Placebo effect large in treatment-resistant depression
Click Here to Manage Email Alerts
In randomized clinical trials, participants with treatment-resistant major depression experienced a large placebo effect across all treatment modalities, according to a meta-analysis in JAMA Network Open.
“Patients with [treatment-resistant depression (TRD)] often receive multiple treatment modalities and have lower rates of response and remission; thus, one would expect them to experience less benefit from the nonspecific effects of treatment,” Brett D. M. Jones, MD, MSc, of the University of Toronto’s department of psychiatry, and colleagues wrote. “This lack of response highlights the importance of quantifying the placebo effect in TRD, how it may differ across treatment modalities and what may contribute to it.”
Jones and colleagues analyzed 50 studies of 3,228 participants (mean age, 45.8 years; 54.8% women) with TRD available on PsychInfo, Web of Science and MEDLINE from inception to June 21. They categorized placebos and shams as pill placebo, liquid placebo, parenteral placebo, sham repetitive transcranial magnetic stimulation (rTMS), sham transcranial direct current stimulation (tDCS) or sham invasive brain stimulation, according to the study.
Overall, the effect size of placebos and shams was large (g = 1.05; 95% CI, 0.91 to 1.18; I2 = 76.19%), with parenteral placebo and sham tDCS having the greatest effect (g = 1.33; 95% CI, 0.63 to 2.04; I2 = 62.28%; and g = 1.32; 95% CI, 0.53 to 2.11; I2 = 52.57%, respectively). Liquid placebo had the least effect (g = 0.45; 95% CI, 0.26 to 1.15; I2 = 0%).
A significantly higher placebo effect was noted for industry-sponsored studies (B = 0.34; 95% CI, 0.09 to 0.59), year of publication (B = 0.03; 95% CI, 0.003 to 0.05) and studies that used an open-label prospective treatment phase before double-blind randomization (B = 0.35; 95% CI, 0.11 to 0.59).
When considering studies with a low risk of bias alone, the overall placebo effect was still large (g = 1.18; 95% CI, 0.98 to 1.38).
The pooled mean response rate was 21.2% among the 42 studies that reported it; the pooled remission rate was 13% among the 25 studies that reported it.
Jones and colleagues conceded that the study was limited by exclusion of RCTs using psychotherapy, electroconvulsive therapy and magnetic seizure therapy, nonconcrete definition of TRD and inability to directly compare two placebo modalities, though it provided information to help interpret the findings past and future studies.
“To better understand the placebo effect, the following improvements are needed: more consistent reporting of data, an agreement on a standard definition of TRD and its possible subgroups, and further assessment and reporting of participants’ expectations and experiences within a clinical trial,” they wrote.
In a related editorial, Cristina Cusin, MD, of Massachusetts General Hospital’s depression clinical and research program, and Luana Colloca, MD, PhD, of the University of Maryland’s department of pain and translational science symptoms, emphasized the need for more research on the placebo effect.
“While this meta-analysis showed that placebo responses are similar across treatment modalities in TRD (placebo, pharmacotherapy, brain stimulation, or psychotherapy), this is not the case for other disorders; for example, a meta-analysis in migraine prophylaxis pointed out that sham acupuncture and sham surgery resulted in higher placebo responses than oral placebos,” Cusin and Colloca wrote. “The fact that placebo response ratios vary across disorders and treatments outlines the need for more mechanistic research to be translated into clinical trial methodology and development.”