After the Crash

The alcoholism treatment establishment uses a drunk driving accident to silence dissent.


At 6 p.m. on March 25, Audrey Kishline was driving west on the eastbound side of Interstate 90 near Seattle when her Ford pickup truck collided head-on with a Dodge coupe occupied by Richard Davis, 38, and his 12-year-old daughter, LaSchell, killing both of them. Kishline had a half-empty vodka bottle on the seat beside her when police found her, unconscious, in her truck. Her blood-alcohol level was 0.26 percent, more than three times Washington's legal limit for drivers. Three months later, she pleaded guilty to two counts of vehicular homicide in Kittitas County Superior Court.

The crash, however tragic and avoidable, would have been no more newsworthy than the thousands of other drunk driving accidents in which Americans are killed each year were it not for the fact that Kishline is the author of the 1994 book Moderate Drinking and founder of Moderation Management, an organization aimed at helping problem drinkers control their alcohol consumption. (I wrote an introduction to the book and served as an adviser to M.M.) To longtime critics of the "controlled drinking" Kishline espoused as an alternative to the abstinence urged by Alcoholics Anonymous and its imitators, the crash was a vindication. The National Council on Alcohol and Drug Dependence (NCADD)–a private group that, like A.A., considers alcoholism a disease that can be controlled only through abstinence–gloated in a press release that Kishline's crash taught a "harsh lesson for all of society, particularly those individuals who collude with the media to continually question abstinence-based treatment for problems related to alcohol and other drugs."

Yet Kishline's one brief statement to the press revealed some facts that ran counter to the NCADD's interpretation. "Two months before the crash," The Seattle Times reported, "she dropped out of the [M.M.] program and joined Alcoholics Anonymous. But it wasn't long before she was consuming so much wine at night she would drink herself to sleep." In other words, Kishline, who belonged to A.A. before founding M.M., had returned. Only then, it appears, did her drinking veer out of control.

Of course, it is as unfair to blame A.A. for the Kishline tragedy as it is to blame M.M. She was apparently experiencing family and financial difficulties that had thrown her life off kilter after seven years of moderate drinking. While M.M. attracted media attention, it had never provided a reliable source of income. Kishline's husband was an itinerant businessman, and she had moved with him four times in the previous seven years. They and their two young daughters ended up living with her in-laws in a small town outside Seattle. But whatever the circumstances of Kishline's relapse, it is a mark of ideological intransigence and intellectual dishonesty that critics such as the NCADD do not note that she was regularly attending A.A. at the time of the crash.

Kishline's story is not just a tale of personal despair and failure. It embodies centuries of American conflict over alcohol in which teetotalers have repeatedly clashed with advocates of moderation. Having failed to impose their vision on the rest of the nation through Prohibition, the forces of abstinence nowadays focus mainly on problem drinkers, insisting that they renounce alcohol rather than try to use it more responsibly. This stubborn position overlooks substantial evidence that the sort of moderation training once advocated by Kishline can succeed where abstinence fails.

The A.A.-style abstinence approach dominates American treatment programs. A 1997 survey of private treatment centers found that 93 percent followed A.A.'s 12 steps (which include admitting that one is "powerless over alcohol") and 99 percent insisted upon abstinence. The belief that alcoholics must abstain is so ensconced in American folklore that contrary research findings tend to produce angry responses.

In 1976, a RAND Corp. study found that 22 percent of alcoholics were drinking without problems 18 months following treatment, compared to 24 percent who were abstaining. Luther Cloud, board member of the National Council on Alcoholism (the NCADD's forerunner), claimed the RAND study would lead to "death and brain damage" among alcoholics. Yet in a four-year follow-up study, the RAND investigators found that many alcoholics continued to drink without problems. Indeed, for some categories of alcoholics, abstainers were at greater risk of relapse than moderate drinkers.

The RAND researchers simply reported on patients in federal alcohol treatment centers, all of which were abstinence-oriented. But in the early 1970s two psychologists, Mark and Linda Sobell, had published research showing that alcoholics they treated with moderation techniques fared better than alcoholics treated in a standard hospital abstinence program. In 1982 three researchers, led by psychologist Mary Pendery, published a rebuttal of the Sobells' work in the prestigious journal Science, charging that the Sobells had misreported their data and that subjects trained to drink moderately often relapsed. The Pendery group's accusations led to a series of investigations by the Canadian Addiction Research Foundation, the U.S. Alcohol, Drug Abuse, and Mental Health Administration, and a congressional subcommittee, all of which exonerated the Sobells of wrongdoing. As a result of the investigations, the Sobells' finding that the abstinence-trained alcoholics had more relapses was actually reinforced.

In a disturbing postscript to this episode, Mary Pendery was shot to death in 1994 by an extremely intoxicated lover who had been treated by the hospital program where Pendery had worked. This incident, which on the face of it did not reflect well on the abstinence approach, did not get anything like the attention attracted by Kishline's crash, which was widely cited as evidence that alcoholics cannot learn to moderate their drinking and so must abstain. A closer look reveals a more complicated story.

At 43, Audrey Kishline had had a drinking problem for years. After treatment with two inpatient alcohol programs, a series of counselors, and A.A., she concluded that there was a better way. In the summer of 1993, she contacted me and several other nontraditionalists in the alcoholism field for assistance in creating her own support group.

The psychologists whose work Kishline consulted contended that there were two groups of alcohol abusers: alcoholics who display the full array of symptoms, including tolerance and withdrawal, and less severely afflicted "problem drinkers," who encounter personal, family, social, work, or legal problems due to drinking. Moderation Management was explicitly presented as a support group for "problem drinkers" who wanted to reduce their drinking to safe levels (quite low ones of, at most, nine drinks weekly for women and 14 for men). In the aftermath of the RAND and Sobell controversies, it was only with such drinkers that behavioral psychologists would dare to attempt moderation therapy, since using that approach with drinkers identified as alcoholics would have invited intense professional criticism and raised the possibility of legal liability. Even so, such psychologists deviate significantly from the disease model, which holds that even a taste of booze sets off a craving in an alcoholic that inevitably leads to excess. Instead, they view problem drinking as a learned response to stressful situations and negative feelings.

You wouldn't think that traditional, 12-step alcoholism counselors would object to moderation for problem drinkers, since they themselves maintain that it is only real alcoholics who need to abstain. But this seeming agreement masks longstanding conflicts, because both sides say it is impossible to know for sure, before the fact, which drinkers need to abstain and which can learn to drink moderately. Thus when Sally Satel, an anti-disease-theory psychiatrist, defended moderation training in The New York Times after Kishline's guilty plea, she undermined her own argument by saying, "The distinction between the problem drinker and the alcoholic, while not razor sharp, exists." Since the line is hard to draw, 12-step advocates argue, therapists should err on the side of caution by urging everyone with a drinking problem to abstain.

When Caroline Knapp wrote about the Kishline case for Salon, she was certain that alcoholics can never moderate, because she had tried and failed. Knapp–author of Drinking: A Love Story, which details her alcoholic degradation and redemption through A.A.–cited no research to support her claim. But what was most remarkable was how similar her language was to Satel's. "If an individual has crossed the line, admittedly fuzzy, into alcoholism," Satel had written, "then the risks of allowing someone to have an occasional drink or two become too high." Knapp wrote, "The line between problem drinking and full-fledged alcoholism may be blurry and difficult to discern–certainly it's difficult for the drinker to accept–but once you've passed a certain point in your abuse, moderation simply ceases to be an option."

The problem drinker/alcoholic dichotomy is a vague approximation of reality. Although the American Psychiatric Association classifies alcohol disorders as either "alcohol abuse" (i.e., problem drinking) or "alcohol dependence," most investigators and clinicians (especially those outside the U.S.) prefer to place drinkers along a scale, based on a combination of social problems (e.g., arrests, fights) and medical symptoms (e.g., withdrawal-induced "shakes"). According to this incremental view, the more severely dependent the drinker, the less likely moderation is–at least without the passage of substantial time.

Even that formulation is overly reductive: Several teams of researchers in the United Kingdom, for example, have found that problem drinkers' beliefs–whether they see themselves as "powerless," for example, or whether they think a single drink will set off a relapse–influence whether abstinence or controlled drinking is the more attainable goal. In other words, the very subjective elements that American alcoholism treatment derides as "denial" can improve the chances of recovery: It is easier to achieve what you believe.

The point is not that moderation training is always better than abstinence. But even if abstinence was the right goal for Audrey Kishline, she didn't manage it. Therapists, whatever their orientation, are reluctant to admit that most alcohol abusers relapse, and that they need training to avoid harming themselves and others when they do. For example, people can learn in therapy to call their spouse when they get drunk to make sure they don't drive while intoxicated. By failing to develop such fallback positions, therapists and support groups are saying, in effect, "If you make a mistake, you might as well give up all restraint–taking a drink, getting drunk, and driving drunk are all equivalent."

When Kishline repopularized the idea that some people could return to nonproblematic drinking, she aroused the ire of the NCADD, which denounced the idea in a July 1995 press release. "Millions of Americans," it said, "have recently seen life-threatening stories in the media that people with alcohol problems don't have to stop drinking completely to get better." In perhaps the high point of national media attention Moderation Management was to receive, U.S. News & World Report featured controlled drinking and M.M. in a July 1997 cover story. The article focused on problem drinkers, who it indicated were a solid majority of those with alcohol problems. Nonetheless, the director of the government's National Institute on Alcohol Abuse and Alcoholism (NIAAA), Enoch Gordis, wrote a letter to the magazine emphasizing that "persons with the medical disorder 'alcohol dependence'" need to abstain.

The NIAAA's own research has contradicted that position. Project MATCH was the largest trial of alcoholism treatment ever conducted. Completed in 1996, less than a fifth of the 952 alcoholics who underwent only outpatient treatment, and a little more than a third of the 774 alcoholics who had hospital treatment followed by outpatient treatment, abstained for as long as a year. So the NIAAA used a different standard of success: It emphasized that the subjects entered treatment drinking, on average, 25 days per month and 15 drinks per occasion; after treatment they drank, on average, five to six days a month and three drinks per occasion. The NIAAA in this case seemed to be endorsing controlled drinking.

In 1992 the NIAAA conducted face-to-face interviews with a representative sample of nearly 43,000 Americans, asking them about their current and past drinking practices. Of the more than 4,500 who had been alcohol dependent at some point in their lives according to the current psychiatric definition, about a quarter had entered treatment as a result. A third of those who had been treated were drinking abusively at the time of the survey, compared to a quarter of those who hadn't been treated. Untreated alcoholics were less likely to be abusing alcohol, but they were twice as likely to be drinking without qualifying as problem drinkers or alcoholics. Abstinence was more common among alcoholics who had been in treatment, but still only a minority (39 percent) were abstaining, while 28 percent were drinking without diagnosable problems. Those who had been treated did, on average, have worse problems to begin with. But all of the drinkers in the analysis had at one time qualified for the diagnosis of alcohol dependence and would certainly have been considered in need of treatment. A plausible explanation for the finding that treated alcoholics were more likely to be drinking abusively is the all-or-nothing message taught by A.A.-style programs: When people who have undergone traditional treatment fall off the wagon, they're convinced that it marks the beginning of a binge, which then becomes a self-fulfilling prophecy.

Thus, federal research whose size and comprehensiveness will not readily be equaled gives the lie to the principal claims of America's treatment establishment: that alcoholics can't recover without treatment; that alcoholics can't reduce their drinking to nonproblematic levels; and that alcoholism treatment reliably produces abstinence.

Ignoring this evidence, abstinence-only advocates are using the Kishline case to close the small cracks that have appeared in the 12-step monolith.

This year, under medical director Alexander DeLuca, the Smithers Center in New York–founded by the late R. Brinkley Smithers, a wealthy recovering alcoholic and A.A. supporter–began to make referrals to M.M. DeLuca did not himself offer moderation treatment at Smithers. Rather, in keeping with the ethical and legal requirement of informed consent, he thought it proper to discuss alternatives such as controlled drinking with patients who were not fully committed to quitting. DeLuca still hoped such patients would ultimately decide to abstain.

Despite these nuances, in July the board of the Smithers Foundation, headed by Brinkley Smithers' widow, Adele Smithers-Fornaci, took out a full-page ad in The New York Times attacking the reconstituted Smithers program (with which her family is no longer associated). "The seductive appeal of controlled drinking to the alcoholic will cause needless loss of life and destruction of families," the ad warned. "This is no more sadly illustrated than in the tragically ironic case of the founder of the Moderation Management program, Audrey Kishline.…Using the Smithers name in conjunction with this type of treatment is an abomination, an insult and a disgrace to the memory of R. Brinkley Smithers." Within days of the ad's appearance, the Smithers Center accepted DeLuca's resignation.