Should We Be Talking “Embedded” Rather Than “Pragmatic” for Real-World Clinical Trials?

?

For many years now—decades, really—real-world design enthusiasts have been extolling the virtues of the pragmatic clinical trial (PCT) as the most scientifically rigorous of all real-world research designs. They’ve foretold a future in which PCTs are a mainstay of evidence generation. (Full disclosure, I’ve been on board.) The reason is simple: because it incorporates randomization, the PCT is less susceptible to the bias and confounding common to studies based on non-randomized data, such as administrative claims and electronic health records (EHRs).

To date, however, the anticipated design benefits of the PCT have largely gone unrealized, as the approach has not been widely utilized—particularly among pharmaceutical manufacturers, who have exhibited a preference for less time- and resource-intensive approaches for RWE generation. The frustrating realization is that the design advantages of the PCT have failed to overcome the practical disadvantages of cost, risk, and time associated with its implementation.??

More recently, a parallel track has emerged in clinical trial innovation, focused on practical matters as opposed to design considerations. The proliferation of EHR systems across healthcare delivery settings has given rise to the opportunity to embed clinical research at the point of care and leverage the EHR for data capture to economize on overall trial burden. This approach has garnered interest from a variety of RWE policy groups, and the U.S. Food & Drug Administration (FDA) has signaled its support in various guidance documents.

I was thinking about this while taking another look at the PRECIS-2 diagram (BMJ 2015;350:h2147), which is more or less divided evenly between five design elements and four practical elements. A couple insights came to mind …

First, historical advocacy for the PCT can be seen as focused almost exclusively on the five design elements on the left side of the circle:

Second, it is only now that the practical aspects of the PCT depicted on the right side of the circle have come to the fore:

This got me wondering whether a shift in focus to embedding trials at the point of care might actually result in more pragmatism in design as well—completing the circle, if you will. I could envision a kind of domino effect of pragmatism playing out:

1. Embedding the trial in routine care improves the likelihood that the trial patients and physician-investigators are more reflective of those in real-world clinical practice.
2. Leveraging EHR systems increases reliance on routinely collected data and decreases requirements for protocol-mandated visits for data capture—taking advantage of smart phones and connected devices for remote data entry further reduces the need for patients to be seen in clinic.
3. This, in turn, results in fewer opportunities for direct enforcement of treatment compliance, allowing this important real-world mediator of treatment benefit to take effect.
4. Blinding of study subjects and physician-investigators to treatment assignment as well as use of placebo are particularly difficult to implement in routine care, thereby making active control and open-label treatment more common.

Given the positive momentum for so-called “embedded” or “point-of-care” clinical trials, along with the potential for such trials to be pragmatic in certain aspects, a course correction by advocates of the PCT may be warranted. Long-running arguments in favor of the scientific merits of PCTs have failed to move the field in this direction, so the focus should be shifted to the practical aspects of trial execution, as this practical focus could indirectly lead to the promised land where scientifically rigorous pragmatic trial designs are more the norm than exception.

Advocates of the pragmatic trial design have a new opportunity to move the field in this direction, but only if willing to extol the practical benefits of point-of-care trials as a means of getting there. We should indeed be talking “embedded” rather than “pragmatic” for real-world clinical trials.