Back to the Top
The following message was posted to: PharmPK
Hello,
Does anyone have a palatability questionnaire that you liked and can
share?
Are Visual-analog scales easier to analyze than responses to graded
questions (ie, 1 is very pleasant to 5 very unpleasant)?
We will be studying adults so don't need those smiley face
questionnaires!
Thanks
Susan Shoaf
Clinical Pharmacology
Otsuka Pharmaceutical Development
& Commercialization, Inc.
Rockville, MD 20850
Back to the Top
The following message was posted to: PharmPK
Susan, If you are doing cross group comparisons you really cannot use
something like a 1-5 scale since a 5 for one person might not be a 5
for another. There has been a lot of criticism of some of the pain
literature because of this issue and the same criticism would apply
here. If inappropriate scaling measures are incorrectly used, you can
skew your results.
Back to the Top
Hello Susan,
I don't have a palatability scale to hand but will try to find someone
at the Food Research Labs, Leatherhead who has used them. I would
expect that palatability will need several different questions and it
is the particular collection of questions that influences validity.
To answer your question on Visual Analogue Scales or 5 point Likert
scales, the risk is that people use VAS very differently. Some swing
from end to end, others hardly move from the centre and perhaps less
than a third use it the way that they are intended, that is evenly
across the whole scale. You can understand the conservative one who
won't move far from the centre because he keeps a reserve of scale in
case the symptom goes off the scale. The swingers probably do
genuinely vary but they may not have anticipated the differences that
occur during the trial. This means that it really is difficult for
anyone to use a VAS completely consistently throughout a study. This
eventually made us give up using them for pain because those
differences obscure the meaning of the results too much. A simple
Likert scale is much easier to use more consistently so you do
actually end up with better statistical results. They are the backbone
of long-validated scales like the Spielberger for anxiety for example.
It is best to avoid an odd number of ratings of a symptom or sign
because there is a tendency to favour the central category, so the 5
point scale you mention could be : None (which does not count for
preferences in this sense), Mild, Moderate, Marked and Severe.
Andrew Sutton, MD(London) FFA,
Consultant in Clinical Pharmacology
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Examples of palatability questionnaires for oral suspensions" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)