We had one TA for Mathematical Analysis class that was smart. Smarter than professors. Freakishly smart. Like, he could just look at the equation that others needed a full blackboard to solve and tell the solution straight away.
He drank. He looked like hell. Legend was that he was frequently found in ditches, completely drunk, looking at the stars.
Sad story.
I remember Feynman mentioning that once, when walking past a bar, he realized that he had considered going in and drinking (he was alone, it was in the middle of the afternoon), so he decided it would probably be better to quit entirely.
As far as I can tell from the summary (has anybody actually read it? I don't have a subscription...), the point the research makes is that people who have tested as having above-average IQs at the age of 10 also have a higher-than-average incidence of problems with alcohol as adults (meaning what, I'm not sure: arrests and/or medical problems, I assume). That could be read as, "smart people don't necessarily have smart coping behaviors".
I'm flabbergasted by the people in thread going, "Smart people have a statistically higher incidence of winding up dead in ditches due to alcohol poisoning! Go us!". But then again, the top level post was summarized as "smart people are drunks", so maybe I'm being naive in expecting people to respond to more than just a summary-of-a-summary in a post about alcohol.
Who actually said that? I'm sorry if the anecdote I mentioned was misinterpreted. It was just one data point. Just like the one about Feynman :).
BTW the original article had 8170 datapoints. And in the free abstract they tell that higher childhood mental ability scores were related not just to incidence of the problems but also with "to higher average intake of alcohol and to drinking more frequently".
As for your anecdote, sad indeed. I was also thinking of the post about Phil Katz's death on here a few weeks back. It's a horrible way to die.