Abstract

The promises of Big Data from a pharmacy perspective are huge. The explosion of electronic data sources and analytic capabilities offer the potential to close the gap between science and practice surrounding medications and provide a personalized answer to the questions: “should I take this medication?” However, medications are complicated, and medication-related data, even more so. First, because medication-related data sources are multiple, and very heterogeneous, providing different types of information about medications, prescriptions, and patients taking them. While medications are contextualized by jurisdiction, prescriptions are contextualized for a given patient, at a given point in time. Concerning medication, the main challenge rests in following the drug catalogues that are continuously changing in each jurisdiction. With regard to prescriptions, the main challenge rests in estimating the exposure to medications for a given patient, both in terms of magnitude, and duration. Second, not all medication data sources are equal, and before making sense of medication data, one must take a close look at the data sources regarding the prescription, dispensation, and/or administration, and the local rules, practices, and particularities. Third, making sense of medication-related data is complicated and impeded by the lack of standards, both in terms of practices and terminologies. Finally, patient-related data, where the core of the analytic potential resides, are rarely available, and a link is still missing to be able to connect the intention associated with medication usage, with the consequences, both desired and unwanted. In this chapter, the specific challenges associated with each of these dimensions will be described.