Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Selecting subjects


From   Svend Juul <SJ@SOCI.AU.DK>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Selecting subjects
Date   Thu, 13 Mar 2008 14:17:06 +0100

Ziad wrote:

Variable 'visit' can have one of the values
 0,2,4,8,12,16,20,24,36, 52 or 999

variable 'week' has the actual week number the patient came in. when
it is unscheduled visit then visit will be 999 and the week will have
the actual number.

Some patients could have come on scheduled and unscheduled visits.
i am interested in knowing
 1. which patients came on all the scheduled visits (it does not
matter if they came in between visits or not)
 2. which patients missed scheduled visits (i.e. they did not come on
the shceduled and on unscheduled ones)
 3. which patients missed scheduled visits but they came on unscheduled
ones?

========================================================================

To avoid confusing myself, I renamed your variable -visit- to -pweek- 
(planned week for visit) and -week- to -aweek- (actual week for visit).
I assume that a missed visit still exist as an observation in the 
dataset.

In the example below, id 1 is perfect, id 2 missed a visit, id 3 had an
unscheduled visit, and id 4 had missed one and had an unscheduled visit.



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