Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: Survival Analysis


From   Steve Samuels <sjsamuels@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Survival Analysis
Date   Mon, 17 Jun 2013 22:07:18 -0400


The time frame of a retrospective review (your 1994 to 2012) is irrelevant to the survival analysis setup, which formally needs a start date, end date, failure indicator, and covariates.

The definition of a survival cohort usually begins with a range of start dates; an example :"all patients presenting at so-and-so clinic with a diagnosis of such-and-such, from dates A through Z, followed to death or to date W").
You may restrict to those whose records received the special review, but the important question is: what determined which records were reviewed?  Unless the reviewed records were  100% or, effectively, a random sample of those eligible, you risk selection bias. So more information on the definition of your cohort would be helpful. 

I am curious about where you are doing this work, as an internet search finds no links to your name except for two Statalist posts this month.


Steve
sjsamuels@gmail.com




On Jun 10, 2013, at 10:00 AM, Reese Andorfer wrote:

Dear All,

I will perform a Cox Regression and I am currently preparing my data;
The time span I'm looking at is from 1994 to 2012, individuals
accounted for which received a special examination within this time
frame.
So, it is know possible that an individual died before 1994 (let's say
starting in 1979 and died in 1986) , but I still have it as an
observation since it's case got reviewed in 1994 or later; and I
actually only care for its survival time and the associated
covariates, however I was wondering now if it would suffice that I
create a variable that indicates the year when the individual "died"
or would you recommend incorporating sth. else.


Cheers,
Reese
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/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/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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