Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Analysis of non normal outcomes in a nested matched case control


From   "roland andersson" <rolandersson@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Analysis of non normal outcomes in a nested matched case control
Date   Fri, 13 Jun 2008 09:11:01 +0200

I have a large cohort of appendicitis patients operated with open and
laparoscopic surgery. The choice of surgical method is related to age,
sex, time period, stage of disease and comorbidity. The relations with
these confounders are very complex and I therefore want to make a
nested case control study to analyse differences in outcomes between
the surgical methods, with matching for age, sex, time period and
stage of disease.

The analysis of dichotomous outcomes is straightforward with
conditional logistic regression. But I need some help with the method
to analyse other outcomes like length of hospital stay. I assume that
I can make some transformation of the length of hospital stay in order
to get a normal distribution of this outcome. Are there linear
regression methods for matched data? What transformation can I use?
How can I transform the result back after the analysis to describe the
difference in length of stay?

Or may I calculate the difference in length of stay between the case
and controls within each matched strata and analyse this outcome?
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index