A comparison of five international clinical trial registers with the South African register for access to information and usability
Introduction: in November, 2005, the South African (SA) National Department of Health (NDoH) mandated that, as from the 1st December, 2005, all new clinical trials to be conducted in the country must be registered on the South African National Clinical Trials Register (SANCTR). The objective was to compare access to the information contained in and the usability of the SANCTR with five other international on-line clinical trials registers. Methods: Access to SANCTR was determined through the use of three search engines using the keywords "South African Clinical Trials." Five high-profile international registers were identified and accessed for comparative purposes. Each register was investigated for information on trials conducted in South Africa using a standardised data extraction form which listed 24 data items. The usability of the various on-line registers was determined through a self-administered questionnaire adapted from the five key usability factors previously defined in literature. Heuristic evaluation was carried out with 10 'experts' (Pharmacy staff and postgraduate students at Sefako Makgatho Health Sciences University (SMU)). Data generated from the heuristic evaluation were analysed using descriptive statistics, univariate and multivariate analyses. Results: The SANCTR website had the highest ranking for access amongst the registers in all three selected search-engines after an internet search using the keywords "South African Clinical Trials". The total number of clinical trials registered varied among the registers. The WHO's International Clinical Trials Registry Platform (ICTRP) recorded 2 599 trials carried out in South Africa, with 2 260 registered in the ClinicalTrials.gov register, 2 196 in the SANCTR and 978, 149 and 174 in the European Union (EU), International Standard Randomised Controlled Trial Number (ISRCTN) and Pan African Clinical Trials (PACTR) registers respectively. The websites ClinicalTrials.gov and ISRCTN provided greater overall information per clinical trial registered and provided information on all 24 clinical trials data items. The PACTR had information on 23 of the 24 data items. The WHO and EU registers each contained 19 data items. The SANCTR provided the least information, only 11 data items. The heuristic evaluation identified ClinicalTrials.gov as the 'best' site, while the PACTR had the lowest rating for layout and design. The EU register and SANCTR were the least easily navigable. The respondents had the least satisfaction while using the 'Search' option in the SANCTR. Users also reported the SANCTR and the PACTR had the lowest overall user-friendliness. Conclusion: The fact that the SANCTR contains less information on SA clinical trials than other registers and is the least user-friendly warrants utmost attention. The study puts forward a case to the regulatory authority (currently the Medicines Control Council) as it takes on a new structure and working arrangements as the South African Health Products Regulatory Authority to optimise the SANCTR to be more user-friendly and contain more complete information on clinical trials conducted in SA.