Dear Paul,
It is like pulling teeth to work with a balanced design in this day and
age, unless you have good access to recrute as many patients as you need ,
particularly after the manuscript was revised. I have experienced this
problem before.
Well, my lowly view is that that analysis can, if fact, be done, but would
be more a sensitivity analysis than a major inference.
As Carlo mentioned, it is important to keep in mind the theorical
hypotheses behind that analysis.
We have done this kind of approach to analyze genetic data and, if you
want, I can aid you with some codes to run the analysis.
All the best,
Tiago
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
