Can Subjects With Major Depression Learn About Key Placebo Response Factors? The Effect of an Educational Placebo Response Video

CenExel is proud to highlight the poster presented at the​ Annual Meeting of the International Society for CNS Clinical Trials and Methodology Conference, February 2018, Washington, DC.

ABSTRACT​

Introduction: The challenge to reliably achieve signal detection in major depressive disorder (MDD) clinical trials cannot be overstated. MDD presents with the heaviest burden of disability among mental/behavioral disorders (NIH, 2017), yet less than 50% of double-blind, randomized, placebo-controlled trials (RCTs) demonstrate statistical superiority for active comparator drug over placebo (Khan et al., 2017). While researchers (Weber et al., 2005) have identified key factors which induce the placebo response (site-subject interactions, subject expectations of benefit, lack of subject understanding of placebo, and subject uncertainty of their role in the trial), the literature reveals few targeted methodological interventions which address the crucial Placebo Response Factors (PRFs). There are strategies addressing rater influence on placebo response (e.g., centralized raters), but they are not subject focused. MDD RCTs employing techniques aimed at educating subjects on PRFs can be additive to other strategies. 

Methods: MDD patients (N=64) participated in this pretest-posttest randomized-control group investigation. Subjects completed a Placebo Awareness Questionnaire (PAQ) containing 5 multiple choice questions to assess their awareness of the key PRFs. Upon completion of the PAQ, subjects were randomly assigned to the Control Group (CG) or Intervention Group (IG). The intervention was a seven-minute video educating subjects about the PRFs. Immediately after the IG subjects watched the video, they completed the same PAQ without access to their first responses. To match the temporality of the IG, the CG subjects completed the PAQ seven minutes after completing the first PAQ. 

Results: A repeated measures ANOVA showed a significant difference between the IG and the CG, demonstrating IG subjects were better able to identify the PRFs after watching the video (p < .001). Secondary analysis indicated this finding was true across age and gender (all p < .05 except for the 20-29 age group). Within-group analyses suggested a differential effect of the intervention on subgroups (e.g., IG females scored higher than IG males post video by 7 points) but not at a significant level. 

Conclusions: Despite MDE subjects prescreening for an RCT, these subjects could not identify PRFs. Subjects who watched a brief video had significantly greater awareness of these issues and their role in reducing placebo response as compared to subjects who did not watch the video. While a limitation of the study is that there were no outcome measures demonstrating improved signal detection, the current study results indicate that sponsors, research sites and rater training vendors should seriously consider implementing PRF training strategies for MDD subjects (e.g., the video used in this investigation or similarly purposed video) in order to potentially enhance signal detection and reduce placebo response. Learning style and attention theories may provide an explanation for why the 20‐29‐year‐old IG subgroup did not score significantly better than the CG of that same age group. It is recommended this study be replicated using other indications, evaluate the duration of improved understanding across study visits, and impact on placebo response.

INTRODUCTION

  • There is a disconcerting high (approximately 50%) placebo effect within major depressive disorder (MDD) double-blind, randomized, placebo-controlled trials (RCTs) (Khan et al., 2017). Walsh et al. (2002) found that over the past 20 years, subjects in depression RCTs responded to placebo by seven percent. ​
  • While various methodological strategies have been implemented or recommended to reduce the placebo effect (e.g., centralized ratings, remote rater monitoring, lessening assessment duration, subject duration of current illness exacerbation, and different lead-in phase procedures), evidence indicates the effect is only increasing as time progresses (Kemp et al., 2010; Loebel et al., 2010) with a profound risk for Big Pharma losing $2.5 Billion per new drug (Mullin, 2014).​
  • There is general consensus (e.g., Alphs et al., 2012; Weber et al., 2005) about the subject-producing causes of the high placebo rate or Placebo Response Factors (PRFs), including: ​
    • Lack of subject understanding of the placebo​
    • Subject expectations of benefit​
    • Subject misconception of expected interactions with research site staff​
    • Subject uncertainty of his/her role in the trial ​
  • Despite the PRFs, research has been arguably scarce in directly targeting how potential interventions specifically aimed at subjects (one of the clear sources of the placebo effect) may help minimize study participants’ response to placebo. As such, interventions focused on educating subjects about the PRFs is warranted. ​
  • The current study explores if a brief video aimed at educating subjects in a major depressive episode (MDE) helps them understand the PRFs, and if so, investigates potential variability among subjects’ gender and age regarding this increased PRF awareness. Examining if subjects can enhance their understanding of the PRFs is merits an empirical answer since arguably these factors are subjective and complex, perhaps particularly for patients who are in a MDE and may not have the cognitive energy and wherewithal to comprehend the PRFs.

METHODS

  • This study implemented a pretest-posttest randomized control group design, as described below.​
  • Subjects completed the site’s consent to treat paperwork and were diagnosed by qualified site staff as experiencing a current major depressive episode (MDE) per the American Psychiatric Association’s (2013) Diagnostic and Statistical Manual of Mental Disorders-Fifth Edition (DSM-5).​
  • Subjects then completed the Placebo Awareness Questionnaire (PAQ) containing 5 multiple choice questions (one correct answer per item) to assess subjects’ awareness of the key factors related to the placebo response. Percent correct answer was computed  for each subject. ​
  • The PAQ was developed specifically for this study to ascertain subject knowledge about the key placebo response factors (see left side Figure 1). ​
  • Upon completion of the Pre-Test PAQ, subjects were stratified based age and gender and randomly assigned to the Control Group (CG) or Intervention Group (IP) in a 1:1 ratio to ensure each group had an equivalent sample of these features. ​
  • Subjects in the IG immediately watched a 7-minute educational video after completing the Pre-Test PAQ. The video consisted of two of the current poster’s co-authors  (one with a DO and the other with a PHD degree) sitting in a chair facing the video camera and teaching (carefully reviewing) the four PRFs. Four chapters were identified in the video respective of the four PRFs. (Video transcripts are provided at the ISCTM poster session or by contacting the poster’s second author at ecohen@hritrials.com).​
  • At the conclusion of the video, IG participants completed the Post-Test PAQ without access to their first responses.
  • To match the temporality of the experimental group, the CG subjects completed the Post-Test PAQ seven minutes after completing their first PAQ.​
  • The order of the Post-Test PAQ questions for all subjects were changed to avoid response bias and ensure subjects read each question carefully (Rubin, 2005). ​
  • CG subjects did watch the placebo response video after completing the Post-Test PAQ to ensure the video intervention and its potential lessons were not withheld from any study subjects.​

 

Figure-1-Placebo-Awareness-Questionnaire

Figure 1- Placebo Awareness Questionnaire (PAQ; green font are correct answers shown for purposes of the poster).

RESULTS

  • There were a total of 64 subjects (see Table 1 for subgroup sample size)Table1-study-sample-size-by-experimental-group
  • As expected, there were no statistical differences of the Pre-Test CG and IG as a whole, as well as by gender or age. All study subjects seemed to have a similar poor understanding of the PRFs. For example, the randomized Pre-Test CG subjects scored M = 46.25 (SD = 19.30) on the PAQ and the randomized Pre-Test IG subjects scored M = 46.88 (SD = 22.40) (a difference of only 0.63 points) – see Figure 2 for an illustration.​
  • After testing for normality, a repeated measures two-way analysis of variance (ANOVA) was conducted (see Figure 2), indicating there was a significant difference (time by group interaction p < .001) between the Post-Test IG and CG overall, such that the IG was significantly better able to identify the factors (i.e., their role) associated with reducing a placebo response after watching the video as compared to the CG (Post-Test CG vs IG: M =  81.88, SD = 19.91 vs. M = 46.25, SD = 22.40). ​
  • The Post-Test CG and IG comparison results were replicated among males (interaction p = .003) and females (interaction p < .001; see Figure 2). Although not significantly different (p = .219), Post-Test IG females scored higher (M = 86.25, SD = 18.93) than their male counterparts (M = 77.50, SD = 20.49) by 9 points (see Conclusions Section). ​
  • The findings were also replicated across age groups (all p < .005), except for the 20-29 age sample (p = .224; see Figure 2). While the 30-39, 40-49 and 50+ age IG samples showed significant improvement on PAQ scores at Post-Test (M = 87.50-90.00) compared to their Pre-Test scores, the 20-29 age group did not (M = 57.50, SD = 16.69). ​

 

PAQ-mean-score

CONCLUSIONS

  • Our thorough review of the literature revealed that the current study is the only one which empirically examines potential differences of subjects in a major depressive episode by gender and age in relation to subjects’ understanding of the commonly cited PRFs known to cause a placebo effect in clinical trials (including MDD studies).​
  • The results of the study indicate that MDE subjects were quite naïve about the placebo response factors and this held true across gender and age. However, the results are encouraging because they indicate that a brief (7-minute) psychoeducational video which carefully addresses the PRFs significantly increased subjects’ understanding of these crucial issues no matter what gender and age (exception within the 20-29 age group). ​
  • Learning style and attention theories may provide a critical explanation for why the Post-Test IG 20-29-year-old subgroup (M = 57.50) did not score significantly better than the Post-Test CG (M = 42.50) of that same age group, although it is noteworthy that the IG scored higher by 15 points. The study’s 20-29-year-olds represents an amalgam of Generations Y (the Millennials) and Z which tend to have a shorter attention span and strong preference for multitasking, using digital technology, social stimulation, and visual graphics when learning new materials (G. Patel, 2017; Seemiller & Grace, 2016; Smith, 2012; Swanson, 2016). The medium of the 7-minute placebo response DVD lecture-oriented video may not have met these requirements. For example, subjects were not allowed to look at their cell phone while watching the video, but research has shown that Generation Y and Z respond to advertisements and also gain knowledge when they are able to use three or more devices simultaneously that provide instant answers (D. Patel, 2017; Wotapka, 2017).​
  • Given the above generational learning preferences, it is recommended that a quick (one-minute) placebo response script that encourages discussion or other forms of medium be explored as a placebo response educational tool.  ​
  • Post-Test IG females scored higher by 9 points than their male counterparts, although this difference was not significant (p=.219). This difference seemed to have been a function of the video since the Pre-Test IG females scored only 1.25 points higher than the Pre-Test IG males. Learning and cognitive style theories may also be applied to explain the Post-Test IG female and male 9 point difference. Males tend to prefer multimodal visual learning content compared to females (Sinha et al., 2013; Wehrwein et al., 2007). The placebo response video was unimodal and auditory: approaches preferred by females. Future tools aimed at enhancing males’ learning of the key placebo response factors should take these learning styles into account.  ​
  • Study limitations: (a) the study does not address whether subjects’ increased PRF knowledge translates to greater drug-placebo separation; and (b) it is unknown if subjects remember the PRFs throughout their trial participation (some sites have begun to read a script explaining these factors to each subject per study visit which should extend their understanding throughout the trial). Future studies should address these matters.​

Howard A. Hassman, DO1, Elan A. Cohen, Ph.D.1, Roberta R. Ball, DO1, Ashok V. Joseph, MD1, Katarzyna Wyka, Ph.D.3, M. Lobb, MS1, and Larry Ereshefsky, Pharm.D., BCPP1,2,4

1CenExel Hassman Research Institute 2CenExel Collaborative Neuroscience Network 3The City University of New York, Graduate School of Public Health and Health Policy 4Retired Professor, The University of Texas

A First-Time Investigation of a Subject Intervention to Reduce the Placebo and Nocebo Effects:​ A Multicenter, Randomized, Single-Blind, All Placebo Study of a​ Placebo-Control Reminder Script for Subjects with Major Depression​

CenExel is proud to highlight the poster presented at the Annual CNS Summit Meeting, November 2018, Boca Raton, FL.

ABSTRACT

Introduction: This investigation is the first known that empirically explores if educating subjects about key causes of the placebo effect significantly reduce placebo and nocebo effects. The key causes are Placebo Response Factors (PRFs), which include participant expectations of benefit, poor placebo understanding, misconception of expected interactions with site staff, and subject role uncertainty.

Methods: In this Institutional Review Board approved, US multicenter, single-blind, all placebo investigation, moderate to severe depressed patients aged 18-65 were randomized to the Control Group (CG) or Intervention Group (IG). IG subjects were read the Placebo-Control Reminder Script (PCRS) which reviewed the PRFs before the primary efficacy scale (self-reported Beck Depression Inventory-II) administration. CG subjects were not read the PCRS. Adverse Events were also collected to assess side effects. Subjects were informed of the 50% chance of being assigned placebo or active drug, yet all subjects received placebo. Given this deception, subjects were provided a Debriefing Form at the end of the study revealing the investigation’s true intent and procedures.

Results and Conclusions: This study is currently enrolling subjects with a requirement of 80 completers, powering the study to detect at least a small effect size (d≥.2). Results will be provided to the CNS Summit Poster Program Committee by middle October 2018. Ideally this proposed poster will be accepted regardless of its findings because the final data outcomes have crucial implications for placebo-controlled clinical trials (e.g., complexities of the placebo and nocebo effects would be discussed if the results indicated insignificant IG and CG separation).

INTRODUCTION

  • The high rate of placebo effect, which is approximately 50% within major depressive disorder (MDD) double-blind, randomized, placebo-controlled trials (RCTs) (Khan et al., 2017), has been found to only be increasing over time (Kemp et al., 2010).

  • While various methodological strategies have been implemented or recommended to reduce the placebo effect (e.g., centralized ratings, remote rater monitoring, data surveillance before subject is randomized, subject duration of current illness exacerbation, and different lead-in phase procedures), no subject targeted interventions aimed at reducing the placebo or nocebo effect were found by the authors of this study to have been empirically investigated. 

  • There is general consensus (e.g., Alphs et al., 2012; Weber et al., 2005) about the subject-producing causes of the high placebo rate or what we term Placebo Response Factors (PRFs), including: 

    • Lack of subject understanding of the placebo

    • Subject expectations of benefit

    • Subject misconception of expected interactions with research site staff

    • Subject uncertainty of his/her role in the trial 

  • While Hassman et al. (2017a, 2017b) found that subjects can enhance their understanding about PRFs compared to study participants who were not educated about the factors, no research could be found that explored if such an understanding reduces the placebo or nocebo effect. 

  • The current study is the first that these authors are aware of that examines whether a Placebo-Control Reminder Script (PCRS; see Figure 1), which reviews the PRFs and read to subjects with major depression, decreases their response to placebo and reporting of side effects (i.e., lessens the nocebo effect). 

METHODS

  • This study implemented a US multicenter (one site in the East and the other in the West Coast), randomized, single-blind, all placebo design aimed to mirror the methodologies typically used in MDD clinical trials, such as implementing conventional inclusion and exclusion criteria, multiple study visits, and evaluation of Adverse Events (AEs) and Serious Adverse Events (SAEs).

  • Also similar to other MDD trials, subjects were informed via the Informed Consent Form they have a 50 percent chance of receiving active medication or a placebo. However, as part of the methodology of the current study, all participants received placebo. 

  • Deception was necessary to assess for the placebo and nocebo effects and all subjects received a Debriefing Form at the end of their participation which revealed the true intent and procedures of the study.

  • The placebo was used as the Investigational Product (IP) because it allowed for specific measurement of the PCRS (the independent variable) to either decrease depression symptoms (the dependent variable) which would entail a placebo effect occurred, or help control for the placebo effect.

  • The Beck Depression Inventory-II (BDI-II; Beck et al., 1996) was used as the primary efficacy scale to assess depressive symptoms. Using this self-report scale was necessary given the single-blind design of the current study. 

  • Subjects digested the IP (total two white capsules at the Screening Visit and Visit 2) at their respective research site rather than at home each day of the week in order to ensure compliance and illuminate the risk that many clinical trials experience regarding study drug adherence at home (Shiovitz et al., 2016). To help rectify (equalize) the expectation by subjects of taking medication at home each day and further mirror clinical trial procedures, subjects were informed the two active medication capsules were developed to be sufficient to treat their depressive symptoms.

Key-inclusion-exclusion-criteria

study-procedures

PCRS-script-BIQ

RESULTS

  • There were 25 subjects within each of the Intervention and Control Groups (N = 50) and a Friedman rank sum test indicated no statistical differences by age/gender between these groups. 

  • After testing for normality, a repeated measures one-way analysis of variance (ANOVA) was calculated (see Table 1), indicating there was a significant difference, F(2, 41) = 700, p < .001, between the Intervention Group and the Controls, such that the Intervention Group subjects were significantly better able to identify the factors (i.e., their role) associated with reducing a placebo response after watching the video as compared to the Control Group participants.

  • Figure 2 illustrates the ANOVA based statistically significant mean differences between the Pre-Test and Post-Test Intervention and Control Group subjects (respectively [with Standard Deviations], 32.04 [4.14], 88.64 [8.96], 31.76 [3.49], 30.80 [2.18]).

  • A Wilcoxon signed rank test indicated significant differences between the Pre-Test and Post-Test Intervention Group (V=325, p<.001), suggesting again that the video intervention helped subjects learn the appropriate placebo response factors, whereas the Control Group subjects showed no statistical difference between their Pre-Test and Post-Test (V=77.50, p=.101).

  • After a Friedman rank sum test was conducted, ꭕ2(3) = 49.62, p < .001 (indicating the differences in the median values of the Intervention and Control Group Pre- and Post-Tests were statistically significant), a pairwise comparison was examined between each variable level to further explore the potential impact of having subjects watch the placebo response video. Figure 3 shows the significant differences between the pairwise comparisons, indicating that subjects learned more about the four placebo response factors as a result of the intervention (i.e., watching the video).

Pre-and-post-test-intervention-and-control-group

CONCLUSIONS

  • Theories abound regarding the ways in which subjects contribute to the placebo effect – notably the commonly cited factors of: their interactions with site staff, expectations of benefit, lack of subject understanding of the placebo, and subject uncertainty of his/her role in the trial.

  • To our knowledge, the current study is the only one which empirically examines how subjects initially are naïve about the above key subject-centered placebo contributing factors and how they may best be educated to enhance their awareness about these factors.

  • The results of the study indicated that subjects were less than clear about the factors which are commonly cited as contributing to the placebo effect in clinical trials and that a brief (7-minute) psychoeducational video focused on these factors significantly increased subjects’ understanding of these crucial issues. 

  • Given that the placebo response factors are centered around subject expectations during a clinical trial, the current results add to the literature concerning the role in which expectations have in generating a placebo effect. For example, Lidtsone et al.’s (2010) seminal publication found that the strength of PD subject expectations for improvement (even though all subjects were blindly given a placebo) predicted increased dopamine release (i.e., reward circuitry) in the ventral striatum and subjects also reported feeling better. The current study results indicate that indeed subject expectations can be managed/harnessed through a brief video regarding the placebo effect in clinical trials. 

  • The limitations of the current investigation include: (a) the study does not address whether subjects’ increased knowledge about the placebo effect key factors translates to greater drug-placebo separation; (b) It is unknown if subjects remember the placebo response key factors throughout their participation in trials (although some sites have begun to read a script explaining these factors to each study subject per study visit which should extend their understanding throughout the trial); and (c) it is unknown if subjects in this current study who were participating in different indication clinical trials (e.g., CNS vs Medical) were more likely to learn the key factors. Future studies should address these matters.

  • The results of the study are even more vital considering that subjects came to the clinic agreeing to participate specifically for a placebo-controlled clinical trial (e.g., many subjects were prescreened over the phone by an experienced clinician and were carefully informed about the details of a placebo-controlled trial). Given that many patients come into research sites with this understanding, it is RECOMMENDED that research sites or sponsors implement a video or similar method (e.g., script read to subjects by raters) to ensure subjects at the outset of their participation clearly understand their role and the explicit expectations of participating in a placebo-controlled study which are intended to reduce their potential biased symptom reporting.

Elan A. Cohen, Ph.D.1, Howard A. Hassman, DO1, David P. Walling, Ph.D.2 , Steven Hoover, Ph.D.2, Katarzyna Wyka, Ph.D.3 , Roberta R. Ball, DO1, Ashok V. Joseph, MD1, Jaclyn M. Lobb, MS,1 Dakota Hazzard-Randolph, MS1, and Larry Ereshefsky, Pharm.D., BCPP1,2,4

1CenExel Hassman Research Institute 2CenExel Collaborative Neuroscience Network 3The City University of New York, Graduate School of Public Health and Health Policy 4Retired Professor, The University of Texas