How much harder is it to take a waist to hip than BMI though?
I obviously can't give you exact numbers since this is all hypothetical but say you have a *small* sample of 1000 people.
First consider how much longer would it take to recruit 1000 people who'd be willing to let you take their waist and hip measurements.
Then consider how you would take those measurements - how do you schedule the participants to come down to get their measurements taken? Do you go out to them to collect that information? How would you get there?
Then consider how much longer that would take, and how much more researchers you'd need, and how much more $$$ you'd need to cover costs.
And we are *only* dealing with 1000 people here.
... Actually before any of that you'd have to find researchers willing to do all that shit ... I was forced into helping with an epi study because I had to do it to finish my degree and it was the worst time of my life just trying to get people to give you their height and weight.
How would you do it for larger samples - say 100000 people?
Like I said, I'd prefer if we could move on from the BMI but it's lasted this long for a reason (because people are shit and practicality).
You have two measures of roughly equal 'difficulty' to obtain
We don't.
Edit. And in the above example we are only looking at collecting BMI/WHR but you'd most likely (I hope) be collecting other data depending on the study, and that just adds more issues to the above.