Registration of Observational Studies of Interventions: Prevalence, Characteristics, and Journal Policies
Bibliographic record
Abstract
Cecilie Jespersen,1,2 Zexing Song,3 An-Wen Chan,3,4 Asbjørn Hróbjartsson1,2 Objective Observational studies of interventions use causal inference to assess the impact of interventions on health-related outcomes.1 Despite concerns about reporting bias, observational studies are not subject to the same registration requirements as clinical trials.2,3 We aimed to determine the prevalence of registration among published observational studies of interventions, assess the association between registration and study characteristics, analyze journal registration policies, and explore authors’ and editors’ attitudes about registration. Design We conducted a meta-epidemiologic cross-sectional study triangulating data from 4 sources. First, we searched PubMed for observational studies published in 2023. Eligible studies were cohort or case-control studies with a control group that assessed causal effects of health interventions. Corresponding registration information was collected. Second, authors of included studies were surveyed to explore reasons for and barriers to registration. Third, editorial policies were sampled from 40 journals: 20 sample-representative journals and the journals ranked in the top 20 in Journal Citation Reports by 2023 Journal Impact Factor across 8 specialty categories. Fourth, 1 editor per journal was invited to share their perspectives on registration. Primary outcomes were the prevalence of registered observational studies of interventions published in 2023 and the estimated association between registration and study characteristics, assessed by multivariable logistic regression. Sample size was estimated based on an expected 15% registration rate. Results Among 1100 screened studies, 200 were included: 69 and 128 cohort studies with prospective and retrospective data collection, respectively, and 3 case-control studies. In total, 28 (14%) were registered, and 17 of these (61%) were prospectively registered (<1 month of their start date) (Table 25-1069). Prospective design and protocol availability were positively associated with registration (retrospective vs prospective cohort: odds ratio [OR], 0.19 [95% CI, 0.07-0.54]; P = .002; no public protocol vs public protocol: OR, 0.04 [95% CI, 0.01-0.23]; P < .001). The survey response rate was 23% (46 responses); 60% of authors supported registration, although many only when registration was deemed relevant. Identified barriers included lack of journal requirements for registration (56%) and limited resources (62%). None of the journal policies explicitly required registration of observational studies of interventions, while 12 (30%) encouraged it. Journals that encouraged registration had a higher 2023 Journal Impact Factor and more frequently encouraged public protocols. Editors had divergent opinions on registration. While some considered it to be worthwhile, just as many questioned the added value. https://assets.underline.io/markdown_image/1/image/99ddb5f59ea342d2117443673974b4db.png Conclusions Only 1 in 7 contemporary observational studies of interventions were registered, although more often in cohort studies with prospective data collection and studies with a publicly available protocol. Authors identified the lack of journal requirements to registration as a key registration barrier, and only one-third of journals had supportive policies. Clearer guidance and journal policies on registration relevance (discriminating hypothesis-testing and hypothesis-generating studies) may reduce the risk of reporting biases in observational studies of interventions. References 1. Hernán MA, Wang W, Leaf DE. Target trial emulation: a framework for causal inference from observational data. JAMA. 2022;328(24):2446-2447. doi:10.1001/jama.2022.21383 2. Williams RJ, Tse T, Harlan WR, Zarin DA. Registration of observational studies: is it time? CMAJ. 2010;182(15):1638-1642. doi:10.1503/cmaj.09225 3. Leducq S, Zaki F, Hollestein LM, et al. The majority of observational studies in leading peer-reviewed medicine journals are not registered and do not have a publicly accessible protocol: a scoping review. J Clin Epidemiol. 2024;170:111341. doi:10.1016/j.jclinepi.2024.111341 1Cochrane Denmark & Centre for Evidence-Based Medicine Odense (CEBMO), University of Southern Denmark, Odense, Denmark, ceciliejespersen@health.sdu.dk; 2Open Patient data Explorative Network (OPEN), Odense University Hospital, Odense, Denmark; 3Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, Canada; 4Women’s College Research Institute, Dept. of Medicine, University of Toronto, Toronto, Ontario, Canada. Conflict of Interest Disclosures An-Wen Chan is a member of the Peer Review Congress Advisory Board but was not involved in the review or decision for this abstract. No other disclosures were reported. Acknowledgments We thank all researchers who participated in the author survey for their valuable contribution to the findings of this study.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.380 | 0.761 |
| Meta-epidemiology (narrow) | 0.001 | 0.002 |
| Meta-epidemiology (broad) | 0.002 | 0.003 |
| Bibliometrics | 0.018 | 0.031 |
| Science and technology studies | 0.003 | 0.005 |
| Scholarly communication | 0.007 | 0.007 |
| Open science | 0.003 | 0.005 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.008 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".