andrea evangelista wrote: > Dear all, my aim is to estimate the efficacy over time of a treatment for > headache prevention. Data consist of long sequences of repeated binary > outcomes (1 if the subject has at least 1 episode of headache , 0 > otherwise) on subjects randomized to placebo or treatment. > > I have fit a logistic regression model with Huber-White cluster sandwich > covariance estimator. > I have put in the model the variables treatment (trt),sex,age and a > restricted cubic spline of time (days) to allow for non-linear treatment > effects. > > I use the functions lrm and robcov from R Design library: > > h<-lrm(head ~ trt*rcs(days)+ age+ sex,x=T,y=T) > h.rob<-robcov(h,id) > > I want to estimate treatment effect over time, then: > > k<-contrast(h.rob,list(day=1:240, trt=1), > list(day=1:240, trt=0)) > > xYplot(Cbind(exp(Contrast), exp(Lower),exp( Upper)) ~ day, data=k) #Plot > of treatment effects (odds ratio). > > The treatment group has a disavantage at the baseline ( for day=1 ,OR=1.16), > however at day=210 I can see a reduction of headache risk (OR=0.58) on > treatment group. > > How can I set to 1 the OR of treatment at the baseline (day=1) with R? In > case, is it corrent?
I don't know of an extremely simple way to do it. But the baseline may be noisy and I'm not sure I would recommend doing what you want. Frank Harrell > > Best regards > > Andrea Evangelista > Italy > > [[alternative HTML version deleted]] > > ______________________________________________ > [email protected] 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. ______________________________________________ [email protected] 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.
