How to run a Repeated measure ANCOVA adjusted for baseline values

[Flo Bzh]

Hi every one, Hope I can find some help here,

In a clinical interventional trial on physical activity training, we would like to compare the effectiveness of a Physical Activity Program (EXERCISE group) versus a Control Group (inactivity) on the physical fitness.
The trial lasts 2 months and the participants (n=36, randomized. EXE n=18 and CONTROL n=18) are evaluated at the beginning and at the end of the study (2 time points).

Analysis: Mean changes in physical performance from baseline using a repeated measures analysis of covariance (ANCOVA) model, adjusted for baseline value, including the effects of the group intervention and time (baseline and post-intervention).

My question:
1/ Does it make sence to adjust on baseline value ? or is it automaticaly assumed because the "Mean changes in physical performance from baseline" take already into acount ''baseline'' ... not sure if I am clear for you.
2/ I don't know how to program it in STATA in the command line panel nor in the ''statistics'' tabs at the top of the software.

can someone explain me both please

Thank you a lot for your help

[Joseph Coveney]

1/ Does it make sence to adjust on baseline value ? or is it automaticaly assumed because the "Mean changes in physical performance from baseline" take already into acount ''baseline'' ... not sure if I am clear for you.

You wouldn't do both. For a fuller discussion on ANCOVA versus repeated-measures ANOVA, Google harrell change score, and take a look at the first two hits that pop up.

2/ I don't know how to program it in STATA in the command line panel nor in the ''statistics'' tabs at the top of the software.

For ANCOVA, you could use anova or regress. For repeated-measures ANOVA, I you're probably better off with xtreg , fe or mixed. Take a look at their help files and user-manual entries for examples.

[Flo Bzh]

Thank you Joseph,
I think you already help me with a similar design and it was really helpfull.

Is the following lines are correct ? in order to analyse theMean changes in physical performance (VO2max) from baseline using a repeated measures analysis of covariance (ANCOVA) model, adjusted for baseline value, including the effects of the group intervention and time (baseline and post-intervention).

Physical performance = VO2max where VO2max1 and VO2max2 are the outcome at besline and at the end of the study respectively.
DeltaVo2max = VO2max2 - VO2max1

anova deltaVO2max Group c.VO2max1

margins Group

I will take a look at "harell change score"

Thank you Jospeh

[Inaamul Haq]

regress deltaVO2max i.Group VO2max1

The deltaVO2max should be adjusted for the baseline values.
A linear regression model is also okay. Although

mixed

will be an option as suggested in #2 above.
An excellent and easy-to-read discussion can be found in "Analysis of Data from Randomized Controlled Trials: A Practical Guide" by Jos WR Twisk".

[Flo Bzh]

thank you both of you !
F