The Man Who Ate Everything (23 page)

September 1 990

Just Say Yes

Give or take a few millennia, men and women have brightened their mealtimes with a glass or two of wine, beer, or spirits for the past twenty thousand years. Civilization dawned with the cultivation of grain, the pressing of olives, and the fermentation of grapes. Yet every sixty years or so, a small band of men and women reject their evolutionary heritage and set out to eradicate alcohol in all its forms.

Today the forces of Prohibitionism again stalk the land, those cheerless folks behind the surgeon general’s warning label on every bottle of alcohol, behind the call for advertising restrictions and excise taxes to force the 60 percent of us who drink moderately to pay for the sins of the
9
percent who overdo. In schools across the country defenseless little children are even taught that alcohol is one of the three “gateway drugs,” as seductive as marijuana and tobacco.

But the fundamental truth is really quite simple, and it has been understood for many years:
People who drink moderate amounts of alcohol on a regular basis have far fewer heart attacks than those who do not drink at all.
And since moderate drinking carries
^y
ery few risks (except on the highway),
moderate drinkers generally live longer than people who do not drink.

Heavy drinking is extremely dangerous. The federal Centers for Disease Control (CDC), in Atlanta, lists thirty-six causes of
death partly or entirely due to drinking, including automobile accidents, homicide, suicide, liver disease, cardiomyopathy, some cancers, and mental disorders. The CDC estimates that 105,095 Americans died in 1987 from alcohol use and misuse.

But seven times as many people die from heart disease, the most common cause of death in America, with 725,110 fatalities in 1988. Over two-thirds of these were cases of coronary heart disease, the closing of the blood vessels that feed the heart. And coronary heart disease (also known as ischemic heart disease and sometimes abbreviated as CHD) is precisely the ailment that moderate alcohol consumption protects us against.

The three major coronary risk factors that doctors warn about—smoking, high blood pressure, and saturated fats— explain only about half the difference in the rate of heart fatalities among various countries and various individuals. The most striking example is France, where the people eat much more butter, cheese, cream, lard, and goose fat than we do but have only one-third the heart attacks. This has become known as
leparadoxe frangais
(which I have explored in “Why Aren’t the French Dropping Like Flies?” in Part One). But the same violation of modern nutritional rules is also common in northern Spain, northerns Italy, Switzerland, and Austria—places where the low incidence of heart attacks is commemorated nightly with dinners full of saturated fat. As the French drink ten times more wine than we do (the people in Spain, Italy, Switzerland, and Austria drink nearly as much as the French), the front-running explanation for solving le
paradoxe francais
is that wine (or alcohol in general) protects the heart.

In 1979 a famous study in the
Lancet
by St. Leger, Cochrane, and Moore took the statistics for coronary heart fatalities in eighteen developed countries and looked for correlations with a wide variety of factors, including the number of doctors available in each country, the dietary consumption of various fats, and the amount of wine and other alcohol drunk by the population. The strongest connection they found was with wine—France, Italy, Switzerland, and Austria showed both the highest wine consumption and the lowest rate of cardiac mortality. Total alcohol consumption was nearly as important as wine. The number of doctors in each country did not seem to matter.

But geographic comparisons like these can only suggest a connection, not prove one. The first clinical evidence came early in this century from pathologists who noticed that the arteries of deceased alcoholics were remarkably clean. More recent angio-gram examinations of live alcoholics are in close agreement. Alcoholics should take no consolation, of course, because any of the CDC’s thirty-six other afflictions, including noncoronary diseases of the heart, is likely to get them early in life.

The most powerful demonstration of alcohol’s protective effects comes from what are called prospective studies. Researchers enroll a large group of people; take extensive information about their medical history, smoking habits, diet, alcohol intake, exercise, and every blood component you can think of; and watch them closely for ten or fifteen years.

The first prospective study of alcohol consumption and its effect on heart disease was conducted in Baltimore in the 1920s; the past two decades have seen a blizzard of them. And nearly every study has found that moderate drinkers suffer far fewer heart attacks—between 21 and 60 percent fewer—than people who do not drink at all. In most, moderate drinkers live longer than either nondrinkers or heavy drinkers.

The consistency of these studies is remarkable—with similar outcomes in Chicago and Albany, in Yugoslavia and Puerto Rico, in Finland and New Zealand and Framingham, Massachusetts, among Japanese men living in Hawaii and Japanese physicians living in Japan, among West Australians, Trinidadians, and British civil servants, among 276,802 men followed for twelve years by the American Cancer Society, among 87,526 women nurses and 51,529 male health professionals in separate studies at Harvard, and among 123,840 patients at the Kaiser Permanente medical centers in the Bay Area. Only one study, in Alameda County, California, failed to find that moderate alcohol consumption protects the heart.

Summing up the results, Walter Willett, M.D., chairman of the Department of Nutrition at the Harvard School of Public Health, wrote in the
New England Journal of Medicine
in January 1991: “At present, the only dietary factor consistently associated with the risk of coronary heart disease in epidemiologic studies is alcohol, which apparently exerts [a] powerful protective effect.”

Several prospective studies found that the more you drink, even to the point of perpetual drunkenness, the lower your chances of becoming a coronary fatality. But in most, the benefits of alcohol follow what now has become the famous U-shaped curve. The odds of suffering a heart attack drop as you take one to three drinks a day, then begin to increase and come up to the starting point somewhere around four or five—which means that people who never drink have about the same chance of suffering a heart attack as those who take four or five drinks a day. At the bottom of the U-shaped curve, moderate drinkers have a much lower risk than either group.

The protective effects of moderate drinking—on coronary disease, stroke, and overall mortality—are even more impressive for women than for men. But women are also more sensitive to a given dose of alcohol. The reasons are that, on average, women weigh less; their bodies have a higher proportion of fat (also known as curves), and this can increase blood levels of alcohol because fat contains fewer blood vessels and less blood than other tissues; and, most important, women have smaller amounts of a gastric enzyme that breaks down alcohol before it enters the bloodstream. If a moderate dose of alcohol for men is two or three drinks a day, then for women moderate drinking means one or one and a half.

The antialcohol forces found it easy to challenge the earliest prospective studies, which typically failed to disentangle the effects of smoking, preexisting heart disease, diet, exercise habits, age, and gender from the effects of alcohol. What if moderate drinkers in America are also more health conscious than non-drinkers, take more exercise, eat less fat, or do not smoke? Most studies conducted after the mid-1970s made statistical adjustments for these baseline risk factors, and the correlation between moderate drinking and lower levels of coronary disease grew even stronger. This outcome should have been obvious from the start—if alcohol were not protective, drinkers would have
more
heart attacks than nondrinkers because drinkers also tend to smoke and to eat fat. So the neo-Prohibitionists were dealt a double blow: many evils that were once blamed on alcohol are, it was discovered, really connected with cigarette smoking.

In the mid-1980s, the antialcohol forces readied their final (one hopes) onslaught against the overwhelming weight of research. What if, they asked, people who never drink are unusual? What if some of the nondrinkers had quit drinking (or never started) for a good reason, such as a long family history of premature coronary disease or their own ill health? These unhealthy nondrinkers would drag down the odds for the entire population of nondrinkers and misleadingly make moderate drinkers look good.

For two or three years, the pro-alcohol camp was thrown into disarray. Some medical journals and newsletters—most already exceedingly uncomfortable with the notion that alcohol protects the heart—announced that the U-shaped curve was a myth. But soon enough, myth again became reality. The four largest prospective studies, all completed after 1987, were careful to separate unhealthy (or high-risk) former drinkers from the population of teetotalers, leaving healthy lifelong nondrinkers. And again, moderate drinkers had the fewest heart attacks. The U-shaped curve was intact.

The facts now seem airtight. The spiraling costs of American health care could be reduced by at least a few dollars a year if researchers no longer felt the need to investigate the issue. Then we could move on to the more interesting questions. Why does alcohol have this protective effect? Is wine better for you than spirits? And what should you do about it?

Remember a few years ago when we learned that cholesterol comes in two types, one good and one bad? HDLs, or high-density lipoproteins, are the good kind of cholesterol because
they lift fatty deposits from the inner walls of your arteries, where they would otherwise cause you untold grief, and carry them to your liver for disposal. (LDLs, or low-density lipoproteins, are the bad kind because they deposit fat there in the first place.) In numerous studies, high levels of HDLs in the bloodstream have been associated with a reduced risk of coronary heart disease. Aerobic exercise increases your HDLs. There is general agreement that alcohol does, too.

But now we learn that HDLs come in at least three (and maybe more) types, or subfractions. The neo-Prohibitionists had been arguing for years that drinking increases only HDL type 3 in our bloodstreams, which performs none of the useful work on the arteries that HDL type 2 does. (The other HDLs are apparently irrelevant to this debate.) Now, to the good fortune of moderate drinkers, several recent studies have shown that alcohol raises the level of both HDL-2 and HDL-3—about 17 percent each. And, in any case, HDL-3 now seems far less lazy than people had once thought.

Just as HDLs were the trendy blood component of the 1980s, something called apolipoproteins will surely become the blood buzzword of the nineties. The latest research tentatively demonstrates that the level of certain apolipoproteins in your bloodstream predicts your chances of having a heart attack even more accurately than the HDLs do. Moderate drinkers can raise their glasses again: alcohol increases the best kind of apolipoprotein, known as apo A-l, just as effectively as it increases the level of HDLs.

The second way that alcohol protects the heart involves the tendency of the platelets in your blood to clump together and form clots, especially around areas on your arteries’ inner walls that have already been damaged by fatty plaque deposits. As very small amounts of alcohol do not raise a drinker’s HDLs but do affect blood clotting, this seems a likely explanation for the protective effects of very light drinking. Both alcohol and aspirin diminish the tendency of these clots to form, which is why doctors regularly recommend that their older patients take an aspirin every other day. The formation of a blood clot involves a complicated series of steps, and nobody has figured out exactly where aspirin or alcohol intervenes in the process, whether they work in the same way or enhance each other’s protective effect. Enough research has been done to suggest that each may reduce clotting in its own way.

Contrary to recent publicity, red wine does not appear to be significantly more or less protective than hard liquor. Few large prospective studies have found a significant difference either way, though wine seems to have a slight edge in some. The effects of wine and hard liquor on HDLs and on blood clotting appear identical; beer usually comes in last in studies that have measured its effects.