Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: age adjusting complex sample survey data

From   Suzy <[email protected]>
To   [email protected]
Subject   Re: st: age adjusting complex sample survey data
Date   Tue, 10 Aug 2004 08:30:10 -0400

Dear Thomas,

I spoke with the Stata staff recently and also a number of other well informed individuals at the CDC who use Stata. Stata does not currently have this capability. In fact, the users of Stata use SUDAAN software for this portion of the analysis. We are waiting patiently for a Stata program! There are a number of "us" out here using Stata and working with complex survey data who have been trying to figure out if age standardization with complex survey data can be done.
Someone told me that we can probably hand calculate our age adjusted prevalence estimates after we obtain our crude estimates by the age-specifed distribution. To expedite this - you can use Episheet by Kenneth Rothman (type in episheet in google. It's the first webpage that comes up). This is a free excel program that allows one to do direct standardization for up to 12 strata ( I believe). This program, by the way, can do much much more!!

If you learn of anything further on this topic - do please pass it along to the statalist or to me direct!.

Hope this helps a bit.

Thomas Torok wrote:

I am using BRFSS (Behavioral Risk Factor Surveillance System) data and
would like to create age-adjusted prevalence estimates for several subgroups
with dissimilar age structures.

I can easily create unadjusted prevalence estimates using svytab and I can
control for age using svylogit to produce adjusted odds ratios, but that
doesn't provide me with adjusted prevalence estimates.

If these data were not complex survey data I would use the "dstdize"
command, but I am not sure if that command can be used in this case.

The CDC has published guidelines for age adjustment using the 2000 US
population ( and provides
code for Sudaan, but I have been unable to find any comparable guidance for



Tom T�r�k
E-mail: [email protected]

* For searches and help try:

*   For searches and help try:

© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index