Analgesic efficacy of propranolol (60 mg twice a day) on facial pain in participants with temporomandibular disorder myalgia was modified by baseline expectations of pain relief (P = .07 for interaction at week 9 of treatment expectation and treatment group). A, For 80 participants with low treatment expectation, the placebo response was low (decreasing between the third and fourth visits), propranolol was efficacious, and the number needed to treat (NNT) was 3.2. B, For 118 participants with high treatment expectation, placebo response increased at each subsequent visit, and no therapeutic effect of propranolol was evident. Adjusted percentages and their 95% confidence intervals (error bars) were estimated with a log binomial generalized estimating equation regression model allowing for repeated visits by study participants, with adjustment for study site, sex, and self-reported race. P < .05 was the threshold for statistical significance.
eFigure. CONSORT Flow Diagram
Data Sharing Statement
Customize your JAMA Network experience by selecting one or more topics from the list below.
Identify all potential conflicts of interest that might be relevant to your comment.
Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.
Err on the side of full disclosure.
If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.
Not all submitted comments are published. Please see our commenting policy for details.
Sanders AE, Slade GD, Fillingim RB, Ohrbach R, Arbes SJ, Tchivileva IE. Effect of Treatment Expectation on Placebo Response and Analgesic Efficacy: A Secondary Aim in a Randomized Clinical Trial. JAMA Netw Open. 2020;3(4):e202907. doi:10.1001/jamanetworkopen.2020.2907
Amid the United States’ chronic pain crisis, novel analgesics are failing to show efficacy in clinical trials.1 High failure rates are attributed to an upward trend in placebo response,2 driven by patients’ heightened expectation of treatment benefit.3 We hypothesized that heightened expectations differentially amplify placebo analgesia, leading to underestimation of the treatment effect in randomized clinical trials.
SOPPRANO (Study of Orofacial Pain and Propranolol) is a double-blind, placebo-controlled, parallel-group, phase 2b randomized clinical trial that enrolled 200 adults aged 18 to 65 years with examiner-verified temporomandibular disorder–associated myalgia from August 1, 2015, to January 31, 2018, at 3 US study sites. Participants were randomized 1:1 to propranolol hydrochloride (60 mg twice a day) or placebo administered for 9 weeks. Using daily pain diaries, treatment response was defined as the proportion with at least 30% reduction in mean pain index (facial pain intensity multiplied by duration) at 9 weeks. Efficacy was further quantified as the number needed to treat (NNT) with 95% CIs. Treatment expectation was determined from participants’ baseline expectation that the study treatment would reduce their facial pain. Ratings of moderate or strong were classified as high treatment expectation, and ratings of none or slight were classified as low expectation. In this planned, intention-to-treat analysis, we tested whether treatment expectation modified the analgesic efficacy of propranolol using a log binomial generalized estimating equation regression model incorporating data from 4 study visits with adjustment for covariates. The generalized score statistic for generalized estimating equation models was used to test for modification of analgesic efficacy. A 2-tailed P < .05 was deemed statistically significant, and values of .05 to .10 were deemed credible. Other results were deemed statistically significant when 2-tailed 95% CIs excluded the null value. A separate logistic regression model assessed the odds of high treatment expectation using continuous measures of clinical and experimental pain, psychological factors, and health-related quality of life at baseline, all standardized to z scores.
SOPPRANO’s study flow diagram and checklist are not yet published.4 The trial protocol and statistical plan are available in Supplement 1. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines, and the study flow diagram is available in the eFigure in Supplement 2. The institutional review boards at each site approved the trial protocol. All participants provided written informed consent.
Among 198 participants in the intention-to-treat sample who rated their treatment expectation, the mean (SD) age was 34 (0.90) years, 155 (78.3%) were women, and 118 (59.6%) had a high expectation of pain relief. Pressure pain sensitivity in the masseter muscle (odds ratio, 1.5; 95% CI, 1.1-2.0) and trapezius muscle (odds ratio, 1.5; 95% CI, 1.1-2.0) was associated with a heightened expectation of pain relief (Table). Among participants with low expectation of pain relief (Figure, A), treatment responders composed 73.5% in the propranolol group and 42.7% in the placebo group. This difference corresponded to an NNT of 3.2 (95% CI, 1.9-11.8; P = .007). In the high-expectation stratum (Figure, B), treatment responders composed 67.0% and 63.6% in propranolol and placebo groups, respectively. The heightened response in the placebo group nullified efficacy (NNT = 29.6; 95% CI, 4.4 to −6.3; P = .73). The P value for this interaction was .07.
In this study, propranolol was superior to placebo, but only among participants whose expectations of treatment were modest. In the presence of heightened expectations, placebo analgesia overwhelmed the efficacy signal, inflated the NNT, and nullified differences between treatment groups. The relatively small sample size limited the power to test if the magnitude of effect of propranolol on temporomandibular disorder–associated pain differed by treatment expectation, yet our findings offer credible evidence of interaction. Few clinical or psychological factors were associated with heightened expectation, with sensitivity to experimental pressure pain being the exception. Greater pain sensitivity may manifest as greater need for and expectation of pain relief. It is well established that treatment expectations are susceptible to verbal suggestion, physician manner, and social observation.5 A more recent influence may be information relayed through direct-to-consumer advertising of prescription drugs. Direct-to-consumer advertising budgets for prescription drugs more than quadrupled in the United States from 1997 to 2016,6 coinciding with the steady increase in placebo response. We recommend assessing treatment expectation to better understand its potential bias on success rates in clinical trials.
Accepted for Publication: February 20, 2020.
Published: April 16, 2020. doi:10.1001/jamanetworkopen.2020.2907
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Sanders AE et al. JAMA Network Open.
Corresponding Author: Anne E. Sanders, PhD, Adams School of Dentistry, Division of Pediatric and Public Health, University of North Carolina at Chapel Hill, CB #7455, Chapel Hill, NC 27599 (email@example.com).
Author Contributions: Dr Slade had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: All authors.
Acquisition, analysis, or interpretation of data: All authors.
Drafting of the manuscript: Sanders, Slade, Tchivileva.
Critical revision of the manuscript for important intellectual content: Slade, Fillingim, Ohrbach, Arbes, Tchivileva.
Statistical analysis: Sanders, Slade, Arbes.
Obtained funding: Slade, Fillingim, Tchivileva.
Administrative, technical, or material support: Fillingim, Arbes.
Conflict of Interest Disclosures: Dr Arbes reported receiving grants from the University of North Carolina at Chapel Hill during the conduct of the study. No other disclosures were reported.
Funding/Support: This study was supported by grants R34-DE022088 and U01-DE024169 from the National Institute of Dental and Craniofacial Research, National Institutes of Health.
Role of the Funder/Sponsor: The sponsor had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Trial Registration: ClinicalTrials.gov Identifier: NCT02437383
Data Sharing Statement: See Supplement 3.
Additional Contributions: We thank SOPPRANO research staff and participants who participated in the trial. We also thank cardiologists Alan Hinderliter, MD (University of North Carolina at Chapel Hill), David Sheps, MD, PhD (University of Florida), and Thomas Cimato, MD, PhD (University at Buffalo) for monitoring participants’ safety.