Dear all,
have you ever been in a situation where subjects in a Case-Control or
Cohort study are not independent?
This is the typical case of population isolates where almost all the
subjects or a very high percentage of them are strongly related.
I have a population of about 30,000 inhabitant, clustered in villages of
400-5000 people each, with strong familial aggregation.
Which strategies do you suggest to study risk factors for a disease in a
population like this?
How do you think non-independence could affect the effect estimates (I
guess power is reduced, but to which extent?)? Would you suggest any way
to face the problem?
Thanks a lot!
Cristian Pattaro
Unit of Genetic Epidemiology
Institute of Genetic Medicine
__________________________________________
EURAC research
Viale Druso/Drususallee 1
39100 Bolzano/Bozen - Italy
Tel. +39 0471 055 527 - Fax +39 0471 055 599
Email: [log in to unmask] <mailto:[log in to unmask]>
Website: www.eurac.edu <http://www.eurac.edu/>
PLEASE NOTE: THE ABOVE MESSAGE WAS RECEIVED FROM THE INTERNET.
On entering the GSi, this email was scanned for viruses by the Government Secure Intranet (GSi) virus scanning service supplied exclusively by Cable & Wireless in partnership with MessageLabs.
In case of problems, please call your organisational IT helpdesk.
______________________________________________________________________
This email has been scanned by the MessageLabs Email Security System.
For more information please visit http://www.messagelabs.com/email
______________________________________________________________________
|