4
$\begingroup$

We say a (randomized) controlled clinical trial for a specific study design. Now my question is, what is controlled refering to?

Different things come up to my mind:

  • Controlled in the sense that the design and the process of the study is fixed before the study begins, so before the data is genereated i.e. before the patients receive the treatment
  • controlled in the sense that in a controlled clinical trial you always have a control group, in opposite to most observational studies, where there not necessarily a control group (in a case-control-study you do have a control group and this is also a observational study)
  • controlled in the sense that you control for unobserved confounders
$\endgroup$

2 Answers 2

3
$\begingroup$

The phrase is usually given the second meaning you list, that there is a control group to compare with the treated group. Ideally, the study is randomized, so the assignment of each individual to control or treatment is determined randomly. The idea here is your third meaning, to control for unexpected confounders.

$\endgroup$
1
  • $\begingroup$ To just encourage some discussion: confounding for what exactly? Selection of treatment? Duration or compliance with treatment? When an RCT inspects a drug that has never before been released, then nobody has taken that drug so is there even a concept of confounding at this point? $\endgroup$
    – AdamO
    Dec 24, 2017 at 18:59
2
$\begingroup$

I'm not sure that it can be pointing to the randomisation because RCT (randomised controlled trial) would then be a tautology. It must allude to the 'control' group. I'm sure the society for clinical trials could provide the answer. Interestingly, they changed their journal from 'controlled clinical trials' to simply 'clinical trials' when they switched publisher

$\endgroup$
2
  • $\begingroup$ +1. However, controlling of unexpected cofounders can be not only by randomization or blindness. $\endgroup$
    – ttnphns
    Dec 23, 2017 at 16:32
  • $\begingroup$ You can have head-to-head comparisons so I am not sure the control group is necessary. $\endgroup$
    – mdewey
    Dec 23, 2017 at 16:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.