| |
 |
|
|
Science Forum Index » Space - Consult Forum » Repeated measurements & logistic regression. How can I inte
Page 1 of 1
|
| Author |
Message |
| Guest |
 Posted: Thu Jan 18, 2007 12:57 pm |
|
|
|
|
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 cluster sandwich
covariance estimator.
Initially, I have put in the model only the variable treatment (trt)
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),x=T,y=T)
h.rob<-robcov(h,id)
For day=1:
summary(h.rob,day=1)
the OR of treatment is 1.16(0.72 to 1.89,95%CI).
For day=210
summary(h.rob,day=210)
the OR of treatment is 0.58(0.33 to 1.01,95%CI).
How can I interpret it? The treatment group has a disavantage at day=1
(OR=1.16), however at day=210 I can see a reduction of headache risk
(OR=0.5 , but I
have a p-value>0.05. Question: has the treatment an effect on the
headache prevention?
Can I detract 0.16 from the OR at day=210? Is it a "swear"?
Best regards,
Andrea Evangelista |
|
|
| Back to top |
|
| |
|
Page 1 of 1
All times are GMT - 5 Hours
The time now is Fri Aug 29, 2008 10:33 pm
|
|