Adherence to RECORD reporting guidelines among observational studies using routinely collected health data published in general medical journals: a metaepidemiologic study

Objectives: The REporting of studies Conducted using Observational Routinely Collected Data (RECORD) tool was developed to address gaps around reporting routinely collected health data. The objective of this study was to assess adherence to RECORD in general medical journals and to evaluate its correlation with study quality.

Methods: We searched PubMed using a filter to identify studies using routinely collected health data published in 8 high impact medical journals between 2016 and 2023. Four journals endorsed RECORD, while 4 did not. For each journal, 24 articles were randomly selected, with 3 studies per year. Study characteristics, RECORD and quality assessments were completed in duplicate and described using proportions and means with SDs. Linear regression was used to estimate the association between journal and study characteristics with adherence to RECORD items.

Results: Studies reported a mean of 70.7% (SD 1.8%) of RECORD items. There was no substantial difference in adherence in RECORD-endorsing journals compared to non-RECORD-endorsing journals (1.8% lower adherence; 95% CI: -5.8, 2.2). Adherence of >80% was reported for RECORD items 1.1, 1.2, 6.1, 7.1, 19.1 and 22.1.

Conclusion: Studies in general medical journals had moderate adherence to RECORD, with no association between journals' endorsement of RECORD and reporting completeness. Other measures to improve adherence to RECORD should be explored, including refinements to the checklist itself. Authors and journals should be aware of and adhere to items required for RECORD reporting to improve the reproducibility of research using routinely collected health data.