It is clear that sampling a person for a serum value in one shot will
not give you a correct account of the population's serum distribution>
 You will not be able to say with the slightest certainty that the
prevalence of hyper-serum is %x.

Some have developed strategies for adjusting for intra-indvidiual
difference,
for instance,
-  Woteki CE. 1991. The importance of within-person variability in
estimating     prevalence. in Monitoring Dietary Intakes, I.
Macdonald, ed.
   Springer-Verlag, New York, pp. 99-109. 38. 
cites a method as:  Adjusted value= MeanX+(Xi-MeanX)(Sb/Sobs)

My own belief is that you *can* use single-shot serum values to
predict disease risk:  but you have to use group these with a value
that has areliable distribution: ex. group serum cholesterol by
subject's bodyweight, which does not vary appreciably among
individuals.

For instance divide your subjects into those with high bodyweight and
low bodyweight and use t-test to compare means of cholesterol levels. 
Although the distribution curve for cholesterol will be off, the mean
will be accurate.
What you cannot do is compare one-shot cholesterol to one-shot dietary
factors.

Is there any criticisms to the above: the t-test should be sufficient
to compare to groups and tell you exactly what the probability of
no-difference due to intraindividual difference is.   Also, would you
need to take into account power and sample size?

Any info, references (journals or books) appreciated.  
I have done a bit of investigating, the best article i have found is
"Statistical methods to assess and minimize the role of
intra-individual ariability" J Chron Dis ol 31, p 399-418  Which
directly addresses this issue of group means.  I am looking for more
current info.
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to