Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: case-control study vs entire dataset


From   "Kieran McCaul" <Kieran.McCaul@uwa.edu.au>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: case-control study vs entire dataset
Date   Tue, 26 May 2009 05:11:25 +0800

..

It's difficult to know what the reviewer means by "a case-control
analysis" based on the information that you have provided.  

If the reviewer is suggesting that you match on some factors before
doing the analysis, then this is clearly wrong.

Matching is not done to control for confounding in a case-control study,
despite what some people believe.  Matching is done for efficiency: if
you know beforehand that a particular factor will be associated with
both exposure and disease, then matching on this factor will give you
more power for a given sample size. 

Matching is an element of the design of a study and is done when the
data is being collected.  If you already have the data, you gain nothing
by matching, other than possibly losing power if data ends up being
discarded after you have matched.

So, with the data already collected, multivariate analysis will allow
you to control for other factors.


______________________________________________
Kieran McCaul MPH PhD
WA Centre for Health & Ageing (M573)
University of Western Australia
Level 6, Ainslie House
48 Murray St
Perth 6000
Phone: (08) 9224-2701
Fax: (08) 9224 8009
email: Kieran.McCaul@uwa.edu.au
http://myprofile.cos.com/mccaul 
http://www.researcherid.com/rid/B-8751-2008
______________________________________________
Man is a credulous animal, and must believe something; in the absence of
good grounds for belief, 
he will be satisfied with bad ones. Bertrand Russell 

-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of moleps islon
Sent: Monday, 25 May 2009 10:51 PM
To: statalist@hsphsun2.harvard.edu
Subject: st: case-control study vs entire dataset

Getting feedback on a manuscript that was submitted from a
retrospective analysis of tumor characteristics and outcome the
peer-reviewer is requesting a case-control analysis. In this
manuscript we've used logistic regression and cox modelling. As far as
I understand using stocc for matching data would entail discarding
data. Is there any advantage of using a control-case instead of
multivariate regression analysis whereby other factors are controlled
for???

Regards,

Moleps
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index