Pragmatic Clinical Trials: Implementation Opportunity, or Just Another Fad?

Pragmatic Trials Are Rare But Likely to Increase

Although lack of detail in trial reporting makes true quantitative analyses difficult,⁶ attempts at quantifying the percentage of published trials that are pragmatic indicate that pragmatic trials are infrequent—less than 2% of trials, according to Chalkicou and colleagues.¹ No doubt one major factor for the paucity of pragmatic trials is that most major funders of research, including the National Institutes of Health (NIH), historically have prioritized explanatory approaches, which by definition are not generalizable to broad clinical practice. Recently, high rates of unwarranted practice variability and increased expenditures have prompted a shift in attention to clinical trials that can provide more valid and reliable information about what works best in our health care system. Pragmatic trials are being emphasized by alternative funders, such as the Patient-Centered Outcomes Research Institute (PCORI) and the Agency for Health Care Research and Quality (AHRQ), under the premise that the results of such trials will be more translatable and usable in clinical practice and by policy makers.⁷

In the field of physical therapy, the increasing emphasis on health services research parallels national trends that prioritize the types of research that can aid decision making in emerging health care reform issues that are immediately relevant. One of the prime examples of this effort is the Center of Excellence in Physical Therapy Health Services and Health Policy Research and Training Grant (CoHSTAR), funded by the Foundation for Physical Therapy. One emphasis of CoHSTAR is on training in, and funding of, pilot work in comparative effectiveness research using more pragmatic approaches.

A more pragmatic approach appears to have gained widespread acceptance; however, some degree of resistance exists among—paradoxically—established clinical trialists within the field, many of whom possess rich funding and publication histories largely in explanatory trials where clinician and patient biases are by definition minimized through stringent entrance criteria, overt intervention oversight, and compliance strategies. Unlike practitioners, researchers indoctrinated into explanatory methods might find it challenging to accept the fact that clinician and patient biases are not viewed as detrimental in a pragmatic trial but rather accepted as part of providers' and patients' responses to treatment and included in the overall assessment of an intervention.

In addition to formal training of the next generation of health care researchers through initiatives such as CoHSTAR, the physical therapy profession has seen a conversion of established researchers who have modified their previous, more explanatory approaches to newer designs that are more pragmatic. In the initial PCORI initiative soliciting large simple pragmatic trials, 2 of the 5 awardees had physical therapists as the principal investigators (PIs). In addition, 4 other physical therapists are PIs in PCORI's broad initiatives focused on comparative effective research designs.

Broadly Inclusive Pragmatic Trials Versus Narrowly Focused Explanatory Trials: Both Problematic?

More pragmatic approaches can be natural extensions of explanatory research efforts and are consistent with the clinical implementation (T3) and health policy (T4) stages of the clinical translational research paradigm. According to the Blue Ribbon Panel on Rehabilitation Research at the National Institutes of Health (NIH), “There are tremendous opportunities to increase the clinical and societal relevance of rehabilitation research throughout the NIH by addressing the gaps in the continuum of translational research and the WHO-ICF [World Health Organization–International Classification of Functioning, Disability and Health] framework.”⁸ Soliciting pragmatic proposals that investigate rehabilitation interventions in multiple settings—including environments where the intervention is intended to generalize—would go a long way to bridging this gap.⁹

Some people have argued that extrapolating the results of broadly inclusive pragmatic trials to patient care may be as problematic as extrapolating the results of narrowly focused explanatory or efficacy trials.¹⁰ Consider the clinical prediction rule (CPR) for manipulation in patients with low back pain. The CPR for manipulation of the lumbar spine was developed by Flynn et al¹¹ as an observational trial and was validated in a subsequent randomized controlled trial by Childs et al.¹² A more pragmatic application by Hancock et al¹³ demonstrated the rule to have no predictive ability to identify responders to manipulation; this application included more relaxed entrance criteria (eg, subjects did not need to meet the same standard of acuity) and a more generalizable definition of spinal manipulation to include nonthrust mobilization techniques. Some have argued that increased pragmatism may have “washed out” the treatment effect; however, from a payer and policy-maker perspective, a more generalizable trial with a broader range of subjects—coupled with a more commonly used intervention (nonthrust mobilization)—provides a much more reliable and predictable effect.

In an effort to more accurately portray daily practice environments with minimal researcher interference, pragmatic trials typically require that data be derived from existing data sources, which usually includes the electronic health record (EHR). Integrating EHR data components that are critical to answering the aims of the trial allows for systematic data collection within the context of everyday practice and obviates the need to create an invasive data collection effort or a research infrastructure separate from the clinical environment. Thus, one promising development arising from pragmatic trial actualization is that EHRs are revamped to collect data necessary to conduct the study, including routine assessment of key variables such as patient-reported outcomes and other key process-of-care information that is deemed critical for clinical decision making.

The Real Barriers to Uptake in the Clinic

More pragmatic approaches that carefully consider the real-world context married to methodological rigor (eg, randomization) should provide information that is useful to providers, patients, payers, and policy makers. But is lack of generalizable published research the major barrier to uptake of research findings into routine health care? It seems doubtful that publishing more pragmatic trials alone will translate to more uptake of research findings into everyday practice. Even with greater numbers of relevant pragmatic trials, passive diffusion and similar implementation strategies will likely be unsuccessful in better aligning evidence with clinical practice behaviors. Methods to promote the uptake of research findings into routine health care in clinical, organizational, or policy contexts will still need to incorporate strategies that take into account more powerful drivers of clinical behavior, including financial alignment to bridge the transition from volume-based to quality-based initiatives.

In our effort to achieve the triple aim of health care reform (improving the patient experience of care, improving the health of populations, and reducing the per capita cost of health care), information from randomized clinical trials will remain a mainstay; however, as a profession, we will need a better balance between both explanatory and pragmatic approaches if we are to address the needs of patients, payers, purchasers, and policy makers.