That's why I am so interested. I could see how getting blind drunk once a week might actually reduce total mortality. Perhaps those who get blind drunk only o

That's why I am so interested. I could see how getting blind drunk once a week might actually reduce total mortality. Perhaps those who get blind drunk only occasionally under planned conditions have fewer accidents than those who drink regularly and therefore spend more time under the influence. I agree that a measurement protocol would be difficult. But the OP was based on self-reporting. They could ask "Have you, in the last week, ever been above the legal driving limit?" or "What is the most you consumed in a single sitting this week?"