You know I initially typed out a long comment agreeing with you.. I was sure there was a lurking variable in that plot, but I looked up the study and sure enoug

You know I initially typed out a long comment agreeing with you.. I was sure there was a lurking variable in that plot, but I looked up the study and sure enough, about half are strictly caffeine related. Which makes their lines a touch misleading in my option, because in the entire other half of the data, caffeine is just one more point in a cocktail of alcohol and narcotics. So including them in the plot is a bit strange. So, there are quite a few caffeine only emergency visits, but if we're just going on caffeine issues, there's not enough in the study to see if there's really such a significant uptick in recent years for purely caffeine related issues. Also, note that a large portion of the visits simply list "adverse effects," not "misuse" (which I would interpret as over consumption). So who knows whether or not these are high doses, or simply people who happen to roll the short end of the stick in regards to caffeine tolerance.