The Interdisciplinary Rehabilitative Research Seminar Series welcomed Steven Z. George, PhD, FAPTA, on January 28 for an engaging presentation titled “The Continuing Trials of Physical Therapy and Rehabilitation Clinical Trials, Part 2 (The Sequel).” George—Laszlo Ormandy Distinguished Professor of Orthopaedic Surgery, Professor of Population Health Sciences, member of the Duke Clinical Research Institute, and Editor‑in‑Chief of Physical Therapy & Rehabilitation Journal—used the “sequel” framing to underscore the iterative nature of science and the need to revisit foundational questions with better methods and stronger data.
A core message of the talk was that clinical trials must answer two essential questions: (1) Does a treatment work (efficacy/effectiveness)? and (2) For whom does it work best (heterogeneity of treatment effects)? George emphasized that while many clinical practice guidelines prioritize overall efficacy and evidence grading, they often fail to account for patient characteristics and subgroup variability—largely because too few studies are designed with sufficient rigor and sample size to support credible subgroup conclusions.
“High‑quality clinical trials don’t just tell us whether a treatment works—they help us understand who benefits most, so we can deliver care that truly fits the patient.”
To illustrate what can hinder progress, George discussed the prevalence of “small crappy trials,” a pointed phrase that gained wider attention through Robert M. Califf, MD, who raised concerns—via a White House Office of Science and Technology Policy panel—about fragmented, underpowered trials that are too small to generate actionable evidence. In the presentation, George noted common features of these studies (often fewer than 100 participants, limited follow-up, and weak statistical approaches) and how mismatches between planning frameworks and interpretation can leave the field with low-confidence conclusions and wasted effort.
As a model for “moving beyond” these limitations, George highlighted the AIM-BACK Trial, a large, cluster-randomized pragmatic study conducted across 17 VHA primary care clinics that enrolled 1,817 Veterans. The trial compared two non-drug care pathways—Sequenced Care Pathway (SCP) and Pain Navigator Pathway (PNP)—and used pain interference and physical function as co-primary outcomes. Importantly, the study also pre-specified subgroup analyses to examine treatment effect heterogeneity, including high-impact chronic pain status and opioid use at trial enrollment.
George shared takeaways that underscore why design matters: overall, there was no pathway superiority, with SCP and PNP demonstrating similar, clinically meaningful improvements in primary outcomes. At the same time, subgroup findings suggested potential opportunities to better target care delivery (including signals among Veterans without high-impact chronic pain assigned to SCP) while opioid use did not appear to meaningfully alter outcomes between pathways.
Closing with a practical call to action, George outlined what it takes to credibly answer “for whom” questions—pre-specifying subgroups based on prior evidence, increasing sample size to support those analyses, using appropriate interaction-based methods, and applying ICEMAN guidance to strengthen the credibility of effect-modification claims. His remarks reinforced a central theme of the seminar: rehabilitation research can deliver more clinically actionable evidence when trials are designed to be both rigorous and patient-centered—and when the field commits to moving beyond studies that are simply too small to answer the questions that matter.