For the variable BMI, why are 65% of cases not in universe for the combined 2000-2013 IHIS? This seems unusually high for top and bottom coding to protect confidentiality.
As noted in the BMI variable’s Universe Tab, from 2000-2013 only sample adults were include in the BMI question universe. You can identify sample adults with the ASTATFLG variable. In order to generate nationally representative estimates from sample adult variables you will need to use the SAMPWT weight variable.
I hope this helps.