Pragmatic Elements: An Introduction to PRECIS-2

Pragmatic clinical trials are designed to study health interventions in real-world settings that are similar or identical to the settings where the interventions will be implemented. Pragmatic trials thus stand in contrast to explanatory, or traditional, trials, which are usually designed to demonstrate the safety or efficacy of an intervention under highly controlled conditions and in carefully selected groups of participants.

Most clinical trials are situated somewhere along the spectrum between pragmatic and explanatory. The Pragmatic–Explanatory Continuum Indicator Summary (PRECIS; revised in 2015 as PRECIS-2) offers a framework to guide study teams to consider in advance the pragmatic or explanatory nature of their trial across 9 domains (Loudon et al 2015). By 2024, the tool had been cited almost 700 times since its publication (Platt et 2024).

Watch the video module: How Do You Design Trials That Are Fit for Purpose?

Table of Contents

Pragmatic-Explanatory Continuum

While pragmatic clinical trials are designed to answer important, real-world clinical questions, trade-offs in flexibility, adherence, and generalizability are inevitable. A pragmatic trial may have both elements that are more pragmatic and elements that are more explanatory. When planning the trial, it is essential to first ask, “What is the best trial design to answer the question of interest?” There is nothing inherently better or worse about a pragmatic trial as compared with a traditional design. Rather, investigators should choose the degree of pragmatism for each element of the design (Figure) that will best answer the research question of interest. To evaluate the degree to which a trial is pragmatic, study teams are encouraged to read the detailed explanations of PRECIS-2, along with examples of how to apply the tool (Loudon et al 2015).

Pragmatic Elements: An Introduction to PRECIS-2

PRECIS-2 Wheel Diagram

In 2016, members of the NIH Pragmatic Trials Collaboratory’s Health Care Systems Interactions Core and colleagues applied the PRECIS-2 criteria to 5 of the NIH Collaboratory Trials. Each trial was found to be “highly pragmatic” across the PRECIS-2 domains. The exercise highlighted the PRECIS-2’s usefulness in guiding decisions about study design but also revealed a number of challenges in applying the tool and interpreting the results (Johnson et al 2016). For example, during the conduct of the PROVEN trial, members of the study team applied the PRECIS-2 domains in a novel manner to assess how dynamic adaptations shifted implementation to either a more explanatory or a more pragmatic approach (Palmer et al 2018).

The PRECIS-2 wheel — PRECIS-2: Kirsty Loudon et al. BMJ 2015;350:bmj.h2147. Copyright 2015 by British Medical Journal Publishing Group. Used by permission.

The Table illustrates key characteristics of interventions that are more pragmatic than explanatory.

Table. Characteristics of Pragmatic Trials

Trial element	Pragmatic characteristic	Example
Research question	Tests whether the intervention is effective in routine clinical practice	In people living with heart disease, what is the best dose of aspirin to use to prevent a heart attack or stroke? For hospitalized patients in non–critical care units, what is the value of chlorhexidine bathing and nasal decolonization in reducing hospital‐associated infections compared with routine care? In adult patients at risk for suicide, is collaborative care management or skills training better than usual care? In people living with hypertension, is chlorthalidone more effective than hydrochlorothiazide at preventing adverse cardiovascular outcomes?
Setting	Embedded in the routine care setting such as primary care, community clinics, hospital units, or healthcare systems	Standalone hemodialysis clinics Physical therapy clinics Federally qualified health center clinics Non–critical care hospital units within a large healthcare system Primary care clinic Clinics within regional healthcare systems serving underrepresented populations Practice-based research networks
Participants	Eligible population requires little selection beyond the clinical indication of interest	People living with heart disease or hypertension Clinics where the patient population has very low rates of guideline-recommended colorectal cancer screening Patients undergoing diagnostic spine imaging Nursing home residents eligible for advance care planning Patients managing multiple chronic conditions
Intervention and comparator	Compares two or more real-world treatments using flexible protocols	Nighttime vs daytime dosing of blood pressure medicine 81 mg vs 325 mg dose of aspirin Dialysis session of 4.25 hours vs standard session length Universal decolonization (antiseptic bath and nasal ointment) vs routine bathing
Outcomes	Endpoints are clinically relevant to participants, funders, communities, and healthcare providers	Clarify the effectiveness of a behavioral treatment to prevent suicide or reduce pain Define how EHRs can be used to optimize guideline-based screening Provide evidence for the best aspirin dose to prevent heart disease Demonstrate improved methods for reducing hospital-associated infections
Clinical importance	Purposely designed for making healthcare decisions in settings in which the intervention will be implemented	Trial results are translated into new national trauma guidelines Trial results show that important clinical questions can be answered directly from patient data Trial results show that epidemiological evidence in patients’ imaging reports can reduce unnecessary diagnostic and therapeutic interventions