Re: st: Analysis of longitudinal data from medical records

[Fred Wolfe <fwolfe@arthritis-research.org>]

There is no simple approach to medication issues in medical records 
in observational studies. Treatment data are often left and right 
censored, altered by dose adjustment and concomitant medications. 
Much depends on the goals of the analysis, the assumptions you are 
making, and the quality and amount of data. Standard data principles 
apply and you have to take it a step at a time. Mixed models do not 
alter these other concerns.

Fred Wolfe
National Data Bank for Rheumatic Diseases

At 09:22 AM 9/1/2005, you wrote:

> Can anyone recommend a good paper or text covering the
> analysis of repeated measures data from medical
> charts/electronic medical records?  I am working on a
> project using data from an EMR and am not sure if
> there is a conventional approach to handling issues
> like medication discontinuation and switching.  I have
> a text entitled "Applied Mixed Models in Medicine" by
> H. Brown & R. Prescott that has been of some help.
> Any other suggestions would be appreciated.
>
> --
> Jim
*
*   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/