The Use of Routinely Collected Data in Clinical Trial Research

RCTs are the gold standard for assessing the effects of medical interventions, but they also pose many challenges, including the often-high costs in conducting them and a potential lack of generalizability of their findings. The recent increase in the availability of so called routinely collected data (RCD) sources has led to great interest in their application to support RCTs in an effort to increase the efficiency of conducting clinical trials. We define all RCTs augmented by RCD in any form as RCD-RCTs. A major subset of RCD-RCTs are performed at the point of care using electronic health records (EHRs) and are referred to as point-of-care research (POC-R). RCD-RCTs offer several advantages over traditional trials regarding patient recruitment and data collection, and beyond. Using highly standardized EHR and registry data allows to assess patient characteristics for trial eligibility and to examine treatment effects through routinely collected endpoints or by linkage to other data sources like mortality registries. Thus, RCD can be used to augment traditional RCTs by providing a sampling framework for patient recruitment and by directly measuring patient relevant outcomes. The result of these efforts is the generation of real-world evidence (RWE).
Nevertheless, the utilization of RCD in clinical research brings novel methodological challenges, and issues related to data quality are frequently discussed, which need to be considered for RCD-RCTs. Some of the limitations surrounding RCD use in RCTs relate to data quality, data availability, ethical and informed consent challenges, and lack of endpoint adjudication which may all lead to uncertainties in the validity of their results.
The purpose of this thesis is to help fill the aforementioned research gaps in RCD-RCTs, encompassing tasks such as assessing their current application in clinical research and evaluating the methodological and technical challenges in performing them. Furthermore, it aims to assess the reporting quality of published reports on RCD-RCTs.