Heart disease and stroke kill fewer Americans than ever before. Between 1972 and 1983, their incidence dropped by nearly a third. But heart disease still kills more Americans than all other diseases combined, and stroke remains a close third, just behind cancer.

Preventing these cardiovascular disasters could save the country an estimated $78.6 billion in 1986.

But scientists disagree about what the best way to prevent cardiovascular disease might be, even though the heart association and the National Heart, Lung, and Blood Institute have been studying prevention for decades. Last month, scientists presented two diametrically opposed prevention strategies during a science writers' forum sponsored by the American Heart Association here..

The first approach is the predictable one -- find an accurate way to identify the individuals at risk for heart disease or stroke and then rely on them to take the appropriate corrective action. Those at risk for heart disease could stop smoking, and keep careful track of their blood pressure and cholesterol levels.

The second idea is to change the world -- to alter the societal circumstances that lead to the high cardiovascular disease rate. Rather than ask individuals to change, make changes in society, said Leonard Syme, a professor of epidemiology at the University of California at Berkeley, and others. There is no good way to tell if a person is likely to get heart disease or have a stroke. Today, physicians basically wait until a patient's blood pressure or cholesterol level rises too high and then start treatment. But not everyone with the classic risk factors for cardiovascular diseases -- high blood presure, high cholesterol and smoking -- gets heart disease or has a stroke.

The American Heart Association suggests that the entire population be on guard against cardiovascular diseases by shunning salt -- which may be associated with high blood pressure -- and eating foods low in cholesterol and saturated fats. But other researchers, such as Syme and Paul Meier, a statistician at the University of Chicago, contend that this is an inefficient way to reach those who are truly at risk.

Molecular biologists believe they will be able one day to reach into the genes themselves to identify who has a high risk for developing heart disease.

"We think we will eventually make predictions with 100 percent accuracy, and we know we can easily get 90 percent accuracy," said Phillipe Frossard, project leader at California Biotechnology Inc. in Mountain View, Calif.

Frossard reported that his group is using molecular genetics to find "markers" for atherosclerosis and high blood pressure. These markers are tiny pieces of DNA that may themselves have nothing to do with heart disease and strokes but that seem to be close to the genes that do control such things as cholesterol build-up in the arteries.

In his search for such genes, Frossard uses the same techniques of molecular biology that recently identified people carrying the genes for Huntington's disease, cystic fibrosis and muscular dystrophy.

Tiny pieces of DNA, the chemical essence of the gene, can be used to probe a person's entire genetic makeup for sections of inherited DNA with the genes necessary to make and break down cholesterol and to deposit cholesterol in arteries. He also studies the insulin gene, because diabetics frequently get heart disease.

So far, Frossard's group has identified 24 genetic markers for heart disease that lie near a total of eight genes that are associated with cholesterol metabolism.

Frossard is looking for a pattern of markers that will be able to predict whether a person has inherited a cholesterol metabolism gene that could, for example, cause high levels of blood cholesterol if the person eats a high-fat diet.

Then, if the battery of markers shows that a person is genetically at risk, Frossard suggests that the person begin taking action to prevent the heart disease long before any symptoms occur.

In testing the heart disease markers, the researchers looked at a group of 500 patients at Muenster University Hospital in West Germany. Each patient had had coronary angiography -- a procedure in which a dye is injected into the coronary arteries, enabling physicians to determine whether a patient's arteries are clogged and, if so, by how much.

Frossard's group, collaborating with Dr. Gerd Assman, who heads the atherosclerosis center at the hospital, looked for genetic markers in these 500 patients and tried to determine which markers were associated with an increased or decreased risk of heart disease.

They found that four of the 24 markers are associated with increased risk and three with decreased risk. Frossard points out that, as a group, the patients with one of the markers indicating high risk had lower cholesterol than the patients with one of the low-risk markers. This finding is consistent, he says, with results by others showing that cholesterol levels by themselves are not terribly good predictors of who will get heart disease. Frossard and his associates now are further testing the associations of these markers with risk of heart disease by looking at 3,000 Americans.

The search for high blood pressure genetic markers is similar but has not yet progressed as far, Frossard said. He has found 14 markers near a total of three genes that seem to be involved with high blood pressure.

Other heart disease researchers at the meeting were enthusiastic about Frossard's work and fully expect it to lead to good predictive tests. "Molecular biology has finally come to cardiology in a big way," says Dr. Kenneth Shine, who is president-elect of the American Heart Association and executive chairman of the department of medicine at the University of California in Los Angeles.

But at least one of the meeting participants is not at all sure that this emphasis on the individual is the best approach.

"I have spent the past 30 years trying to identify risk factors for coronary heart disease," said UC-Berkeley's Syme. "My rationale has been that once people are informed of their risk, they will change their behavior to lower their risk."

Syme, however, said he has become disillusioned. He learned the hard way that "a lot of people either can't or won't change their behavior to lower their risk."

"The experience," said Syme, "that really changed my life was my involvement in MRFIT." This was a 10-year, $115 million study, whose initials stand for Multiple Risk Factor Intervention Study, that was meant to test the hypothesis that changing life styles can reduce the risk of heart disease. It was was sponsored by the National Heart, Lung, and Blood Institute and was completed in 1982.

The MRFIT researchers began by recruiting nearly 1,300 middle-aged men who, by all accounts, were at very high risk for heart disease. They had high blood cholesterol and high blood pressure, and many of them smoked. Moreover, they were motivated to change.

The men were divided into two groups. One group was followed by the MRFIT investigators but was essentially left to its own devices to reduce its risk factors. The other group, called the special intervention group, was in the hands of extensively trained MRFIT researchers. The men were taught by every means the MRFIT group thought feasible to change their diets, control their blood pressure and stop smoking. They were closely monitored and encouraged to turn their lives around.

The MRFIT study, it turned out, was inconclusive. At the end of the study, the heart disease rate was essentially the same in the special intervention group as it was in the "usual care" group. The customary explanation is that the usual care group changed its behavior, too, and so its risk factors ended up nearly the same as those of the special intervention group. But Syme has a different view.

What struck Syme, he said, is that "people in the special intervention group changed their behavior so modestly. Our work with that group was as good an opportunity as we will ever have to work with people on a one-on-one basis to change behavior. We had a state-of-the-art fantastic intervention plan. We worked with them for six or seven years. And the changes were not bad -- 40 percent stopped smoking, for example. But it ought to have been 100 percent."

Syme concluded that it is not only extremely costly but also inefficient to try to convince people to change their behavior simply because you tell them that they personally are at risk. A better way to reduce heart disease deaths would be to change the forces in society that put people at risk, he said. "I'm not arguing against a one-on-one approach, but it seems to me that we have to change the world at the same time as we get people to change."

Changing the world to prevent disease is not a new idea, Syme pointed out. It is essentially the way infectious diseases were conquered in this country in the first half of the century. Better hygiene -- clean water, sewage systems, less crowding, better food storage procedures, for example -- reduced the incidence of disease. And it is also the way that highway fatalities have been reduced. Safer cars and highways, seat belts to some extent, and lower speed limits made the difference. "But in general we have not applied this approach to heart disease," Syme said.

To demonstrate the concept's feasibility, Syme, together with Dr. June Fisher of San Francisco General Hospital, and their colleagues are studying bus drivers in San Francisco and are working on changing smoking behavior in Richmond, Calif., a small town between Berkeley and Oakland.

So far, said Syme, they have data on 1,500 bus drivers, and it turns out that these bus drivers have twice the incidence of high blood pressure as the rest of the population of their age, race, and physical characteristics. But their blood pressure was no different from the rest of the population's before they started driving buses. There is something about driving a bus in San Francisco -- and in many other cities too, Syme is learning -- that leads to high blood pressure.

The central problem, said Syme, is that bus drivers have "a very tough life." They are asked to meet schedules that are simply impossible. "There is no way you could meet those schedules even if you drove around San Francisco on Sunday morning in your racing car," Syme said. Yet they are given demerits if their buses are late. Consequently, they "pretend they can meet the schedules." To do this, they give up their rest stops and food breaks in a frantic attempt to be on time.

The bus drivers also "feel misunderstood. People are always mad at them," Syme said. To top it all off, they tend to work split shifts, which makes their day essentially 12 hours long. After work each day, the drivers spend several hours "unwinding" -- drinking to relax before they go home and fall into bed.

What Syme and his colleagues hope to do is to change the circumstances of the bus drivers' lives, which should reduce their blood pressure. Before starting the project, he spoke to the management of the bus company and received pledges of cooperation. He hopes to show that risk of heart disease for the drivers can be reduced more easily by changing their environment than by teaching them to relax, for example, and give up salt.

The researchers are even more ambitious in their work with the residents of Richmond, Calif. They want to make smoking a socially unacceptable behavior. They will provide assistance to citizens who want to stop smoking but, said Syme, "that's not our goal."

Of course, Syme's approach has its critics. He has been told that it sounds dictatorial, for example. His response is "not at all." Society already lacks free choices, he said. For example, adolescents constantly are bombarded with cigarette advertisements, and all the junk food at the grocery store is at eye level while the more healthful food is on the bottom shelves.

A second criticism is that it is a pretty tall order to try to change the world. Wouldn't it be better to try to get people to change themselves? To this Syme said: "My experience is that it is extraordinarily difficult to try and get people to change their ways."

There is no resolution to these conflicting views on how to reduce the death rate from heart attacks and strokes, and perhaps it is better this way.

After all, said some heart disease researchers, the most effective strategy in the short run might be a combination of the two. Provide a blood test that can pinpoint who is at risk and, at the same time, work on the much slower process of changing the world.