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

FW: st: age adjusted incidence rates in STATA

From   Nita Forouhi <[email protected]>
To   "'[email protected]'" <[email protected]>
Subject   FW: st: age adjusted incidence rates in STATA
Date   Fri, 4 Mar 2005 15:23:09 -0000

Is anyone able to answer a query about age adjusted incidence rates please?

I have a cohort of 1000 people free of disease xx at baseline, and at mean
f/up of 10 yrs, 61 develop the disease - so i want to present the cumulative
10-yr incidence of disease xx per 1000 pyar, and then present the same
adjsted for age, and for age and sex. So far i have not worked out how stata
can do that. So if crude cumulative 10yr IR is 8 per 1000 pyar, what is the
age adjusted and age/sex adjusted incidence in that cohort?

Does anyone know?

It was suggested that dstdize command could do this - but i have specific
questions on that as below; presumably poisson command might be used too? 

1. for dstdize - How can i tell STATA that it should use person-time data
(pyar) rather than population count?

2. Do i have to use an external reference standard population (say the
European standard population - or suggestions for others welcome), or can
the combined population (or person time) of the study itself by each age
category be used instead?

so say, in another example, crude incidence in cohort is: 4.1 per 1000 pyar
(total pyar =
39570), and in each of 3 age categories the incidence rates and person time
is as below (computed using strate command, for outcome = chd-death:

Age category	incidence rate per 1000		person-time	(yrs)
age 40-49		1.4					17700	
age 50-59		4.9					15500
age 60-69		9.3					6370	

Then how can i tell stata to give me age-standardised rate? 

If dstdize can be used, pls explain how exactly with pyar; also could
poisson or other options be used instead? which is better?

many thanks


*   For searches and help try:

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