Speed Gibson

The National Cancer Institute (as Bob from the Lung Assn keeps reminding us) estimates that 38,000 deaths a year are attributable to second-hand smoke. That's a remarkably precise figure for something so difficult to measure. I find that validating the NCI's overall credibilty may be as challenging.

Since this is a statiscal estimate, let us analyze it accordingly. By stating a value with a precision of 1,000, this implies that 38,000 is a demonstrably better statistical estimate than 37,000 or 39,000 would be. You don't have to be a doctor, scientist or actuary to understand why no such demonstration is even remotely possible. The variables involved simply vary too much, and most of these are masked by many more significant factors.

Do you add Dura Lube or STP to your engine oil? Assuming that their friction-reducing claims are true, why wouldn't you? Because even with any nominal brand of oil, the rest of the car will age far faster than the engine. The wheels will fall off long before STP can make a significant difference.

So it is with second-hand smoke. How much exposure do you get, both time and parts per million? What's your family medical history? Do you exercise? How old are you? What do you weigh? Did you ever smoke? Are you married? Do you eat healthy? How much do you drink? Trying to find the needle in this formidable haystack of competing factors is absurd.