[remram]

Those are very small variations. I would look for confounding factors, e.g. income levels, mental health, etc that could easily explain a few percent difference.

Looking at the full paper, specific factors were accounted for but the list seems kind of short for such a small statistical result. The paper acknowledge that racial minorities have a higher incidence but the data doesn't seem to contain income information... yeah.

Honestly I'm mostly surprised it came out so small.