BMI is mostly useless metric when comparing genetically diverse cohorts. Weight gain is mostly in fat and that is proportional to fat cell count. There is a say

BMI is mostly useless metric when comparing genetically diverse cohorts. Weight gain is mostly in fat and that is proportional to fat cell count. There is a saying "nerve cells are born and die, but fat - lives forever". The truth to it is that at an adult age fat cell count is mostly constant and mostly genetic. As you get fatter you don't produce more fat cells, they just get fatter, pun intended. There is huge genetic diversity between geographic regions / ethnic lineages in this regard. On one end you have northern european / african lineages, on the other end you have far east lineages, with other lineages somewhere in between, with northern european / african lineages having the largest fat cell counts, east asians the lowest. Furthermore, north european / african lineages tend to have fat distribution much more biased towards subcutaneous fat, whereas east asian lineages are biased towards abdominal fat, so many of the problems associated with high body fat (not insulin resistance) are seen at lower body fat percentages in east asian lineages. On top of that, body fat percentage does not map to BMI. BMI may roughly linearly scale with body fat percentage around the "healthy" region, however there will be huge offset between genetic cohorts, including sex. You should expect east asian BMIs to be lower across the board given similarly "unhealthy" diets.