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

RE: st: associating mortality with potential predictors when the rateis high

From   uri goldbourt <>
Subject   RE: st: associating mortality with potential predictors when the rateis high
Date   Wed, 01 Aug 2007 09:27:25 -0400

How about when your endpoint is stroke incidence, your purpose is to
identify risk factors or rather quantify their association with the endpoint
and it is, say, a 40-year follow up of middle aged individuals. Over 40
years many will die from other causes, free of stroke, but the stroke risk
associated with their risk profile has not accounted for.

By the way, I purchased your book three months ago and have enjoyed reading
it and keep getting back to it. I have several ideas how that could be
expanded or intensified, so to speak, but then you may have not intended to
go into all the depths of epidemiological analysis in a book probably meant
to be introductory.
(my stata knowledge lags way behind my epidemiology and the book was


-----Original Message-----
[] On Behalf Of Svend Juul
Sent: Wednesday, August 01, 2007 3:15 AM
Subject: Re: st: associating mortality with potential prdeictors when the
rate is high

Uri G wrote:

When a follow up on a cohort reveals that, say, 70% have
experienced an endpoint, the use of most survival models
(DH cox's etc) is unreliable (failure to consider competing
risks aside).
Does stata have programs using more appropriate models?


What makes you believe that most survival models are
unreliable in this situation? Could you give a reference
(complete, please!)



Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6
DK-8000  Aarhus C, Denmark
Phone: +45 8942 6090
Home:  +45 8693 7796

*   For searches and help try:

*   For searches and help try:

© Copyright 1996–2022 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index