Pragmatic Randomized Clinical Trial — Effectiveness in Real-World Care

PRECIS-2 (Pragmatic-Explanatory Continuum Indicator Summary, version 2) is a validated 9-domain rating wheel that helps trial designers characterize and communicate how pragmatic or explanatory each element of their trial design is. Domains include eligibility, recruitment, setting, organization, flexibility of the intervention and comparator, follow-up intensity, primary outcome, and analysis approach. Using PRECIS-2 at the design stage ensures that design choices are coherent with the trial's purpose and supports transparent reporting. Most journals publishing pragmatic trials now expect a PRECIS-2 diagram in the methods section.

Blinding is often not feasible or not desirable in pragmatic trials — particularly for behavioral, surgical, or complex health system interventions — because the trial is designed to reflect routine care. When blinding is impossible, the risk of performance and detection bias should be acknowledged and outcome assessors should be blinded where possible. The absence of blinding does not disqualify a trial from being a legitimate pragmatic RCT, but it must be reported transparently.

These concepts overlap but are not identical. A large simple trial is characterized by a very large sample, minimal data collection beyond the primary outcome, and broad eligibility — all features consistent with pragmatism. A pragmatic RCT can be small if the setting and design choices reflect routine care. The pragmatic label refers to the philosophy of design (real-world applicability), whereas 'large simple trial' refers more to the operational strategy of maximizing sample size and minimizing complexity.

Generally, regulatory agencies such as the FDA and EMA require controlled, explanatory-style evidence for initial marketing approval because they need to establish efficacy and safety under defined conditions. However, pragmatic data are increasingly accepted for label extensions, post-marketing commitments, and health technology assessments. Some agencies have published guidance on real-world evidence, and pragmatic RCT data can complement — though rarely replace — explanatory trial packages for regulatory purposes.

The appropriate comparator is the best available current treatment or usual care — whatever the patient would realistically receive if not enrolled in the trial. Using a placebo comparator in a pragmatic trial is usually inappropriate because it does not answer the comparative effectiveness question that clinicians and patients face. Head-to-head comparisons of active treatments are the norm, and the choice of comparator should be justified based on current clinical practice in the target setting.