The more accurate headline would be "American mensa members suffer from more mental and physical disorders" because the data was taken exclusively from American Mensa Society members. I'm actually quite curious why they did this because this seems to be a somewhat obvious source of potentially severe bias. It implicitly assumes that the distribution of mensa members is representative/identical to the distribution of all people in the 98th percentile, and that seems unlikely to be true.

I think any study that focuses on IQ essentially needs to use data from an area with global compulsory IQ tests. And there are places within Scandinavia that hit the mark there: compulsory military enlistment that comes with an IQ test creates a perfect test bed for data.