Seems silly to not, at least, split the population into those who made a choice to fast and those who couldn't eat. What an incredibly low signal to noise ratio

Seems silly to not, at least, split the population into those who made a choice to fast and those who couldn't eat. What an incredibly low signal to noise ratio. What causes people to pursue studies like this? Or stop so, objectively, early in the analysis? Money runs out?