Proper R command for comparison analysis of treatments

Question

I have to find out if the treatment group affects the blood sugar level (2 treatments are being compared). I want to correct for baseline variables (gender and weight of patient).

bslevel = continuous outcome parameter of blood sugar level
ttgrp = a factor variable of treatment group of patient 
gender = another factor variable: gender of patient
wt = a continuous baseline variable: weight of patient

I am not sure how to analyze this in R. Whether to use lm or aov and should I use '*' rather than '+' ?

I tried following but I am not clear which of following to use:

aov(bslevel ~ ttgrp + gender + wt, data=mydata)
aov(bslevel ~ ttgrp*gender + wt, data=mydata)
lm(bslevel ~ ttgrp + gender + wt, data=mydata)
aov(bslevel~ttgrp+Error(SubjectID/ttgrp),data=mydata)

Thanks for your help.

Answer 1

If you are interested in only the effect of treatment, then

aov(bslevel ~ gender + wt + ttgrp, data = mydata)

or

lm(bslevel ~ gender + wt + ttgrp, data = mydata)

will give you the same results.

If you are interested in examining whether there is an interaction between gender and treatment group, then

aov(bslevel ~ gender + wt + ttgrp + gender * ttgrp, data = mydata)

or

lm(bslevel ~ gender + wt + ttgrp + gender * ttgrp, data = mydata)

would work. The * indicates the interaction between two variables.

lm(bslevel ~ gender + wt + ttgrp + gender * ttgrp, data = mydata)   

gives the same results as

lm(bslevel ~ wt + gender * ttgrp, data = mydata)

because we cannot have an interaction between A*B without the main effects of A & main effects of B.

For studies with repeated measures, suppose wt is measured multiple times for each subject, then we would use

aov(bslevel ~ wt + Error(subject/wt), data = mydata)