Date: Wed, 22 Oct 2003 09:27:46 -0400
From: Jay Kaufman <[email protected]>
Subject: st: Re: Population attributable percentage
Anthony Gichangi wrote:
>
> There exist several ways to compute the 95 % CI for the populatioon
> attributable risk depending
> on the type of study. A good reference is .
>
> Comparisons of Confidence Intervals for Attributable Risk
> Hoi M. Leung; Lawrence L. Kupper
> Biometrics, Vol. 37, No. 2. (Jun., 1981), pp. 293-302.
This article also contains a method that can be readily applied
to the PAR%, as demonstrated in their examples:
Am J Epidemiol. 1998 Apr 15;147(8):783-90.
Confidence limits made easy: interval estimation using a substitution
method.
Daly LE.
Daly's approach has been criticized because it does not take into account
variation in the estimated prevalence of exposure. See Letter to the Editor
by Greenland S, American Journal of Epidemiology 1999; 149(9):884, which
gives a formula that does.
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
