Age and Insurance adjusted Proportion estimate--PROC SURVEYREG

Emma8 — Sun, 12 Apr 2020 14:12:39 GMT

Hi. I want to produce the age- and Insurance adjusted prevalence of Diabetes by Race.

I used this link:

https://www.cdc.gov/nchs/tutorials/NHANES/NHANESAnalyses/AgeStandardization/Task1b_SAS92.htm

My questions are: can I use the proportion of each group from the study cohort (my_data) instead of "Standard Proportions for 20-year Age Groups Based on the 2000 U.S. Census Standard Population"? The same for insurance, see below code. Do I have to adjust for the interaction term as in the link (race*age .3966 .3718 .2316 0 0 0 0 0 0 0 0 0;)?

Please note: Race is coded as 1 or 2; Age_group is coded as 1, 2, or 3; and Insurance is coded as 1 or 0.

Thank you.

proc surveyreg data = mydata nomcar;
weight study_weight;
class Race Age_group;
model DIABETES= Race Age_group Insurance/noint solution vadjust=none CLPARM;
ESTIMATE 'White' Race 1 0 Age_group0.25 0.40 0.35 Insurance 0.97 0.03;
ESTIMATE 'Black' Race 0 1 Age_group 0.25 0.40 0.35 Insurance 0.97 0.03;
run;

Re: Age and Insurance adjusted Proportion estimate--PROC SURVEYREG

StatDave — Mon, 13 Apr 2020 14:05:49 GMT

Direct and indirect standardization methods for rates and proportions are now specifically available in PROC STDRATE. See the discussion and examples in the documentation for that procedure.

topic Re: Age and Insurance adjusted Proportion estimate--PROC SURVEYREG in Statistical Procedures

Age and Insurance adjusted Proportion estimate--PROC SURVEYREG

Re: Age and Insurance adjusted Proportion estimate--PROC SURVEYREG