Radical Thinking: Scientific Rigor and Pragmatism

Document Type

Journal Article

Publication Date



Statistics in Biopharmaceutical Research




Benefit-risk; Clinical trials; Desirability of Outcome Ranking (DOOR); Intention-to-Treat; Pragmatism


There is an innate desire to conduct research faster and cheaper. The desire is magnified by todays’ business and political pressures. Though understandable, such desires can be dangerous threatening our objectivity and ability to reason, resulting in studies with lower integrity, replicability, and applicability. Modern ideas include: conducting observational rather than randomized studies rationalized by the increasing access to real world data and the belief that modeling can replace randomization, using surrogate rather than clinical outcomes, using surrogate rather than clinical diseases, using PP (Per Protocol) analyses instead of ITT (Intention-to-Treat), using assumptions regarding treatment effects in place of data to estimate those effects, and using adaptive designs that promise efficiencies. These trends are often labeled as “innovation.” But is this progress…or regress…a fancy way of lowering the usual evidentiary standard and introducing greater uncertainty? Is it time for a market correction for these “products” that are often presented with a degree of commercialism rather than scientific objectivity? Indeed, our traditional approaches are in need in improvements. Rather than compromising on scientific rigor, can we redirect our motivation to find BETTER answers to the most important questions for patients and clinicians? Though often misunderstood, one such movement is pragmatism, the concept of truly getting closer to understanding the effects of interventions as they are experienced by patients and the value of diagnostics in real world practice. I propose that we place increased interest on questions of a pragmatic origin to match their clinical importance and utility. I present ideas for more pragmatic assessments of the effects of interventions and the yield of diagnostics.