 |
 |
Thank you for your help !
I did not mention age because I assume obviously that grey matter volume is highly correlated with age.
Concerning sex and ICV :
- I assume that grey matter volume is correlated to head size (ie ICV) ;
- I assume that differences between male and female grey matter volumes are related to differences of head size between these two subgroups.
-> so I could use either variable, hence my question... Unless gender could influence grey matter volume through other mechanisms than head size ?
I will follow your advices,
Best,
F
Hi Franz,I agree with Rosalia. Sex and age are used as covariates because they usually correlate to gray matter volume. You can test this in your data, but it will depend on the distribution of the variable. Then you should analyze with and without ICV. You use sex, age and ICV in the model because those account for some of the variance and you want to remove that effect from your drug effect. It is not a rule to use ICV but because it may be correlated to gray matter volume, you need to make sure it is not explaining your drug effect. Check for linear and non-linear effects in your data and I suggest to do both (with and without ICV) and report both even if one is not significant. If the drug effect disappears with ICV included, then you would need to investigate or discuss further.Kind regardsEduardoEduardo A. Garza-Villarreal, M.D., Ph.D.
- CONACYT Research Fellow, Investigaciones Clínicas, Instituto Nacional de Psiquiatría, Mexico.
- Visiting Senior Researcher, Center of Functionally Integrative Neuroscience, University of Aarhus, Denmark.
On Tue, Aug 23, 2016 at 11:07 AM, Rosalia Dacosta Aguayo <[log in to unmask]> wrote:HI Franz,As far as I know, according to literature gender, age and intracranial volume are used as covariates or regressors not only intracranial volume and gender.I think you are assuming different things: let's go further with your assumption: by the same way you say that just taking total intracranial volume could account for gender differences...by it could not be supposed that total intracranial volume can account for age differences regarding gray matter atrophy?Account only for TIV is just one part of the history. I would recommend you inspect your data and see if they are differences regarding age, and gender (separately) and then take your own decission based on your sample. But TIV does not account for all the differences you can explain, the add information to the regression model...and this is why in the literature it is recommended to use those variables as covariates.Hope this helps,Kind regards,Rosalia2016-08-23 17:43 GMT+02:00 Franz Hozer <[log in to unmask]>:Dear all,
I'd like to test for influence of lithium treatment on grey matter volume using standard multiple regression.
According to literature, gender and total intracranial volume are systematically used as covariates.
My question is as follows : why use these 2 covariates together ? Could we suppose that gender differences in the size of brain structures are related to differences of ICV between male and female, so only one of these 2 covariates should be used ?
Thank you for your help,
Franz
--Rosalía Dacosta-Aguayo, Post-Doctoral Researcher.
Biomedicine, PhD
Clinical Neuropsychologist
Department of Neurosciences.
Group of Neurodegenerative Diseases
Gaixotasun Neuroendekapenezko Taldea
Neurozientziak Arloa
[log in to unmask]g
T. +34 943 006 128
IIS BIODONOSTIA - OSI DONOSTIALDEA
Pº Dr. Beguiristain s/n * 20014 Donostia-San Sebastián SPAIN * T.+34 943
006 012 * F. +34 943 006 250 * [log in to unmask] *
www.biodonostia.org