Pragmatic Clinical Trials and Real-World Evidence: An Introduction

Pragmatic clinical trials (PCTs) are designed to bridge the gap between traditional randomized controlled trials (RCTs) and everyday clinical practice. PCTs are designed to test the effectiveness of interventions in real-world settings (Dal-Ré, Janiaud, & Ioannidis, 2018), which reflect routine healthcare practices, providing data directly applicable to a broad patient population. Generally, an intervention’s efficacy (interventions tested under optimal conditions) and safety are determined using the RCT design, and then effectiveness tested using the PCT design. The origin of PCTs can be traced back to concerns about the applicability and generalizability of results from traditional randomized clinical trials (RCTs) (Ford & Norrie, 2016). While meticulously designed to establish the efficacy and safety of interventions under optimal conditions, traditional RCTs often fell short in informing everyday clinical practice. These trials generally have highly selective study eligibility criteria and controlled conditions that do not reflect the reality of diverse patient populations and healthcare settings. While these conditions are ideal for isolating the effect of an intervention, they often lead to overestimating efficacy and safety and a potential underrepresentation of adverse effects (Ford & Norrie, 2016). Furthermore, the small sample sizes and narrow participant selection criteria limit the external validity of these trials (Ford & Norrie, 2016). This paper builds upon the other manuscripts in this series, discusses the differences between the RCT and PCT, introduces the basics of the PCT, and provides an example of the research method.

What are the Differences Between Explanatory (RCT) and Pragmatic Clinical Trials?

In healthcare research, RCT and pragmatic trials represent two approaches to evaluating interventions’ efficacy, safety, and effectiveness. Each has unique attributes and applications, contributing valuable insights to clinical decision-making. Explanatory trials, also known as RCTs, are designed to test the efficacy and safety of an intervention under optimal conditions (Roland & Torgerson, 1998; Van Norman, 2021). RCTs recruit study participants who meet strict eligibility criteria, are often conducted in specialized research centers, and adhere to tightly controlled protocols. The primary goal of an RCT is to determine whether an intervention works in theory, and outcomes are measured to understand the physiological, biological, or impact of an intervention (Roland & Torgerson, 1998; Van Norman, 2021). For instance, an RCT might investigate whether a new cancer drug can shrink tumors in patients with a specific type of cancer and no other health conditions. However, these results may not fully capture the impact of an intervention on patients’ quality of life, functional capacity, symptom burden, or impact on patients with tumors and other health conditions and concomitant medications (Gaudry et al., 2017).

PCTs, on the other hand, evaluate an intervention’s effectiveness in real-world conditions (Roland & Torgerson, 1998; Van Norman, 2021). These trials recruit a broad range of participants, reflecting diversity in age, ethnicity, comorbidities, and other relevant factors. PCTs are typically conducted in healthcare settings, such as primary care clinics or community hospitals (Roland & Torgerson, 1998; Van Norman, 2021). They allow for more flexibility in terms of protocol parameters, acknowledging that variations in practice are a reality of clinical care. PCTs aim to determine whether an intervention works in practice, and the benefits of a treatment are often measured using patient-centered outcomes, such as quality of life or functional status (Roland & Torgerson, 1998; Van Norman, 2021). These measures provide a more holistic view of the intervention’s impact, helping to inform shared decision-making between patients and healthcare providers. A PCT might assess whether a new diabetes medication reduces hospital admission among patients with varying degrees of disease control and comorbid conditions.

The choice between an RCT and PCT study design hinges on the research question and the intervention’s developmental stage. An RCT might be the first step to establishing the efficacy and safety of a newly developed product or intervention. Once confirmed, a PCT can assess its effectiveness in a real-world setting. However, it’s worth noting that the dichotomy between the RCTs and PCTs is not absolute. Many trials fall along the spectrum between these extremes, combining elements of both approaches to answer complex research questions (Lyngbakken et al., 2021). For example, a PCT might address research inquiries directly affecting clinical or health policy decisions. Such trials may involve the comparison of the effectiveness of various existing treatment alternatives in real-world clinical scenarios (i.e., acupuncture versus chiropractic manipulations for chronic back pain). Alternatively, they may seek to answer research questions related to clinical procedures, such as delivering treatment/intervention, monitoring treatment/intervention, dosage administration, or intervention interactions. To illustrate, consider the comparison of blood pressure management strategies for hypertension. One approach might involve pharmacists at local chain pharmacies monitoring blood pressure using a digital device that transmits results to an electronic case report form and is sent to their primary providers when patients come to pick up their medication refills. Alternatively, blood pressure readings could be taken at a patient’s primary care provider’s office whenever new prescriptions are written. The approaches present different avenues for assessing the management of hypertension using a PCT design and represent a decentralized clinical trial (DCT) where some or all trial activities occur at locations other than at traditional research centers (Van Norman, 2021). A future article in this series focuses on the DCT.

PCTs have gained popularity in various fields due to the limitations of traditional placebo-controlled trials in addressing a wide range of clinical questions (Hohenschurz-Schmidt et al., 2023). PCTs are valuable for evaluating the effectiveness of a therapy compared to established care in diverse populations and non-academic settings, regardless of the underlying mechanisms (Hohenschurz-Schmidt et al., 2023). PCTs are particularly relevant for therapies with well-established safety and efficacy profiles. Additionally, PCTS are often designed for therapies widely used in clinical practice despite limited evidence of efficacy and safety, such as complementary and alternative therapies, where potential harm is considered low (Hohenschurz-Schmidt et al., 2023). PCTs can overcome the limitations of strict eligibility criteria by better reflecting the realities of clinical practice. This includes patients with multiple comorbidities, high levels of disability, or socioeconomic barriers to treatment participation. Therefore, PCTs provide more realistic estimates of effect sizes and enhance the translation of research findings into clinical practice (Hohenschurz-Schmidt et al., 2023).

The Pragmatic-Explanatory Continuum Indicator Summary

While the lines distinguishing explanatory and pragmatic trials may often appear unclear, there is an instrument that can measure the extent of pragmatism in clinical trials. The Pragmatic-Explanatory Continuum Indicator Summary (PRECIS) is an instrument that assists investigators in determining the positioning of a trial on the continuum between pragmatic and explanatory (Kirsty et al., 2015). Essentially, the instrument helps to clarify whether a trial is more real-world-oriented (pragmatic) or more controlled and explanatory. The PRECIS evaluates trials across several domains, including eligibility criteria, to provide an overall indicator of a study’s “pragmatic-ness” in that each domain is scored from 1 (very explanatory) to 5 (very pragmatic) (Elder & Munk, 2014; Thorpe et al., 2009). See Table 1 for the domains of the PRECIS-2.

Table 1.

Summary of the PRECIS-2 Tool – Domains of Pragmatic Trial

Dimensions	Pragmatic Assessment
Eligibility:	How similar are the trial participants to those receiving the intervention in usual care?
Recruitment:	How much extra effort is put into recruiting participants compared to usual care?
Setting:	How different are the trial settings from the usual care setting?
Organization:	How different are the organization’s resources, provider expertise, and care delivery model in the intervention arm compared to usual care?
Flexibility (delivery):	How does the intervention’s flexibility differ from what is expected in usual care?
Flexibility (adherence):	How does the flexibility in monitoring and encouraging participant adherence differ from what is expected in usual care?
Follow-up:	How does the intensity of measurement and follow-up in the trial differ from typical follow-up in usual care?
Primary outcome:	How directly relevant is the trial’s primary outcome to the participants?
Primary analysis:	To what degree are all data part of the analysis of the primary outcome?

Adapted from: Loudon, 2015 Link to the PRECIS tool kit https://www.precis-2.org/.

PRECIS was developed in 2009 and revised in 2015 and is now referred to as PRECIS-2 (Lyngbakken et al., 2021). The revision aimed to provide a more comprehensive and user-friendly tool for investigators to help them decide on the design elements of their studies (Norton, Loudon, Chambers, & Zwarenstein, 2021). The PRECIS-2 evaluates trials across nine domains (Kirsty et al., 2015), offering a framework for study teams to prospectively consider the pragmatic aspects of their trial design (Norton et al., 2021).

Historically, the development of PRECIS and its subsequent revision as PRECIS-2 have played a crucial role in refining the design of randomized trials (Norton et al., 2021; Riddle et al., 2010). It has helped guide researchers in making design choices that align with their intended purpose, whether to test intervention efficacy and safety under ideal conditions (explanatory) or to evaluate effectiveness in everyday practice (pragmatic) (Thorpe et al., 2009). PRECIS instruments contribute to clinical trials by encouraging investigators to think about the implications of their design choices and facilitating the alignment of trial design with the intended use (Kirsty et al., 2015).

A Pragmatic Clinical Trial Design Example

The cluster randomized clinical trial is an example of a PCT (Polit, 2022). For details, see the article in this series, A Primer to Cluster Randomized Trial (Capili & Anastasi, 2023). Another example of a PCT, as proposed by Porzsolt and colleagues, is as follows (Porzsolt, Eisemann, Habs, & Wyer, 2013). For this PCT example, all patients from a primary care clinic diagnosed with hypertension and prescribed anti-hypertension medication are included in the study. Including all patients minimizes the risk of selection bias (Porzsolt et al., 2013). The treatment administered is not randomly selected but rather based on the preferences of the providers and patients. For illustrative purposes, patients in Group A attend a virtual group educational session, Group B has a personalized session with a nurse, and Group C receives the usual care used by the clinic (Zwarstein, 2008). Patients are categorized into high, intermediate, and low-risk groups to account for different baseline risks and their impact on blood pressure and level of adherence to medications. In this example, high-risk patients are characterized by multiple comorbid conditions, such as diabetes, a history of stroke, and the prescription of three or more antihypertensive medications. Conversely, intermediate-risk patients present with one comorbid condition and are prescribed no more than two antihypertensive medications. Lastly, low-risk patients are those without any additional comorbid conditions and are only prescribed a single antihypertensive medication.

High- intermediate and low-risk criteria depend on the outcome under investigation and the available evidence (Porzsolt et al., 2013). Subsequently, different treatments/interventions are compared within similar risk groups; therefore, this study design offers the advantage that the outcomes of low-risk, intermediate-risk, and high-risk patients who receive the same intervention can be compared. Groups A and B of the PCT study examine the clinical outcomes of two new interventions tested in the primary care clinic using real-world conditions, wherein all the patients and healthcare providers can participate. In contrast, in the RCT model, testing interventions require the selection of the interventionists (i.e., healthcare providers to conduct the intervention), with specific requirements to serve in the role and, as mentioned previously, strict eligibility criteria for the study participants.

Including Group C, the usual care group, is crucial in minimizing selection bias by ensuring that the outcomes of all patients are reported. Group C encompasses all eligible patients seen in the primary care clinic during the study period, except those already included in Groups A and B. If the pragmatic controlled trial fails to establish a difference between the two interventions being investigated (Groups A and B), the results of these treatments should be compared with Group C. This extra measure is essential to validate the specificity of the outcome. In other words, the outcome data should demonstrate that Groups A and B, despite having similar effectiveness, yield superior results compared to the average outcomes of Group C.

Summary: Pros and Cons of RCTs and PCTs

Regarding the pros and cons, RCTs provide information about an intervention’s biological or physiological effects, allowing investigators to understand how and why an intervention may work. Additionally, RCTs control for factors that could influence the results, offering high internal validity. However, the strict control of variables and limited participant diversity can limit the generalizability of findings to broader, real-world populations (Lurie & Morgan, 2013). On the other hand, PCTs offer high external validity, meaning their results are more generalizable to a broader population and more diverse clinical settings. The diversity ensures that the trial’s findings apply to a wide range of patients (Gamerman, Cai, & Elsäßer, 2019). Furthermore, PCTs account for real-world variables like adherence to treatment, healthcare provider variability, and other factors that can impact the success of an intervention. PCTs provide valuable insights into how an intervention performs in ‘real-life’ conditions, which can inform clinical decision-making and health policy.

However, designing and conducting PCTs is not without its challenges. Ensuring an adequate sample size, managing the complexity of real-world settings, and accounting for numerous confounding factors require careful planning and execution. Moreover, the less controlled environment of PCTs can introduce more variables, potentially making it harder to establish cause-and-effect relationships (Lurie & Morgan, 2013).

Summary

In conclusion, explanatory and pragmatic trials play crucial roles in healthcare research. Understanding their differences and applications allows us to design and conduct trials with the most relevant and actionable evidence to guide clinical practice. The growing emphasis on evidence-based medicine and personalized healthcare has underscored the importance of PCTs. By providing a more accurate reflection of routine healthcare practices, PCTs contribute valuable data for making informed healthcare decisions, thus playing a crucial role in improving patient outcomes and healthcare delivery.