Stephen Smith looks at the real risks behind that glass of red.

It's enough to drive a person to drink: one day, alcohol's good for you. The next, it's bad for you.

British researchers who tracked the the behaviour of almost 1.3 million women for years concluded that having a daily glass of wine, a bottle of beer, or other alcoholic drink raised the overall risk of cancer by 6 per cent.

It was sobering news, enough to prompt some imbibers to put a cork in the pinot noir. "I had lots of my own friends and colleagues say, 'Gee, I guess I should stop drinking'," said Eric Rimm, a Harvard School of Public Health researcher who has studied the health effects of alcohol.

But wait a minute. What about the studies from Harvard - published in The New England Journal of Medicine over the past decade and a half - that found that women over 50 who have one to three drinks a week live longer? And men who have a drink at least three or four days a week suffer fewer heart attacks?

And consider this: a study published last month showed that a glass of wine a day might protect against a precursor to esophageal cancer.

There is no neat answer on whether alcohol is good or bad for you. In the end, the decision to drink - or not - may come down to a sense of personal risk. Someone already at peril for heart disease might decide that a glass of merlot a day is a good thing. Someone else, whose mother and sister died from breast cancer, might figure that same glass of merlot is not worth the added risk.

And even age and gender can factor into that calculus. For example, as women age, their risk of suffering a life-threatening cardiovascular condition climbs precipitously.

It is against that background that studies, pro and con, proliferate. There has never been a gold-standard medical evaluation of alcohol - the kind used to assess pharmaceuticals' safety and efficacy - but some specialists have begun asking whether that time has arrived.

Even those who study alcohol's health effects concede it is easy to suffer whiplash when bombarded with competing medical studies.

"What's the take-home message from all of these studies? It's fair to say no study is going to advocate increased alcohol use," said Dr Wendy Chen, a breast cancer specialist at the Dana-Farber Cancer Institute. "No doctor is going to say, 'to protect your heart, have a six-pack a day.' "

But some doctors do acknowledge the potential benefit of a drink a day - sometimes pointing to official dietary guidelines, the official word on what we should be putting into our bodies.

The US Government guidelines ever-so-timorously note that a drink a day for women middle-aged and older - two for men - might offer benefits, while there is no similar evidence for the young. (Australia's official guidelines - updated just last month - do not rule out imbibing. They say that for healthy men and women, drinking no more than two standard drinks on any day reduces the lifetime risk of harm from alcohol-related disease or injury.)

An even more elemental risk assessment may be at work in deciding whether to drink: Which do you fear more? A tumour or a heart attack?

"As an oncologist, of course, I think if I was going to go, I'd rather have heart disease," says Dr Peter Shields, the deputy director of Georgetown University's Lombardi Comprehensive Cancer Centre. "I think the cardiologists, they do a pretty good job of figuring out how to keep the plumbing to your heart working."

The recent British study was among the largest on alcohol's health effects. The scientists followed the women in their study for an average of seven years, finding that those who downed 10 grams of alcohol a day - roughly equivalent to a small glass of wine - suffered higher rates of certain, but not all, cancers. It did not appear to matter whether their drink of choice was spirits, beer or wine - the risk was the same. The researchers looked for other factors that might explain the elevated cancer risk but kept returning to alcohol.

Much of the increased cancer risk was associated with breast tumours. Based on the British results, the researchers calculated that 30,000 American women each year develop cancer as a result of moderate alcohol consumption.

There is, scientists say, a plausible biological explanation for that. Laboratory evidence shows alcohol can stimulate production of sex hormones, including estrogen, and doctors know that estrogen is a culprit in some breast cancers.

The study also found that cancers of the mouth and throat increased among smokers who drank. The alcohol, they theorise, acts as an accelerant.

"And the more you drink, the higher your risk of cancer," said Naomi Allen, the cancer epidemiologist at Oxford University who presided over the study. "The less you drink, the less at risk you are of developing cancer. It is as simple as that."

Or, maybe not.

Rimm, the Harvard researcher, said alcohol can suppress folate, a vitamin that keeps cells healthy. So it is possible a healthier diet richer in fruit and vegetables might protect against the deleterious consequences of alcohol.

Allen countered that most research suggests diet does not influence a woman's chances of developing breast cancer.

Still, she said, her research group intends to explore the interplay between what women eat and drink and their cancer risk.

When it comes to alcohol's potential for preventing heart disease, scientists point to findings that show alcohol boosts levels of the healthy form of cholesterol, HDL.

The Harvard studies have estimated that having a few drinks a week reduces heart disease risk as much as 30 to 40 per cent.

"There are few other lifestyle factors or medications that lower risk of heart disease by that much," Rimm says.

Some specialists question whether it is truly possible to disentangle the supposed health effects of alcohol from all the other things going on in people's lives.

"It's important that you don't over-describe the findings as definitive, because they're not," said Dr Steven Nissen, chairman of the Department of Cardiovascular Medicine at the Cleveland Clinic.

If nothing else, the confusion offers a cautionary tale about diet research: it is the rare single study that should dictate a wholesale change in the food we eat or the drinks we quaff.

Dr Nissen and Dr Robert Brewer, chief of the Alcohol Program at the US Centres for Disease Control and Prevention, said one thing was certain: alcohol's potential for harm is beyond dispute. Each month, Dr Brewer said, almost one of every three drinkers in the US drinks to excess, contributing to 79,000 deaths each year.

"The real bottom line is if you drink and you care about your health," Brewer said, "it's always better to drink less than to drink more."