understanding HSD use can help triallists make decisions about using HSD in future trials. However, it is unknown to what extent HSD use has been reported in trial publications, despite the development of guidelines such as ESMO-GROW and CONSORT-ROUTINE extension for reporting HSD use in trials. This study expands on work previously conducted by Lensen and colleagues (1). It aims to provide insights into how HSD use is reported in main result publications that present main trial results, before and after release of the CONSORT-ROUTINE extension. Methods : This was a systematic review of the reporting of HSD use in primary results publications of trials that accessed HSD between 2017-2018. Of 90 trials identified by Lensen and colleagues, those that had published primary outcome results were included in the review. Trials were excluded if (1) primary results were not yet due to be reported; (2) not yet published; (3) they were published prior to June 2017; (4) they were published in 2017, but HSD was accessed in 2018 and (5) the primary publication only reported HSD use in secondary, interim or Study Within a Trial (SWAT) analysis. Eligible publications were identified using ISRCTN, ClinicalTrials.gov, EU Clinical Trials Registries, PubMed and Embase. The reporting of HSD use was compared against expectations for reporting outlined in the CONSORT-ROUTINE extension. Results : Forty-nine primary publications from 46 trials were included in the review. Overall, none of the included publications reported all the information suggested in the CONSORT-ROUTINE. However, there has been an improvement in the reporting of HSD use, since the publication of the CONSORT-ROUTINE guidelines. Conclusions : Reporting of HSD use has improved over time. However, it still does not meet the expectations set out in the CONSORT-ROUTINE extension. Triallists should be encouraged to provide further information in publications about the use of HSD as per the CONSORT-ROUTINE extension guidelines. This would allow greater transparency in reporting, facilitating effective HSD use in future trials.