Throughout the last century, pharmaceutical regulators all over the world have used various methods to test medical treatments. From 1962 until 2016, the Randomized Clinical Trial (RCT) was the reference test for most regulatory agencies. Today, the standards are about to change, and in this chapter we draw on the idea of the data journey to illuminate the trade-offs involved. The 21st Century Cures Act (21CCA) allows for the use of Electronic Health Records (EHRs) for the assessment of

more »

... t treatment indications for already approved drugs. This might arguably shorten the testing period, bringing treatments to patients faster. Yet, EHR are not generated for testing purposes and no amount of standardization and curation can fully make up for their potential flaws as evidence of safety and efficacy. The more noise in the data, the more mistakes regulators are likely to make in granting market access to new drugs. In this paper we will discuss the different dimensions of this journey: the different sources and levels of curation involved, the speed at which they can travel, and the level of risk of regulatory error involved as compared with the RCT standard. We are going to defend that what counts as evidence, at the end of the journey, depends on the risk definition and threshold regulators work with.