Harm effects in non-registered versus registered randomized controlled trials of medications: a retrospective cohort study of clinical trials

Chang Xu ¹, Shiqi Fan ², Luis Furuya-Kanamori ³

¹Proof of Concept Center, Eastern Hepatobiliary Surgery Hospital, Third Affiliated Hospital, Second Military Medical University, Naval Medical University, Shanghai, China. xuchang2016@runbox.com.
²Proof of Concept Center, Eastern Hepatobiliary Surgery Hospital, Third Affiliated Hospital, Second Military Medical University, Naval Medical University, Shanghai, China.
³UQ Center for Clinical Research, The University of Queensland, Herston, Australia.
⁴Department of Endocrinology and Metabolism, Division of Guideline and Rapid Recommendation, Chinese Evidence-Based Medicine Centre, Cochrane China Centre, MAGIC China Centre, West China Hospital, Sichuan University, Chengdu, China.
⁵Department of Epidemiology and Biostatistics, University of Arizona, Tucson, AZ, USA.
⁶Statistical Research and Innovation, Global Biometrics and Data Management, Pfizer Inc, New York, NY, USA.
⁷Division of Biostatistics, School of Public Health, University of Minnesota, Minneapolis, MN, USA.
⁸Department of Health Sciences, University of York, York, UK.
⁹Norwich Medical School, University of East Anglia, Norwich, UK.
¹⁰Department of Pediatrics, Faculty of Medicine and Dentistry, Department of Psychiatry, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada.

⁰Proof of Concept Center, Eastern Hepatobiliary Surgery Hospital, Third Affiliated Hospital, Second Military Medical University, Naval Medical University, Shanghai, China. xuchang2016@runbox.com.

Bmc Medicine +

|

October 11, 2024

View abstract on PubMed

Summary

Trial registration enhances transparency in clinical research. Non-registered and retrospectively registered medication trials show lower harm estimates than prospective ones, suggesting potential underreporting of adverse events.

Area Of Science

Clinical research methodology
Drug safety and pharmacovigilance
Biostatistics

Background

Trial registration is crucial for transparency and accountability in clinical research.
Selective or non-reporting of findings can introduce bias.
Investigating the impact of registration status on reported harm effects is essential.

Purpose Of The Study

To assess the influence of trial registration status on the estimated harm effects in randomized controlled trials of medication interventions.
To compare harm estimates between prospectively registered, retrospectively registered, and non-registered trials.

Main Methods

Searched PubMed for systematic reviews and meta-analyses of medication harm trials (2015-2020).
Included meta-analyses with at least five trials, varying registration statuses, and adverse event data.
Used Directed Acyclic Graph (DAG) method for confounder harmonization and hierarchical linear regression to compare harm estimates (ratio of odds ratios).

Main Results

Analyzed 629 meta-analyses comprising 10,069 trials; 74.3% were registered, 23.9% unregistered.
Non-registered (OR=0.82) and retrospectively registered (OR=0.75) trials showed lower harm estimates compared to prospectively registered trials.
No significant difference in harm estimates was found between retrospectively registered and non-registered trials.

Conclusions

Medication harms may be understated in non-registered trials.
Retrospective registration did not demonstrably reduce selective or absent reporting of harms.
Prospective trial registration is strongly recommended to ensure accurate reporting of medication-related harms.

Keywords:

Harm effects Randomized controlled trials Registration Selective reporting