Mantel-Haenszel test? Numerical vs categorical data

Hello, and thanks in advance for any help!

I'm testing my hypothesis that certified physicians are more likely to be younger than the overall physician population age in the state. However, I have no way of gathering numerical data on the overall physician age- All I can get is age-ranges. SO, I'm stuck- I've got 360 recorded ages for a group of certified physicians, but only 5 categories of ages for the population total.

Categorical data on physician ages in the state
<35- 2323
35-44- 4748
45-54- 4865
55-64- 4741
>=65- 4660

If I choose to make my data categorical, and thus be comparing two sets of categorical data, do I use a Mantel-Haenszel test? I am admittedly lost.