Measures of interaction contrast (biological interaction) - ic, ici and icp

Publikation: Bidrag der ikke har en tekstformSoftwareprogram

It has become more common to investigate not only single factors effect on a outcome but also to look at the interaction between factors, as is facilitated by the last decade’s massive increase in computer resources to gather and analyses large databases.

Several approaches haves been promoted.For the analyzes of binary (or count) data, a fairly well established position holds, that loglinear models estimating measures of relative risk type represent a convenient choice, whereas interactions are often best interpreted if estimated on a linear/additive scale.

This code implement the procedure described in Hosmer & Lemeshow (1992) and Alfredsson et all (2005) purely in Stata.These two sources however both uses odds ratios as approximations for relational risks. This code on the other hand opens up for using proper estimates of relational risks.

To install type "ssc install ic" in the Stata command


Andersson T, Alfredsson L, Källberg H, Zdravkovic S, Ahlbom A.

Calculating measures of biological interaction

Eur J Epidemiol. 2005;20(7):575-9.


Hosmer, David W., Lemeshow, Stanley,

Confidence Interval Estimation of Interaction,

Epidemiology, Vol. 3, No. 5 (Sep., 1992), pp. 452-456

Udgivelsesår9 mar. 2015
StatusUdgivet - 9 mar. 2015

Se relationer på Aarhus Universitet Citationsformater

ID: 85445854