The Underappreciated Placebo Effects and Responses in Randomized Controlled Trials on Neck Pain: A Systematic Review With Meta-analysis.

OBJECTIVE:To quantify placebo effects and responses in randomized controlled trials (RCTs) on neck pain and explore how they would influence the treatment of neck pain. DATA SOURCES:We searched MEDLINE (PubMed), EMBASE (Ovid), CINAHL (EBSCO), Physiotherapy Evidence Database (PEDro), and World Health Organization International Clinical Trials Registry Platform from the inception of August 15, 2021, to identify relevant RCTs. STUDY SELECTION AND DATA EXTRACTION:The abstracts and full texts of potential studies were independently screened, and data extraction was also independently performed by 2 researchers. Scales of the score measuring neck pain and the scores both at baseline and the endpoint were extracted. DATA SYNTHESIS:A total of 60 RCTs were included. The mean improvement in the pain score after placebo treatment was 15.65 (mean difference [MD]=-15.65, 95% confidence interval; CI [-19.19, -12.12]; P<.05), which we defined as the placebo response. In the active groups, it was 25.91 (MD=-25.91, 95% CI [-29.15, -22.68]; P<.05), and in the no-treatment groups, it was 5.80 (MD=-5.80, 95% CI [13.28, 1.69]; P=.13). Using the 3 MDs from the 3 groups, the placebo effect was calculated to account for 38.0% of the pain score improvement in the active group. CONCLUSIONS:The pain scores of patients with neck pain were reduced after treatment with placebos, but the magnitude of pain score reduction was not clinically significant enough. The 38.0% amount of pain score reduction in patients treated with active interventions was caused by placebo. Interventions with considerable clinically significance for neck pain were still required.