An interactive hypertext database of synoptic summaries of cardiovascular clinical trials

Tags: Poster
Cucherat M, Boissel J, Haugh M, Strang N, Gueyffier F

In cardiology there is a large number of clinical trials and this number increases each year. This makes it not feasible to remember and use all the information without help. A centralized system providing synthetic information for use in clinical pharmacology, evidence-based medicine and teaching could provide the necessary help.

We are developing a database of cardiovascular clinical trials summaries which will provide rapid access to a synthetic information about the characteristics and results of the trials. These trials are grouped together by pathology and by treatment comparison (experimental treatment versus control treatment); e.g., trials comparing aspirin versus placebo in secondary stroke prevention.

The main trial characteristics (design, treatment and patient characteristics) are presented in tables (similar to those in a meta-analysis report). The results for each trial are reported for all relevant outcomes (number of events/patients, relative risk, the 95% confidence interval, and the p value). An estimation of the common treatment effect is given as well as the p value of the heterogeneity test. These results are also presented in the form of a graph. Two types of summary exist. In the first, only direct comparisons between treatments for the same pathology are presented. In the second indirect comparisons of different treatments for the same pathology are presented (or the results obtained for the same treatment in different pathologies or in different patient populations).

The database has been developed using HTML, which allows interactive hypertext navigation with common Internet browsers, such as Netscape or Internet Explorer. This database will be available on Internet or CD-ROM. An evaluation version can be consulted at:

www.spc.univ-lyonl.fr/~mcu/ssct/index.htm.

Finally, it should be stressed that this database can only be considered as a reference for specialized users who need to refer to the raw results of the clinical trials. It is not a dissemination tool for evidence for use in medical practice. The absence of discussion of the information could lead to misinterpretation of the results by those without the necessary training.