> Correlate that to health metrics at a wide scale. It'd be much more powerful than any currently existing study given the userbase likely to opt-in. Apple di

> Correlate that to health metrics at a wide scale. It'd be much more powerful than any currently existing study given the userbase likely to opt-in. Apple did run something like this in their heart research study. But not specifically counting drinks, just more survey like questions.