Date:         Tue, 28 Dec 2004 09:47:53 -0800
Reply-To:     cassell.david@EPAMAIL.EPA.GOV
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         "David L. Cassell" <cassell.david@EPAMAIL.EPA.GOV>
Subject:      Re: How can I select matched controls for 1:m case-control study
In-Reply-To:  <cqs16t$1lod$1@mail.cn99.com>
Content-type: text/plain; charset=US-ASCII

zhang yewu <zhangyw@NCMIH.BJMU.EDU.CN> wrote:
> I have two data set, one is for patients(cases), and the another for
> non-patients(controls). the variables of two data sets are same. I
should
> pick out m matched nopatients for each patient. In each matched
subgroup, a
> patient and m non-patients should be same address, sex, and their age
> difference shoud be within 6 months.
>
> how can I make it?

I think that you should start by looking at the case-control matching
macros that are freely available from the Mayo Clinic website at this
URL:

    http://www.mayo.edu/hsr/sasmac.html

One or more of their matching macros will do just what you want.

HTH,
David
--
David Cassell, CSC
Cassell.David@epa.gov
Senior computing specialist
mathematical statistician