Statalist The Stata Listserver

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

RE: st: ptrend for adjusted ORs

From   "Svend Juul" <SJ@SOCI.AU.DK>
To   <>
Subject   RE: st: ptrend for adjusted ORs
Date   Mon, 28 Aug 2006 17:56:43 +0200

After the exchange below, Jennifer asked:
Are you referring to the OR and P-value that is given after the ptl after the second logistic?
Yes, that's it. In the second command, -ptl- is handled as a continuous variable; it has the actual values 1, 2, 3, 4. Of course, the validity of the trend test depends on the shape of the association between -ptl- and -low-  -- as is the case for any trend test.
Jennifer wrote:
I am trying to calculate the ptrend for an adjusted OR.
I know that the command ptrend or ptrendi works for my
crude OR, but is there a way to include the other
variables that I am adjusting with.
Try this:
   webuse lbw.dta, clear
   xi: logistic low i.ptl i.race i.smoke
   xi: logistic low ptl i.race i.smoke
The first -logistic- command displays the 4-category
-ptl- as 3 dummy variables. In the second you get a
single OR, and it can be interpreted as a trend test,
adjusted for race and smoke.
Svend Juul
Institut for Folkesundhed, Afdeling for Epidemiologi
(Institute of Public Health, Department of Epidemiology)
Vennelyst Boulevard 6 
DK-8000 Aarhus C,  Denmark 
Phone, work:  +45 8942 6090 
Phone, home:  +45 8693 7796 
Fax:          +45 8613 1580 

*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index