Calculating BMI percentile using user written command "zanthro"

[Sam Lin]

Dear Statalist,

Does anyone know whether the bmi zscore/percentile calculated from the user written command (egen zanthro) provides similar results to CDC's SAS scode and reference data?

http://www.cdc.gov/nccdphp/dnpao/gro...ources/sas.htm

Any assistance on this would be appreciated.

Best,

[Megan Lehnerd]

Hi Sam,

Sorry, I don't have an answer, but I've been trying to track down more info about using the zanthro and zbmicat commands. Do I need to install an additional program to have those commands work?

Thanks!

Megan

[Clyde Schechter]

@ Sam: That is certainly what it is supposed to do if you specify that you want CDC norms when you invoke it. I have only used -zanthro()- a couple of times, and I have not encountered any results that didn't seem right, but I have never explicitly compared them to the output of CDC's SAS code.

@Megan: -zanthro- is a user written command. To use it you have to first install it (-net describe dm0004_1, from(http://www.stata-journal.com/software/sj13-2)-). One note of caution: there are a large number of ancillary data sets that get downloaded with the programs, and you need them. So don't forget to get those also: it's a separate step in the installation. Also, they may not end up in the right place. So after you get those files, check to make sure that they have landed in the same directory as _gzanthro.ado (which, in my Windows set up is c:\ado\plus\_). If they didn't, you need to move them there. Otherwise when you try to use -zanthro- you will get a bunch of "file not found" errors.

[Sam Lin]

Clyde Schechter Thanks for your valuable input. Sorry for the very late reply. I actually ended up comparing my results to the SAS result that a colleague ran for me. The result match despite the decimal were off by a little bit.

Megan Lehnerd: Hope you found the "zanthro" command useful.