2
$\begingroup$

Medical student doing his first research here. This is a randomized controlled trial. Basically the study compares the effectiveness of skin glue over skin sutures. Since the length of each scar will be different in different patients, and I have the closure time as the primary outcome, how do I go about this?

2) Alternatively, can I calculate a sample size expecting a 50% difference in the primary outcomes? If so, how? Any link to a calculator for the same?

$\endgroup$
1
$\begingroup$

If you randomise them then the effect of scar length will, in the long run, balance out. You should record it though and you can ultimately do a regression analysis with scar length as a covariate which might add extra precision and would help to correct for any residual imbalance between the groups.

If your outcome is closure time you do not want to base your sample size on a difference between proportions as this will through away data.

$\endgroup$

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.