Harvard Mental Health Letter

The prevalence and treatment of mental illness today

The first large survey of mental illness and its treatment in the United States since the early 1990s shows that almost half of adult Americans at some time, and nearly a quarter in any given year, have had a psychiatric disorder. More of them are getting treatment today than in the early 1990s, but the treatment is still usually delayed and inadequate.

The study, called the National Comorbidity Survey Replication, was conducted in 2001–2003 with funding from the National Institute of Mental Health and a number of academic institutions and foundations.

Interviewers used a standard format to question a representative sample of more than 9,000 adults. At some time in their lives, nearly 46% had at least one psychiatric disorder (as defined by the American Psychiatric Association's diagnostic manual). The rate was highest for anxiety disorders, including panic disorder, generalized anxiety, social anxiety, phobias, and post-traumatic stress disorder (29%). Next came impulse control disorders, including attention deficit hyperactivity disorder, conduct disorder, and oppositional defiant disorder (25%). Twenty-one percent had had a mood disorder and 15% had been dependent on or an abuser of alcohol or other drugs.

The most common individual psychiatric disorders were major depression (17%), alcohol abuse (13%), social anxiety disorder (12%), and conduct disorder (9.5%). Women were more likely to have had anxiety and mood disorders, men more likely to have had impulse control disorders. Different disorders often went together, especially anxiety and depression. About 28% of the population suffered more than one psychiatric disorder.

In the previous year, 26% of those interviewed had had a psychiatric disorder. Again, anxiety disorders were the most common (18%), followed by mood disorders (9.5%), impulse control disorders (9%), and substance abuse and dependence (4%).

Psychiatric disorders began early in life — in half of cases before age 14 and in three-fourths of cases before age 24. On average, anxiety and impulse control disorders first appeared at age 11, substance abuse at age 20, and depression at age 30.

Study authors define a "severe" disorder as one involving a suicide attempt, psychosis, severe drug dependence, serious violence, substantial disability or limitation, or being "out of role," that is, unable to function normally in family life, at work, and in personal relationships, for a month or more. By this definition, 22% of psychiatric disorders were severe, and 6% of the population had a severe psychiatric disorder in the previous year.

These numbers may be an underestimate. Some people must have failed to recall symptoms (especially chronic mild to moderate depression) or failed to report them because of shame and stigma. Homeless and institutionalized persons were excluded from the survey. The rate of response was 71%, and people who declined to participate probably had a higher than average rate of psychiatric illness.

Interviewers went on to ask: "Have you ever been treated for problems with your emotions or nerves or your use of alcohol or drugs?" About 80% of people with a psychiatric disorder had eventually sought treatment, but often only after a long delay — the average was 10 years after symptoms first appeared. Major depression and panic disorder were usually treated fairly quickly, but fewer than 7% sought treatment for social anxiety disorder, post-traumatic stress disorder, and attention deficit disorder within the first year. And nearly half of those with impulse control or drug problems had never sought help at all.

About 17% of the interviewees, including 41% of those with a psychiatric disorder, said they had used mental health services in the previous year. Women were more likely to use these services than men, and whites more than blacks and Latinos with similar symptoms.

Family doctors, nurses, and other general medical professionals provided treatment for 23%; psychiatrists for 12%; other mental health professionals such as social workers and psychologists for 16%; counselors or spiritual advisers for 8%; and complementary and alternative practitioners (including self-help groups) for 7%. (The total is more than 41% because some people received treatment from more than one source.)

Most of this treatment was inadequate, at least by the standards applied in the survey. The researchers defined minimum adequacy as a suitable medication at a suitable dose for two months, along with at least four visits to a physician; or else eight visits to any licensed mental health professional. By that definition, only 33% of people with a psychiatric disorder were treated adequately, and only 13% of those who saw general medical practitioners.

A comparison with the original National Comorbidity Survey, conducted in 1991–1992, showed that Americans have been increasing their use of mental health services. The proportion of the population receiving treatment in the previous year rose more than 50% during the decade, mostly because of more visits to psychiatrists and other physicians.

It may be surprising to learn that 46% of the American population has been mentally ill at some time. But more than 99% of us will have a significant physical illness at some time in our lives, and even mild to moderate psychiatric disorders can be as harmful as chronic physical illness. Major depression, for example, causes more disability and misery than most medical disorders. And many psychiatric disorders are life-threatening — consider the relationship between alcoholism and accidental death, or between depression and suicide. Also, unlike most physical illnesses, mental illness usually begins in youth and affects people in the prime of life.

Treatment has become more widespread since the early 1990s because of greater public awareness, more effective diagnosis, less stigma, more screening and outreach programs, and greater availability of medications. Most important, according to the survey researchers, has been the growing willingness of general practitioners to prescribe psychoactive medications, especially antidepressants.

Still, at the beginning of the 21st century nearly 60% of people with psychiatric disorders were getting no treatment. And partly because most treatment was still inadequate, the overall rate of mental illness did not change between 1991–92 and 2001–2003. According to survey researchers, one reason may be that many physicians lack the time, training, and experience needed to persuade patients to keep taking medications and make return visits.

Some researchers point out that the problem may not be as serious as it seems. People often recover spontaneously from psychiatric disorders, as they do from physical illnesses. And, as with physical illnesses, sometimes there is no reliable treatment. But it can be hard to determine when treatment will be unnecessary or ineffective. The question is whether we need to detect mild symptoms earlier so that they won't get worse, or concentrate resources on the more severe (and less common) types of chronic mental illness. Survey researchers also suggest that we need more outreach and voluntary screening, more education about mental illness for the public and physicians, and more effort to treat substance abuse and impulse control disorders.


Kessler, RC et al. "Lifetime Prevalence and Age-of-Onset Distributions of DSM-IV Disorders in the National Comorbidity Survey Replication," Archives of General Psychiatry (June 2005): Vol. 62, No. 6, pp. 593–602.

Kessler, RC et al. "Prevalence, Severity, and Comorbidity of 12-month DSM-IV Disorders in the National Comorbidity Survey Replication," Archives of General Psychiatry (June 2005): Vol. 62, No. 6, pp. 617–27.

Kessler, RC et al. "Prevalence and Treatment of Mental Disorders, 1990 to 2003," New England Journal of Medicine (June 2005): Vol. 352, No. 24, pp. 2515–23.

Wang, PS et al. "Failure and Delay in Initial Treatment Contact After First Onset of Mental Disorders in the National Comorbidity Survey Replication," Archives of General Psychiatry (June 2005): Vol. 62, No. 6, pp. 603–13.

Wang, PS et al. "Twelve-Month Use of Mental Health Services in the United States: Results from the National Comorbidity Survey Replication," Archives of General Psychiatry (June 2005): Vol. 62, No. 6, pp. 629–40.

For more references, please see www.health.harvard.edu/mentalextra.