Hirano Yoko, Asami Yuko, Kuribayashi Kazuhiko, Kitazaki Shigeru, Yamamoto Yuji, Fujimoto Yoko
Medical Affairs, Pfizer Essential Health, Pfizer Japan Inc., Tokyo, Japan.
Medical Affairs, Pfizer Essential Health, Pfizer Japan Inc., Tokyo, Japan.
Value Health Reg Issues. 2018 May;15:1-5. doi: 10.1016/j.vhri.2017.05.002. Epub 2017 Jun 20.
Many pharmacoepidemiologic studies using large-scale databases have recently been utilized to evaluate the safety and effectiveness of drugs in Western countries. In Japan, however, conventional methodology has been applied to postmarketing surveillance (PMS) to collect safety and effectiveness information on new drugs to meet regulatory requirements. Conventional PMS entails enormous costs and resources despite being an uncontrolled observational study method. This study is aimed at examining the possibility of database research as a more efficient pharmacovigilance approach by comparing a health care claims database and PMS with regard to the characteristics and safety profiles of sertraline-prescribed patients.
The characteristics of sertraline-prescribed patients recorded in a large-scale Japanese health insurance claims database developed by MinaCare Co. Ltd. were scanned and compared with the PMS results. We also explored the possibility of detecting signals indicative of adverse reactions based on the claims database by using sequence symmetry analysis. Diabetes mellitus, hyperlipidemia, and hyperthyroidism served as exploratory events, and their detection criteria for the claims database were reported by the Pharmaceuticals and Medical Devices Agency in Japan.
Most of the characteristics of sertraline-prescribed patients in the claims database did not differ markedly from those in the PMS. There was no tendency for higher risks of the exploratory events after exposure to sertraline, and this was consistent with sertraline's known safety profile.
Our results support the concept of using database research as a cost-effective pharmacovigilance tool that is free of selection bias . Further investigation using database research is required to confirm our preliminary observations.
近年来，西方国家利用许多使用大规模数据库的药物流行病学研究来评估药物的安全性和有效性。然而，在日本，传统方法已应用于上市后监测（PMS），以收集新药的安全性和有效性信息，以满足监管要求。尽管传统的PMS是一种非对照观察性研究方法，但它需要巨大的成本和资源。本研究旨在通过比较医疗保健索赔数据库和PMS中舍曲林处方患者的特征和安全性概况，探讨数据库研究作为一种更有效的药物警戒方法的可能性。
扫描由MinaCare有限公司开发的大规模日本健康保险索赔数据库中记录的舍曲林处方患者的特征，并与PMS结果进行比较。我们还通过使用序列对称分析，探讨了基于索赔数据库检测不良反应信号的可能性。糖尿病、高脂血症和甲状腺功能亢进作为探索性事件，其在索赔数据库中的检测标准由日本药品和医疗器械管理局报告。
索赔数据库中舍曲林处方患者的大多数特征与PMS中的特征没有明显差异。暴露于舍曲林后，探索性事件的风险没有升高的趋势，这与舍曲林已知的安全性概况一致。
我们的结果支持将数据库研究作为一种具有成本效益且无选择偏倚的药物警戒工具的概念。需要使用数据库研究进行进一步调查以证实我们的初步观察结果。
