Robot dogs and cloning are not the only developments anticipated in "Sleeper" that have come to pass. Near the beginning of Woody's Allen's 1973 science-fiction comedy, a doctor remarks that steak, cream pies, and deep-fried foods once "were thought to be unhealthy -- precisely the opposite of what we now know to be true." We seem to hear similar news every month or so.
Our latest "Sleeper" moment came when researchers announced that drinkers are less prone to obesity than teetotalers. The idea that one holiday indulgence could help protect us from the consequences of another has understandable appeal this time of year. But the hope that the things we like are also good for us springs eternal in a society where many people are irrationally anxious about pursuing pleasure for its own sake.
How else are we to understand the reaction to a 2003 study that suggested eating dark chocolate can lower blood pressure? "You can sin with perhaps a little less bad feeling," one physician told the Associated Press, as if nibbling a Godiva truffle should be up there on the tablets with perjury, theft and murder.
In the drinking study, published on Dec. 5 in BMC Public Health, two researchers analyzed data on 8,236 nonsmokers from the National Health and Nutrition Examination Survey, which includes direct measurement of body mass index (weight in kilograms divided by height in meters squared). Current drinkers were 27 percent less likely than abstainers to be obese, defined as having a BMI of 30 or more.
That average conceals a more complicated picture: Although subjects who reported consumption of one or two drinks a day were substantially less likely to be obese than abstainers, those who said they had three drinks a day were about as likely to be obese, while those who said they had four or more drinks a day were substantially more likely to be obese. Since alcohol consumption was self-reported, the actual levels may be higher, but the trend of risk falling and then rising with the amount of drinking seems clear.
As one skeptic pointed out in a Health Day story about the study, this association is counterintuitive, since "alcohol is very energy-dense," containing seven calories per gram, compared to nine for fat and four for protein and carbohydrates. Yet other studies, based on self-reported height and weight, have yielded similar results.