I think there is some misunderstanding here. It is when a refractometer is used to measure ABV that it needs to be calibrated for the family of beers. If you have a lab with a densitometer or pycnometer but don't want to have to do all that work for each gyle you come up with a calibration curve for each beer in the profile an can, using that, quickly estimate ABV, in the cellar, on each batch. Of course a modern lab isn't equipped with a densitometer. It is equipped with an Alcolyzer. Just put a sample bottle with a few mL in the carousel and ABV, TE, OG, ADF, RDF, AE... are all sent to your computer.
The problem with the use of refractometers WRT wort is that most of the time they read within 1 °P. But sometimes they are in error by more than 1, as much a 2 Bx or even more (an example would be the gent who read 2.6 on his tea with no sugar in it) and the only way to know is by checking against a hydrometer and if you do that then what's the point in having the refractometer? But if you are OK with potential error of 1 Bx most of the time and occasionally being off by more than that, a refractometer may be fine for you.