Placebo Response and Media Attention in Randomized Clinical Trials Assessing Cannabis-Based Therapies for Pain: A Systematic Review and Meta-analysis

Abstract

Importance: Persistent pain is a common and disabling health problem that is often difficult to treat. There is an increasing interest in medicinal cannabis for treatment of persistent pain; however, the limited superiority of cannabinoids over placebo in clinical trials suggests that positive expectations may contribute to the improvements.

Objective: To evaluate the size of placebo responses in randomized clinical trials in which cannabinoids were compared with placebo in the treatment of pain and to correlate these responses to objective estimates of media attention.

Data sources: A systematic literature search was conducted within the MEDLINE and Embase databases. Studies published until September 2021 were considered.

Study selection: Cannabinoid studies with a double-blind, placebo-controlled design with participants 18 years or older with clinical pain of any duration were included. Studies were excluded if they treated individuals with HIV/AIDS or severe skin disorders.

Data extraction and synthesis: The study followed the Preferred Reporting Items for Systematic Review and Meta-analyses reporting guideline. Data were extracted by independent reviewers. Quality assessment was performed using the Risk of Bias 2 tool. Attention and dissemination metrics for each trial were extracted from Altmetric and Crossref. Data were pooled and analyzed using a random-effects statistical model.

Main outcomes and measures: Change in pain intensity from before to after treatment, measured as bias-corrected standardized mean difference (Hedges g).

Results: Twenty studies, including 1459 individuals (mean [SD] age, 51 [7] years; age range, 33-62 years; 815 female [56%]), were included. Pain intensity was associated with a significant reduction in response to placebo, with a moderate to large effect size (mean [SE] Hedges g, 0.64 [0.13]; P < .001). Trials with low risk of bias had greater placebo responses (q1 = 5.47; I2 = 87.08; P = .02). The amount of media attention and dissemination linked to each trial was proportionally high, with a strong positive bias, but was not associated with the clinical outcomes.

Conclusions and relevance: Placebo contributes significantly to pain reduction seen in cannabinoid clinical trials. The positive media attention and wide dissemination may uphold high expectations and shape placebo responses in future trials, which has the potential to affect the outcome of clinical trials, regulatory decisions, clinical practice, and ultimately patient access to cannabinoids for pain relief.

Figure 1.. Flow Chart of Articles Included in the Meta-analysis

Figure 2.. Publication Year of Articles Included in the Meta-analysis

Number of publications per year for all 53 articles that met the inclusion criteria in the present meta-analysis from 1973 to 2021.

Figure 3.. Association Between Placebo and Change in Pain Intensity Ratings

The overall treatment response to placebo was statistically significant. The blue squares to the right of the midline represent improvements in pain intensity after treatment, squares on the midline represent no change, and squares to the left of the midline represent worsening of pain intensity after treatment.

Figure 4.. Association Between Media Attention and Effect Sizes of Individual Articles