The Naked Scientists

The Naked Scientists Forum

Author Topic: How can we test the placebo effect?  (Read 905 times)

Offline thedoc

  • Forum Admin
  • Administrator
  • Sr. Member
  • *****
  • Posts: 511
  • Thanked: 12 times
    • View Profile
How can we test the placebo effect?
« on: 28/10/2015 14:50:02 »
Paul asked the Naked Scientists:
   Dear Chris,
My iPad keeps calling you Christ! You are bringing science to the world. Hallelujah !
I have just bee n reading the BBC app article on the placebo effect in the USA. It would be good to be able to compare that with the effect of placebos in tests done in the UK, Germany, Japan, China, and some country where they don't trust the health care system at all. Perhaps you could throw in a country where witch doctors still carry a lot of sway.
I was wondering if it were possible to do a test with the real pharmaceutical and the placebo at the same time? A trial list is told to take two pills each morning, one is the real pharmaceutical and the other is the placebo. Each has a different marker, whether ir be a dye or whatever. The urine and faeces are collected and it is then worked out if the pharmaceutical has reacted in the body or whether the placebo has reacted in the body. This doesn't necessarily determine whether the patient will " feel" better.
It is a pity some sort of nano robot couldn't sit in the gut or wherever this pharmaceutical is supposed to be working and monitor the effects.
If one volunteer is a pessimist and another is an optimist, that is likely to affect the result.
I can see where the organizers need to be clinical and honest and not say it is possible "cure".
What do you think?
« Last Edit: 28/10/2015 14:50:02 by _system »


 

Offline cheryl j

  • Neilep Level Member
  • ******
  • Posts: 1460
  • Thanked: 1 times
    • View Profile
Re: How can we test the placebo effect?
« Reply #1 on: 03/11/2015 03:57:58 »
I think the question makes no sense. The placebo is not supposed to "react in the body" at all. That's the point. It is, though, unfortunate that we confuse "controls" with placebos. There can be a number of reasons why subjects in the control group experience or demonstrate some effect - they get better because physiological healing processes that would have occurred anyway, they don't want to seem "negative" or disappoint their health care provider for whatever reason and report an improvement that hasn't happened, they have become more health conscious in general because of interactions with the medical system and start taking better care of themselves. These types of results have nothing to do with the placebo effect in terms of "mind of matter."
 

Offline evan_au

  • Neilep Level Member
  • ******
  • Posts: 4105
  • Thanked: 245 times
    • View Profile
Re: How can we test the placebo effect?
« Reply #2 on: 03/11/2015 09:45:08 »
Quote from: Paul
do a test with the real pharmaceutical and the placebo at the same time?
It is possible that this question originates from a misunderstanding of the way a placebo is used in a clinical trial.

Yes, a placebo is often used at the same time as the proposed pharmaceutical (and sometimes also at the same time as the current best pharmaceutical, but this increases the cost of the trial by 50%). But the different substances are given to different people. The people are randomly assigned to the pharmaceutical group or the placebo group.

In some cases, they run half of the trial this way, and then reverse who is on the placebo half way through the trial, to see if the benefits also swap over. But this doubles the duration and cost of the trial.
Quote
The urine and faeces are collected and it is then worked out if the pharmaceutical has reacted in the body or whether the placebo has reacted in the body.
By the design of a placebo-controlled trial, the patient, people administering the medicine, and those assessing the patient outcomes do not know who is on which medication. So you do not want to administer a test to work out who is on which medication!

The team who randomises the patients keep a record of who is on which treatment. At the end of the trial, the trial is "unblinded", and the outcomes compared for the pharmaceutical group and the placebo group.

Sometimes, the results are analysed at the midpoint of the trial (without telling the patients involved in the trial who is on which medication). In some cases, the results have been so clear-cut that the trial has been terminated early; either because the new pharmaceutical has bad side-effects, or because it is clearly superior to the placebo. In such cases, it is not considered ethical to give patients a treatment which is clearly inferior.

Unfortunately for the pharmaceutical industry, frequently the new, expensive medication does not perform much better than the placebo, so you need large, expensive studies to show any benefit at all.

See: https://en.wikipedia.org/wiki/Placebo-controlled_study
 

The Naked Scientists Forum

Re: How can we test the placebo effect?
« Reply #2 on: 03/11/2015 09:45:08 »

 

SMF 2.0.10 | SMF © 2015, Simple Machines
SMFAds for Free Forums