<b>2. The Use of Medical Information in Foreign Countries: Examples in Pharmacoepidemiology Studies</b> / 薬剤疫学

ABSTRACT

In this manuscript, the use of medical information in foreign countries is viewed by taking pharmacoepidemiology studies conducted in US, Sweden and Taiwan as examples. The first example is the cohort study on the association between pioglitazone and bladder cancer using diabetes registry in Kaiser Permanente Northern California(KPNC) which suggested the causal relationship between pioglitazone and bladder cancer. The success of the study was due to the diabetes registry constructed taking long time in KPNC from various data sources including claims data, data of electronic health record and data in pharmacies. The second example is the study on safety of H1N1 vaccine in Sweden. In this study, the vaccine registry was newly developed on the website where almost all of the vaccinated subjects were registered as reimbursement was subject to registration. The web-based registry was linked with the existing common healthcare registers and the investigators estimated hazard ratios of rare outcomes like Bell's palsy and paraesthesia associated with vaccination. The third example is the Large Linked Database(LLDB) to monitor safety of H1N1 vaccination in Taiwan. The LLDB used the technology of IC Card Data Center and the information on diagnosis and vaccination was collected on the daily basis. The risk of various outcomes as in the second study and adverse outcomes associated with pregnancy were monitored. In all of those examples, new mechanisms such as diabetes registry, web-based vaccine registry and the LLDB were established purposefully. The record linkage is the key element to enhance the value of medical information. (Jpn J Pharmacoepidemiol 2012; 17(2) 109-116)