PREVENTIVE MEDICINE, the rallying cry of today's health care professionals, has been stymied for years by the odd fact that people don't listen to what they are told. Despite findings that link cigarette smoking to cancer, heart disease and a raft of other problems, 54 million Americans continue to smoke. Despite the evidence that damns diets rich in fats and cholesterol, Americans continue to overeat. It can't happen to me, we think when we hear the latest scary statistic. So we continue to drink, to use drugs, to let industries pollute our air.

But now physicians may at last have a way to shake us from our shortsighted smugness. They can personalize the statistics. Through "genetic markers"--windows on an individual's genetic makeup, which can be anything from blood type or hair color to the presence or absence of enzymes in the blood--they can predict someone's predisposition toward developing a disease. Doctors today can say, for example, "Your tests reveal high levels of a critical enzyme called AHH. If you continue to smoke, you run a risk of lung cancer that is 36 times the risk of a smoker with low levels of AHH."

That's a powerful incentive to quit smoking; it lends new irony to the Epicurean philosophy that guides us in many of our harmful habits: "Eat, drink, and be merry, for tomorrow we die." And it is just one example among dozens of the way that genetic soothsayers can predict our individual fates. In this remarkable book, Genetic Prophecy: Beyond the Double Helix, authors Zsolt Harsanyi (a medical geneticist) and Richard Hutton (a science writer) describe in thoughtful, well-paced detail the stunning medical revolution brought to us by the men and women who read our genes. More impressive than any crystal ball gazers, genetic prophets can, with some relatively easy tests, tell us each, individually, how we ought to live and how we most likely will die.

One of their most potent tools is the HLA antigen, a biological signature whose story is told here in clear, compelling prose. Out of the 90 or more HLA antigens known, each of us carries, in every one of our cells, 10. Our own 10 HLA antigens are arranged in five characteristic pairs, in combinations as unique as our fingerprints. They sit on the cell's surface and serve as an ID card that the white blood cells continually monitor. If the right HLA antigen sequence appears during an ID check, the white blood cells recognize it and leave the cell alone. But if the wrong sequence appears--because the cell is from a transplanted organ or an invading pathogen-- the white blood cells mount an attack.

In recent years, scientists have learned to read HLA antigens almost as well as white blood cells do. Particular HLA antigens have been found to correlate highly with particular diseases: among patients with a rare arthritic condition called ankylosing spondylitis, for instance, 95 percent have the antigen known as HLA-B27. Now, doctors feel confident in using the HLA antigen system to predict an individual's susceptibility to any one of 80 or more diseases. Persons with type A blood and the HLA antigens B5 and Cw4 run a risk of bladder cancer 15 times that of the general population. Those carrying HLA antigen Cw6 run five times the normal risk of developing psoriasis. And those who inherited different HLA antigens from each parent seem to live significantly longer than do those with matched pairs.

Another powerful use of genetic prophecy relates to diseases that run in families. If many of one's relatives suffer from cancer or heart disease or ulcers, the threat of that disease looms ominously in one's own future. Siblings of juvenile diabetics, for instance, are known to run a greater-than-average risk of also developing diabetes; this means that each sibling must deal with the possibility (about one chance in eight) that he will develop the disease and pass on the susceptibility, in turn, to his own children. But doctors can now fine tune that prediction. Through HLA typing, they can determine whether the second child runs a slim chance of developing diabetes-- one chance in 1000 or one chance in 50, depending on the antigen involved--or a much greater risk of about one chance in two.

A similar story can be told for families prone to duodenal ulcers. Here, the enzyme pepsinogen-I can help signal which relatives are most likely to develop the disease: those with high levels of the enzyme have a 40 percent chance of developing ulcers, compared to 14 percent for those with low levels from the same family. And in families with a strong history of breast cancer--where sisters and daughters often submit to annual mammograms and, in extreme cases, preventive mastectomies--a genetic marker called GPT can help doctors determine which women are most likely to benefit from these potentially dangerous procedures. When breast cancer runs in their family, women who lack the GPT marker can be assured that their risks of developing cancer are no higher than the general population's. Those with the GPT marker can learn that they run one chance in eight of developing cancer before the age of 35, one chance in two before age 50, and a staggering nine chances in 10 before age 80.

The book is packed with stories like these--tales about how our genes can determine our susceptibility to alcoholism, anemia, arthritis, lung disease, obesity and a variety of cancers. It details the uncomfortably strong links found between genes and depression, IQ, schizophrenia, and even such personality traits as gregariousness. The stories made me want to run to my nearest genetic clinic (the appendix lists four in the District of Columbia and one in Bethesda, Md.) for a complete series of predictive tests. What if I had one of the truly troublesome HLA antigens, like HLA-A1, which increases your chances of having adrenal gland hyperfunction, chronic pancreatitis, recurrent herpes infections, pre-eclampsia of pregnancy, and Hodgkin's disease? But then I began to wonder: If I had something like HLA-A1, did I really want to know?

We're all aware of the fact that we're going to die, which in itself is a thought too terrible to bear for very long; how many of us can bear knowing precisely what we will die of? Because of the intimate relationship between genes and environment, none of these predictions based on genes alone can be certain. Mightn't the very fact of uttering a prediction change the environment enough to make the prophecy come true? Quite apart from the psychic burden of knowing one's own fate is the political dynamite of other people knowing it, too. How can we expect employers, insurance companies, and the government to use this awesome information? And how can we perpare ourselves for a time, not too many years hence, when all of us--even fetuses--are screened for our HLA types and other measures of genetic acceptability?

To their credit, the authors raise all these thorny questions--and, again to their credit, they come up with no easy answers. As they suggest, the possibility for abuse probably does not warrant the throttling of new knowledge, since by learning to interpret the genes, scientists are learning less how to control our lives than how to extend them. The best protection we have against abuse, say the authors, is an informed public ready to participate in meaningful debate about how guidelines and safeguards can best be applied. This book is a good start in that direction.