A Trained Eye

Shy? Or Something More Serious?

By Christopher Lane
Special to The Washington Post
Tuesday, November 6, 2007

If anyone in my parents' generation had argued that shyness and other run-of-the-mill behaviors might one day be called mental disorders, most people would probably have laughed or stared in disbelief. At the time, wallflowers were often admired as modest and geeks considered bookish. Those who were shy might sometimes have been thought awkward -- my musically gifted mother certainly was -- but their reticence fell within the range of normal behavior. When their discomfort was pronounced, the American Psychiatric Association called it "anxiety neurosis," a psychoanalytic term that encouraged talk-related treatment.

All that changed in February 1980, when the APA classified the broadly defined "avoidant personality disorder" and "social phobia" (later dubbed "social anxiety disorder") as diseases. The professional group also listed 110 other new disorders in its revised diagnostic manual, with the result that the total number of mental illnesses on the books almost doubled overnight. It was a dramatic example of the modern medicalization of behavior.

Bashfulness, once prized as a virtue, became a sign for medical concern. According to the 1994 National Comorbidity Survey, as much as 12.1 percent of the U.S. population might have social anxiety disorder and a staggering 28.8 percent suffer from some kind of anxiety disorder.

As a result of statistics like these and the disease criteria listed in the updated Diagnostic and Statistical Manual of Mental Disorders, or DSM, large numbers of people swallow daily doses of Paxil, Prozac and Zoloft for conditions that many experts now consider medical problems stemming from a chemical imbalance. After examining prescription rates for these three antidepressants alone, David Healy and Graham Aldred of the North Wales Department of Psychological Medicine at Britain's Cardiff University reported in the International Review of Psychiatry that just over 67.5 million Americans had taken at least one of them in the 15-year period ended in 2002. More than 18.5 million of those had received a prescription for Paxil, the first antidepressant to receive FDA approval for social anxiety disorder.

Is shyness really such a debilitating and widespread trait, or have psychiatrists merely made it seem that way? The psychiatric literature on social anxiety disorder is vast and well intentioned, tied to a host of drug trials and clinical studies aimed at lessening suffering. Chronic anxiety can be a serious problem needing treatment. But did substituting social anxiety disorder for anxiety neurosis blur an important distinction between ordinary shyness and that kind of paralyzing distress?

My own research over the past three years, including several days' intensive work in the APA archives, suggests so. I was able to review hundreds of unpublished letters and memos written by members of the task force assembled to define new disorders -- and by mental health experts who'd heard and read about the changes and hinted at a process bordering on caprice.

I found one document acknowledging that field trials for some disorders involved just one patient, treated by the person advocating for inclusion of the disorder in the DSM. I discovered another memo warning that the calculations used to set the diagnostic thresholds across the board threatened to define as mentally ill far too many people who had no symptoms of disease.

Proposals also surfaced for the approval of problems as vague and questionable as "chronic complaint disorder" and "chronic undifferentiated unhappiness." The symptoms of the first were listed, quite matter-of-factly, as whether people grumbled too much about the weather or said "Oy vay" too many times.

Among the dozens of letters from observers are questions about why a team of "kindred spirits" (the chairman of the task force's term) was put in charge of such sweeping changes. Documents indicate that the task force met for four years before one participant observed that those representing different -- long-esteemed -- perspectives in psychiatry, including psychoanalysis, weren't invited to contribute. Some warned that the revisions risked turning American psychiatry into a laughingstock.

In interviews with others and me, the chairman in question, Robert Spitzer, a professor of psychiatry at Columbia, has since characterized the psychoanalytic community as getting "very uptight," as if its near-total exclusion from the process were nothing to worry about.

Were the dissenters right to voice concern?

When the third edition of the DSM appeared in 1980, fanfare heralded the supposedly rule-driven, evidence-based diagnoses. But in 2005, Theodore Millon, a consultant to the task force, conceded in the New Yorker, "There was very little systematic research, and much of the research that existed was really a hodgepodge -- scattered, inconsistent and ambiguous."

Behind the scenes, some of the framers of avoidant personality disorder had indeed fretted in their exchanges and even on their written votes that the difference between it and ordinary reticence was a "minefield," and not (as they had hoped) a "borderline" or "continuum."

Definitional questions like these became minefields for American psychiatry: While some on the task force reckoned that diagnosis should be limited to those chronically impaired by anxiety, others thought it fine to gauge impairment by whether a person with a supposedly avoidant personality preferred traveling to work by car or on public transportation.

The manual went on to list dislike of eating alone in restaurants as the prime symptom of social phobia, with fear of hand-trembling a close second and avoidance of public restrooms third. With the inclusion of more and more behaviors -- public-speaking anxiety, concern about dealing with people in authority, even dating anxiety -- the diagnostic category ballooned until it overlapped with common shyness, as several key studies suggested, including a 1990 article in Behaviour Research and Therapy by University of Pittsburgh psychiatrist Samuel M. Turner and his colleagues.

It's not difficult to explain why. By 1987, a revised edition of the 1980 DSM had removed the phrase "a compelling desire to avoid" fear-inducing situations, requiring only "marked distress." Signs of this could include concern about saying the wrong thing -- a fear afflicting almost everyone on the planet.

With these elastic guidelines, the "illness" became widely diagnosed (some estimates, such as one in the December 2000 Harvard Review of Psychiatry, put it just a notch behind depression and alcoholism).

The line between shyness and social anxiety disorder has only gotten murkier. In the 1990 article, Turner and his colleagues wrote, "Interestingly, the central elements of social phobia, that is discomfort and anxiety in social situations and the associated behavioral responses . . . are also present in persons who are shy."

Four years later, Murray Stein, a specialist in anxiety at the University of California at San Diego, and his team published an influential article about the disorder's vague threshold. The piece drew from a single study -- a random telephone survey of 526 urban Canadians -- with results suggesting that social anxiety among them ranged from 1.9 percent to 18.7 percent, depending on the diagnostic threshold used. To most ears, that would sound sufficiently open-ended to be valueless.

But not to the drug companies. To them, articles by Stein and others that portrayed experts as troubled that almost one American in five might suffer from social anxiety disorder offered the potential for increased sales. SmithKline Beecham, maker of Paxil, for example, put more than $92 million behind a campaign aimed at convincing people that their shyness might in fact be a disorder treatable with drugs.

While many people want a more open dialogue about the widespread risks of overdiagnosis, including the potential side effects of medication, a group of psychiatrists is pressing ahead, discussing the inclusion in the next edition of the DSM (due out in 2012) of more eye-popping additions: apathy, overuse of the Internet and excessive shopping. The first of those would-be disorders was discussed as a candidate in the April 2005 issue of NeuroPsychiatry Reviews.

Meanwhile, researchers at the University of Minnesota have just wrapped up a trial on whether the antipsychotic drug Seroquel, prescribed for bipolar disorder, could benefit people anxious about speaking to large audiences. How long before a clever marketer pens a new ad campaign: "Think it's nerves about your work presentation? It may be public-speaking anxiety disorder"? ¿

Christopher Lane, a professor of English at Northwestern University, is the author of "Shyness: How Normal Behavior Became a Sickness" (Yale, $27.50), from which this article is adapted. Comments:health@washpost.com.

© 2007 The Washington Post Company