Psychiatry Practice Changers 2014

Bret S. Stetka, MD; Cristoph U. Correll, MD


December 08, 2014

In This Article

Treatment Meta-analysis: The Efficacy of Psychiatric Interventions

In part to address the continued public and professional controversy over the effectiveness of psychiatric treatments, an extensive meta-analysis published earlier this year in JAMA Psychiatry[13] set out to assess the efficacy and safety of pharmacotherapies and psychotherapies in treating mental illness. Performing a "meta-review" of 62 meta-analyses on 20 psychiatric disorders containing 693 trials and 113,833 participants, the authors calculated effect sizes for efficacy outcomes derived from the latest meta-analysis that compared either: (1) psychopharmacologic or psychotherapeutic interventions with control groups, (2) both treatment modalities head to head, or (3) combination vs monotherapy strategies. A prior meta-review of 94 meta-analyses that pooled together 16 medications for eight psychiatric disorders had a comparable medium effect size as 48 medications for 20 medical diseases (0.49, confidence interval [CI]: 0.41-0.57 vs 0.45, CI: 0.37-0.53).[14]

Most acute effect sizes were medium (0.4-0.5, number needed to treat (NNT): 5-9), maintenance effect sizes were somewhat larger (0.6-0.8, NNT: 3-6), and effects of adjunctive strategies in partial responders/nonresponders were small (0.2-0.3) or insignificant. Effect sizes in psychotherapy studies tended to be larger (0.58, CI: 0.40-0.76) than in pharmacotherapy studies (0.40, CI: 0.28-0.52), but this indirect difference was generally not confirmed in head-to-head studies.

The largest effect size treatments in psychiatry (>0.75) were methadone for treatment retention in opiate dependence, electroconvulsive therapy for major depressive disorder (MDD), benzodiazepines for sleep duration, antipsychotics for relapse prevention in schizophrenia, selective serotonin reuptake inhibitors (SSRIs) for relapse prevention in generalized anxiety disorder and social phobia, amphetamines for response in adult attention-deficit/hyperactivity disorder, as well as cognitive behavioral therapy (CBT) for obsessive compulsive disorder, bulimia and binge eating disorder, habit reversal for trichotillomania, and family therapy for anorexia nervosa.

By contrast, interventions for alcohol dependence yielded small effect sizes as did psychotherapy for dysthymia. Effects were nonsignificant for antipsychotics or antidepressants for anorexia nervosa, SSRIs for trichotillomania, psychodynamic therapy for schizophrenia, and dialectic behavioral therapy for borderline personality disorder. In head-to-head studies, drug treatment was significantly superior to psychotherapy for schizophrenia (psychodynamic) and dysthymia (CBT), while psychotherapy was superior to drug treatment for MDD relapse prevention and bulimia. Adding drug treatment to psychotherapy was significantly more effective than psychotherapy for MDD, social phobia, and bulimia. Conversely, adding psychotherapy to drug treatment was superior to drug treatment for panic disorder and bulimia.

Taken together, results from this large meta-review of meta-analyses that covered all major psychiatric disorders in adults reinforced that effect sizes of psychiatric treatments are not worse than those of other medical fields. Moreover, although methodologies differed across evaluated meta-analyses, some of which biased psychotherapeutic interventions towards having larger effect sizes, results indicate that both medications and psychotherapy have their strengths and often worked even better in combination than each modality on its own.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.