To provide context for those still using BMI as the sole 'fat or not' discriminator — the CDC published an n=9894 longitudinal study* about ten years ago; generally summarizing figure 1, the median starts around 0.53 +/- gender variance at 20-29, peaks at 0.62 +/- at 70-79, and then begins decreasing from there; however, they found that BMI fails to represent the 'fat' levels improperly (to our detriment) for people in their 40s (figure 2), as older human bodies tend to lose fat in areas (i.e. arms) that have no significant bearing on overall health, but gain fat in the abdomen (which other studies have shown does correspond to increasing cardiovascular issues). They show median waist-to-height-ratio data as 'monotonically increasing abdominal adipose tissue throughout the years of adulthood but decreasing mass in non-abdominal regions', which bodes very poorly for clothing manufacturers — because not only do you have to account for nine body shapes in women, but you also have to account for age skewing the waist-to-length ratios of the body shape further.

It would be particularly interesting to repeat this sizing study using the garment length to identify where it falls in 'height' median for women, and then identifying what 'age' median the garment's waistline is calibrated for. I can certainly guess what the results will be from personal experience on a per-retailer basis, and it would be a useful way to mathematically identify 'underserved niches' in today's market to target with appropriately-fit clothing (without a body scan).

* doi:10.1371/journal.pone.0172245 (2017) https://stacks.cdc.gov/view/cdc/44820

“BMI fails to represent the 'fat' levels improperly” - what is that supposed to say. I’m just done trying to understand things that the writer didn’t even read.