With regard to major depressive disorder, data from the Baltimore Epidemiologic Catchment Area study—a survey involving more than 175,000 household residents– revealed no evidence of an “epidemic of depression” in the period from 1981-2004.
This timeframe is the same period that witnessed a marked increase in the use of antidepressant medication in the U.S. If antidepressant treatment was truly “driving” an epidemic of major depression, we would expect to see a sharply rising incidence and/or prevalence of this condition—but neither has been detected for the U.S. population as a whole.
Further confirmation of the “non-epidemic” of major depression is provided by eight-year NSDUH data. These show that the percentage of adults who had a major depressive episode in the past year remained stable between 2005 (6.6 percent) and 2013 (6.7 percent).
What are Reliable Measures?
Some who argue that actual psychiatric illness is on the rise in the U.S. point to increasing rates of psychiatric disability determinations—for example, increased numbers of those “disabled” by mental disorders who qualify for Supplemental Security Income (SSI) or Social Security Disability Insurance (SSDI).
But SSI/SSDI determinations are in no sense a reliable measure of the actual incidence (new cases) or prevalence (total cases at a given time) of psychiatric disorders—nor are they necessarily an indicator of treatment success or failure.
Disability determinations by the Social Security Administration (SSA) are largely administrative hearings—not medical-psychiatric evaluations—and may depend, for example, on how well-prepared the applicant is when presenting his or her “case” to the evaluator—who may or may not have medical or psychiatric expertise.
Similarly, rising rates of medication prescription—though of potential concern from the clinical and societal standpoint—cannot serve as a reliable proxy for actual incidence or prevalence rates. Prescribing patterns may be subject to fluctuations owing, for example, to rates of direct-to-consumer advertising or to public and professional awareness of a particular condition.
Neither can office-based rates of psychiatric diagnoses serve as reliable indices of illness incidence or prevalence, since clinicians differ widely in their application of DSM diagnostic criteria, which are often ignored.
So, for example, rising rates of office-based diagnosis of bipolar disorder in children and adolescents—admittedly, a matter of societal and medical concern—do not necessarily indicate an increase in actual cases of bipolar disorder in the general population.
Similarly, rates of treatment for a given disorder may be misleading with respect to actual prevalence of the illness, since psychiatric treatment is widely subject to socioeconomic variables like insurance coverage and access to mental health practitioners.
In sum: though important for several reasons, none of these measures—disability rates, prescribing patterns or rates of office-based diagnosis—is a valid means of determining an illness’s incidence or prevalence. None is a substitute for applying consistent diagnostic criteria and/or using structured clinical interviews in comparable populations over long periods of time.
Ideally, clinical assessments should be made at the time of the person’s illness, rather than inferred retrospectively via survey responses—but on a large scale, this undertaking would be daunting. Unfortunately in the U.S., we have very few carefully collected, prospective data with which to track prevalence and incidence of psychiatric disorders, from childhood through adulthood.