RESUMO
<B>Objective</B>: Databases, such as the Medicaid recipient database in the USA and the General Practice Research Database (GPRD) in the UK, take on an important role as resources for balancing the benefits and risks of medicines in Europe and the United States.<Br>Their record sizes are several ten million and a few million each. They are actually used for epidemiological studies. However, in Japan, a database that can be used for such studies is insufficient. We attempted to create a database of pre-marketing clinical trial data for antihypertensive drugs. These data have been managed by a Controller Committee. <Br>(<B>Design</B> : not applicable )<Br><B>Methods</B>: The database is made from the data and the documents, including electronic and paper media. The creation process was as follows : computerizing documents, item-name identification, defining the integrated database, protocol review, batch processing, and logical/validation checking.<Br><B>Results</B>: The database has 13 datasets and consists of 56 trials and 12,389 subjects. Overall, 15 trials involved beta-blockers which is the largest drugs as the investigational drug, and 43 trials compared the same group of antihypertensive drugs.<Br><B>Conclusion</B>: A database that can be used for quantitative evaluation of various hypotheses has been built. It is possible to completely analyze all of the data in this large-scale database to conduct, for example, individual patient data (IPD) meta-analyses.
