Dear Statalists,
I would like to transform the estimates from a GEE model into estimates easy to interpret. I am analyzing the data from a therapeutic intervention in a pilot study with 26 individuals, randomized to either the treatment or a placebo. The outcome is a set of inflammatory proteins, which expression has been normalized as an arbitrary log2 scale. There are longitudinal measurements at weeks 0, 1, 2 and 3.
To evaluate the impact of treatment on each protein, I have used GEE models. For each protein, the code looks like this:
And the model output
The interaction term "treatment#c.week" indicates that this protein increases over time. In order to put it in context with the estimates from models for the other proteins, I would like to translate this 0.39 coefficient into something like this:
"Subjects in the placebo arm experience a X % (or X-fold) greater protein increase per week".
But, having a log2 transformed outcome, I am struggling to come up with the correct formula.
Thanks!
I would like to transform the estimates from a GEE model into estimates easy to interpret. I am analyzing the data from a therapeutic intervention in a pilot study with 26 individuals, randomized to either the treatment or a placebo. The outcome is a set of inflammatory proteins, which expression has been normalized as an arbitrary log2 scale. There are longitudinal measurements at weeks 0, 1, 2 and 3.
Code:
* Example generated by -dataex-. For more info, type help dataex clear input long id float log2_biomarker long treatment float week 1 -.7054 1 0 1 -1.2324 1 1 1 -1.1534 1 2 1 -1.8024 1 3 2 -1.8788 1 0 2 -.6176 1 1 2 -.9384 1 2 2 -1.0344 1 3 3 .1669 2 0 3 .3913 2 1 3 -.2084 2 2 3 .0628 2 3 4 1.8445 1 0 4 .7555 1 1 4 -.679 1 2 4 -.8782 1 3 5 -.7671 2 0 5 -.296 2 1 5 -.2317 2 2 5 -.3516 2 3 6 -.7551 2 0 6 -.6648 2 1 6 -.461 2 2 6 .3682 2 3 7 -.9501 1 0 7 -.1625 1 1 7 -1.5157 1 2 7 -.1009 1 3 8 .7412 2 0 8 .658 2 1 8 2.391 2 2 8 2.5516 2 3 9 -.2701 2 0 9 .446 2 1 9 .2264 2 2 9 -.4364 2 3 10 -.8445 2 0 10 -.9837 2 2 10 -.4747 2 3 11 -.6203 1 0 11 -.774 1 1 11 -.0134 1 2 11 -.5443 1 3 12 .2051 1 0 12 .0964 1 1 12 1.629 1 2 12 .0134 1 3 13 -.6022 1 0 13 -1.2432 1 1 13 -1.1496 1 2 end label values id id label def id 1 "R01", modify label def id 2 "R02", modify label def id 3 "R03", modify label def id 4 "R04", modify label def id 5 "R05", modify label def id 6 "R06", modify label def id 7 "R07", modify label def id 8 "R08", modify label def id 9 "R09", modify label def id 10 "R10", modify label def id 11 "R11", modify label def id 12 "R12", modify label def id 13 "R14", modify label values treatment grupo label def grupo 1 "FMT", modify label def grupo 2 "Placebo", modify
To evaluate the impact of treatment on each protein, I have used GEE models. For each protein, the code looks like this:
Code:
xtgee log2_biomarker treatment##c.week, family(gaussian) link(identity) corr(ar 1)
Code:
GEE population-averaged model Number of obs = 104
Group and time vars: id week Number of groups = 26
Family: Gaussian Obs per group:
Link: Identity min = 4
Correlation: AR(1) avg = 4.0
max = 4
Wald chi2(3) = 11.38
Scale parameter = 1.018093 Prob > chi2 = 0.0098
----------------------------------------------------------------------------------
log2_biomarker | Coefficient Std. err. z P>|z| [95% conf. interval]
-----------------+----------------------------------------------------------------
treatment |
Placebo | .1010699 .379534 0.27 0.790 -.6428031 .8449428
week | -.1841974 .125257 -1.47 0.141 -.4296967 .0613018
|
treatment#c.week |
Placebo | .3919614 .1771402 2.21 0.027 .044773 .7391497
|
_cons | .063295 .268371 0.24 0.814 -.4627026 .5892926
----------------------------------------------------------------------------------
"Subjects in the placebo arm experience a X % (or X-fold) greater protein increase per week".
But, having a log2 transformed outcome, I am struggling to come up with the correct formula.
Thanks!
