st: additive interaction after multiple imputation

["M. Bot" <M.Bot@uvt.nl>]

I was wondering if there is a way to calculate additive interactions (also
referred to as biological interactions) and their 95% confidence intervals
in STATA version 10.1. I am using a cox regression model in which I adjust
for several confounders. I found the syntax written by Andersson et al
(2005) in the paper entitled 'Calculating measures of biological
interaction' in the European Joural of Epidemiology. They use the delta
approach described by Hosmer and Lemeshow to calculate 95% confidence
intervals, based on the covariance structure in the model. This approach
worked well on my data at first, but after I used multiple imputation and
micombine, I cannot derive the covariance matrix (vce) anymore. 
Does anyone knows a solution for this, or an alternative method for
calculating the confidence intervals? Thank you for your help!

Best wishes,

Mariska Bot

PhD Student
Tilburg University
Center of Research on Psychology in Somatic diseases
Tilburg, The Netherlands


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/