wouterjohannes
2012-Jul-09 12:18 UTC
[R] additive interaction for a dichotomous dependent variable (i.e. risk difference)
Dear all, For my research I want to test additive interaction for a dichotomous dependent variable. Can anyone help me to estimate this in R? Wacholder describes this procedure in the American Journal of Epidemiology in 1986 (Binomial regression in GLIM: estimating risk ratios and risk differences). He, however, describes macros for GLIM and not for R. Others (Uher et al. Journal of Affective disorders. 2011) also use his method and describe it as follows: It has been proposed that gene?environment interactions (G?E) should be conceptualized as departures from additivity of risks between genetic and environmental factors, as such departures most likely correspond to biological causal mechanisms involving both genetic and environmental factors (Rothman et al., 2008; Schwartz, 2006). To follow this recommendation, we tested G?E in a generalized linear model from the binomial family with identity link estimating risk differences for binary outcomes (Wacholder, 1986). My question is: how can I test for additive interaction in R? In the glm package the binomial family and identity link do not seem to go together. Thanks in advance. Best regards, Wouter Johannes -- View this message in context: http://r.789695.n4.nabble.com/additive-interaction-for-a-dichotomous-dependent-variable-i-e-risk-difference-tp4635842.html Sent from the R help mailing list archive at Nabble.com.
sarahw
2012-Nov-16 01:39 UTC
[R] additive interaction for a dichotomous dependent variable (i.e. risk difference)
Did you ever get a response to this or resolve this yourself? Many thanks! -- View this message in context: http://r.789695.n4.nabble.com/additive-interaction-for-a-dichotomous-dependent-variable-i-e-risk-difference-tp4635842p4649687.html Sent from the R help mailing list archive at Nabble.com.
Uwe Ligges
2012-Nov-18 18:38 UTC
[R] additive interaction for a dichotomous dependent variable (i.e. risk difference)
On 16.11.2012 02:39, sarahw wrote:> Did you ever get a response to this or resolve this yourself? >You sent a message to the mailing list, but apparently not to the original poster. Uwe Ligges> Many thanks! > > > > -- > View this message in context: http://r.789695.n4.nabble.com/additive-interaction-for-a-dichotomous-dependent-variable-i-e-risk-difference-tp4635842p4649687.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > R-help at r-project.org mailing list > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >
Frank Harrell
2012-Nov-18 22:52 UTC
[R] additive interaction for a dichotomous dependent variable (i.e. risk difference)
In many situations the interactions indicated by additive risk models are false in the sense that they merely reflect restrictions so that risk are in [0,1]. I tend to prefer models that have no restrictions on the parameters. In fact, I think that a test of goodness of fit of a model can be made on the basis of comparing with another model and seeing which of the two have the smallest explained variation by interaction terms. Frank wouterjohannes wrote> Dear all, > > For my research I want to test additive interaction for a dichotomous > dependent variable. Can anyone help me to estimate this in R? > > Wacholder describes this procedure in the American Journal of Epidemiology > in 1986 (Binomial regression in GLIM: estimating risk ratios and risk > differences). He, however, describes macros for GLIM and not for R. Others > (Uher et al. Journal of Affective disorders. 2011) also use his method and > describe it as follows: > > It has been proposed that gene?environment interactions (G?E) should be > conceptualized as departures from additivity of risks between genetic and > environmental factors, as such departures most likely correspond to > biological causal mechanisms involving both genetic and environmental > factors (Rothman et al., 2008; Schwartz, 2006). To follow this > recommendation, we tested G?E in a generalized linear model from the > binomial family with identity link estimating risk differences for binary > outcomes (Wacholder, 1986). > > My question is: how can I test for additive interaction in R? In the glm > package the binomial family and identity link do not seem to go together. > > Thanks in advance. > > Best regards, > > Wouter Johannes----- Frank Harrell Department of Biostatistics, Vanderbilt University -- View this message in context: http://r.789695.n4.nabble.com/additive-interaction-for-a-dichotomous-dependent-variable-i-e-risk-difference-tp4635842p4649989.html Sent from the R help mailing list archive at Nabble.com.