Relton et al 1 propose an interesting study design for pragmatic
trials which may overcome some of the drawbacks of other RCTs. However,
the cmRCT does seem to have several limitations:
• Like most pragmatic trials, it does not control for non-specific
effects of the experimental treatment 2.
• It is open to selection bias: only patients in the randomized
arm(s) need to consent to the experimental treatment. They are thus
selected and possibly different from the pool of all patients.
• The comparator treatment is ‘care as usual’; this means trialists
have little control over what exactly patients elect for (self-)treatment.
Thus the experimental treatment might be tested against a package of care,
the exact contents of which is unknown. Essentially one might compare one
unknown to another.
These drawbacks might mean that the cmRCT generates false positive
results. I can, for instance, imagine a pure placebo, like homeopathy,
coming out of such a test smelling of roses.
1) Relton C, Togerson D, O'Cathain A, Nicholl J. Rethinking pragmatic
randomised controlled trials: introducing the "cohort multiple randomised
controlled trial" design. BMJ 2010; 340:963-967.
2) Ernst E, Lee MS. A trial design that generates only ''positive''
results. J Postgrad Med 2008; 54(3):214-216.
Competing interests: No competing interests