any help to solve this is greatly appreciated:

The research question is to assess the superiority of Treatment A for estimulate bone growth in comparison with Treatment B, the current standard.

I'm planning to compare two treatments to estimulate bone growth in maxilary sinuses. For that, I will treat each patient with treatment A in one sinus and treatment B in the other.

Hence, my design is To randomize X patients to three groups with

  • Group 1: one sinus with Treatment A vs the other sinus with nothing
  • Group 2: one sinus with Treatment A vs the other sinus with Treatment B
  • Group 3: one sinus with Treatment B vs the other sinus with nothing

My first tought was to compare within each group with a paired T-test. Then I considered to compare the means with an one-eay ANOVA of repeated measures, with two measures for each patient. Is the ANOVA use right?

Which is the better way to compare means between treatments with repeated patients?

Any comment, answer or criticism is welcome.


My thought is to approach this with a mixed effects regression model with a random intercept for each person. I image the data set up as follows: id condition outcome 1 A 3.2 1 C 1.2 ... 30 A 4.2 30 B 6.2 ...

with two rows per person and a variable indicating whether that sinus received treatment A, B, or the control (C).

You could then model this in R with the nlme package as:

lme(outcome ~ condition, random = ~ 1|id)

  • $\begingroup$ Thanks, your solution is a better approach to handle this data. $\endgroup$ May 31 '17 at 23:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.