fetching data ... 
Background: Transparent trial conduct requires prospective trial registration, the publication of the protocol of a randomized controlled trial (RCT) before enrolment of the first participant. Its main purpose is to minimize the risk of bias through unjustified or hidden modification of trial design during the course of study, and to ensure that the analyses are performed following a specific hypothesis and would not be driven by the data. Prospective trial registration also allows to track the status of studies in terms of completion and publication, allowing a better assessment of publication bias. Both the Declaration of Helsinki [1] and the International Committee of Medical Journals Editors (ICMJE) [2] support and require prospective trial registration since 2005.
Objectives: Part 1: To estimate the proportion of RCTs in rheumatology that are prospectively registered, and analyze the time trends and the factors associated with prospective registration. Part 2: To evaluate the reasons for non-adherence with prospective registration and explore the researcher perceived effectiveness of different proposed mechanisms aimed at enhancing adherence with prospective registration.
Methods: Part 1. We retrieved all primary reports of RCTs in rheumatology published between January 2009 and December 2022, and which recruited their first patient after 2005. We searched MEDLINE through PubMed, using regular expression pattern analysis of a registration number in metadata, abstract or full text. We extracted data concerning trial registration from the International Clinical Trials Registry Platform of the World Health Organization using trial registration number. We performed multivariable logistic regression to identify factors independently associated with prospective trial registration. Part 2. We sent an online survey to the corresponding authors and contact researchers of published RCT who failed to prospectively register the trial, inquiring about possible reasons for nonadherence with prospective registration and exploring their opinion about potential solutions.
Results: Part 1. Of 1109 primary reports of rheumatology RCTs – half of which published in journals endorsing the ICMJE recommendations – 409 (37%) were not prospectively registered: 127 (12%) were not registered and 282 (25%) were retrospectively registered. The proportion of RCTs prospectively recorded increased over time at a rate of 3 percentage points per year (p<0.001), with less than one-third of trials prospectively registered in 2009, against 75% in 2022 (Figure 1). In ICJME journals, these proportions were 52% in 2012, 81% in 2016, and 85% in 2022 (Figure 1). In the adjusted model, factors independently associated with a higher likelihood of prospective registration were: larger sample size, recruitment in multiple countries, evaluation of pharmacologic interventions, being published in an ICMJE journal and in a journal with higher impact factor. Part 2. The main reasons reported by investigators for retrospective registration were a lack of knowledge (53% of 51 participants), or organizational problems (40%). Among investigators respondents, 70% stated that linking the obtention of ethic approval to trial registration would be the best option to ensure prospective registration.
Conclusion: Despite large improvement, adherence with prospective registration is still unsatisfactory in rheumatology. Investigators considered that linking registration to ethics approval was the preferred option for improving adherence.
REFERENCES: [1] The World Medical Association. Declaration of Helsinki - Ethical Principles for Medical Research Involving Human Subjects.
[2] De Angelis C, Drazen JM, Frizelle FA, Haug C, Hoey J, Horton R, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. Lancet. 2004;364(9438):911-2.
Time evolution of prospectively registered published RCTs for all journals (in grey), those following ICMJE recommendations (in green) and those not following ICMJE (in red)
Acknowledgements: NIL.
Disclosure of Interests: None declared.