My lab did not perform any of the indicated analyses. Labs are blinded to the subject IDs and associated information. You would need to cull the data from CDC database.
I believe there is laboratory data for NHANES 2007-8, 2009-10, 2011-12, and 2013-14. These cycles can be added, but there are only 4 cycles. There are limited persons for whom serum vitamin D was taken and even less for the glucose tolerance test. It sounds as if you would like to do an association study with glycemic markers and serum vitamin D. I suggest you work with someone who can combine 2001-2014.
There are alot of clinical markers available in the NHANES data. However, you have to remember that these individual data points represent randomized draws. Thus, whatever you choose to use, you'll need to use the accompanying weights to generalize to the overall US population. Additionally, these are cross-sectional data, so that you cannot stack the NHANES results over time, nor can you assume they are from the same individuals.