proc surveyreg/ surveylogistic and age standardization

hey. thanks for the help.

i am attempting to age adjust data from a complex survey(nyc hanes) using proc surveyreg and then want to obtain a age adjusted OR using proc surveylogistic.


i am using this document as a example of what to do: http://www.bettycjung.net/Pdfs/Sasnhanes.pdf

i'm using 2000 census distribution to standardize.

so here is the code from the proc survey reg:

***********************************
PROC SURVEYREG DATA=ALL(WHERE=(20<=RIDAGEYR<=74));
CLUSTER SDMVPSU;
STRATA SDMVSTRA;
CLASS RACE AGE1;
WEIGHT WTMEC4YR;
MODEL HIGHTC = RACE AGE1 AGE1*RACE /SOLUTION;
ESTIMATE "MEX AMER" INTERCEPT 1 RACE 1 0 0 0 0 AGE1 0.4332 0.4062 0.1606
AGE1*RACE 0.4332 0.4062 0.1606 0 0 0 0 0 0 0 0 0 0 0 0;
ESTIMATE "NH BLACK" INTERCEPT 1 RACE 0 1 0 0 0 AGE1 0.4332 0.4062 0.1606
AGE1*RACE 0 0 0 0.4332 0.4062 0.1606 0 0 0 0 0 0 0 0 0;
ESTIMATE "NH WHITE" INTERCEPT 1 RACE 0 0 1 0 0 AGE1 0.4332 0.4062 0.1606
AGE1*RACE 0 0 0 0 0 0 0.4332 0.4062 0.1606 0 0 0 0 0 0;
ESTIMATE "OTHER" INTERCEPT 1 RACE 0 0 0 1 0 AGE1 0.4332 0.4062 0.1606
AGE1*RACE 0 0 0 0 0 0 0 0 0 0.4332 0.4062 0.1606 0 0 0;
ESTIMATE "OTHER HISPANIC" INTERCEPT 1 RACE 0 0 0 0 1 AGE1 0.4332 0.4062 0.1606
AGE1*RACE 0 0 0 0 0 0 0 0 0 0 0 0 0.4332 0.4062 0.1606;
ESTIMATE "NH BLACK - NH WHITE" RACE 0 1 -1 0 0
AGE1*RACE 0 0 0 0.4332 0.4062 0.1606 -0.4332 -0.4062 -0.1606 0 0 0 0 0 0;
RUN;

****************************************************

here is the code from the proc surveylogistic:


PROC SURVEYLOGISTIC DATA=ALL(WHERE=(20<=RIDAGEYR)) ;
CLUSTER SDMVPSU; STRATA SDMVSTRA;
CLASS SEX(ref='M') AGE1(ref=FIRST) RACE(ref='Non-Hispanic White');
WEIGHT WTMEC4YR;
MODEL HIGHTC(EVENT='1') = SEX AGE1 RACE ;
RUN;
************************************
I am just using the code from the pdf for simplicity's sake.

What I don't understand is how the proc surveylogistic is utilizing age adjusted data to run ' MODEL HIGHTC(EVENT='1') = SEX AGE1 RACE '.

1. Did the proc surveyreg make permanent changes to the dataset ALL(WHERE=(20<=RIDAGEYR))?

2. What did proc surveylogistic call to utilize or age adjusted data?

My eyes are like jelly right now.

Thanks for the help.