Trial publication after registration in Clinical Trials Registry of India (CTRI): a cross-sectional analysis of randomized controlled trials

Tags: Poster
John D1, Stanley A2, Gudi N3, Rao MVV4
1Campbell Collaboration, 2Achutha Menon Centre for Health Science Studies, 3Independent Researcher, 4National Institute of Medical Statistics

Background: The Clinical Trials Registry-India (CTRI) is a free and online public record system for registration of clinical trials being conducted in India. From July 2009 to December 2017 11,052 trials have been registered in CTRI.

Objectives: We carried out a comprehensive examination of the rates of publication and reporting of results within two years for all completed randomized controlled trials (RCTs) registered in the CTRI database.

Methods: We included trials registered in the CTRI between 15 June 2009 and 15 June 2015. We extracted data from these trials to identify various aspects (e.g. sponsor; condition studied; study purpose, type, or phase; number of study arms; intervention model; randomization status; enrolled participants; population; registration; enrolment; trial start and end dates). We used a three-step process to identify publication status; identifying publications from CTRI database; searches using scientific title, short title, and principal investigator (PI) name on PubMed, Google Scholar, IndMed and South Asian Database of Clinical Trials databases, and mails to the PIs seeking details of unpublished studies. Relevant publication and non-publication details were also extracted for analysis.

Results: We identified 2949 trials as per inclusion criteria from 5721 RCTS out of a total of 9099 trials registered under CTRI. Publication details (DOI, citation, ISSN) of 80 (2.7%) studies were mentioned in CTRI. From 2949 studies, 1950 studies were searched and we found 560 (28.7%) studies had been published. We selected 70 studies randomly from within the 560 studies, and revealed that none of the trials had updated the 'primary completion date' or 'study completion date' on the CTRI website. A total of 1390 (71.3%) of the 1950 studies remain unpublished, 420 studies were available on MEDLINE, we identified no studies through contacting principal investigators.

Conclusions: Publication outputs is poor, and there is noticeable variation in the dissemination of clinical trials results across RCTs registered in CTRI.

Patient or healthcare consumer involvement: This study is being conducted in collaboration with National Institute of Medical Statistics (NIMS) and Dr Rao, Director, NIMS is a co-author. NIMS hosts the CTRI database and the study findings will be used to improve reporting of trials registered in CTRI in future.