Alcohol consumption lies on a spectrum, from nondrinkers on one end to patients dying of alcohol-related liver disease on the other. Novelist Jim Harrison once claimed that his ambition was to be a “problem drinker,” rather than an alcoholic.
Dr. Vaillant was the motive force behind the Harvard Medical School’s Study of Adult Development, which began in 1940. The study was divided into three groups: the middle-class College sample, the so-called “Core City Longitudinal Study”, and a later group, the Clinic sample, comprised solely of people who had been admitted to a clinic for detoxification, and the only one of the three groups that included women.
So what did George Vaillant discover in the longest formal study of drinking behavior ever undertaken in the United States? One of the first observations Vaillant drove home was that the amount of alcohol consumed is NOT a reliable indicator of alcoholism. Not only is “amount of alcohol ingested” an unreliable predictor of alcoholic drinking—so is “frequency of intoxication.” Both attributes, Vaillant found, proved to be very poor “discriminators” when it came to distinguishing alcoholics from “problem” drinkers.
So, if how much you drink, and how often you drink, are not reliably predictive of alcoholism, what, IS a reasonable predictor? The key item on Vaillant’s questionnaires turned out to be: “ Admits problem controlling alcohol use.”
Vaillant showed that “multiple alcohol-related problems result not from ingesting large amounts of alcohol but from being unable consistently to control when, where, and how much alcohol is consumed.” Alcoholism, Vaillant concluded, is “defined by the number, not by the specificity, of alcohol-related problems.”
This insight dovetailed neatly with the definition of alcoholism the National Council on Alcoholism had adopted way back in 1976: “ The person with alcoholism cannot consistently predict on any drinking occasion the duration of the episode or the quantity that will be consumed. ”
For George Vaillant and other pioneers of the disease model, the conceptual view had shifted long ago: “As with coronary heart disease, we must learn to regard alcoholism as both disease and behavior disorder.” Genetic loading, says Vaillant, “is an important predictor of whether an individual develops alcoholism,” while a difficult childhood environment “is an important predictor of when an individual loses control of alcohol.”
Based on his lifelong study, Vaillant discloses that chronic alcoholics are involved in at least half of all vehicle accident deaths, which comes as no surprise. Half of all emergency room patients with severe multiple fractures are alcoholics, and alcoholics are six times as costly to keep when hospitalized, says Vaillant. Alcoholics—22 million of them in the U.S. alone, by a recent federal estimate—commit suicide far more often than non-alcoholics do. 10 to 30 per cent of all suicides may be alcoholics, according to statistics cited by Vaillant. And this does not include suicides by people addicted to drugs other than alcohol.
“Progressive” alcoholics—those who never abstain or go through periods of stable social drinking—were twice as likely to smoke two packs of cigarettes a day, compared to other drinkers. Indeed, alcoholics almost never stop smoking. Vaillant believes that heavy drinking acidifies the urine in a way that increases the urinary excretion of nicotine. Hence, there is a need to smoke more while you are drinking more, lest blood nicotine levels fall below the comfort range.
And along with cigarettes, says Vaillant, comes depression: “Like smoking and alcohol abuse, depression is associated with premature mortality; andif alcoholics were three times as likely to be dead, they were also three times as likely to be depressed. ” Depressed men and women, then, drink more, smoke more, and take more drugs than non-depressed people. It is never hard to anticipate the chorus of complaint that such reasoning elicits. Everybody has emotional vicissitudes, but not everybody resorts to abusing drugs. Everybody gets the blues, but not everybody drowns them in drink. With courage and discipline, people can resist the easy notion of a whiskey, neat, for every problem.
To combat that notion, Vaillant also documents the dismal stats concerning “treated” alcoholism: “For short periods, middle-class individuals respond well to treatment in the medical model, but that response may be short-lived and reflect premorbid variables rather than the efficacy of specific treatment.” It turns out that if you simply control for two “premorbid” variables—marriage and employment—you can account for most of the difference. Alcoholics who are solidly married and have jobs fare better in treatment. But even they don’t always do very good for very long.