A number of years ago, I was seated next to a physician at a dinner party. In the course of our conversation, I mentioned that my mother had had breast cancer."Oh,well, you'll most likely get it, too," he said. For a moment, I sat, stunned, not quite sure what piece of his comment to respond to first. The incorrect conclusion? (among other things, my mother had post-menopausal cancer, which is a different disease, with a far less clear genetic link, than the pre-menopausal version.) The lack of professional approach or tact? The arrogance? 

"Well," I managed, after a pause, "I think it's a bit more complex than that." 

"Well," he conceded gruffly, "it is multi-factorial." 

Indeed. Which presents a challenge to researchers and doctors--and to any of the rest of us who want very much to know which thing to do or food to eat so that we can avoid the misfortune of cancer, heart disease, or other life-threatening diseases. It's a challenge not only because pinning down any particular cause or high-risk behavior is then harder to do, but also because it increases the odds of correlation not equalling causation. Researchers can note that people who eat four rutabagas a day have a lower rate of cancer. But those people might also all live outside cities, drink well water, eat 3 pounds of pasta a week, receive 30 minutes of sunshine a day growing those rutabagas, and sleep at least 7 hours every night. Which factor, or combination of factors, is important for avoiding cancer? Or could the answer be "none of the above"? Maybe the truth is, those people all have good genes. Or good luck. 

Clearly, research into links between risk factors and disease is important. The discovery of the link between smoking and lung cancer has saved untold lives by discouraging people from continuing, or taking up, smoking. But sometimes it seems the intensity of our drive to "prove" culprits or preventative factors, no matter how small, has more to do with our psychological needs than any clear medical result. 

Take, for example, a piece in last week's Science Times, which cast doubt on previous studies linking a moderate level of drinking with a reduced risk for heart disease, diabetes, and dementia. Critics said that all the studies had been "observational," not "randomized, controlled clinical studies" (in which people would be given alcohol, or not, without knowing which one they were getting, in order to test the impact). "The moderate drinkers [the studies observed] tend to do everything right--they exercise, they don't smoke, they eat right and they drink moderately," complained one critic. "It's very hard to disentangle all of that." 

Disregard, for the moment, that the above-mentioned critic is being funded by an alcohol and substance abuse prevention foundation, while the researchers he was criticizing are being funded by a non-profit group supported by the alcohol industry. Looking at it strictly from a "scientific" perspective, there really are problems with getting definitive, quantitative answers about the role a moderate amount of alcohol may play in preventing or causing disease. Consider the blind study requirements: take a group of non-drinkers and surreptitiously feed half of them regular, daily doses of alcohol? And then let them drive home? 

And even then ... can you be sure that all the other potential factors or links are normalized in the study group, so that the only differing factor, over a long-term study, is the daily consumption of alcohol? Reading researchers' concerns about all the different studies that are being considered--abstainers may not all abstain for the same reasons, and those reasons may influence their risk for disease; moderate drinkers seem to be "socially advantaged in ways that have nothing to do with their drinking"-- another question began to nag at me. Why are we so fixated on proving this issue, one way or another? 

Nobody is proposing that consuming one or two glasses of wine a day will cause or prevent disease at the risk level that smoking or not smoking does. At best, the studies would likely prove that a glass of red wine can (or can't) nudge the odds a little in your favor. Which might make drinkers or teetotalers feel better, depending on the outcome. But my guess is that the amount it nudges the odds, one way or another--given that most diseases are not 100% understood yet, and seem to have multi-factorial causes--would be small. 

So why do we care so much about finding out? Scientific curiosity? Maybe. But I suspect that public interest in the subject, at least, is driven equally by a fear of things outside our control. Cancer can hit regardless of what you do? That's too scary to contemplate. It reminds me of my nephew Tyler, who at age four asked his father if he could teach him what he needed to know so that he wouldn't have to die. Interwoven with our curiosity to discover how the world works is, I believe, a hope that those discoveries will ultimately give us some measure of control over it all. Prove to me what works, and even though it takes a lot of effort to eat and drink and exercise and monitor exactly according to formula, I can rest easier at night, knowing my odds of survival are better. 

And yet, there may well be an ironic twist to this control-oriented approach. A couple of years ago, I read an article noting that the French, with all their cigarettes, liquor, and rich foods, had the same life expectancy as their higher-strung, more health-and-exercise-fanatic American cousins. The studies weren't randomized, controlled clinical trials (of course), but the researchers posited that the stress Americans accumulate by being so fanatical about their lifestyle habits (and by taking such a higher-stress approach to their work lives) might just negate the benefits of any individually healthy elements. 

So maybe proving whether or not that one drink is actually preventing disease (and to what precise percentage degree) is less important than not worrying so much about it in the first place. Maybe, just maybe, the critic's complaint about previous alcohol studies is the best answer, sitting right in front of the researchers' eyes: "exercise, don't smoke, eat right and drink moderately." 

Of course, a general prescription of moderation and not worrying overmuch about every last provable detail requires making peace with a certain level of uncertainty and lack of control over our lives. Which is to say, ironic as it sounds, letting go of our need for control might just be the single best thing we can do to gain the very control we seek. 

I'm not sure how you'd test that theory objectively and conclusively, of course. But rest assured ... someone out there is working on it. 

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.