ABSTRACT
Retrospective databases are a rich source of data for pharmacoeconomic analyses. Data from claims and electronic medical/health records constitute pre-existing data sources that are readily available, less expensive than a randomized controlled trial and often more representative of patient populations. There are other, less conventional methods of developing databases, such as polling patients or healthcare providers that may be considered. This chapter presents the theory behind retrospective database analyses and practical aspects of procuring these data sets. It also discusses some of the potential pitfalls and ways to avoid biases inherent in working with these databases. Multiple examples employing these data sets are presented.
