Early in my career I was involved in trials of antidepressant and anxiolytics where there was a placebo run in phase and placebo responders were eliminated. That approach is no longer used and there is only one paper I can find that looks at the difference between trials that use this approach and those that do not. I am looking for any references or FDA research or policies that I may not be aware of that addresses this issue.
Lee S, Walker JR, Jakul L, Sexton K. Does elimination of placebo responders in a placebo run-in increase the treatment effect in randomized clinical trials? A meta-analytic evaluation. Depress Anxiety. 2004;19(1):10-9. PubMed PMID:14978780.
This practice clearly would introduce a severe bias in the results. Just because some people respond to placebo early on does not mean that placebo resolves their problem. They easily can "relapse" in a week or two, reporting it worked at first but .... They also may, eg, report pain dropped from "terrible" to "bad" on placebo, then report a further drop to "moderate" as placebo run-in silently shifts to actual medication. If you leave out those who report a placebo effect, you are leaving out people with a different psychological make-up from the people you kept. Major journals probably would reject the manuscript out-of-hand as a biased sample.
From the above paper it probably does not because there is no difference in the results. There is also an associated nocebo effect with people dropping out because of side effects from placebo. I will query journal editors about this and see if it is in fact their policy. I think the fact that there has been no explanation of the phenomenon and rigid adherence to current methods leaves a large gap in the explanation about why we do trials the way we do them.
The FDA has attempted to look at the efficiency question. There are FDA documents on this process like the 2013 document: "Enrichment Strategies for Enrichment Strategies for Clinical Trials" where placebo lead in is described as one strategy.
Is this the paper you are referring to ?
https://www.ncbi.nlm.nih.gov/m/pubmed/14978780/
Seems to me that the selection bias that Ted explained well is the reason for lack of similar research.
Yes it is.
In their rationale the FDA points out that no clinical trial is a random sample of the population or even the problem under study. Selection criteria for inclusion/exclusion is the most obvious bias that limits generalizability.
- Badges
- Science topic
- Similar topics
- Biological Science
- Kinesiology
- Running