Fig. 1

The role of routinely collected health data (RCD) in randomized controlled trials in various phases of a clinical trial (based on the Consolidated Standards of Reporting Trials [CONSORT] flow diagram [11]). (1) During enrollment, RCD sources can be screened retrospectively for eligible patients, but they can also be used prospectively as targeted screening and recruitment tools. (2) Informed consent could be given both for data use and for trial participation, so that when patients decline to participate in the randomized component of the trial, their usual care can still be followed. (3) Allocation can be facilitated by RCD through point-of-care randomization. Patients who are not allocated to an intervention but select care on the basis of personal and clinical preferences can be observed with RCD. (4) During the follow-up phase, RCD allows patients who would otherwise be lost to follow-up to be tracked, and thus less missing data may be encountered. (5) Long-term follow-up, such as in registries, may be possible with RCD even after formal completion of the primary study phase. (6) RCD allows analysis of both nonrandomized and randomized patients and direct supplementation of information to the randomized part of the trial