The NIMH trials summarized. Is there an effective paradigm of care for mental health?

By Robert Whitaker author of Anatomy of an Epidemic: Magic Bullets, Psychiatric Drugs, and the Astonishing Rise of Mental Illness in America

In the past 15 years, the NIMH has funded a number of major, multicenter trials of drug treatments for mental disorders in adults and children, and although these studies have not been placebo-controlled, they still provide insight into how well drug-treated patients are faring over longer periods of time. In June, researchers will publish the one-year outcomes in the TEOSS trial, which assessed the merits of antipsychotics for early onset schizophrenia spectrum disorder, and thus it is now possible to summarize the results from five NIMH studies that looked at outcomes for patients treated with medications for 12 months or longer.


1) CATIE. In this study of antipsychotics for schizophrenia, 74% of the 1,432 patients stopped taking the assigned medication within 18 months, mostly because of “intolerable side effects” or the drug’s “inefficacy.” The atypical antipsychotics did not produce better results than the standard antipsychotic.

2) STAR*D. In this study of antidepressants, depressed patients who failed to respond to an initial medication were switched to another, with this process then repeated several times. In the initial stage of the trial, fifty-one percent of the 3,671 patients “remitted” at some point, which meant their depressive symptoms cleared. Then, during a one-yer followup study, 737 patients (20% of the original cohort) reported at some point that they were still doing well. But drop-out rates were high, and by the end of the 12-month followup, there were only 108 patients — 3% of the original cohort — still in the trial who had remitted and not relapsed.

3) STEP-BD. This large, 22-site study enrolled 4,360 bipolar patients from 1999 to 2005, and researchers conducted multiple randomized trials and naturalistic investigations to assess their outcomes. In regards to drug therapy, there were two primary findings. First, antidepressants were not found to be beneficial for bipolar patients. Second, in a one-year naturalistic follow-up study involving 1,742 patients, 409 (23%) remained well and in the trial throughout the 12 months. The remaining patients either dropped out (32%), or suffered one or more new mood episodes (45%).

Children and adolescents

4) MTA Trial. In this ADHD study, stimulants were basically compared to behavioral therapy, and at the end of 14 months, those treated with stimulants were doing better. Their core ADHD symptoms had abated to a greater degree, and there was a hint that their readings skills were better too.

The study then entered a second phase, in which the researchers periodically assessed how the children in the study were doing and whether they were taking a stimulant, and at the end of three years, “medication use was a significant marker not of beneficial outcome, but of deterioration. That is, participants using medication in the 24-to-36 month period actually showed increased symptomatology during that interval relative to those not taking medication.” In addition, those on stimulants had higher “delinquency scores, and they were also now shorter and weighed less than their non-medicated counterparts.

At the end of six years, the results were the same. Continued medication use was “associated with worse hyperactivity-impulsivity and oppositional defiant disorders symptoms,” and with greater “overall functional impairment.”

5) TEOSS Trial. In this study of antipsychotics as a treatment for early onset spectrum disorder, 54 out of 116 youth (ages 8 to 19) responded to the drug treatment in the initial eight weeks. The 54 responders were then entered into a 44-week “maintenance” study, and at the end of that period, only 14 youth were still on the study medication, with the remaining 40 dropping out, or going off the medication because of “inefficacy” or “intolerable” side effects. Thus, 12% of the initial cohort responded to an antipsychotic and were still taking the medication at the end of one year.

To sum up, the NIMH studies documented the following long-term findings:

  • 26% of schizophrenia patients in CATIE were able to stay on their assigned antipsychotic for 18 months.
  • 3% of the depressed patients in STAR*D remitted and then stayed well and in the trial throughout the 12 month followup. 23% of the bipolar patients in STEP-BD stayed well and remained in the study during a one-year follow-up.
  • Medication usage in the MTA ADHD study was a marker for deterioration at the end of three years, and associated with worse outcomes at the end of six years.
  • 12 percent of the early onset schizophrenia spectrum patients in TEOSS responded to an antipsychotic and were still taking the medication at the end of one year.

Now that these outcomes are in, our society can better address this question: Do these results tell of a successful paradigm of care?

First posted at Psychology Today and Mad In America. Published here with the permission of  the author.

About Monica Cassani

Author/Editor Beyond Meds: Everything Matters