<%BANNER%>

Factors Affecting Placebo Acceptability

Permanent Link: http://ufdc.ufl.edu/UFE0042983/00001

Material Information

Title: Factors Affecting Placebo Acceptability Deception, Outcome and Disease Severity
Physical Description: 1 online resource (40 p.)
Language: english
Creator: KISAALITA,NKAKU R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ANALGESIA -- ETHICS -- PLACEBO
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A burgeoning body of evidence supports the efficacy and elucidates the mechanisms of placebo analgesia. Considerable debate persists, however, concerning the ethics of their use, with many of the present arguments being primarily philosophically-based. Given the magnitude of placebo analgesic effects and the frequency of physician-administered placebos, the present study empirically investigated the acceptability of an analgesic placebo treatment among lay individuals. A total of 103 participants (M age = 22.67, SD = 4.57) completed a web-based study examining placebo attitudes and consequences by responding to vignettes depicting patients receiving a placebo analgesic. We experimentally manipulated 1) placebo treatment instructions (level of deception), 2) treatment outcome, and 3) patients? pain severity. Participants rated vignettes on outcome measures of deception, physician-patient relationship, and patient mood. Participants then characterized a range of placebo acceptability through ratings of deceptiveness, effectiveness, and negative consequences. Results showed that placebos described as ?medication shown to be a powerful analgesic in some people? were equally deceptive as those described as ?standard drug treatment.? Placebo deceptiveness was primarily determined by the nature of its administration. However, ratings of patient mood and physician approval were determined as much by treatment instruction as by treatment outcome. Furthermore, consistent with recent placebo studies, negative mood ratings were largely unaffected by placebo administration if they resulted in analgesia, indicating that an analgesic response mitigated the negative consequences of deceptive administration. Participants suggested they would tolerate moderate effectiveness and considerable negative consequences in an acceptable placebo, though results suggest that lay individuals may not have a sophisticated conceptualization of placebo effectiveness. Studies designed to alter individuals? understanding of placebo effectiveness and mechanisms are needed to identify additional factors determining placebo acceptability.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by NKAKU R KISAALITA.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Robinson, Michael E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-10-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042983:00001

Permanent Link: http://ufdc.ufl.edu/UFE0042983/00001

Material Information

Title: Factors Affecting Placebo Acceptability Deception, Outcome and Disease Severity
Physical Description: 1 online resource (40 p.)
Language: english
Creator: KISAALITA,NKAKU R
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: ANALGESIA -- ETHICS -- PLACEBO
Clinical and Health Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, M.S.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: A burgeoning body of evidence supports the efficacy and elucidates the mechanisms of placebo analgesia. Considerable debate persists, however, concerning the ethics of their use, with many of the present arguments being primarily philosophically-based. Given the magnitude of placebo analgesic effects and the frequency of physician-administered placebos, the present study empirically investigated the acceptability of an analgesic placebo treatment among lay individuals. A total of 103 participants (M age = 22.67, SD = 4.57) completed a web-based study examining placebo attitudes and consequences by responding to vignettes depicting patients receiving a placebo analgesic. We experimentally manipulated 1) placebo treatment instructions (level of deception), 2) treatment outcome, and 3) patients? pain severity. Participants rated vignettes on outcome measures of deception, physician-patient relationship, and patient mood. Participants then characterized a range of placebo acceptability through ratings of deceptiveness, effectiveness, and negative consequences. Results showed that placebos described as ?medication shown to be a powerful analgesic in some people? were equally deceptive as those described as ?standard drug treatment.? Placebo deceptiveness was primarily determined by the nature of its administration. However, ratings of patient mood and physician approval were determined as much by treatment instruction as by treatment outcome. Furthermore, consistent with recent placebo studies, negative mood ratings were largely unaffected by placebo administration if they resulted in analgesia, indicating that an analgesic response mitigated the negative consequences of deceptive administration. Participants suggested they would tolerate moderate effectiveness and considerable negative consequences in an acceptable placebo, though results suggest that lay individuals may not have a sophisticated conceptualization of placebo effectiveness. Studies designed to alter individuals? understanding of placebo effectiveness and mechanisms are needed to identify additional factors determining placebo acceptability.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by NKAKU R KISAALITA.
Thesis: Thesis (M.S.)--University of Florida, 2011.
Local: Adviser: Robinson, Michael E.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-10-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0042983:00001


This item has the following downloads:


Full Text

PAGE 1

1 FACTORS AFFECTING PLACEBO ACCEPT ABI LITY: DECEPTION OUTCOME AND DISEASE SEVERITY By Nkaku R obert K isaalita A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENT S FOR THE DEGREE OF MASTER OF SCIENCE UNIVERSITY OF FLORIDA 2011

PAGE 2

2 2011 Nkaku Robert Kisaalita

PAGE 3

3 To my family member s friends, and colleagues who have steadfastly supported me in this journey towards completing this academic and professional m ilestone

PAGE 4

4 ACKNOWLEDGMENTS I wanted to thank my advisor, Dr. M ichael Robinson, for his patience and guidance through out this process. Furthermore, I would like to thank Daniela Roditi for her contributions in writing the journal article associated with this manuscript. Also, I would like to recognize members of my supervisory committee: Dr. David Jani c ke, Dr. Vonetta Dotson, and Dr. Glenn Ashkanazi Finally, I want to recognize my friends, family members, and colleagues for their support throughout this rewarding endeavor.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIG URES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ............................. 9 ABSTRACT ................................ ................................ ................................ ................... 10 CHAPTER 1 INTRODUCTIO N ................................ ................................ ................................ .... 12 2 METHODS AND MATERIALS ................................ ................................ ................ 14 Participants ................................ ................................ ................................ ............. 14 Procedure ................................ ................................ ................................ ............... 14 Part 1: Placebo Vignettes ................................ ................................ ................. 14 Part 2: Acceptability Survey ................................ ................................ .............. 16 Measur es ................................ ................................ ................................ ................ 16 Statistical Analyses ................................ ................................ ................................ 17 Principle Axis Factoring ................................ ................................ .................... 17 Factor ial Repeated Measures Analyses of Variance (ANOVA) ........................ 18 3 RESULTS ................................ ................................ ................................ ............... 21 Extent Deception ................................ ................................ ................................ .... 21 Extent Deception Main Effects ................................ ................................ ......... 21 Extent Deception Interactions ................................ ................................ ........... 21 Physician Approval Factor ................................ ................................ ...................... 22 Physician Approval Main Effects ................................ ................................ ...... 22 Physician Approval Interactions ................................ ................................ ....... 22 Nega tive Mood Factor ................................ ................................ ............................. 23 Negative Mood Main Effects ................................ ................................ ............. 23 Negative Mood Interactions ................................ ................................ .............. 23 Acceptability Survey ................................ ................................ ................................ 24

PAGE 6

6 4 DISCUSSION ................................ ................................ ................................ ......... 31 Extent Deception ................................ ................................ ................................ .... 31 Physician Approval ................................ ................................ ................................ 32 Negative Mood ................................ ................................ ................................ ........ 33 Acceptability Survey ................................ ................................ ................................ 33 Implications ................................ ................................ ................................ ............. 34 Conclusions ................................ ................................ ................................ ............ 36 LIST OF REFERENCES ................................ ................................ ............................... 37 BIOGRAPHICAL SKETCH ................................ ................................ ............................ 40

PAGE 7

7 LIST OF TABLES Table page 2 1 Placebo vignettes and a cceptability s urvey o utcomes ................................ ........ 19 3 1 Placebo v ignette o utcom e m eans (VAS = 0 100) ................................ ............... 25 3 2 Placebo v ignette s main e ffec t, interactions and c ontrasts (VAS = 0 100) .......... 26 3 3 Acceptability s urvey m eans (VAS = 0 100) ................................ ........................ 29

PAGE 8

8 LIST OF FIGURES Figure page 2 1 Placebo v ignettes b ase. ................................ ................................ ..................... 20 3 1 Acceptability s urvey g raph. ................................ ................................ ................. 30

PAGE 9

9 LIST OF ABBREVIATION S ANOVA Analysis of variance B Got better/improved EA Entirely acceptable EP Enhanced placebo EU Entirely unacce ptable FD Full deception MA Marginally acceptable NC Remained the same/no change NP Non progressive pain 2 Partial eta squared PAF Principle axis factoring PP Progressive pain RA Random a ssignment VAS Visual analogue scale W Worsened

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillmen t of the Requirements for the Degree of Master of Science FACTORS AFFECTING PLACEBO ACCEPT ABI LITY: DECEPTION OUTCOME AND DISEASE SEVERITY By Nkaku Robert K isaalita May 2011 Chair: Michael E. Robinson Major: Psychology A burgeoning body of evidence s upports the efficacy and elucidates the mechanisms of placebo analgesia. Considerable debate persists, however, concerning the ethics of their use, with many of the present arguments being primarily philosophically based. Given the magnitude of placebo ana lgesic effects and the frequency of physician administered placebos, the present study empirically investigated the acceptability of an analgesic placebo treatment among lay individuals. A total of 103 participants ( M age = 22.67, SD = 4.57) completed a web based study examining placebo attitudes and consequences by responding to vignettes depicting patients receiving a placebo analgesic. We experimentally manipulated 1) placebo treatment instructions (level of deception), 2) treatment outcome, and 3) physician patient relationship, and patient mood. Participants then characterized a range of placebo acceptability through ratings of deceptiveness, effectiveness, and negative consequences.

PAGE 11

11 ature of its administration. However, ratings of patient mood and physician approval were determined as much by treatment instruction as by treatment outcome. Furthermore, consistent with recent placebo studies, negative mood ratings were largely unaffecte d by placebo administration if they resulted in analgesia, indicating that an analgesic response mitigated the negative consequences of deceptive administration. Participants suggested they would tolerate moderate effectiveness and considerable negative c onsequences in an acceptable placebo, though results suggest that lay individuals may not have a sophisticated conceptualization of placebo effectiveness. mechanisms are need ed to identify additional factors determining placebo acceptability.

PAGE 12

12 CHAPTER 1 INTRODUCTION Placebo effects have the potential to augment the efficacy of all active medical treatments and procedures. 22 Despite significant advances in understanding the e ffect, considerable debate persists concerning placebo acceptability and ethics. 11, 24, 27 Opponents of placebo comparisons have cited a number of moral and ethical arguments against their use including violation of personal autonomy and the integrity of t he physician patient relationship, psychological harm, and deception. 4, 22, 26, 31 Conversely, placebo proponents state that under certain circumstances placebo comparisons are the only means to examine behaviors and evaluate treatment efficacy, and that a degree of deception is necessary to invoke the effect. 2, 8, 12, 19 There appears to be a discrepancy between research and clinical use of placebo. Reports indicate that physicians use placebo treatments, 9, 10, 15, 25, 30 with one study reporting that o ver half of surveyed physicians administered placebo interventions on a regular basis. 32 This issue towards placebo use, although one focus group study found that placebo use was associated with negative attitudes towards medical research. 1 The use and consequences of deception are focal points of the placebo ethics debate. 13, 23 Despite the claims of a number of ethicists, empirical evidence of the negative consequences of deception remains inconsistent. 17 To date, the literature pertaining to placebo acceptability and deception has been based largely on philosophical and theoretical tenants, with relatively few data driven empirical studies. 2, 5, 13, 18, 21 Chung et al. 7 found that placebo for pain relief caused no negative effects on mood or placebo responses even after the use of placebo was revealed to participants.

PAGE 13

13 Another study by Martin & Katz 20 found no significant differences in mood nor the magnitude of placebo analgesia between two groups differentiated by whether or not they were informed of potential deception while consenting to study participation. A that patients wer e generally open to receiving placebo interventions, though they lacked understanding about how placebos worked. 6 The current study attempted to expand upon the sparse empirical literature on the acceptability of placebo interventions for pain. In this vi gnette analog study, systematic manipulation of 1) the instructions used in administering the placebo (hypothesized to represent a range of deceptiveness), 2) the outcome of the treatment, and 3) the to examine how these factors influence the acceptability and ethics of a placebo treatment for pain. We hypothesized that placebo in scenarios in which patients were completely deceived by treatment instructions, received worse treatment outcomes, and exp erienced a more serious pain condition would be rated as less acceptable; conversely, placebo in scenarios in which patients were randomly assigned to treatment, had reduced pain outcomes, and experienced a less serious pain condition would be rated as mor e acceptable. Finally,

PAGE 14

14 CHAPTER 2 METHODS AND MATERIAL S Participants Participants were comprised of 103 adults (76 females and 27 males; M a ge = 22.61, SD = 4.52) with the only inclusion criteria being that participants were 18 years old or older. The majority of participants were students as 59 participants (57.3 percent) me graduate Almost a third of the sample ( N racial breakdown of the sample was as follows: White ( N =64); Black ( N =9); Hispanic ( N = 9); Asian or Pacific Islander ( N =13); Indian (N=4) and Other ( N =4). The s tudy was advertized through flyers posted throughout the University of Florida campus. Upon study completion participants provided a name and postal address to receive a $15 gift card in the mail. Procedure This study was reviewed and approved by the Univ ersity of Florida Institutional Review Board and informed consent was obtained from each subject. The questionnaire took approximately 45 60 minutes to complete and responses were anonymous. The study was completed entirely online. Participants were provid ed the online URL for the study in addition to a unique username and password. The study was comprised of two separate sections: the 1) Placebo Vignettes and the 2) Acceptability Survey. Part 1: Placebo Vignettes The Placebo Vignettes consisted of 18 vigne ttes, each describing the outcome of a physician administered placebo intervention for patients experiencing pain. The base

PAGE 15

15 vignette c an be found in Figure 2 1 The order in which the vignettes were presented was randomized for each participant. Three fact ors varied per vignette: the treatment to the patients, and the outcome/effectiveness of the placebo treatment. There were two levels of pain progressiveness. The patient s in each vignette were being treated for either non progressive pain/pain that is stable (NP) or progressive and progressive pain, respectively. Although participants may have been responding to any of the descriptors in responding to a vignette, for purposes of simplicity only the terms non progressive pain and progressive pain will be used in this manuscript. Th previous placebo analgesia studies and has been proposed to be an ethically permissible placebo suggestion. 27, 33, 34 For the full deception (FD) instruction, patients r pain. For the random assignment (RA) instruction, patients were informed by their manipulat ion of degree of deception. The purpose of this manipulation was to study/investigate the individual differences in the consequences of learning that one has been deceived.

PAGE 16

16 There were three levels for outcome/effectiveness of the placebo treatment. Upon co unaffected by the treatment (NC), or worsened (W). Part 2: Acceptability Survey The Acceptability Survey, the final section of the study, was completely distinct from the Placebo Vignettes in structure and response format. Participants were asked to rate three levels of acceptability regarding placebo use: an entirely unacceptable placebo (EU), a marginally acceptable placebo (MA), and an entirely acceptable placebo (EA). E ach level of acceptability was rated using a combination of three separate Visual Analogue Scales (VAS) (0 100) measuring the relative contribution of 2 1). Following the VAS ratings, participants were asked to re Measures Visual analogue scales (vas) VAS measurements were used in the Placebo Vignettes and in the Acceptability Survey. After reading a vignette, VAS ratings were following eight questions served as outcome measures : (1) the extent that there was deception (Extent Deception); (2) rating of trust in the physician (Dr. Trust); (3) rating of confidence in the physician (Dr. Confidence); (4) rating of willingness to see the

PAGE 17

17 physician (Dr. Willingness); (5) rating of moral character of the ph ysician (Dr. Moral Character); (6) rating of how angry the patient would be if they knew they had received a placebo (Pt. Anger); (7) how anxious the patient would be if they knew they had received a placebo (Pt. Anxiety); (8) how depressed the patient wou ld be if they knew they had received a placebo (Pt. Depression). Previous research in our lab has shown that pain related negative emotions, such as anger, depression, and anxiety, typically load highly on to a latent negative mood factor; thus it was hyp othesized that these three outcomes measure would load highly on to a latent negative mood factor. 14, 28 Furthermore we had no separate a priori hypotheses about individual mood measures. It should be noted that in the Placebo Vignettes, deception is used a both an independent variable (i.e. by proxy through treatment instruction levels) and a dependent measure (i.e. Extent Deception VAS ratings). For Acceptability Survey VAS descriptions refer to the previous section and Table 2 1. Statistical Analyses Pri nciple Axis Factoring intended to be analyzed separately a priori. However because of the potential multi collinearity of the Placebo Vignettes items, a Principle Axis Factoring (PAF) with oblique rotation was conducted for seven of the eight vignette VAS outcome variables to reduce the number of subsequent analyses and to determine if the three mood related measures loaded on to a single latent negative mood factor. An initial PA F was run that included all eight outcome measures and showed that the Extent Deception measure had low loadings on all factors (highest loading of .451); Thus, the Extent Deception

PAGE 18

18 measure was excluded from this analysis because of its marginal factor loa dings and because we had a priori interest in examining its unique relationship with placebo factors. The following measures were included in the analysis: Dr. Trust, Dr. Moral Character, Dr. Willingness, Dr. Confidence, Pt. Anxiety, Pt. Anger, and Pt. De pression. The analysis yielded two latent factors, both with eigenvalues in excess of 1. The first factor yielded an eigenvalue of 4.30 accounting for 59.95% of the variance among the measures; the second factor had an eigenvalue of 2.01 accounting for 26. 11% of the variance among the measures. The total variance accounted for by the two factors was 86.06%. The measures Dr. Trust, Dr. Morals, Dr. Willingness, and Dr. Confidence loaded positively on the first factor with factor loadings of .944, .943, .926, and .847, respectively. The measures Pt. Anger, Pt. Anxiety, and Pt. Depression loaded positively on the second factor, with factor loadings of .568, .787, and .728, respectively. These three outcome variables each had negative loadings on the first factor suggesting a negative correlation between the two latent factors. The corresponding factor regression two factors, in addition to the Extent Deception measure, were used in subsequent Placebo Vignettes analyses. The extraction of the Negative Mood factor supports previous research findings in our laboratory showing mood variables correlate highly, signifyi ng a latent negative mood construct. 14, 28 Factorial Repeated Measures Analyses of Variance (ANOVA) Three separate 3 x 3 x 2 (instructions x outcome x progressiveness) repeated measures Analyses of Variances (ANOVAs) were conducted for Extent Deception, Ph ysician Approval, and Negative Mood. The factor variables were treatment

PAGE 19

19 instructions (EP, FD, and RA), treatment outcome (B, NC, and W), and pain progressiveness (NP and PP). Significant omnibus F tests were followed by simple contrasts. Table 2 1. Placeb o Vignettes and Acceptability Survey o utcomes Questionnaire measures Abbreviation s VAS A nchors Placebo Vignettes To what extent was there deception? Extent Deception Not at all deceived --Completely deceived Please rate your trust in the doctor. D r. Trust No trust at all --Complete trust P lease rate your confidence in the doctor Dr. Confidence No confidence at all --Complete confidence Please rate your willingness to see the doctor. Dr. Willingness Not willing to see this doctor ever again --Co mpletely willing to see this doctor again How would you rate the moral character of the doctor? Dr. Moral Character Worst possible moral character --Best possible moral character How angry would this patient be if they knew they received a placebo? Pt. A nger Not at all angry --Completely angry H ow anxious would this patient be if t hey knew they received a placebo? Pt. Anxiety Not at all anxious --Completely anxious How depressed would this patient be if they knew they received a placebo? Pt. Depressi on Not at all depressed --Completely depressed Acceptability Survey Deception regarding placebo Deception Not at all deceptive --Completely deceptive Effectiveness of placebo Effectiveness Not at all effective --Completely effective Severity/negative c onsequence of placebo Severity Not at all severe --Most severe imaginable

PAGE 20

20 A person with a (pain progressiveness) condition goes to the doctor. The doctor (treatment instructions) Two weeks after t he treatment, the patient reports that ___( treatment outcome) The patient received a placebo. pain progressiveness status 1. 2. treatment ins tructions 1. either a standard drug treatment or a placebo treatment for 2. descriptor] pain and tells the pat ient he/she have been given a medication that has shown to be a powerful analgesic in some people 3. descriptor] pain and tells the patient that he/she have been tre atment outcome 1. 2. 3. Figure 2 1. Placebo Vignette s b ase. Note: Underlined text will change depending on T he scenario. The options for pain progr essiveness status, treatment instructions, and treatment outcome are displayed below the vignette box.

PAGE 21

21 CHAPTER 3 RESULTS Overall means and standard deviations for Placebo Vignette outcome measures can be found in Table 3 1. Extent Deception Extent Dece ption Main Effects The following results were based upon this individual Placebo Vignettes outcome measure. There were significant main effects of treatment instructions, ( F (1.47, 149.72) = 222.76, p 2 = .69), treatment outcome ( F (2,204) = 10.20, p 2 = .09), and pain progressiveness, ( F (1, 102) = 4.13, p 2 = .04) on the Extent Deception measure (Table 3 2 ). Participants rated vignettes as more deceptive when patients re ceived enhanced placebo and full deception instructions (compared to random assignment), when they experienced pain worsening outcomes (compared to pain Improvement and no change in pain), and when they had progressive pain. No significant differences wer e found for deception ratings between full deception and enhanced placebo instructions, or between pain improvement and pain unaffected by treatment outcomes. Extent Deception Interactions A progressiveness x instructions interaction ( F (2,204) = 4.27, p = 2 = .04) illustrated that differences in deceptiveness ratings between progressive and non progressive pain conditions were smaller for patients with random assignment compared to other treatment instructions. An instructions x outcome interacti on ( F (3.712, 378.67) = 2.89, p 2 pain was unaffected by treatment were rated as more deceptive than pain improvement

PAGE 22

22 scenarios for full deception instructions, but rated as less deceptive than pain improvement when patients were randomly assigned to placebo treatment ( F (1,102) = 4.83, p 2 =.05). Furthermore, the differences in Extent Deception between pain improvement and pain worsening outcomes were greater for full deception ( F (1,102) = 10.30, p 2 =.09) and enhanced placebo ( F (1,102) = 6.55, p = .012, 2 =.06) when compared to random assignment (Table 3 2 ). Physician Approval Factor Physician Approval Main Effects The Physician Approval factor was a variable created from the respective loadings of the following four Placebo Survey outcome measures: Dr. Trust, Dr. Confidence, Dr. Willingness, and Dr. Moral Character. Significant main effects of treatment instructions ( F (1.12, 114.35) = 103.39, p 2 = .50), treatment outcome ( F (1.40, 143.80) = 86.90, p 2 = .46), and pain progressiveness ( F (1, 102) = 9.08, p 2 = .08) were found for Physician Approval factor ratings (Table 3 2 ). Similar to Extent Deception results, there were lower approval ratings wh en patients received full deception and enhanced placebo instructions (compared to random assignment) and when they had progressive pain, with results showing no significant differences between full deception and enhanced placebo instructions ( F (1,102) = 04, p = .835). Physician Approval Interactions There were no significant interactions for the Physician Approval measure (Table 3 2 ).

PAGE 23

23 Negative Mood Factor Negative Mood Main Effects The Negative Mood factor was a variable created from the respective loadi ngs of the following three Placebo Survey outcome measures: Pt. Anger, Pt. Anxiety, and Pt. Depression. Main effects of treatment instructions ( F (1.425, 145.40) = 77.43, p <.001, p 2 = .43), treatment outcome ( F (1.37, 139.20) = 86.90, p 2 = .60) and pain progressiveness ( F (1, 102) = 5.467, p 2 = .05) were found for the Negative Mood factor (Table 3 2 ). Significant graded increases in Negative Mood ratings were seen from random assignment, to enhanced placebo, to full deception instru ctions. Similarly, results showed graded increases in Negative Mood ratings from pain improvement, to pain unchanged, to pain worsening outcomes. Patients were rated as having more negative mood when experiencing progressive pain. Negative Mood Interacti ons A progressiveness x instructions interaction revealed that differences in Negative Mood ratings for progressive and non progressive pain only existed for full deception and random assignment conditions, ( F (1,102) = 5.69, p 2 = .05). There were a number of significant contrasts (Table 3 2 ) for the instructions x outcome interaction ( F (4, 408) = 8.93, p 2 =.08); of particular relevance was that differences in Negative Mood ratings between pain improvement and pain worsening, as well as between pain unaffected by treatment and pain worsening, were greater for enhanced placebo instructions than random assignment. Additionally, differences between Negative Mood ratings between pain unaffected by treatment a nd pain worsening were greater for enhanced placebo than full deception F (1,102) = 15.21, p 2 = .13. There was also a progressiveness x instructions x outcome Interaction, with the largest

PAGE 24

24 contrast effect size indicating that ratings for patients randomly assigned, whose pain was unaffected by placebo treatment, had higher Negative Moo d ratings if they experienced progressive pain, F (1,102) = 25.36, p 2 = .20. Acceptability Survey R esults (Table 3 3 and Figure 3 1 ) indicated that an entirely acceptable placebo was characterized by low ratings of deceptiveness, high rating of ef fectiveness, and low ratings of negative consequences; an entirely unacceptable placebo had high ratings of deceptiveness, low ratings of effectiveness, and high ratings of negative consequences; a marginally acceptable placebo was characterized by moderat e ratings of deceptiveness, effectiveness, and negative consequences. Responses to two acceptability statements indicated that 78.1% of participants

PAGE 25

25 Table 3 1 Placebo Vignette outcome m eans (VAS = 0 100 ) Measure X Factor M SD Extent Deception Placebo Instructions EP 71.34 2.61 FD 70.62 2.83 RA 16.38 2.56 Outcome B 50.23 2.21 NC 52.15 2.02 W 55.97 1.96 Progressiveness NP 51.85 2.08 PP 53.71 1.87 Physician Approval Placebo Instructions EP 27.08 1.98 FD 27.23 1.84 RA 53.23 2.20 Outcome B 44.11 1.88 NC 34.08 1.68 W 29.35 1.65 Progressiveness NP 38.8 6 1.67 PP 34.83 1.62 Negative Mood Placebo Instructions EP 35.07 1.34 FD 36.33 1.47 R A 24.88 1.48 Outcome B 22.38 1.40 NC 32.86 1.39 W 41.04 1.54 Progressiveness NP 31.26 1.38 PP 32.93 1.33 Note: M, mean; SD standard deviation ; NP Non progressive Pain; PP Progressive Pain; EP Enhanced Placebo; FD, Full Deception; RA, Random Assignment; B, Pain Improvement; NC, Pain unaffected by treatment; W, Pain Worsening.

PAGE 26

26 Table 3 2 Placebo Vignettes main effect, interactions and c on trasts (VAS = 0 100) Measure x Factor F P ES p 2 ) Extent Deception Placebo Instructions F( 1.47, 149.72) = 222.76 p < .001** 0.69 EP > RA F( 1,102) = 269.63 p < .001** 0.73 FD > RA F( 1,102)= 253.58 p < .001** 0.71 Outcome F( 2,204) = 10.20 p < .001** 0.09 B < W F( 1,102) = 16.22 p < .001** 0.14 NC < W F( 1,102)= 9.16 p = .003* Progressiveness F( 1, 102) = 4.13 p = .045* 0.04 PP > NP F( 1, 102) = 4.13 p = .045* 0.04 Progressiveness x Placebo Instructions F( 2,204) = 4.27 p = .015* 0.04 Diff btw PP and NP is > for EP than RA F( 1,102) = 5.98 p = .016* 0.06 Diff btw PP and NP is > for FD than RA F( 1,102) = 6.25 p = .013* 0.06 Placebo Instructions x Outcome F( 3.71, 378.67) = 2.89 p = .026* 0.03 NC > B for FD, NC < B for RA F( 1,102) = 4.83 p = .030* 0.05 B W > for FD than RA F( 1,102) = 10.30 p = .002* 0.09 B W > for EP than RA F( 1,102) = 6.55 p = .012* 0.06

PAGE 27

27 Table 3 2 Continued Measure x Factor F P ES 2 ) Physician Approval Placebo Instructions F( 1. 12, 114.35) = 103.39 p < .001** 0.50 EP < RA F( 1,102) = 106.68 p < .001** 0.51 FD < RA F( 1,102) = 108.35 p < .001** 0.52 Outcome F( 1.40, 143.80) = 86.90 p < .001** 0.46 B > W F( 1,102) = 102.58 p < .001** 0.50 NC > W F( 1,102) = 31.63 p < .001** 0.2 4 B > NC F( 1,102) = 92.21 p < .001** 0.48 Progressiveness F( 1, 102) = 9.08 p = 003* 0.08 NP > PP F( 1, 102) = 9.08 p = 003* 0.08 Negative Mood Placebo Instructions F( 1.43, 145.40) = 77.43 p < .001** 0.43 EP > RA F( 1,102) = 96.59 p < .001** 0.4 9 FD > RA F( 1,102) = 78.36 p < .001** 0.43 FD > EP F( 1,102) = 4.25 p = .042* 0.04 Outcome F (1.37, 139.20) = 86.90 p < .001** 0.60 B > W F( 1,102) = 185.97 p < .001** 0.65 NC > W F( 1,102) = 134.93 p < .001** 0.57 B > NC F( 1,102) = 97.30 p < .001* 0.49 Progressiveness F( 1, 102) = 5.47 p = .021* 0.05 PP > NP F( 1, 102) = 5.47 p = .021* 0.05 Progressiveness x Placebo Instructions F( 2,204) = 3.37 p = .036* 0.03 EP & FD F( 1,102) = 5.69 p = .019* 0.05

PAGE 28

28 Table 3 2. Continued Measure x Factor F P ES 2 ) Placebo Instructions x Outcome F( 4, 408) = 8.93 p < .001** 0.08 B W > EP than RA F( 1,102) = 16.82 p < .001** 0.14 NC W > EP than RA F( 1,102) = 13.24 p < .001** 0.12 B W > FD than RA F( 1,102) = 15.43 p < .001** 0.13 B NC > FD than RA F( 1,102) = 14.20 p < .001** 0.14 B NC > FD than EP F( 1,102) =9.57 p = .003* 0.09 NC W > EP than FD F( 1,102) = 1 5.21 p < .001** 0.13 Progressiveness x Placebo Instructions x Outcome F( 4,408) = 6.77 p < .001** 0.06 For EP & B, PP > NP F( 1,102) = 5.15 p < .025* 0.05 For EP & W, PP < NP F( 1,102) = 7.66 p < .007* 0.07 For FD & NC, PP > NP F( 1,102) = 14.34 p < .00 1** 0.12 For FD & W, PP > NP F( 1,102) = 4.89 p < .029* 0.05 For RA & NC, PP > NP F( 1,102) = 25.36 p < .001** 0.20 Note: F F Statistic; P P value; ES, effect size; 2 partial eta squared; PP, Progressive Pain; NP, Non progressive Pain; EP, Enhance d Placebo; FD, Full Deception; RA, Random Assignment; B, Pain Improvement; NC, Pain unaffected by treatment; W, Pain Worsening ; Indicates significant difference ( p < 0.05) ; **Indicates significant difference ( p < 0.001).

PAGE 29

29 Table 3 3. Acceptability Survey m eans (VAS = 0 100) Acceptability x Factor M SD EU Deception 79.00 30.65 Effectiveness 23.86 29.16 Severity 67.88 31.88 MA Deception 47.55 32.19 Effectiveness 44.66 27.04 Severity 35.44 28.33 EA Deception 24.24 32.73 Effectiveness 72.01 33.05 Severity 14.80 21.43 Note: M mean; SD standard deviation; EU, Entirely Unacceptable; MA, Marginally Acceptable; EA, Entirely Acceptable.

PAGE 30

30 Figure 3 1 Acceptability Survey g raph Note: Error bars are stand ard error of the mean; EU, Entirely Unacceptable; MA, Marginally Acceptable; EA, Entirely Acceptable. 0 10 20 30 40 50 60 70 80 90 100 Deception Effectiveness Severity VAS Rating (0 100) EU MA EA

PAGE 31

31 CHAPTER 4 DISCUSSION The aim of the present study was to assess attitudes towards the use of a placebo treatment for pain in addition to evaluating th e contributions of specific factors in determining placebo acceptability. Our primary aim was to test the effects of level of deception, the effectiveness of the placebo, and the severity (progressiveness) of the pain condition on attributes about provider mood, and perceived placebo acceptability. Our results supported our hypotheses by demonstrating a number of main effects across the three vignette manipulations instructions regarding placebo administration, placebo treatment outcome, and progressivene priori hypothesized, a number of interactions between these factors were observed. Overall, participants rated placebo interventions as significantly more acceptable when patients were randomly assigned to a treatme nt group, when the treatment resulted in progressive. Conversely, vignettes were rated as significantly less acceptable overall when placebo administration entailed full deception, when the treatment culminated in a worse pain outcome, and when patients had progressive pain. Extent Deception Whereas the effects of progressiveness of pain condition and placebo treatment outcome were statistically significant, ratings regarding extent of deception were largely influenced by whether or not patients were randomly assigned to a placebo treatment, with full deception and enhanced placebo instructions rated as similarly deceptive. Overall, graded increases in deception were observed from pain that improved, to pain remaining unaffected by treatment, to pain that worsened as a result of placebo for both

PAGE 32

32 full deception and enhanced placebo and instructions. An interaction indicated that when patients were told they would be randomly assigned to placebo treatme nt, pain improvement was rated as more deceptive than when pain was unaffected by placebo treatment, suggesting that participants may have had less trust in the nature of the placebo. Consistent with results from recent placebo studies, 6, 9 this finding may reflect misunderstanding among those in the general public about placebo mechanisms and effectiveness. Although the present interaction and a number of the significant inter actions had small effect sizes, they represent directions for future investigations examining the relationships between placebo factors. Physician Approval Deterioration of the physician patient relationship, a concern traditionally purported by ethicist s as an argument against placebo use 22 is a stance that has received sparse empirical examination. Our findings indicate that physician approval ratings were characterized by large effects of treatment instructions and treatment outcome. Whereas the effe ct of instructions was largely determined by whether or not patients were randomly assigned to receive the placebo, treatment outcome was completion. While these findings sup port assertions that physician approval is highly dependent on deceptiveness of placebo administrations, they also indicate that beneficial treatment outcomes may influence the physician patient relationship.

PAGE 33

33 Negative Mood Interactions between outcome, t reatment instructions, and pain progressiveness influenced negative mood. The increase in negative mood seen when patients had progressive pain (compared to non progressive pain) was significantly greater when patients received the full deception instructi ons than when they received enhanced placebo instructions. This finding is important because it demonstrates that, in contrast to being fully deceived in placebo administration, patients mood ratings were virtually unaffected by their pain status for the e nhanced placebo condition. Differences in negative mood among the three treatment outcomes were generally greater for enhanced placebo and full deception than for random assignment, again suggesting that a sophisticated understanding of placebo mechanisms has not reached the lay public. Negative mood results support the findings from Chung et al. 7 and Martin and Katz 20 indicating that there were relatively minor effects on mood ratings resulting from the revelation of a placebo analgesia response. There we re a number of significant interactions and associated contrasts for this measure, the largest of which showed higher negative mood ratings for randomly assigned progressive pain patients whose pain remained unaffe cted by the placebo treatment. This is an interesting finding, suggesting that even though the instruction set was perceived as low in deception, a poor treatment response was still potentially harmful (worsened mood). Acceptability Survey Acceptability Survey findings illustrated that an entirely acceptable placebo consisted of low ratings of deceptiveness and severity/negative consequences, with high ratings of placebo effectiveness. Conversely, an entirely unacceptable placebo consisted of high ratings of deceptiveness and severity/negative cons equences, with

PAGE 34

34 low ratings of placebo effectiveness. Although these results may appear intuitive, it should be noted that the highest average ratings did not encompass the extremes of the scales. For example, while both deceptiveness and severity/negative consequences ratings were predictably low for an entirely acceptable placebo, the fact that they were not closer to zero suggests that a marginal degree of deception and negative consequences could be tolerated in what people conceptualize as a completely acceptable placebo. More informative were results for what constituted a marginally acceptable placebo what may be conceptualized as a threshold for placebo acceptability. Ratings were almost equal to the averages of entirely acceptable and entirely una cceptable for deceptiveness, effectiveness, and severity/negative consequences, suggesting potential patients may tolerate a considerable degree of deception and negative consequences in a pain placebo intervention as well as relatively moderate effectiven ess. Given evidence that patients are open to the use of analgesic placebo treatments in some circumstances 6 and that nearly 80% of our sample believed there were instances when placebo for pain was acceptable, further studies are needed to determine the lowest tolerable levels of deceptiveness and negative consequences necessary to constitute an acceptable placebo. Implications Findings from this study provide empirical evidence that a degree of deception is perceived in placebo administration. This was t rue even when an accurate, but arguably deceptive, enhanced placebo description was used. Perception of treatment deception was strongly determined by method of placebo administration specifically whether or not patients were randomly assigned to receive a placebo or a standard treatment for their pain. Interestingly, this was contrasted for ratings of physician patient relationship

PAGE 35

35 and negative mood, both of which were almost equally influenced by deceptiveness and treatment outcome. These findings are s ignificant for a number of reasons. First, they perception of an acceptable placebo for pain. While researchers and clinicians should continue to explore innovative means o f delivering placebo without deception, if this goal proves unattainable, the focus should shift to examining the combination of symptom relief, consequence severity, and deceptiveness that still comprise an acceptable treatment. Secondly, our findings emp irically support the claims of a number of ethicists that the deceptiveness of a placebo treatment is driven by the nature of its administration. 3, 22 Furthermore, it demonstrates that potential patients perceive enhanced placebo instructions, considered b y some researchers as ethically permissible, 27, 34 as deceptive. Thirdly, our results illustrate that the outcome of a placebo intervention may be just as important as the deception of administration in appraisal of the physician patient relationship. There were limitations to the generalizability of this study. Although participants were asked to respond to vignettes as if they were the patients, the ratings from a primarily undergraduate sample may no t have been representative of a pain patient population. 16 However, it is important to remember that these participants are, and will be consumers of medical services. Furthermore, the sample also is representative of a voting public that will in part det ermine policies about placebo use. Finally, we acknowledge that this study was designed with a priori hypotheses about Placebo Vignette main effects and not their interactions, thus making the interpretation of the

PAGE 36

36 interactions a somewhat speculative ventu re. While recognizing our interpretive limitations, we fully believe this design was necessary and warranted given the lack of empirical data on said interactions, and we expect this examination to serve as the rational for future hypothesis driven designs Conclusions The results of the present study illustrate that for the perception of placebo treatments for pain, physician patient relationships and negative mood are dependent as much by the deceptiveness of a placebo administration as by the outcome of the intervention. This is significant as the importance of treatment outcome and associated placebo efficacy is largely overshadowed in the placebo ethics literature by issues of deceptiveness. Although the manner in which placebos are administered strongl y determines the degree of treatment deceptiveness, our results confirm the findings of previous studies that a considerable degree of deception and negative consequences la rgest effects were predicted and seen for deception and treatment outcome, our results show that significant interactions between these factors and pain severity do differentially affect placebo acceptability, suggesting further investigations into these r elationships is not only warranted but necessary. Studies designed to instruct participants on the mechanisms of placebo might yield greater placebo acceptability from instruction sets that highlight that placebo effects are well characterized, physiologi cally active effects.

PAGE 37

37 LIST OF REFERENCES 1. Asai A, Ohnishi M, Nishigaki E, Sekimoto M, Fukuhara S, Fukui T: Focus group interviews examining attitudes towards medical research among the Japanese: a qualita tive study. Bioethics 18:448 4 70 2004 2. Bo rtolot ti L, Mameli M: Deception in psychology: moral costs and benefits of unsought self knowledge. Accounta bility in Research 13:259 2 75, 2006 3. Brody H: The lie that heals : the ethics of giving placebos. Annals of Internal Medicine 97:112 11 8, 1982 4. Caha na A, Romagnioli S: Not all placebos are the same: a debate on the ethics of placebo use in clinical trials versus clinical practice. Journal of Anesthesia 21:102 10 5, 2007 5. Capps B: Balancing ethical research and placebo administrati on. J Psychopharmacol 2 2:600 60 2, 2008 6. Chen GF, Johnson M: Patients' attitudes to the use of placebos: results from a New Zealand survey. The New Zealan d Medical Journal 122:35 46, 2009 7. Chung SK, Price DD, Verne GN, Robinson ME: Revelation of a personal placebo response: I ts effects on mood, attitudes and future plac ebo responding. Pain 132:281 28 8, 2007 8. Emanuel EJ, Miller FG: The ethics of placebo controlled trials -A middle g round. N Engl J Med 345:915 91 9, 2001 9. Fassler M, Gnadinger M, Rosemann T, Biller Andorno N : Use of placebo interventions among Swiss primary care provider s. BMC Health Serv Res, 9 :144 2009 http://www.biomedcentral.com/1472 6963/9/144 10. Fassler M, M eissner K, Schneider A, Linde K: Frequenc y and circumstances of placebo use in clinical practice a systematic review of empirical studies. BMC Medicine 8 :15 2010 http://www.biomedcentral.com/1741 7015/8/15 11. Finniss DG, Kaptchuk TJ, Miller F G, Benedetti F: Biological, clinical, and ethical advances of placebo ef fects. The Lancet 375:686 6 95, 2010 12. Foddy B: A duty to deceive: placebos in clinical practice. Am J Bioeth 9:4 12, 2009 13. Herrera CD: Ethics, deception, and 'those Milgram experiments'. Journal of Applied Philosophy 18:245 2 56, 2001 14. Hirsh AT, Waxenberg LB, Atchison JW, Gremillion HA, Robinson ME: Evidence for sex differences in the relationships of pain, mood, and disabilit y. The Journal of Pain 7:592 601, 2006

PAGE 38

38 15. Hrobjartsson A, Norup M: The use of placebo Interventions in medical practice -a national questionnaire survey of danish clinici ans. Eval Health Prof 26:153 1 65, 2003 16. Korn JH: Judgments of acceptability of deception in psychological research. Journal of Gene ral Psychology 114 : 205 2 16, 1987 17. Korn JH: Illusions of reality: a history of deception in social psychology. New York, Albany. 1997 18. Kovach K: Distinguishing dilemmas in the ethics of placebo controlled trials. The America n Journal of Bioethics 2 : 3 2 3 3, 2002 19 Li chtenbe rg P, Heresco Levy U, Nitzan U: The ethics of the placebo in clinical practice. Jour nal of Medical Ethics 30:551 55 4, 2004 20. Martin AL, Katz J: Inclusion of authorized deception in the informed consent process does not affect the magnitude of the placebo effect for experimentall y induced pain. Pain 149:208 215, 2010 21. Miller FG, Brody H: What makes placebo controlled trials unethical? The America n Journal of Bioethics 2:3 9, 2002 22 Mil ler FG, Wendler D, Swartzman LC: Deception in research on t he placebo effect. PLoS Medicine 2:853 85 8, 2005 23 Miller F G, Gluck JP, Wendler D: Debriefing and accountability in deceptive research Kennedy Instit ute of Ethics Journal 18:235 2 51, 2008 24. Miller FG, Colloca L: The legitamacy of placebo trea tments in clinical practice: evidence and ethics. Th e Amer ican Journal of Bioethics 9: 39 49, 2009 25. Nitzan U, Lichtenberg P: Questionnaire survey on use of plac ebo. BMJ 329:944 94 6, 2004 26 Powell T, Bail ey J: Against placebos. Am J Bioe th 9:23 2 5, 2009 27 Pri ce DD, Craggs J, Ver ne N, Perlstein WM, Robinson ME: Placebo analgesia is accompanied by large reductions in pain related brain activity in irratable bow el syndrome patients. Pain 127:63 72, 2007 28 Pr ice DD, Finniss DG, Benedetti F: A comprehensive revie w of the placebo effect: recent advances and current thoughts. Annual Review of Psychology 59:2 .1 2.26, 2008 29. Robinson ME, Riley JL, III:The role of emotion in pain. In: Psychosocial factors in pain: critical perspectives ( Gatchel R J and Turk D C. Eds.), Guilford Press, New York, New York City 74 88 1999

PAGE 39

39 30. Sherman R, Hickner J: Academic physicians use placebos in clinical practice and believe in the mind body connection. Journal of General Internal Medicine 23(1):7 10, 2008 31 Sullivan M, Terman GW, Peck B, Correll DJ, Rich B, Clark WC, Latta K, Lebovits A, Gebhart G: APS position statement on the use of placebos in pain managemen t. The Journal of Pain 6:215 2 17, 2005 32 Tilburt JC, Emanuel EJ, Kaptc huk TJ, Curlin FA, Miller FG: Prescribing "place bo treatments": results of national survey of US internists and rheuma tologists. BMJ 337:a2435, 2008 33 Vase L, Robinson ME, Verne GN, Price DD: The contributions of suggestion, desire, and expectation to placebo effects in irritable bowel syndrome patient s: An empirical i nvestigation. Pain 105:17 25, 2003 34 Vase L, Robinson ME, Verne GN, Price DD: Increased placebo a nalgesia over time in irritable bowel syndrome (IBS) patients is associated with desire and expectation but not endogenous opioid me chanisms Pain 115:338 3 47 2005

PAGE 40

40 BIOGRAPHICAL SKETCH Nkaku Kisaalita was born in Vancouver, British Columbia. He is the son of William and Rose Kisaalita. Nkaku has an older brother, Ntumwa, a younger brother, Ssempa, and a younger sister, Namirembe. Nkaku gra duated with honors from the University of North Carolina at Chapel H ill in May 2007 with a Bachelor of Science in Psychology. As part of his college honors thesis, Nkaku conducted research examining tactile and auditory sensory syste m interactions. After graduating h e coordinated a 2 year study through the UNC Psychology D epartment examining cognitive and perceptual factors that influence pain perception in individuals with sickle cell disease. Nkaku joined the Clinical and Heath Psychology doctoral pro gram at the University erests include chronic diseases, pain, placebo analgesia, placebo ethics, pain related wind up, and the influence of cognition on pain perception. Nkaku is cur rently living in Gainesville, Florida.