Comparing the change over time between two groups?

Husum

New Member
#1
Hi there.

Hope you can help me.
In a randomized (treatment vs. controls) control trial with two groups, I want to test the effect of the treatment over time.
The assumption is that treatment X will reduce variable Y to a certain degree.
Variable Y is measured at randomization and 1 year (also every 3 months, but that's not the endpoint).
What I want is to test if treatment X leads to a bigger reduction in variable Y than controls.
Ideally, it could be reported as a percentage decrease, but I need to know which test to use to compare this percentage difference between the two groups.

T-test on the percentage change? Is it that simple?

Thanks.
 
Last edited:

Karabiner

TS Contributor
#2
You achieve more statistical power if you include the baseline measurement
and perform a mixed analysis of variance with a repeated-measures factor ("time")
and a between-subjects factor ("group"). The interaction will tell you whether the
groups changed differently over time. You could use percentage differences to
further describe what happend.

With kind regards

Karabiner