A neighbor confesses she's an alcoholic to no one's surprise. The boss gives up his latest scheme to quit smoking just as everyone at the office bet he would. A colleague's gambling gets out of hand, as expected. And was there ever any doubt that the boy down the street would get hooked on drugs?

Addiction is now America's number-one health hazard, exceeding cancer, respiratory illness and even heart disease in deaths, disabilities and total cost to the nation, according to a study commissioned by New York Gov. Hugh L. Carey and made public this month.

The report estimates that half a million Americans are addicted to heroin, more than a million abuse hypnotic drugs and tranquilizers and 13 million abuse or are addicted to alcohol. Other research concludes that approximately 9 million Americans are compulsive gamblers and one in three adults smokes cigarettes regularly.

To suggest that people who become addicts are usually "the type" seems better suited to over-the-fence gossip than clinical psychology. But some scientists say tracking the types of personalities prone to addiction may provide the best clues for combating the nation's premier health problem.

"Recent research shows that common personality traits may exist in persons prone to different kinds of addictive behavior," says Gloria Leon, a clinical psychologist who has pioneered addiction-prevention studies at the University of Minnesota.

"If you can discover whether people are high risks for developing an addictive disorder, you can warn them to be more vigilant before serious and harmful problems occur."

Psychologists such as Leon are the first to admit that a lot of "ifs" remain in identifying those kinds of personality patterns. In the past decade, for instance, researchers only have begun to pinpoint what different types of addicts and addictions have in common. But, says Leon, researchers have stumbled upon one indicator, a simple pencil-and-paper test known as the MacAndrew Scale, which has proven remarkably effective in distinguishing some types of addicts and potential addicts from "normal" random samplings.

"So far, the MacAndrew Scale has been demonstrated to be unique in its ability to identify existing and potential alcoholics, drug addicts and pathological gamblers," says Leon. "It's about 78 percent reliable in those categories, and to a lesser extent it has also tagged addictive smoking in males and massive obesity in women, two other kinds of substance abuse."

The 49-question test is part of the Minnesota Multiphasic Personality Inventory (MMPI), a widely used psychological profile exam, designed originally to nab alcoholics who lie about their alcohol abuse. Eight years ago, it surprised the scientific community with its reliability also as a predictor.

"The MMPI was so popular at the University of Minnesota in its early years that all incoming freshmen had to take it as a college-entrance test," says Leon. "A follow-up study in the mid-'70s showed that those early MacAndrew scores identified with 85 percent accuracy freshmen who were found 13 years later to be alcoholics. None of them had an alcohol problem at the time of the test."

Other subsequent studies have discovered similar results in detecting drug addiction. "Heroin and opiate drug abusers consistently score in the high-risk range before and after successful rehabilitation," says Leon, suggesting that the addiction can change, or be overcome, but the addictive personality remains the same.

Clinical psychologist John Graham, who has spearheaded addiction-prediction research at Kent (Ohio) State University, where he is director of clinical training, more recently handpicked 50 known compulsive gamblers who had no history and showed no signs of alcoholism or other addictions. "Every one of them" tallied as addicts on the test.

"We know now that a high score is a reliable indicator of addictive behavior or of a predisposition to addiction," says Graham, who along with colleagues, admits that the MacAndrew Scale is a "strange" test. "The problem is that we don't really know how it works or what it's measuring."

MacAndrew scores are tabulated on the rate of agreement or disagreement with the test's 49 items, described by Graham and Leon as having no apparent relation to addiction. For example, addicts and potential addicts, says Leon, tend to agree with statements like:

* I am easily bored.

* I vomit often.

* I often got into trouble with the principal in grade school.

* I sweat easily.

* I believe in the second coming of Christ.

* I enjoy reading crime news.

* I have had periods when I couldn't remember what I'd done.

* I enjoy a game more when I can bet on it.

* I enjoy big noisy parties.

* I stole things as a child.

"We're not sure what attitudes attribute to this common ground," says Leon, who points out that some addictive behaviors--inexplicably--are not identified by the test.

In a recent cigarette-smokers' study, for instance, men who smoked two or more packs a day scored well into the high-risk range while their female counterparts slipped by with normal scores.

Similar studies show that the test doesn't reliably detect or forecast eating disorders such as moderate obesity and anorexia nervosa, or "positive addictions" such as fanatic jogging.

But persons suffering from massive obesity (weighing at least 100 percent above their ideal body weight) report a mean score that is borderline addictive, even a year after undergoing intestinal bypass surgery. That, says Leon, suggests compulsive eaters share at least some personality characteristics with persons exhibiting other kinds of addiction.

"Generally," she summarizes, "not all persons with addictive tendencies score high on the MacAndrews, but most of those who do score high have addiction problems or should be alert to the likelihood."

"In most cases, addiction problems also start pretty early in life, usually before age 21," says Graham, whose current study applies the MacAndrew Scale to high-school students.

"High scores at that level are finding students who are using alcohol and drugs to the point where it is starting to interfere with their lives. They often are disciplinary problems, have lower grades, are involved in almost no extracurricular activities, are often unhappy or depressed and sometimes have problems at home.

"We're trying to steer these students into comprehensive alcohol- and drug-addiction programs offered in the schools and communities."

Leon agrees that early intervention may be the best route to prevention, but cautions that a little knowledge can be dangerous.

"Alerting people that they're potential addicts may have the disadvantage of offering a self-fulfilling prophesy to some people, but if you could warn a compulsive eater that something in her personality makes her prone to obesity, she might monitor her eating habits more diligently and get treatment that recognizes her problem as an addiction.

"A heavy smoker recognized as a high risk could join an anti-smoking program that approaches the problem as more than just a bad habit.

"Right now, prevention of addiction is such an alien concept," says Leon, "but it may be the best answer we have."