My question is: Does anybody used Stata to test for additive interactions?
Details:
I would greatly appreciate to receive comments on the use of additive
interactions in statistical models.
Although I have been working on several studies in which I assessed (and
sometimes found) interactions between risk factors, I always tested for
interaction using multiplicative assumptions.
For example, let A and B be to risk factors in a case-control study:
I usually test for the interaction between A and B by inserting the
interaction term in the logistic model:
xi: logit depvar i.A*i.B
This is straighforward.
However, many colleagues believe that multiplicative interactions are just a
subset of interactions. Testing for additive interactions would allow to
test for any kind of interactions, and provide us with a finer
power-profile.
There are different ways to test for additive interactions:
1.
The simplier is to calculate the Sinergy Index (SI). SI is the ratio
Observed/Expected, where Observed is the OR for the joint effect of A and B
(OR_AB) and Expected is obtained from the main effects of A and B as
follows: (OR_A+OR_B-1). If SI is significantly different from 1, then we
have an additive interaction. I believe it could be possible to go this way
with Stata, by using "lincom" after my usual logistic model (i.e., xi: logit
depvar i.A*i.B, or). SI, in the end, is just a combination of ORs. Has
anybody ever tried this?
SI, however, has been criticised and other methods are often prefered
2. Additive relative risk model (as implemented in Epicure)
Is there a way to do this in Stata?
I hope this was clear enough. Any suggestion/trick/experience on the issue
would be most welcomed.
Thanks,
Andrea
Andrea Baccarelli, MD, MPH, PhD
Genetic Epidemiology Branch
Division of Cancer Epidemiology and Genetics
National Cancer Institute, NIH, DHHS
6120 Executive Blvd. - EPS 7110
Bethesda, MD 20892-7236
Tel. (301)-496-5786
Fax (301)-402-4489 [email protected]
