Data completeness and quality: the results of the Australian New Zealand Clinical Trials Registry 2011 data audits

Article type
Authors
Tai F1, Ooi W1, Hunter K1, Ko H2, Askie L1
1Australian New Zealand Clinical Trials Registry, Australia
2Singapore Health Services (SingHealth), Singapore
Abstract
Background: The Australian New Zealand Clinical Trials Registry (ANZCTR) is an on-line register of >6500 clinical trials undertaken in Australia, New Zealand, and elsewhere. As a World Health Organization International Clinical Trials Registry Platform (WHO ICTRP) Primary Registry, it is required to demonstrate that registered trials contain clinically meaningful and complete information, and are registered only once within each member registry. A Data Completeness and Quality Audit (DCQA) and Trial Duplication Audit (TDA) are undertaken annually to fulfil this requirement.

Objectives: To assess the data completeness and quality of trial information registered with the ANZCTR, and identify any duplicate registrations.

Methods: Both the DCQA and the TDA included trials registered from June 18, 2010 to August 23, 2011, inclusive, with 1497 trials having been assessed. Of these, 1330 were interventional trials and 167 were observational studies. The DCQA process involved reviewing each trial’s data field input and marking those trials not meeting ANZCTR requirements as ‘incomplete’. Up to three reviews were conducted, with the ANZCTR Manager consulted if necessary. The TDA process consisted of reviewing the input of a select number of data fields for each registered trial. A full review was conducted for any suspected duplicates, with up to three reviews being conducted.

Results: For the DCQA, the top five fields marked ‘incomplete’ were ‘Description of intervention(s)/exposure’ (7.21%), followed by ‘Ethics committee address’ (4.66%), ‘Secondary Outcome’ (4.05%), ‘Primary Outcome’ (3.64%), and ‘Contact person for scientific queries address’ (3.47%), with reasons including lack of detail in responses and formatting issues. In the TDA, 17 trial records were initially suspected of being duplicates. Further review proved these to be otherwise.

Conclusions: These results highlight the effectiveness of the ANZCTR’s registration processes in ensuring the quality and completeness of data, preventing duplicate registrations, and identifying areas where future user inputs may be improved.