Medical Advisory Service of Health Insurance Fund, Lützowstr. 53, D-45141 Essen, Germany
published online 22 February 2010.
Article Outline
To the Editor:
I greatly appreciate the work of the pragmatic–explanatory continuum indicator summary (PRECIS) group in framing, structuring, and illustrating the pragmatic–explanatory continuum [1]. In complete support and hopefully correct interpretation, I would like to sharpen the arguments by three short comments:
1.It is implicit from the presentation but should be made explicit: there is no such thing as “a pragmatic trial” or “an explanatory trial.” Schwartz and Lellouch's notion of “attitudes” was well chosen, and from the continuum and the illustrative wheel, it is clear that every trial will be positioned somewhere between the extremes and has its pragmatic and explanatory elements.
2.From the context, it is quite obvious what the authors have in mind while stating “explanatory trials test causal research hypotheses.” However, it should be made very clear that answers “that help users choose between options of care” of course are answers to questions of causal relationships—in a way that A changes the risk of B (by an amount of X). It is the “noise of practice” that differs pragmatic from explanatory trials, not the general aim of identifying and quantifying causal effects.
3.The point mentioned above is of major importance because design follows question and purpose. For causal questions, the randomized trial is the standard and this is not questioned by the term “pragmatic.” At least in Germany we are confronted with suggestions that “pragmatic” means the use of other research designs, nonrandomized observational research, or registers. By not mentioning such design aspects as one of the 10 items, the PRECIS-authors seem to share my position that such interpretations are misleading. A clear statement that such suggestions have nothing to do with “pragmatism” in the Schwartz/Lellouch sense would be useful.
Finally, the progress of work should address the use of placebo groups in a “pragmatic trial.” There is still much nonsense in “usual” or “standard” care, which can only be identified by placebo-controlled trials. A finding from a “pragmatic trial” that a new treatment is as good as usual care is logically correct but hardly of value for decision makers as long as they cannot be sure that usual care actually is effective. At least in such situations, placebos may have their place also on the pragmatic side of the continuum.
Back to Article Outline
Reference
- Thorpe KE, Zwarenstein M, Oxman AD, Treweek S, Furberg CD, Altman DG, et al. A pragmaticeexplanatory continuum indicator summary (PRECIS): a tool to help trial designers. J Clin Epidemiol. 2009;62:464–475
PII: S0895-4356(09)00351-5
doi:10.1016/j.jclinepi.2009.10.006
© 2010 Elsevier Inc. All rights reserved.
