프라그마틱 홈페이지 is a non-commercial open data platform and infrastructure that facilitates research on pragmatic trials. It is a platform that collects and shares clean trial data and ratings using PRECIS-2 which allows for multiple and varied meta-epidemiological studies that compare treatment effects estimates across trials that have different levels of pragmatism, as well as other design features.
Background
Pragmatic studies are increasingly recognized as providing real-world evidence to support clinical decision-making. However, the use of the term "pragmatic" is not uniform and its definition as well as assessment requires clarification. Pragmatic trials are intended to inform clinical practices and policy choices, rather than prove a physiological or clinical hypothesis. A pragmatic study should try to be as similar to actual clinical practice as is possible, including its recruitment of participants, setting up and design, the delivery and execution of the intervention, as well as the determination and analysis of outcomes as well as primary analyses. This is a major difference between explanatory trials as defined by Schwartz & Lellouch1, which are designed to prove a hypothesis in a more thorough way.
Truly pragmatic trials should not be blind participants or the clinicians. This could lead to a bias in the estimates of treatment effects. The trials that are pragmatic should also try to enroll patients from a variety of health care settings so that their results are generalizable to the real world.
Additionally, pragmatic trials should focus on outcomes that are vital to patients, like quality of life or functional recovery. This is particularly relevant when it comes to trials that involve the use of invasive procedures or potential dangerous adverse events. The CRASH trial29 compared a two-page report with an electronic monitoring system for patients in hospitals with chronic heart failure. The trial with a catheter, on the other hand utilized symptomatic catheter-related urinary tract infections as its primary outcome.
In addition to these characteristics pragmatic trials should reduce the procedures for conducting trials and requirements for data collection to cut down on costs and time commitments. Furthermore, pragmatic trials should seek to make their findings as relevant to actual clinical practice as possible by making sure that their primary method of analysis follows the intention-to treat approach (as described in CONSORT extensions for pragmatic trials).
Despite these requirements, many RCTs with features that challenge pragmatism have been incorrectly self-labeled pragmatic and published in journals of all types. This can result in misleading claims of pragmatism, and the usage of the term should be standardized. The creation of a PRECIS-2 tool that offers a standardized objective assessment of pragmatic features is a first step.
Methods
In a pragmatic study the aim is to inform clinical or policy decisions by demonstrating how an intervention can be integrated into routine treatment in real-world contexts. This is different from explanatory trials that test hypotheses about the causal-effect relationship in idealized conditions. Therefore, pragmatic trials might have lower internal validity than explanatory trials and might be more susceptible to bias in their design, conduct, and analysis. Despite these limitations, pragmatic trials can provide valuable information to decision-making in healthcare.
The PRECIS-2 tool evaluates the degree of pragmatism within an RCT by assessing it on 9 domains ranging from 1 (very explicative) to 5 (very pragmatic). In this study, the recruitment, organisation, flexibility: delivery, flexible adherence and follow-up domains were awarded high scores, but the primary outcome and the method for missing data were not at the practical limit. This suggests that a trial can be designed with well-thought-out pragmatic features, without compromising its quality.
However, it is difficult to determine the degree of pragmatism a trial is since pragmatism is not a binary attribute; some aspects of a trial may be more pragmatic than others. Additionally, logistical or protocol modifications made during a trial can change its pragmatism score. Koppenaal and colleagues discovered that 36% of 89 pragmatic studies were placebo-controlled, or conducted prior to the licensing. They also found that the majority were single-center. They are not in line with the norm and are only called pragmatic if their sponsors agree that these trials aren't blinded.
Furthermore, a common feature of pragmatic trials is that the researchers attempt to make their findings more meaningful by analysing subgroups of the sample. This can lead to unbalanced results and lower statistical power, increasing the risk of either not detecting or incorrectly detecting differences in the primary outcome. In Related Home Page of the pragmatic studies that were included in this meta-analysis this was a significant problem since the secondary outcomes were not adjusted to account for variations in the baseline covariates.
Furthermore, pragmatic trials can also have challenges with respect to the gathering and interpretation of safety data. This is due to the fact that adverse events are generally reported by the participants themselves and prone to reporting delays, inaccuracies or coding deviations. It is crucial to improve the accuracy and quality of the results in these trials.
Results
While the definition of pragmatism doesn't require that all clinical trials are 100% pragmatist There are advantages to including pragmatic components in trials. These include:
Enhancing sensitivity to issues in the real world as well as reducing cost and size of the study and allowing the study results to be faster implemented into clinical practice (by including patients who are routinely treated). However, pragmatic trials be a challenge. For instance, the right type of heterogeneity could help the trial to apply its results to different patients and settings; however the wrong type of heterogeneity may reduce the assay's sensitivity, and thus decrease the ability of a trial to detect minor treatment effects.
Several studies have attempted to classify pragmatic trials using a variety of definitions and scoring methods. Schwartz and Lellouch1 have developed a framework that can differentiate between explanation studies that confirm a physiological hypothesis or clinical hypothesis, and pragmatic studies that guide the selection of appropriate therapies in real world clinical practice. The framework was composed of nine domains assessed on a scale of 1-5 with 1 being more informative and 5 was more pragmatic. The domains were recruitment setting, setting, intervention delivery with flexibility, follow-up and primary analysis.
The original PRECIS tool3 was based on a similar scale and domains. Koppenaal et al10 created an adaptation to this assessment called the Pragmascope that was easier to use in systematic reviews. They discovered that pragmatic systematic reviews had higher average scores across all domains, but lower scores in the primary analysis domain.
This difference in the main analysis domain could be due to the fact that the majority of pragmatic trials process their data in the intention to treat way, whereas some explanatory trials do not. The overall score for pragmatic systematic reviews was lower when the domains of organisation, flexible delivery and follow-up were merged.
It is important to note that a pragmatic trial doesn't necessarily mean a poor quality trial, and there is an increasing rate of clinical trials (as defined by MEDLINE search, however this is not specific or sensitive) that employ the term "pragmatic" in their title or abstract. These terms could indicate a greater understanding of pragmatism in abstracts and titles, but it's unclear whether this is evident in the content.
Conclusions
In recent years, pragmatic trials are gaining popularity in research as the importance of real-world evidence is becoming increasingly acknowledged. They are clinical trials that are randomized which compare real-world treatment options rather than experimental treatments under development, they have patient populations that are more similar to those treated in routine care, they employ comparisons that are commonplace in practice (e.g., existing drugs) and depend on participants' self-reports of outcomes. This approach has the potential to overcome the limitations of observational research which include the biases associated with reliance on volunteers and the lack of availability and the variability of coding in national registry systems.
Other advantages of pragmatic trials are the ability to use existing data sources, as well as a higher likelihood of detecting meaningful changes than traditional trials. However, these tests could still have limitations which undermine their reliability and generalizability. For example the participation rates in certain trials might be lower than anticipated due to the healthy-volunteer effect as well as financial incentives or competition for participants from other research studies (e.g. industry trials). Practical trials are often limited by the need to enroll participants on time. In addition some pragmatic trials lack controls to ensure that the observed differences are not due to biases in trial conduct.
The authors of the Pragmatic Free Trial Meta identified 48 RCTs that self-described themselves as pragmatic and that were published up to 2022. The PRECIS-2 tool was employed to assess the pragmatism of these trials. It covers areas like eligibility criteria and flexibility in recruitment as well as adherence to interventions and follow-up. They discovered 14 trials scored highly pragmatic or pragmatic (i.e. scoring 5 or above) in at least one of these domains.
Trials with high pragmatism scores tend to have more lenient criteria for eligibility than conventional RCTs. They also have populations from many different hospitals. The authors argue that these characteristics can help make pragmatic trials more effective and relevant to everyday practice, but they do not guarantee that a pragmatic trial is free of bias. The pragmatism characteristic is not a fixed attribute the test that does not possess all the characteristics of an explanation study may still yield valuable and valid results.