Mental Illness by the 12-Month Numbers: Who Had What In The Past Year?

Posted on July 27, 2012 by


It’s hard for me to imagine wanting a job where you go around knocking on strangers’ doors and asking if they have a mental illness–or maybe two or three.

But that’s precisely what researchers in the National Comorbidity Survey by Harvard University, the very first large-scale study of the prevalence of mental illness in the United States, did.

“Hi. I’m from Harvard. Have you ever suffered from bipolar disorder?” is how I imagine the conversation going, but the researchers got a lot of useful data, so perhaps they were more adept than I would be at feeling out the answers.

These conversations first occurred between the fall of 1990 to the spring of 1992 as the National Comorbidity Survey (NCS) did, in their words, “the first nationally representative mental health survey in the U.S. to use a fully structured research diagnostic interview to assess the prevalences and correlates of DSM-III-R [Diagnostics and Statistical Manual of Mental Disorders] disorders.”

And apparently it was such a big hit–that they did it again. Same questions, different year. The National Comorbidity Study Replication Study (NCS-R) took place between February 2001 and April 2003, when 9,282 interviews were conducted, this time using criteria from the DSM-IV. Aside from lifetime prevalencee of mental disorders, the study determined 12-month prevalence, or the percentage of participants who identified symptoms in the 12 months leading up to the interview. ["Hi. I'm from Harvard. Have you been drinking heavily this past year?" I guess is how that might play out.]

And since I promised you yesterday less talk and more visual, this is what I have to show for myself about the frequency of the occurrence of mental disorders from the NCS-R, occurring within a year of the interview: