You're right, de la Chapelle is the more relevant condition in this discussion.

In general, though, chromosomal abnormalities make lines based on the presence or absence of particular chromosomes seem arbitrary.

I get that you're trying to simplify this by drawing a bright line. I'm sympathetic to that urge. Quick and easy sorting systems are often the most useful approach, even if there are a few false sorts on the margins.

But the goal is not always to get as bright a line as possible. Sometimes there are other considerations that are more important, like as GP noted, trying not to bias survey participation.

EDIT: Tone.