All Downloads are FREE. Search and download functionalities are using the official Maven repository.

file.newsgroup.med.58966 Maven / Gradle / Ivy

There is a newer version: 0.500
Show newest version
From: [email protected] (Russell Turpin)
Subject: Re: Placebo effects

-*-----
In article <[email protected]> [email protected] (Daniel Prince) writes:
> Is there an effect where the doctor believes so strongly in a 
> medicine that he/she sees improvement where the is none or sees 
> more improvement than there is?  If so, what is this effect 
> called?  Is there a reverse of the above effect where the doctor 
> doesn't believe in a medicine and then sees less improvement than 
> there is?  What would this effect be called?  Have these effects 
> ever been studied?  How common are these effects?  Thank you in 
> advance for all replies. 

These effects are a very real concern in conducting studies of new
treatments.  Researchers try to limit this kind of effect by 
performing studies that are "blind" in various ways.  Some of these
are:

  o  The subjects of the study do not know whether they receive a 
     placebo or the test treatment, i.e., whether they are in the
     control group or the test group.

  o  Those administering the treatment do not know which subjects 
     receive a placebo or the test treatment.

  o  Those evaluating individual results do not know which subjects
     receive a placebo or the test treatment.

Obviously, at the point at which the data is analyzed, one has to 
differentiate the test group from the control group.  But the analysis
is quasi-public: the researcher describes it and presents the data on
which it is based so that others can verify it.  

It is worth noting that in biological studies where the subjects are
animals, such as mice, there were many cases of skewed results because
those who performed the study did not "blind" themselves.  It is not
considered so important to make mice more ignorant than they already
are, though it is important that in all respects except the one tested,
the control and test groups are treated alike.

Russell




© 2015 - 2024 Weber Informatics LLC | Privacy Policy