This Month’s Expert: Psychiatric Epidemiology by Ron Kessler, PhD

This Month’s Expert Psychiatric Epidemiology Ron Kessler, PhDTCPR: Dr. Kessler, you are one of the preeminent researchers into the prevalence of mental illness in the United States. The numbers generate controversy. For example, we often hear that 20% to 30% of Americans have a mental illness. What is that figure based on?

Dr. Kessler: The government periodically conducts big national surveys like the National Comorbidity Survey (NCS) and the National Comorbidity Survey Replication (NCS-R). In addition, every month, the government conducts a household survey called the Substance Abuse and Mental Health Services Administrations (SAMHSA) Survey.

TCPR: And how do those surveys arrive at the 20% to 30% figure?

Dr. Kessler: This means that between 20% and 30% of adults in the U.S. population in a given year meet criteria for any of the disorders assessed in those surveys. The disorders surveyed run the gamut from the common anxiety, mood, and substance disorders, to behavioral disorders, nonaffective psychoses, and various other things. So the estimate is that one out of every four Americans meets criteria at some time in a given year for a broadly defined psychiatric disorder.

TCPR: There’s an even more alarming statistic often reported in the media—namely, that “half of the U.S. population has some type of mental illness.” Can that really be true?

Dr. Kessler: This statistic is related to lifetime prevalence. What it really means is that half of the population meets criteria for a psychiatric disorder at some time in their lives. Although it seems huge at first, imagine if I were to say to you that 99.9% of the population of the U.S. was physically ill at some time in their lives. You wouldn’t blink an eye, because there is illness, like a sprained finger, for instance, and then there is real illness, like terminal cancer.

TCPR: And the same is true of mental disorders.

This article originally appeared in The Carlat Psychiatry Report -- an unbiased monthly covering all things psychiatry.
Want more, plus easy CME credit?
Subscribe today!

Dr. Kessler: Yes. When you look at every single disorder in DSM-IV, including adjustment disorders and things of that sort, it is not all that surprising that half the people in the population have had a psychiatric disorder. The criteria for some of the disorders are not that difficult to meet. For example, there are lots of people who have snake phobias that really don’t get in the way of life that much, but if you go by the book, they meet criteria for a mental illness.

TCPR: So, the definition of mental illness is so broad in this case, it’s no wonder half the country meets the criteria.

Dr. Kessler: Yes. I think a more useful set of numbers is the proportion of the population that has a really seriously impairing mental disorder, such as nonaffective psychosis, bipolar disorder, or schizophrenia. To the best of our estimates, at some time in their lives, 1% of the population meets criteria for schizophrenia; 1% for bipolar I disorder, 1% for bipolar II, another 2% to 3% for a bipolar spectrum disorder; and 2% to 3% meet a broadly defined criteria for a nonaffective psychosis, such as delusional disorder or atypical psychosis.

TCPR: So when you add that up, only about 6% of the population meets criteria for a major mental disorder in their lives.

Dr. Kessler: Yes, and furthermore, in any given year, only about 2% to 3% of this group have an active episode of these really severe disorders. Then another 3% to 4% of people have a serious but not severe illness, and another 20% have a disorder that may or may not need treatment.

This Month’s Expert Psychiatric Epidemiology Ron Kessler, PhD

TCPR: Are the people who conduct these surveys psychiatrists?

Dr. Kessler: No. But we try to validate these measures by having psychiatrists go back and blindly re-interview some of these people, and the experienced clinicians often come to the same conclusion. Of course, data collection is never perfect with psychiatry, because we can’t just say: “Well I took the temperature and it is 99.7.” We don’t have those kinds of cut-and-dry tests in psychiatry.

TCPR: Are the people assessed in these surveys the same kinds of serious patients that might make their way into my office for treatment?

Dr. Kessler: In general, yes. The data show that about 13% to 15% of the U.S. population is in some type of mental health treatment in a given year. But if you focus on the 5% or 6% of the population with major mental illness, 70% to 80% of them have been in treatment this past year. So there is a pretty strong relationship between the severity gradient and whether or not people are in treatment.

TCPR: By “in treatment,” do you mean under the care of a psychiatrist?

Dr. Kessler: “Treatment” can mean care by a psychiatrist, a primary care doctor prescribing antidepressants, or a social worker or marriage counselor, for example. But essentially, we find that people who have more severe disorders are more likely to be getting intensive treatment for their illnesses, such as more frequent visits, and specialty care from a psychiatrist or a clinical psychologist rather than a primary care doctor or a minister.

TCPR: If only 5% or 6% of people are seriously mentally ill, but 13% are in treatment, who are all these other people?

Dr. Kessler: We find that quite a few people who have had some kind of treatment don’t meet criteria for any disorder. These might be people who got divorced recently and see a minister, or perhaps someone who was sexually assaulted and sees a social worker, but doesn’t meet criteria for a mental disorder. These people typically have a small number of visits—one to three per year, versus say 10 to 12 for someone who is severely ill.

TCPR: You have been following the data from the NCS and NCS-R for many years. What changes have you seen?

Dr. Kessler: The NCS was done in the early 1990s and the NCS-R a decade later. We found that despite all the turbulence of the 1990s, there wasn’t much change in prevalence of disorders in a given year, between the early 1990s and the early years of the 21st century. However, we found a massive difference in patterns of treatment.

TCPR: More people got treatment in the early 2000s than in the early 1990s?

Dr. Kessler: Many more—the number of people who got treatment for mental disorders increased by 50% to 60% over that decade. This is the decade when all of a sudden there was direct to consumer advertisement—little sad faces saying they were depressed, and ads about social phobia, panic attacks, and OCD. And there were also campaigns at NIMH and various consumer organizations educating people that mental illness is an illness like any other, and that if you have a “chemical imbalance,” you should go get help—and many people did.

TCPR: For which illnesses did treatment increase the most? Was it mostly for depression, for example?

Dr. Kessler: Interestingly, the increase in treatment occurred mostly in mild or moderate disorders, and mostly for people going to see their primary care doctors. There was no shift in seeing specialists among people with severe disorders. And so, from the start to the end of the decade, among the people with really severe mental illness, 70% to 80% of them were still in treatment, 20% to 30% were still not in treatment. And even though the number of people getting treatment has been going up since the early 1990s, the number of those who are getting adequate treatment has been going down.

TCPR: How do you decide what is adequate treatment?

Dr. Kessler: We use the standard treatment guideline recommendations for saying whether a person had what we consider at least minimally adequate treatment. So, for example if a person has bipolar disorder and he or she is getting a benzodiazepine, we don’t consider that adequate treatment. Or, if a person was treated with “psychotherapy” and says she had two 15-minute sessions with her primary care doctor, we don’t consider that psychotherapy, because there is not any evidence from the literature that psychotherapy in that short of a duration is effective. We have found that adequate treatment only occurs for about one out of every four people in treatment.

TCPR: Do you have any suggestions for psychiatrists to ensure we’re giving our patients adequate treatment?

Dr. Kessler: The lack of adequate treatment happens more often among patients being treated by primary care doctors than specialists. It’s a little unfair to the primary care doctors, since they’re often the first ones to broach mental illness with their patients. By the time a psychiatrist sees a patient, he or she has probably gotten to the point of accepting that there is a mental illness. But for both types of doctors, collaboration regarding a patient’s treatment is always good.

TCPR: Thank you, Dr. Kessler.

This Month’s Expert: Psychiatric Epidemiology by Ron Kessler, PhD

This article originally appeared in:

The Carlat Psychiatry Report
Click on the image to learn more or subscribe today!

This article was published in print 3/2011 in Volume:Issue 9:3.


APA Reference
Kessler,, R. (2013). This Month’s Expert: Psychiatric Epidemiology by Ron Kessler, PhD. Psych Central. Retrieved on April 20, 2019, from


Scientifically Reviewed
Last updated: 5 Oct 2013
Last reviewed: By John M. Grohol, Psy.D. on 5 Oct 2013
Published on All rights reserved.