EVALUATING A SUFFERER'S PAIN: THE ROLE OF CONTEXTUAL AND ATTRIBUTIONAL VARIABLES by Linda M. Lundquist B.Sc. , The University of Northern British Columbia , 1997 THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in PSYCHOLOGY © Linda Marie Lundquist, 1999 THE UNIVERSITY OF NORTHERN BRITISH COLUMBIA June, 1999 UNIVERSITY OF NORTHERN BRITISH COlUMBIA UBRAP.:Y All rights reserved. This work may not be reproduced in whole or in part, by photocopy or other means, without the permission of the author. Prince George, BC ii Abstract This study examined the effects of context related information and observers' attributional style on evaluations of pain severity, the pain sufferer, and treatment choice for shoulder pain patients. Ninety-six undergraduate participants read information containing the independent variable manipulations before viewing a videotape of one of two shoulder pain patients undergoing a physiotherapy assessment. Participants then rated their perceptions of the pain patient and completed a questionnaire assessing their attributional style (AS) for others' misfortunes. This study used both a 2 (coping: yes or no) X 2 (medical evidence: present or absent) X 2 (target patients) X 3 (pain level: high, moderate, and low) design with repeated measures on the last variable (with AS as a covariate) and a 2 (coping: yes or no) X 2 (medical evidence: present or absent) X 2 (target patients) mixed design (also with AS serving as a covariate). Significant main effects and an interaction emerged indicating that a contextual factor, the perception of the patient's coping, influenced judge's perceptions of the pain sufferer. In addition , observers were sensitive judges of levels of pain severity. However, contrary to expectation , there were no effects for AS or medical evidence. Thus, while judges can differentiate between differences in the amount of pain a patient is in, pain patients appear to be facing biasing factors which may result in differential evaluations and treatment choices . Ill DEDICATION In loving memory of my father, Robert Jonas (Bob) Lundquist 1940-1992 iv ACKNOWLEDGMENT My most heartfelt gratitude to those who have helped me along the way: Foremost I would like to acknowledge my supervisory committee, particularly Nancy Higgins and Ken Prkachin for their hours of guidance and help. I would also like to thank Alex Michalos and Tom Strong for their advice and feedback. Additionally I thank Glenda Prkachin for her help in making the stimulus tapes, and John Barton for his construction of the dividers that allowed me to test participants in groups. Finally I would like to note the appreciation I feel toward my friends and family for their continued support. v TABLE OF CONTENTS Abstract ....... ............. ................................................................ ... ......... ................ ........ ii Dedication ........ .. ........... .............................. ........... ..... .... ...................... .......... ............. 111 Acknowledgement .... ........................................................ ................ .............. ............. iv Table of Contents .. ...... .... .. .... ... ...... ..... .... ............... .... ...... ..... .... ..... .. ...... .... .. ...... .......... v List of Tables .... ...... ............ .... ......... ........................ ...... ........... ...... ........ .................. . viii List of Figures ............ ................................... ... ........................ ............. ...................... 1x CHAPTER ONE Evaluating a Sufferer's Pain: The Role of Contextual and Attributional Variables ...... 1 Background Literature ................................................................ ..................... 1 Controllability and Observer's Perceptions of Pain Sufferers ............... 2 Biases in the Assessment of Pain Severity .................. ........................ 5 Additional Biases in the Assessment of Pain Sufferers ........................ 7 Medical Evidence ..................................................... ..... ......... .... 7 Attribution a I Style .............. ................. ..... ....................... .. ..................... 8 Unsupportive Attributional Style ......... ...... ...................... .......... 12 Previous Experience with Chronic Pain ................... ................. ... ....... 13 The Impact of Observers' Perceptions of Sufferers' Pain ................... 13 Methodological Differences Between Past Research and the Current Research ...................... ..... ............................................. .................... 14 Use of Videotapes .......................... .. ............................ 14 Levels of Pain Severity ............................ ..................... 15 Controllability ................................. ............................... 15 Gender ....... ........... ... .. ...... .... .. .... .... ...... ... ... .... ............ .. 15 vi Dependent Variables .............. .. ... .............. .............. ........... ............... 16 Hypotheses ................................................. ..... ............ ......... ............. 18 CHAPTER TWO Method ................ ... ...... ... .. ...... ..... ..... ..... ...................... .... ...... .................... ...... ....... . 20 Participants ........................................................................ ................. ...... ..... 20 Materials ................. ...... .. ... ...... ................. .... ... .. ....... ....... ...... .. ...~ ......... ... . .. ... 20 RMQ ....... ... .. ..... .. ............................ ..... ... ... ..... ...... ......... ... ..... ........ ...... 20 Videotapes ............. ... ........... ............ ..... ... ....................... .................... 21 Dependent Variable Assessment ........... ............................ ................ 24 Procedure ............................................................................ .......................... 26 CHAPTER THREE Results .. ...... ............ ... .... ........... ....... .. ......... ..... .................... ......... .. ...... ........ ...... ..... 28 Overview ....... .. ... ...................................................... .. .... ................ ............... 28 Pain Severity ..................................................... ........................... 31 Distress and Disability ........................................ .......................... 34 Responsibility, Anger, Sympathy and Support ....... ...... ............... 34 Effects of Family History and Gender on Ratings of Pain ........... . 36 Effects of Family History and Gender on Ratings of Distress and Disability ..... ................. .............................................. .. ................. 36 Effects of Family History and Gender on Ratings of Responsibility, Anger, Sympathy and Support ............................ ......................... 36 Effects of Coping, Evidence, Patient, Family History and Gender on Treatment Choice ........ .. ........... .... ........................ .. .... .... .. .. .... 36 vii Effects of Distress, Disability, Responsibility, Anger, Sympathy and Support on Treatment Choice ................................. .. ............ .. .. ... 36 CHAPTER FOUR Discussion ................................................................................................................ 39 Coping ........ .................. .. ... ...... ........ ... .. .. ........................ ....... ...... ..... ............ . 39 Attributional Style .......................................................................................... 41 Medical Evidence .......................................................................................... 42 Pain Severity ............................................................................................. .... 42 Treatment Choice .......................................................................................... 44 Family History .......................... ..... ............. ......... ..... ............... ........... .. .......... 45 Gender .... ................................... .... ........... ...... .. .... ......... ..... ............. ... .......... 45 ································································ 46 Implications ···:································ Limitations and Suggestions for Future Research ...... .... .................. .. ....... .... 47 REFERENCES ......................................................................................................... 49 APPENDIX A Reasons for Misfortunes Questionnaire ....... ..... ... ........ ......... ... 53 APPENDIX 8 Vignettes and Dependent Variables Questionnaire .................. 57 APPENDIX C Personal Information Questionnaire ... ............. ...... .. .. ... ......... ... 61 vi ii LIST OF TABLES Table 1 Main Effects and Effect Sizes from the Schwarzer and Weiner (1991) Coping and Controllability Manipulations .. ................ .. .. .. ............ .. .. .......... .. .. ............... 4 2 Correlations Among the Dependent Variables and Attributional Style ............ 29 3 Mean Attributional Style Scores for the Independent Variables .. .. .. .. .. .. .. .. .. .. .. 30 4 Univariate Simple Effects for the Coping Manipulation .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 35 5 Discriminant Function Analysis Classification Results .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. .. . 38 ix LIST OF FIGURES Figure 1 Weiner's Attributional Model of Helping Behaviour ... .............. ................. .... ..... 10 2 Interaction Between Participant's Mean Pain Severity Ratings for Patient A and Patient Bin the Low, Moderate and High Displayed Levels of Pain Conditions ...... ...... ...... .... .... ....... .. ..... .... .. .. ....... .. ... .............. .... ........... ....... ........ 32 1 CHAPTER I Evaluating a Sufferer's Pain: The Role of Contextual and Attributional Variables Background Literature It is impossible to know another's pain directly. Nevertheless, decisions about and evaluations of the suffering of others happen routinely every day. The most obvious examples occur in health-care settings where physicians, nurses and other professionals evaluate different forms of evidence to draw conclusions about such things as the severity of injuries, treatment progress and the effectiveness of treatment. Less obvious examples occur when parents evaluate how to respond to a child who has experienced a cut or scrape or a stomach ache or a headache, between supervisors and employees and between peers in innumerable social contexts. Understanding such processes is important from a number of perspectives. From a purely scientific perspective it is important to understand how accurate or inaccurate an observer's perception may be and to understand the forces that may bias an observer's judgements. From a practical perspective, it is important to understand when people are likely to make decisions about sufferers -- decisions that happen all the time --that may promote healing, recovery, or positive adjustment and when people are likely to make judgements that may prolong suffering. From a theoretical perspective, arguments are made that certain reactions (for example, expressing sympathy) that are dependent on the perception that another person is suffering may have more or less irreversible effects on the subsequent behaviour of the afflicted individual. From a humanistic or moral perspective, it is important to 2 understand how people evaluate others who are suffering in order to promote more humane treatment and to eliminate unfairness. Prkachin and Craig (1995) presented a model of pain communication that attempted to organize the general psychological and social processes that become involved in an episode of pain. From this perspective, pain involves a social and communicative process by virtue of the fact that pain is accompanied by a number of behavioural changes that "encode" the experience into effective signs that are "broadcast" into a receiving social world. Such signs are "decoded" by observers who then make judgements and decisions about the sufferer. Decoding, or the perception of pain expression, has been the subject of increasing interest in the last decade as researchers have attempted to describe and explain the principles and the stimuli that observers use to estimate suffering and formulate decisions. Much of this research has focused on the identification of stimuli that influence the perception of people in pain (e.g., Prkachin, Berzins, and Mercer, 1994). Controllability and Observers' Perceptions of Pain Sufferers However, perceptions do not occur in a contextual vacuum. The circumstance(s) surrounding an event affect how the event is viewed. For example, observers' reactions to another's illness are influenced by the causal inferences they make about the illness sufferer. Illness sufferers are deprecated if the onset of the physical symptom is viewed as being under their control (Marteau & Riordan, 1992). In their study, Marteau and Riordan (1992) provided participants with details of various ailments and the afflicted individual's actions prior to the onset of these ailments. These details differed in the patients' health habits prior to their ailments (either 3 following or not following a health relevant action) . Patients who were presented as not following a health relevant action (i.e., having onset controllability) were seen as less concerned about their situation, less likely to follow advice given to them, less likely to understand their condition and less enjoyable to work with by physicians and nurses (Marteau & Riordan , 1992). Another type of controllability, coping , has also been implicated in influencing judge's perceptions of sufferers. Schwarzer and Weiner (1991) found that maintenance controllability (coping) , which is the individual's present effort to control the ailment, also influences observer perceptions. In their research, Schwarzer and Weiner (1991) examined participants' affective reactions toward and intentions to help individuals with health-related stigmas. These stigmatized individuals were presented as either having or not having onset control over the stigma, and as either coping or not coping. Both types of information , onset controllability and coping , influenced participants' affective reactions and helping judgements (see Table 1 for a list of the variables used in Schwarzer & Weiner, 1991 , and the effect sizes for both the controllability and the coping manipulations) towards the stigmatized individuals. Individuals who were presented as having onset-control or not coping received more negative evaluations and lower ratings of willingness to offer support from judges. In a related vein , perceptions of patients' ability to cope were found to influence nurses' ratings of patients undergoing minor abdominal surgery (Salmon & Manyande, 1996). When making personality judgements, the nurses offered more positive evaluations of patients that they perceived as coping with their pain than patients they deemed unable to cope. The belief that patients were able to cope with their pain led 4 Table 1 Main Effects and Effect Sizes from the Schwarzer and Weiner (1991) Coping and Controllability Manipulations IV F(2,81) o} Blame Controllability Coping 110.4 16.8 .72 .27 Sympathy Controllability Coping 9.5 8.8 .17 .16 Anger Controllability Coping ns 58.4 no effect .58 Social Support Controllability Coping 6.8 71.7 .02 .38 DVs Note. All Q's < .01. Based on the criteria outlined by Kirk (1 996), a} values of .01 0, .059 and .138 were taken as corresponding to small, medium and large effect sizes, respectively. DV = dependent variable, IV =independent variable. 5 to evaluations of patients as compliant and cooperative, whereas the belief that they were unable to cope with their pain led to evaluations of these patients as dependent, demanding and unpopular (Salmon & Manyande, 1996). This has implications for the amount of pain a patient exhibits; patients value popularity with nurses and may restrict their pain expression to appear to be in less pain and more effectively coping in order to attain a more positive evaluation (Salmon & Manyande, 1996). Likewise, perceptions of people in pain may be influenced by similar variables. Presumably, some perceptions (e.g ., of following a health relevant action) can also affect evaluations positively. This thesis focuses on the process whereby contextual variables related to social-psychological attribution theory may influence the perception of pain in others. Biases in the Assessment of Pain Severity Not only will perceptions of onset controllability of physical symptoms negatively affect evaluations about sufferers themselves, they also have a negative impact on observers' perceptions of the severity of the sufferer's pain (Tait & Chibnall, 1994). Participants in the Tait and Chibnall (1994) study (48 undergraduate subjects) were presented with information concerning the presence or absence of supporting medical evidence, the valence, either positive or negative, of the relationship between the participant and the person in pain , and the degree of control , either present or absent, of the pain patients. All of these manipulations were performed by presenting the participants with written scenarios describing the patient's circumstances. The medical evidence manipulations varied. However, most of the manipulations involved the suggestions that tests performed by professionals (e.g., CT scan) either confirmed 6 or disconfirmed the patient's claim of chronic (lasting longer than 12 months) low back pain . The relationship valence manipulations also varied, but the positive valence manipulation mostly concerned the suggestion that the sufferer was a close friend (e.g. , a member of the participant's church and a valued friend). The negative valence manipulations suggested that the sufferer was like persons the participants were told to suspect of fraud , who were suing the participants, or who were suing someone the participant knew. The degree of control manipulation had patients being struck by something that they could not avoid (e.g., struck from behind in a car accident) and acquiring arthritis (these encompass the uncontrollable conditions), or disregarding warnings (e.g. , lifted a heavy box despite a legitimate warning not to do so) and causing a car accident (these encompass the controllable conditions). The dependent measures examined were participants' ratings of the pain patient's pain severity, distress and disability. Tait and Chibnall (1994) also found that a lack of medical evidence resulted in lower ratings of patients' levels of pain severity, emotional distress and disability relative to the presence of medical evidence. In addition , those patients presented as having a positive relationship with participants were accorded higher levels of pain severity, emotional distress and disability relative to those patients presented as having a negative relationship with the participant. Moreover, the controllability manipulation had effects on perceptions of pain intensity (with those in the uncontrollable condition given ratings of higher pain severity than those in the controllable condition) and emotional distress (with those patients in the uncontrollable condition being rated as having more emotional distress than those patients in the 7 controllable condition). There were no effects of controllability on participants' disability ratings . These findings indicated that observers are, among other things, systematically underestimating the pain expressed by individuals who they deem are in control of the onset of their symptoms, and this introduces a source of bias in the assessment of pain . Additional Biases in the Assessment of Pain Sufferers Medical Evidence. Another potentially biasing factor in the assessment of pain (not limited to pain severity judgments , but also including general perceptions) identified by Tait and Chibnall (1994) and also in their follow-up study (Chibnall & Tait, 1995), is the presence or absence of objective medical evidence explaining or justifying the pain condition . In their 1995 study, Chibnall and Tait attempted to rectify problems associated with their previous study, including nonequivalence of their manipulations. They modified all of their medical evidence manipulations to the results of a nerve conduction study and CT scan as either confirming or disconfirming the patient's claim of injury. They also modified their relationship valence manipulation to a coworker who the participants either liked or disliked . Likewise , their controllability manipulation was modified to a car accident that the patient either caused or did not cause. As in their 1994 study, all manipulations were performed through the use of written scenarios. Based on the responses of 80 undergraduates, Chibnall and Tait (1995) found the same results as in their 1994 study for the effects of medical evidence and relationship valence on pain severity, emotional distress and disability. However, controllability information had no effect on any of these variables. 8 As was shown by Chibnall and Tait (1995), the absence of medical evidence leads to reduced ratings of pain, a finding that also generalizes to health care professionals. Without medical evidence, health care professionals may view the sufferer's pain as psychogenic and the patient as malingering, despite the elusiveness of specific causes for some chronic pain (Rudy, Turk, & Brena, 1988; Turk, 1994; Turk & Flor, 1987). One possibility as to why pain patients are disparaged is that their motives may be suspect. People tend to respond to pain displays in others with active assistance such as solicitousness, concern, or monetary rewards. These consequences are thought to serve as positive reinforcers for the continuation of pain behaviours (Fordyce, 1976; Turk & Okifuji, 1997). There are also negative reinforcers , such as avoidance of activity, which may strengthen the pain behaviours. With these types of reinforcement contingencies, the pain behaviour may continue even in the absence of pain (Fordyce, 1976; Turk & Okifuji, 1997). Attributional Style Additionally, the influence of these attributional variables on observers' perceptions could be moderated by a related personality variable, namely, attributional style. As noted by Anderson and Weiner (1992), attributions are partially the product of an individual's history and culture; therefore, the attributional styles brought to the situation by attributors must also be considered . Attribution theory is concerned with how an individual answers questions beginning with "why?" (Kelley, 1971; 1973). In the context of person perception, how an individual explains another's negative outcomes (e.g., falling on a bus) will influence his/her affects and actions towards the target person (Kelley, 1971; 1973; Weiner 1980a; 1980b; 1986; 1993; 1995a). For 9 example, if the cause for a person falling on a bus is that the person appears intoxicated by alcohol (considered controllable) rather than due to that individual's apparent blindness (considered uncontrollable) that individual will be met with more disgust, less sympathy and less willingness to help than the latter individual (Weiner, 1980a). These findings are in accordance with Weiner's (1980a; 1980b; 1986; 1995a) attributional model of helping behaviour, which proposes an ~ ~ behaviour" sequence in which the causal inferences an individual makes regarding a person's need for help lead to feelings of sympathy or anger which mediate judgements about the likelihood of helping and actual helping behaviour. Higher levels of sympathy and help are accorded to those with an onset-uncontrollable cause of need ; whereas those with an onset-controllable cause are met with increased levels of anger, lower sympathy and decreased help (see Figure 1 for a schematic representation of this sequence). This model is well supported by empirical studies (e.g. , Betancourt, 1990; Higgins & Shaw, 1999; Meyer & Mulherin , 1980; Reisenzein , 1986; Schmidt & Weiner, 1988; Weiner, 1980a; 1980b; Williams, 1993). While the Tait and Chibnall (1994) study addressed the of the sequence, the entire ~ ~ affect part ~ behaviour sequence has been applied to health-related scenarios and has received further support. Weiner, Perry and Magnusson (1988) examined the relationship between perceived responsibility for health-related stigmas and resultant affective reactions and behavioural predictions of judges. Those stigmas deemed uncontrollable (in this case physical problems) elicited sympathy and not anger and high judgements of the degree to which participants would be willing to help the individual. The stigmas that were deemed 10 Figure 1 Weiner's Attributional Model of Helping Behaviour 11 Attribution _ _ _ _ ___. Affect - - - - - - - - + Behaviour Sympathy Anger Help Uncontrollable I Low I Controllable I High I 12 controllable (mental and behavioural problems) were met with anger, not sympathy and low judgements of help (Weiner et al. , 1988). Unsupportive Attributional Style. As such, an individual's habitual tendency to attribute causal controllability over negative outcomes to a sufferer, a variable that has been termed unsupportive attributional style (AS; Higgins & Morrison, 1998), is a potential biasing factor in the perception of pain in others. Specifically, unsupportive AS refers to a tendency to view the negative outcomes of others as internal to and personally controllable by the victim . That is, individuals with an unsupportive AS need little situational pressure to attribute the causes of negative outcomes to internal, controllable aspects of the victim , whereas individuals with a supportive AS view the negative outcomes of others as more external and less controllable by the victim (Higgins & Morrison, 1998). Unsupportive and supportive attributional styles are measured using the Reasons for Misfortunes Questionnaire (RMQ; Higgins, 1992) which is based on the causes participants generate for negative outcomes that they imagine have happened to a hypothetical other, and their subsequent ratings of each cause. A more complete discussion of the RMQ appears in the Method section. For example, in an examination of the attribution a I model of helping behaviour, attributional style interacted with situational (context) information to influence helping behaviour (Higgins & Shaw, 1999). Unsupportive AS individuals were more likely to help a victim whose reason for need was uncontrollable than someone whose need was controllable. However, supportive AS individuals offered similar levels of help irrespective of the controllability of the reason for need. This provides evidence that individuals with different attributional styles about others' negative outcomes did not 13 respond similarly in helping situations that differed in causal controllability over the negative outcome. Given that observing a person in pain is quintessentially a helping situation , it is reasonable to expect that an observer's attributional style may influence her or his perception of behaviour in relation to the sufferer. Previous Experience with Chronic Pain An additional factor thought to influence evaluations of pain severity is previous experience (specifically, family history) with chronic pain (Prkachin, Solomon , Hwang , & Mercer, 1995). Tait and Chibnall (1994) identified, post hoc, that a substantial percentage of their sample had prior experience with chronic pain . As this was identified after the experiment was conducted , the researchers could not examine this experience, but speculated that this exposure may influence evaluations of pain. This speculation was substantiated by Prkachin et al. (1995) who identified systematic differences between participants with no family history of chronic pain, participants with a family history of chronic pain and health care professionals (specifically, nurses). In comparing these groups, the researchers found that those with a family history of pain rated sufferer's pain as more severe than those without a family history, and health care professionals rated pain as less severe than those without a family history of chronic pain . The Impact of Observers' Perceptions of Sufferers' Pain For the current research, the helping behaviour examined was towards a person in pain, so it was necessary to examine how observers perceive pain in others. The decisions an observer makes about the qualities of the pain (i.e., its severity and nature) are problematic (Prkachin & Craig , 1995). Lay-persons and professional 14 health care workers alike tend to underestimate pain relative to the judgements made by the pain sufferers themselves (Grossman , Sheidler, Sweeden , Mucenski, Piantadosi , 1991 ; Prkachin et al. , 1994). This bias is problematic due to the profound impact observers have on the pain behaviours of pain sufferers (Fordyce , 1976; Prkachin , et al , 1995). Pain behaviours of an individual afflicted with chronic pain have been proposed to be affected by reinforcement deriving from expressions of concerns by amateurs (friends and family) and professionals (health care professionals) (Fordyce, 1976). This concern may strengthen the future pain behaviours of the sufferers themselves and , indirectly, be responsible for the type of treatment chosen by a health care professional (Prkachin et al. , 1995). Methodological Differences Between Past Research and the Current Research Use of Videotapes. There are several methodological differences between past research that has examined the current variables of interest and the current research . First, in the Chibnall and Tait (1995) study, participants were asked to read vignettes describing fictitious pain sufferers, including their "self-reports" of pain, presented as high or low. In the low self-reported level of pain condition , participants' estimates of the sufferers' pain did not differ from the ratings present in the vignettes. However, in the high self-reported level of pain condition , participants' ratings were significantly lower than those presented in the vignettes. The present study was based partially on the research of Chibnall and Tait (1995) and sought to replicate these results, but more closely approximated a clinical reality by examining responses made by participants who have viewed videotapes of real pain sufferers undergoing a 15 physiotherapy assessment. A clinical setting allows for the observation of behaviour, and to some health care professionals, this observation is the most influential factor in decision making (Ferrell, Eberts, McCafferey, & Grant, 1991 ). Therefore, the ability of the participant to view the expression of pain on the patients face was deemed an important addition. Levels of Pain Severity. The present research also expanded on the Chibnall and Tait (1995) study by including a "moderate" level of pain severity. This addition was included as previous work with these videotapes found that lay-persons were able to differentiate between gross variations in expressed pain, but were unable to respond to more subtle variations (Prkachin et al., 1994). As such, the "moderate" level of pain was included to investigate how a less clear display of pain is treated with regard to perceptions of the severity of pain, subjective distress, disability, responsibility, feelings of anger and sympathy, likelihood of offering social support and choice of treatment for the pain patient. Controllability. Another difference from Chibnall and Tait's (1995) study was the omission of the onset-controllability manipulation in favour of a coping manipulation. This decision was based on the effect sizes for the dependent measures used in the Schwarzer and Weiner (1991) research that were also used in the present research (i.e., affect and social support) which showed equal or stronger effects for coping , and thus coping was the variable manipulated in the present research . Gender. Tait and Chibnall (1994) examined the impact of the participant's gender on ratings of pain severity, emotional distress and disability and found no effects. However, gender is known to affect judgements of social support (help). 16 Females generally offer increased levels of help in social support situations; men in dramatic emergency situations (Kessler, Mcleod , & Wethington , 1985). The helping situation in this study was one of social support, and thus gender was considered in the present research . Dependent Variables The dependent variables that were examined in the present study were the participants' evaluations of patients' pain severity, emotional distress, disability and responsibility, as well as participants' affective reactions (sympathy and anger) to the patients, willingness to extend help (social support) and treatment choice. The inclusion of pain severity ratings as a dependent variable was of central concern for understanding factors that contribute to observer perceptions of pain . The decision to include emotional distress and disability was based on the rationale of Chibnall and Tait (1995; and Tait & Chibnall , 1994) that emotional distress and disability, in combination with pain severity, are conventions utilized by society in determining whether or not a pain patient may be exempted from the responsibilities of a functioning member of society. The present study extended the work of Chibnall and Tait to include an assessment of the perceived responsibility of the patient for their pain. The concepts of responsibility and controllability (be it onset or maintenance controllability) are closely related (Schwarzer & Weiner, 1991 ). However, there are some conceptual distinctions. According to Weiner (1995b) , controllability refers to characteristics of the cause, whereas responsibility is a judgement made about a person. Responsibility inferences are based initially on causal judgements (in this case, controllability 17 judgements) and then shift to consideration of the person . In order for an individual to be considered responsible for a negative outcome, they must first be considered to be in control of the cause of the outcome. This distinction is important as an individual can be considered to be in control over the cause of an event but not be responsible if there are perceived mitigating circumstances to negate responsibility. For example, consider the scenario in which a small child finds a loaded gun. If, upon picking up the gun , the child pulls the trigger and the bullet causes damage , is the child responsible? The child had control of his/her actions, but as there were mitigating circumstances (the child 's age and lack of understanding of firearms) the child would not be considered responsible. Therefore, including responsibility as a dependent variable is necessary for a more complete understanding of how observers make attributions, and their subsequent affect and behavioural judgements towards a pain sufferer. The inclusion of treatment choice as a dependent variable was added as it is a natural extension of the helping situation in the context of patient pain . The helping situation used related to social support (i.e., how much participants predict they would be willing to support the patient), although controllability judgements have been shown to affect treatment choices as well. For example, Brewin (1984) investigated an attributional model of psychotrophic drug prescription and found that even if pain levels were equal, medical students were more willing to prescribe psychotrophic drugs to patients whose problems were onset-uncontrollable than to those with onsetcontrollable problems. 18 Due to the influence that observers have on pain sufferers, it is important to understand the mechanisms that contribute to observers' decisions to offer social support (help) to pain patients. Therefore, because of the known influence of affect (sympathy and anger) on helping judgements and behaviour, these variables also were also included in the present research. Hypotheses The central hypotheses for the present study were as follows : 1. Displayed levels of pain in the target patients would influence the ratings made by the participants. It was expected that higher levels of pain would be attributed to the tests in which the patients are exhibiting high levels of pain relative to the low level of displayed pain . The effect of the moderate level of displayed pain is unknown. 2. It was expected that there would be a main effect for coping information. Specifically, participants led to believe that the patients were coping would show lower levels of anger, increased levels of sympathy, higher predictions of social support and selection of the less painful treatment choice relative to participants led to believe that patients were not coping. 3. There would also be a main effect for attributional style in that those participants with an unsupportive AS would view the patients as more in control of their pain than those participants with a supportive AS , irrespective of coping information. 4. A main effect was also anticipated for medical evidence. The absence of medical evidence was expected to lead to decreased evaluations of patients' pain severity. 19 5. Family history was also anticipated to affect the pain severity variable . Participants with a positive family history would produce higher judgements of pain relative to participants with a negative family history. 6. A main effect of gender was anticipated in that females were expected to offer greater levels of social support, as this was a non-emergency situation . 20 CHAPTER II Method Participants Ninety-six undergraduate students participated in this study in exchange for either credit in their psychology classes or for entry into a lottery. Of these participants , 63 were female and 33 were male. The age range was from 17 - 52 years (M = 23.22 , SO= 6.64). Materials The materials used in this study included the RMQ (Higgins, 1992; Appendix A), videotapes of patients with shoulder pathology undergoing a physiotherapy assessment, written vignettes to accompany the videotapes which included the context manipulation and measures designed to assess the dependent variables (Appendix B). Additionally, participants answered a personal information questionnaire (Appendix C) in the final part of the study as part of the postexperimental inquiry. RMQ. The RMQ was chosen for this study as it was expressly designed as a measure of person perception , whereas all other measures of attributional style address self-perception (e.g., Peterson, Semmel , von Baeyer, Abramson & Seligman's 1982 "Attributional Style Questionnaire", revised by Peterson & Villanova in 1988). Based on McAuley, Duncan and Russell's (1992) Causal Dimension Scale II, the RMQ assesses supportive and unsupportive AS based on the causes participants generate for six different negative outcomes that they imagine have happened to a hypothetical other, and their subsequent ratings of each cause on 12 21 scales pertaining to the dimensions of locus of causality, stability, personal and external control. These dimensions are measured on Likert-type scales from 1 - 9, with higher scores representing higher internal, stable, personal and external control judgements. An example of a negative outcome and an item assessing personal control appears below. 1. Cancer. One likely c a u s e : - - - - - - - - - - - - - - - - - - - - - - - - - - Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Individuals' attributional styles were derived by averaging the locus and personal control ratings for the six negative outcomes. Videotapes. The videotapes that were used in the present research were excerpts taken from a videotape of patients with shoulder pathology undergoing range of motion tests at a physiotherapy session (for a full discussion of patient characteristics and videotape construction, see Prkachin & Mercer, 1989). These range of motion tests included two types: 1) active (performed by the patient); and 2) passive (performed by the physiotherapist) . In a study which used these videotapes to examine the communication of pain, it was found that observers systematically underestimated the patients' pain (relative to the patients' self reports), particularly in the active tests (Prkachin et al., 1994). While the observers were poor judges for the passive tests, they were less poor than for the active tests, and as a result, only passive tests were used in the current study. 22 The facial pain displays of these patients were quantified according to intensity and duration by Facial Action Coding System (FAGS) analysis (see Prkachin & Mercer, 1989). This allowed for an objective measure of displayed pain severity. Patients' pain levels were considered similar if they had comparable FAGS scores. These FAGS scores are a composite of pain related facial movement intensity and duration. So comparable patients also had to exhibit approximately the same number of facial movements and of roughly equal duration for their overall scores to be considered similar. Additionally, observers tend to respond differently to pain in males and females, in that females' pain is denigrated relative to males' pain, although the reasons for this discrepancy are not well understood (Holden, Gladstein, Trulsen, & Wall , 1994). Therefore, for the present research, target patients of the same gender (males) were used . Previous use of these videotapes quantified no pain as a FAGS index range of 0, low pain as 1-10, and moderate to severe pain as any index score over 10 (Deyo, 1999). The target patients in the current research were chosen to loosely fit these criteria, however, these criteria had to be modified somewhat for current purposes to exclude active tests and females, exclusions not used by Deyo (1999) . The addition of these two new criteria limited the selection available in the videotapes, and resulted in low pain defined as a FAGS index of 0- 9, moderate pain as an index of 10 - 15, and high pain as above 15. The range of FAGS index scores in the original videotapes was from 0- 120. There were six different videotapes constructed. Each tape consisted of: 1) two practice patients (to allow participants to familiarize themselves with the procedure); and 2) one of the two targets. For each target, there were three tests, one each for 23 low, moderate and high displayed levels of pain; the sequencing of these tests was counterbalanced to control for order effects (resulting in six different videotapes) . The excerpts were 10 seconds long, and consisted of a close-up view of the patient's face. Between these excerpts was 10 seconds of a black screen which provided participants the opportunity to make their severity judgements before viewing the next excerpt. The context manipulations, which consisted of coping and medical evidence information, were presented in written vignettes to the participants before they viewed the videotapes (see Appendix B). The vignettes began with a brief introduction to the patient, including the nature of the patient's pain, and also included information about the videotapes themselves (e.g., that they were filmed during the patient's first visit to the physiotherapist). The vignettes then proceeded to ask the participant to imagine that they were the patient's physician and to base their pain severity ratings on the highest amount of pain that the patient appeared to be experiencing. The vignettes then went on the explain the patient's condition in more detail, including the length of time that the patient had experienced this pain (18 months), and the coping and medical evidence manipulations. Each participant received only one version of the coping information, and one version of the medical evidence information for each patient. This resulted in four different versions of the vignettes. The manipulations appear below with medical evidence preceding coping information: A neurologist has examined patient A, and a nerve conduction study and CT scan were ordered. These tests show clear evidence (no evidence) of nerve damage or any well recognized pain syndrome. 24 After being injured , this patient was advised by his physician to take antiinflammatory drugs on a prescribed schedule, to apply a cold pack daily, and to engage in activity to distract himself from the pain. It is clear from assessment of the patient that he (COPING- has followed this advice quite closely, but his pain persists. NOT COPING- has not followed this advice very closely, and his pain persists). Dependent Variable Assessment. Pain severity ratings were measured by a Likert-type scale from 0 (no pain) to 10 (unbearable pain). The question to assess distress was taken from Chibnall and Tait (1995) . It included a description of what is meant by "emotional distress" and measured participants' responses on a Likert-type scale from 0 (no distress) to 10 (extreme distress). The questions to assess disability were based on those used by Tait and Chibnall (1994). For their purposes, they modified the Pain Disability Index (PDI ; Tait, Chibnall & Krause, 1990; Tait, Pollard, Margolis, Duckro & Krause, 1987), a measure designed to assess one's own painrelated disability. The researchers modified the scale to assess others' pain-related disability, a modification maintained for this study. The PDI assesses the areas of functioning in family/home responsibilities , recreation, social activity, occupation , sexual behaviour, self-care and life support activities on Likert-type scales from 0 (no disability) to 10 (total disability) and respondents' answers were summed to arrive at a single disability score. An example from this scale follows. Recreation? This category includes hobbies, sports, and other similar leisure time activities. 0 no disability 1 2 3 4 5 6 moderate disability 7 8 9 10 total disability 25 Responsibility, as assessed by the question "If you were a physician looking after this patient, how much of the responsibility for the pain do you believe is the patient's?" was measured on a 0 (not responsible) to 10 (completely responsible) scale. The question to assess anger, which read "If you were a physician looking after this patient, how much anger would you feel towards him?" was also measured on a 10 point scale from 0 (no anger) to 10 (a lot of anger). The question to assess sympathy was the same as for anger, except that the word "anger" was substituted with "sympathy" in both the question and the rating scale. Two helping judgements were used , one concerning social support, and one concerning treatment choice. The social support question read: "If you were a physician looking after this patient, would you be willing to extend support (such as spending time talking and listening to) this patient?". This question was also rated on a Likert-type scale from 0 (not willing) to 10 (completely willing). A second helping judgement focused on participants' choice of one of two treatments for the pain patient. The treatment choices were set up in an identical fashion, describing how the therapy worked, who performed the therapy, its duration and purpose, the level of discomfort that the patient would endure, and the treatment effectiveness. Everything was equivalent between the two choices with the exception of the level of discomfort, the therapies either included pain (repetitive exercise) or not (ultrasound) . An example of the more painful choice is provided below: Treatment A: Activation therapy. This therapy addresses the shoulder pain through repetitive exercise. The therapy is delivered by a physiotherapist. The patient is prescribed a 30 minute exercise routine, which is performed twice a 26 week for four weeks. The purpose of the exercises is to promote healing by strengthening the muscles that support the shoulder joint. The exercises themselves are uncomfortable, but the overall effectiveness of the therapy is excellent in most cases. The personal information questionnaire was standard, except for the inclusion of family history with chronic illness questions. The questions concerning family history with chronic illness were prepared following the format of Prkachin et al. (1995) , defining chronic pain as lasting longer than the time required to heal and at least six months. Additionally, participants were provided with a definition of what "having lived with" and "spending significant amounts of time with" meant for the purpose of the study. Participants also were given questions pertaining to the specific family member, the location of the pain , and its intensity and duration. The scale on which participants rated their family members' pain intensity was the same Likert-type scale from 0 (no pain) to 10 (unbearable pain) used to assess the target patient's pain. Following the personal information , participants were asked about their perceptions of the purpose of the study as part of the post experimental inquiry. Procedure Participants completed the study either individually or in small groups of up to four people. At the front of the testing room was a television and videocassette recorder to allow participants to view the videotape of the pain patients. Upon entering the testing room participants were briefly introduced to the study and informed consent was obtained. Half of the participants received the RMQ before viewing the videotape and the other half viewed the videotape before responding to 27 the RMQ . Before viewing the videotape, participants read the brief scenario that included the manipulation and after each video excerpt participants had ten seconds to make their pain severity ratings . After viewing the complete videotape, participants were asked to respond to questions pertaining to their evaluations of patients' levels of distress, disability and responsibility, as well as their affect, behavioural predictions of offering social support to the patient in the videotape and selection of a course of therapy from two choices. The order of the two therapy choices was counter-balanced to control for order effects. Upon completion of their ratings based on the videotape and the RMQ, participants completed a personal information questionnaire and questions pertaining to their thoughts about what the study was examining. After that, participants were given a partial debriefing sheet and their participation was complete. 28 CHAPTER Ill Results Overview A repeated-measures analysis of covariance (ANCOVA) with AS as a covariate was used to test the effects of the coping and medical evidence manipulations as well as to identify any effect(s) due to the two patients on observers' ratings of the patient's pain severity. Two separate multivariate analyses of covariance (MANCOVAs) were used to test the effect(s) for the dependent variables of: (1) distress and disability; and (2) responsibility, anger, sympathy and support. Separate analyses were deemed appropriate as separate evidential bases exist for these variables and there were no compelling reasons to combine them, as a combination was not necessary for the present purposes. Correlation coefficients among the dependent variables and AS appear in Table 2. The independent variables did not appear to influence attributional style scores (see Table 3 for mean AS scores for the different levels of the independent variables) . The effects of family history with chronic pain (positive or negative) and gender were examined in the same manner as above with a repeated-measures ANCOVA with AS as a covariate to identify effects on pain severity ratings , and separate MANCOVAs to examine distress and disability on the one hand, and responsibility, anger, sympathy, and support on the other. A logistic regression was used to examine the relationships between AS, coping, medical evidence, patient, family history, gender and the binary outcome variable of treatment choice. In addition , a hierarchical discriminant function analysis 29 Table 2 Correlations Among the Dependent Variables and Attributional Style Distress Disability Resp Anger Sympathy Support AS Distress 1.00 Disability 0.41** 1.00 Resp -0.12 0.09 1.00 Anger -0.01 -0.11 0.38** 1.00 Sympathy 0.21* 0.40** -0.28** -0.40** 1.00 Support 0.11 0.36** -0 .13 -0 .17 0.56** 1.00 AS -0.05 -0.17 0.07 -0.01 -0 .05 0.11 Note. Resp =Responsibility. *p_ < .05 , **p_ < .01. 1.00 30 Table 3 Mean Attributional Scores for the lndegendent Variables ~ IVs Level Mean so Medical evidence present absent 5.58 5.69 .73 .60 Coping yes no 5.67 5.60 .66 .68 Patient A B 5.72 5.55 .61 .71 31 with AS as a covariate was used to examine the applicability of distress, disability, responsibility, anger, sympathy and support as predictors of treatment choice. All significant multivariate effects were examined using univariate analyses and subsequent significant univariate effects were followed up with t-tests when 2 necessary. Measures of effect size (eta-squared (11 )) for multivariate and univariate analyses , and population point biserial correlation (Ppb) coefficients fort-tests) were also included for all significant effects. Based on the criteria outlined by Kirk (1996), 2 11 values of .01 0, .059 and .138 were taken as corresponding to small, medium and large effect sizes, respectively, and Ppb values of .1 0, .24 and .37 were taken as corresponding to small, medium and large effect sizes. Pain severity. The results of the repeated measures ANCOVA revealed no effect for AS , and a main effect of pain level on ratings of the severity of the patient's pain , E(2 ,87) = 96.94, Q = .000 , 11 2 = .690 . The results of follow up tests indicated that, as expected , participants gave higher ratings of pain severity to high pain (M = 5.56) than moderate pain (M = 4.39) , !(95) = 6.77, Q < .01, PID2 = .57 , and low pain (M = 2.56), !(95) = 12.97, Q < .01 , PID2 = .80. Participants also rated moderate pain higher than low pain , !(95) = 8.01 , Q <.01 , PID2 = .63. This effect was moderated by patient. The results revealed a significant pain level by patient interaction , E(2,87) = 17.93, Q = .000, 11 2 = .292. Follow up tests, as shown in Figure 2, revealed that participants' ratings of Patient A's high pain (M = 6.02) were significantly higher than their ratings of Patient B's high pain (M = 5.10) , !(94) = 2.34 , Q = .02, PID2 = .23 . For their ratings of moderate pain , participants' ratings 32 Figure 2 Interaction Between Participant's Mean Pain Severity Ratings for Patient A and Patient B in the Low. Moderate and High Displayed Levels of Pain Conditions 33 10 • • 9 en C> 8 c 7 0::: 6 ca +-' >. +-' ·c Q) 5 (f) 4 ca a.. 3 > Q) c 2 6.02 2.43 1 0 Low Moderate Pain Level High Patient A Patient 8 34 of Patient A's pain (M =3.94) were significantly lower than their ratings of Patient B's pain (N = 96 , M = 4.83) , !(94) = -2 .50, Q = .014, PQ.Q = -.26. There were no differences in ratings of pain severity between Patient A and Patient B for low pain. These results were investigated further by examining the FAGS scores of these patients. Patient A's FAGS scores for high, moderate and low pain were 37 .14, 10.5 and 0, respectively. Patient B's FAGS scores for high , moderate and low pain were 30.00 , 13.02 and 0, respectively. Thus, it seems that participants were able to correctly distinguish between subtle variations in displayed levels of pain . There were no effects for medical evidence or coping on participants' ratings of pain severity. Distress and disability. Again, there was no effect for AS, however, significant multivariate effects on ratings of patients' levels of distress and disability were identified for the coping manipulation , E(2,86) = 4.32, Q = .01 , YJ 2 = .09 . Univariate effects revealed that the coping manipulation significantly influenced participants' ratings of patients' emotional distress, E(1 ,85) = 8.65 , Q = .00 , YJ 2 = .09. Participants in the coping condition (D.= 49) rated the patients' emotional distress as significantly higher (M = 4.84) than those in the not coping condition (D.= 47, M = 3.66) . There were no significant effects on participants' ratings of the patient's disability. Responsibility. anger. sympathy and support. AS was again insignificant, although a multivariate effect of the coping manipulation was revealed , E(4,84) =8.12, Q =.00, YJ =.279. As shown in Table 4, the results from subsequent univariate tests 2 revealed effects on responsibility, anger, sympathy and support ratings. As also shown in Table 3, the comparisons between the coping and not coping conditions on 35 Table 4 Univariate Simple Effects for the Coping Manipulation Variables Simple Effects DVs IV E(1 ,87) 11 Responsibility Coping Not Coping 9.53** Anger Coping Not Coping Sympathy Social Support 2 M (SO} PJ2Q .09 4.22 (2.44) 5.68 (2.21) .30 27.23** .24 0.83 (1 .51) 2.82 (2.13) .48 Coping Not Coping 9.52** .09 5.02 (2.11) 3.70(2.15) -.32 Coping Not Coping 4.88* .05 6.08 (2.20) 5.02 (2.52) -.23 Note. Coping n= 49, Not coping n =47. *Q = .05, **Q < .01 . 36 these variables indicated that when participants believed that the patient was not coping with their pain they reported significantly higher ratings of patient responsibility and anger towards the patient as well as lower ratings of sympathy and support for the patient than those participants who believed that the patient was coping with their pain. Effects of family history and gender on ratings of pain. There were no effects of AS, family history and participants' gender on ratings of the patient's pain severity. Effects of family history and gender on ratings of distress and disability. A multivariate effect for gender was identified , E(2,90) = 4.45, Q = .01 , 11 2 = .09. The results of the univariate follow up tests indicated that the difference was observed in disability scores, E(1 ,91) = 6.69 , Q = .011 , 11 2 = .068. Females (D.= 63) gave significantly higher ratings of pain-related disability (M =29.16) than did males (D. =33, M = 23.79) . There were no significant differences in participants' ratings of the patient's distress, and no effect for the AS variable. Effects of family history and gender on ratings of responsibility. anger. sympathy and support. No significant interactions or main effects were identified . Effects of coping, evidence. patient. family history and gender on treatment choice. The independent variables did not account for any significant changes to treatment choice (Wald statistics and the overall l test were non-significant). Effects of distress. disability, responsibility. anger, sympathy and support on treatment choice. A discriminant function was calculated with a l(1) =5.83 , Q =.01 . A statistically significant separation between the two treatment choices was identified, E(1 ,94) =6.04, Q =.01. Of the predictors, only the sympathy variable was able to 37 separate the two treatment choices, with 31 (58.5%) of those participants who selected activation therapy being correctly classified, and 29 (67.4%) of those participants who chose ultrasound therapy being correctly classified (see Table 5 for the Discriminant Function Analysis Classification results). The classification function coefficient (Fisher's linear function coefficient) for the sympathy variable for the activation group was .834, with a constant of -2 .31, and for the ultrasound group was 1.07 with a constant of -3.35. As no other predictor variables were included in the analysis after this first step (based on the criterion that a variable's E value had to exceed 2. 71 to be included), this classification could not be improved upon by the inclusion of any of the remaining variables. Participants who chose ultrasound therapy (the less painful alternative to activation therapy) offered higher levels of sympathy (D.= 43, M sympathy rating = 4.97) than those participants who chose activation therapy (D. =53, M sympathy rating =3.88). 38 Table 5 Discriminant Function Analysis Classification Results Actual Group No. of Cases Predicted Membership Group 1 Group 2 Group 1 activation 53 31 58.5% 22 41 .5% Group 2 ultrasound 43 14 32 .6% 67.4% 29 Note. Percentage of "grouped" cases correctly classified based on participants' sympathy ratings: 62.5% 39 CHAPTER IV Discussion Consistent with past research, a contextual factor was found to influence observers' perceptions of a pain patients' symptoms. Specifically, a patient's level of coping (i.e., present efforts to control an ailment by following their physician's advice) , was found to influence observers' judgements of the amount of distress the patients felt, the patients' responsibility for their plight, observers' affect toward the patient and willingness to extend support to the patient. Observers gave higher ratings of distress to patients who were following their physicians' advice than to those who were not. This finding extends Tait and Chibnall's (1994) results on the effect of onset controllability on ratings of distress. In their research, Tait and Chibnall (1994) found that patients who had no control over the onset of their pain (e.g., car accident, struck 01 elilt>tJOtlat (mstless "trran cuuse .. · M. nom ~ wele' glven ol over their pain (e.g., ran a red light and struck another patients who had onset c .sent study of a significant effect for pain-related disability car). The absence in the all (1994), who also failed to identify a control-related again mirrors T a it and Ct disability effect. Coping participants' ratings of responsibility, anger, sympathy The effect of copir pected directions and lend further support to Weiner's and support were all in tt attributional model of helping behaviour. Participants (1980a; 1980b; 1986; 19 thad followed his physician's advice quite closely (i.e., who were told that the pc: er ratings of responsibility to the pain patient and the coping condition) ga" 40 expressed lower anger, higher sympathy and a greater likelihood of offering social support than did participants who were told that the patient had not followed his physician's advice very closely (i.e., the not coping condition). These findings may be explained in part by the findings of Marteau and Riordan (1 992), who found that patients in their controllable condition were viewed as less concerned about their situation, less likely to follow advice given to them , less likely to understand their condition and less enjoyable to work with by physicians and nurses. Logically, then, participants should rate distress as higher when the participant is presented as coping. That is, if a patient is following their physician's advice, they are actively trying to reduce their ailment; therefore, they must be concerned about their situation, and it is logical to assume that this could lead to higher distress. In the present study, those patients who were presented as not coping may have been viewed as too unconcerned to act on their own behalf, which would logically lead to perceptions of lower distress. This logic is also supported by participants' responsibility ratings, they seem to be assuming that patients who are not coping are more responsible for their ailment, instead of thinking that there may be mitigating circumstances which prevent the patients from coping. Following the attributional theory of helping behaviour, not coping would also be expected to lead to higher negative affect, lower sympathy and less willingness to help (Weiner 1980a; 1980b; 1986; 1995a). Why this did not extend to lower disability and severity ratings is an interesting question. Participants appear to be denigrating the person, but not the person's pain experience. They may be less likely to help these patients, but they are 41 at least aware that pain levels and resulting disability between the two groups (coping versus not coping) are equivalent. Attributional Style The lack of any significant effects for attributional style was surprising in light of previous research which has found effects of these variables on related outcomes. For example, Higgins and Shaw (1999) found an interactional effect for AS and controllability information on social support. Those participants with a supportive AS were just as likely to offer support whether the persons' reason for need was under their control or not. However, participants with an unsupportive AS offered greater support when the reason for need was out of the control of the needy person than when the reason for need was in their control. The Higgins and Shaw (1999) study used a "real-world" helping situation in that participants were telephoned on behalf of another student who needed help for reasons that were controllable or uncontrollable by the needy individual. Participants were then asked to telephone this "needy individual" (actually a confederate) if they were willing to help. This methodology had participants actually help the needy person, whereas in the current study, participants had to predict whether or not they would help. Using a "real-world" helping scenario Higgins and Shaw (1999) found an interactional effect of AS on helping behaviour, whereas in the current study a paper and pencil helping scenario was used and no relationship between AS and helping was found. This suggests that people may respond differently to laboratory situations than real life situations. 42 Medical Evidence Additionally, the lack of any significant effects for medical evidence was also surprising due to the findings of previous research. In both Chibnall and Tait (1995) as well as in Tait and Chibnall (1994) the presence of medical evidence led to increased judgements of emotional distress and pain-related disability relative to the judgements given in the absence of medical evidence . However, both of these studies used paper and pencil manipulations exclusively. That is, those studies had participants read descriptions of the pain patient, and based their judgments on that information alone, whereas in the current study, patients not only read information about the person, but also that information was embellished by participants actually viewing the pain behaviour of the person that they were judging on a videotape. While the discrepancies in the findings may be the result of other unidentified factors, the current findings seem to imply that findings based on the presentation of scenarios may not be valid indicators of how people will actually react in real life situations. To increase external validity and the ability to apply laboratory findings to real situations, representations of the manipulations should be as true to life as possible. Participants respond differently to a hypothetical other than to a person that they have actually viewed. It is conceivable that participants would respond differently to patients that they actually interact with, a situation which would even more closely approximate reality. Pain Severity Participants in the current study were highly sensitive judges of the severity of pain. Not only were they able to differentiate between the levels of pain (high, 43 moderate and low), but they were able to identify the subtle differences in the levels of pain between the two patients. Although the current study used excerpts from videotapes used by Prkachin et al. (1994), participants in the two studies responded quite differently. Participants in the current study were able to distinguish between very subtle variations in the patients' facial expressions of pain, something that the participants in the Prkachin et al. (1994) study were not able to do. This may have been due, in part, to the excerpts chosen. As previously mentioned, participants in the Prkachin et al. (1994) study were better at judging the pain expressed on the videotapes of passive tests (performed by the physiotherapist) than they were at judging the active tests (performed by the patient), and thus, for the current research, only passive tests were selected for use. Additionally, the current sample (N = 96) was substantially larger than the sample used in Prkachin et al. (1994) (N = 5), and presumably more heterogeneous. The difference in performance between the two samples may also be partially due to the measurement tools used. Prkachin et al. (1994) had participants rate the amount of pain they observed on a 15 point verbal descriptor scale (Heft, Gracely, Dubner, & McGrath, 1980). On this scale, each of the 15 points has a corresponding verbal descriptor ranging from "slightly unpleasant" to "very intolerable". The scale used in the current research was a 10 point scale with the descriptor "no pain" at one end, "moderate pain" at the midpoint, and "unbearable pain" at the opposite end. While the current scale is somewhat more crude and limited, it was designed to be comparable to the scales used by Chibnall and Tait (1995), and Tait and Chibnall (1994), as the current study was meant as a replication/expansion of these studies and the use of 44 similar scales was meant to facilitate comparisons with these studies. As the more restricted scale (1 0 versus 15 point) is less sensitive and gave participants fewer options, this may have resulted in exaggerations of their perceptions in differences of displayed levels of pain. Another possibility for this discrepancy could be the number of pain ratings made by participants, 280 in the Prkachin et al. (1994) study, and 3 in the current study. Furthermore, before rating the pain of the target patient (Patient A or Patient B), participants were provided with a chance to practice the procedure by viewing two practice patients, each displaying a high, moderate and low level of pain that had comparable FACS scores to the target patients. This may have primed the participants so that they were expecting high, moderate and low levels of displayed pain, and possibly served to increase their sensitivity to subtle variations. With only a few ratings to make, participants would have had the opportunity to think about their ratings and compare them to their previous ratings. Undoubtedly, sensitivity to variations in pain levels is the result of many interacting factors, and identification of the mechanism(s) responsible for this finding would be worthwhile. Typically, people are inaccurate judges of pain (usually underestimation of the severity of pain, e.g., Prkachin et al., 1994) so any method of increasing people's accuracy in pain assessments would be of great benefit. Treatment Choice Interestingly, participants' sympathy ratings were the only variable that was shown to predict treatment choice reliably. Higher ratings of sympathy were associated with the selection of the less painful treatment choice (ultrasound over 45 activation) , conversely, lower ratings of sympathy were associated with the selection of the more painful alternative (activation over ultrasound). It is not likely that judges were being vindictive towards those they felt were responsible for their plight, as there was no relationship between negative affect and treatment choice. However, there was a relationship between their level of supportive affect (sympathy) , either high or low and their treatment selection . Family History The lack of an effect for family history was unexpected due to the findings of Prkachin et al. (1995), who reported a moderate effect for family history (with those participants' with a positive family history rating pain as more severe than those without a family history of chronic pain). Contrary to expectation , patients with a positive family history of chronic pain did not rate pain differently than those patients without such a family history; both groups were equally sensitive to subtle variations when assessing pain . These results are likely the by-product of the same mechanisms that resulted in the sensitivity of the pain assessments; whatever influenced participants to rate the subtle differences in pain levels differently worked equally well for participants with or without a family history of chronic pain. Gender Although the present study examined a non-emergency helping situation, females did not offer higher levels of social support than did males (contrary to expectations based on the findings of Kessler, Mcleod , & Wethington, 1985). However, females did offer higher judgements of pain related disability than did males (contrary to the findings of Tait & Chibnall , 1994, who did not identify any gender 46 related differences for disability ratings) . Some possible explanations for this finding were considered. For example, if this finding had to do with the general conception of females as more nurturing than males (e.g., Hopkins & Klein, 1993) it may be I assumed that females would have higher sensitivity to disability. However, this finding would expect to extend to other ratings such as distress, sympathy and anger. It was also considered that perhaps this finding might be explained by some personal variable (such as more females in the sample having direct experience with chronic shoulder pain) of the participants' . Although the hypothesis that females in the current sample had more experience with shoulder pain was confirmed (14% of the females, n =9, compared to 6% of the males, n =2) it is unlikely that these small numbers could have exerted a significant effect. There may have been some other personal variables not examined in this study which could account for these findings, however, it should be noted that results from analyses based on gender should be interpreted cautiously due to the inequality of the sample sizes (males n = 33, females n = 63), and thus these findings may simply be artifacts. Implications There are a number of important practical, theoretical and methodological implications that follow from the results of this study. Practically, pain patients who do not present themselves as actively working towards healing themselves face more negativity from observers than patients who present themselves as actively working towards healing themselves. For non-coping patients, peoples' perceptions of their lack of coping may lead to anger, less willingness to lend these patients social support (that is, spend time talking with and listening to them), lower feelings of sympathy, and 47 lower judgements about the patients' level of disability. In short, observers tend to denigrate these patients relative to patients who are perceived to be coping with their pain . Interestingly, there was no relationship between negative judgements of responsibility and anger and observers' ratings of the amount of pain and pain-related disability suffered by the patient. Thus, it seems that participants are not influenced by negative emotional or social judgements when assessing suffering and disability. However, when deciding on a treatment, observers were influenced by the level of sympathy they felt toward the pain patient. Thus the relationship between emotion and treatment involves observers' levels of affinity, not malice. As such, if a patient wants to ensure that they receive the least painful of all possible treatments with equivalent outcomes, they must make certain that they have the sympathy of the individual making that decision. Of course, it is impossible to know what a health care professional who has the power to decide on a treatment would choose based on these results, but it would be worthwhile to examine whether these results also hold true for that population. Limitations and Suggestions for Future Research Not only can these findings not be generalized to health care professionals, one must also keep in mind that the present sample consisted of primarily first year undergraduates, who all knew they were participating in a psychology experiment regarding a pain patient. Thus, it might not be a completely accurate assessment of how a typical observer would make the various ratings in this study. Also, the age range was from 17 - 52, however, the mean was 23 and the mode was 19. It would be worthwhile to examine the developmental differences in the influence of context in 48 pain perception , to identify age-related effects. Future studies should also include more males in the sample (this study only had 33 males, compared to 63 females}, and should include female pain patients as well to see if there are differential effects of context information on perceptions of the pain of males and females . An additional limitation to the current study was built into the videotapes used . These videotapes only included a close-up of the patient's face. However, observers may also use other information in real life situations, such as the patient's posture and demeanor when making the type of judgements that were required for this study. While the use of these videotapes is a closer approximation of reality than simply reading about a fictitious pain sufferer, as in the typical paper and pencil measures, more realistic representations of the manipulations are necessary to aid in our understanding of the various influences on the perceptions of pain. Equally important is a further understanding of ways to improve judges' accuracy in assessing severity of pain . As is evident in the current findings , people can be sensitive to variations in displayed pain level, and identifying the factors which contribute to this sensitivity can have important implications for people in pain. Appropriate treatment and care begins with a full and accurate understanding of a pain patient's condition, including the actual amount of pain that they are experiencing , and being able to identify subtle differences in displayed levels of pain is a step toward increasing accuracy in pain assessment. 49 References Anderson, C., & Weiner, B. (1992) . Attribution and attributional processes in personality. In G. V. Caprara & G. L. Van Heck (Eds.), Modern personality psychology: Critical reviews and new directions. New York, NY: Harvester Wheatsheaf, pp. 295-324. Betancourt, H. (1990). An attribution-empathy model of helping behavior: Behavioral intentions and judgments of help-giving . Personality and Social Psychology Bulletin . 16, 573-591 . Brewin, C. R. (1984). Perceived controllability of life-events and willingness to prescribe psychotropic drugs. British Journal of Social Psychology, 23, 285-287. Chibnall, J. T. & Tait, R. C. (1995) . Observer perceptions of low back pain: Effects of pain report and other contextual factors. Journal of Applied Social Psychology, 25, 418-439. Deyo, K. S. (1999) . Developmental differences in the ability to decode facial expression of pain. Unpublished Master's Thesis, University of Northern British Columbia. Ferrell , B., Eberts, M., McCafferey, M. , & Grant, M. (1991). Clinical decision making and pain. Cancer Nursing. 14, 289-297. Fordyce, W. E. (1976). Behavioral concepts in chronic pain and illness . In P. 0. Davidson (Ed .), The behavioral treatment of anxiety, depression and pain . New York: Bruner/Mazel, pp. 147-188. Grossman , S. A., Sheidler, V. R. , Sweeden , K., Mucenski, J., & Piantadosi, S. (1991 ). Correlations of patient and caregiver ratings of cancer pain . Journal of Pain and Symptom Management. 6, 53-57 . Heft, M., Gracely, R., Dubner, R. , & McGrath , P. (1980). A validation model for verbal descriptor scaling of human clinical pain . Pain. 9, 363-373. Holden, E. W ., Gladstein, J., Trulsen, M. , & Wall, B. (1994). Chronic daily headache in children and adolescents. Headache. 34, 508-514. Hopkins, H. R., & Klein , H. A. (1993). Multidimensional self-perception: Linkages to parental nurturance. Journal of Genetic Psychology, 15, 465-473. Higgins, N. C. (1992). Cross situational consistency attributions about the causes of others' misfortunes: A critical evaluation of attributional style. Doctoral dissertation, Simon Fraser University. 50 Higgins, N. C., & Morrison , M.A. (1998). Construct validity of unsupportive attributional style: The impact of life outcome controllability. Social Indicators Research. 45, 319-342. Higgins, N. C. , & Shaw, J. K. (1999). Attributional style moderates the impact of causal controllability information on helping behavior. Social Behavior and Personality, Kessler, R. C., Mcleod, J. D., & Wethington, E. (1985). The cost of caring: A perspective on the relationship between sex and psychological distress. In I. G. Sarason & B. R. Sarason (Eds.), Social support: Theory research and applications (pp. 491-506). Dordrecht, The Netherlands: Martinus Nijhoff. Kelley, H. H. (1971). Attribution in social interaction. In E. E. Jones, D. E. Kanouse, H. H. Kelley, R. E. Nisbett, S. Valins, & B. Weiner (Eds.), Attribution : Perceiving the causes of behavior. Morristown , NJ; General Learning Press, pp. 1-26. Kelley, H. H. (1973). The process of causal attribution . American Psychologist. 28 , 107-128. Kirk, R. E. (1996) . Practical significance: A concept whose time has come. Educational and Psychological Measurement. 56, 746-759 . Marteau, T. M., & Riordan D. C. (1992). Staff attitudes towards patients: The influence of causal attributions for illness. British Journal of Clinical Psychology, 31, 107-110. McAuley, E., Duncan, T. E., & Russell, D. W. (1992) . Measuring causal attributions: The revised causal dimension scale. Personality and Social Psychology Bulletin. 18, 566-573. Meyer, J. P. & Mulherin, A. (1980). From attribution to helping: An analysis of the mediating effects of affect and expectancy. Journal of Personality and Social Psychology, 39, 201-210. Peterson, C., Semmel, A., von Baeyer, C. , Abramson , L. Y. , & Seligman, M. E. P. (1982) . The Attributional Style Questionnaire. Cognitive Therapy and Research . Q, 287-300. Peterson, C., & Villanova, P. (1988). An expanded attributional style questionnaire. Journal of Abnormal Psychology, 97, 87-89. Prkachin, K. M., Berzins, S., & Mercer, S. R. (1994) . Encoding and decoding of pain expressions: A judgment study. Pain. 58, 253-260. Prkachin, K. M., & Craig , K. D. (1995). Expressing pain: the communication and interpretation of facial pain signals. Journal of Nonverbal Behavior. 19, 191-205. 51 Prkachin , K. M. , & Mercer, S. R. (1989). Pain expression in patients with shoulder pathology: validity, properties and relationship to sickness impact. Pain, 39 , 257-265. Prkachin , K. M., Solomon , P. S., Hwang , T. , & Mercer, S. R. (1995). Does experience affect judgements of pain behaviour?: Evidence from relatives of pain patients and health-care providers. Manuscript in preparation . Reisenzein , R. (1986). A structural equation analysis of Weiner's attributionaffect model of helping behavior. Journal of Personality and Social Psychology, 50 , 11 23-11 33. Rudy, T. E. , Turk, D. C., & Brena, S. F. (1988) . Differential utility of medical procedures in the assessment of chronic pain patients. Pain. 34 , 53-60. Salmon , P., & Manyande, A. (1996). Good patients cope with their pain: Postoperative analgesia and nurses' perceptions of their patients' pain. Pain. 68, 6368 . Schmidt, G. & Weiner, B. (1988). An attribution-affect-action theory of behavior: replications of judgments of help-giving. Personality and Social Psychology Bulletin. 14, 610-621. Schwarzer, R., & Weiner, B. (1991). Stigma controllability and coping as predictors of emotions and social support. Journal of Social and Personal Relationships, 8, 133-140. Tait, R. C., & Chibnall , J. T. (1994) . Observer perceptions of low back pain . Journal of Applied Social Psychology, 24 , 415-431. Tait, R. C., Chibnall , J. T., & Krause, S. J. (1990). The pain disability index: Psychometric properties. Pain. 40 , 171-182. Tait, R. C., Pollard, C. A., Margolis, R. B., Duckro, P. N., & Krause, S. J. (1987). The pain disability index: Psychometric and validity data. Archives of Physical Medicine and Rehabilitation. 68, 438-441. Turk, D. C., & Flor, H. (1987). Pain> pain behaviors: The utility and limitations of the pain behavior construct. Pain. 31, 277-295. Turk, D. C. (1994). Perspective on chronic pain: the role of psychological factors. Current Directions in Psychological Science. 3, 45-48. Turk, D. C. , & Okifuji , A. (1997). Evaluating the role of physical, operant, cognitive, and affective factors in the pain behaviors of chronic pain patients. Behavior Modification. 21, 259-280 . 52 Weiner, B. (1980a) . A cognitive (attribution)-emotion-action model of motivated behavior: An analysis of judgments of help-giving . Journal of Personality and Social Psychology, 39, 186-200. Weiner, B. (1980b). May I borrow your class notes? An attributional analysis of judgments of help giving in an achievement related context. Journal of Educational Psychology, 72 , 676-681 . Weiner, B. (1986). An attributional theory of motivation and emotion. New York: Springer-Verlag. Weiner, B. (1993). On sin versus sickness: A theory of perceived responsibility and social motivation. American Psychologist. 48, 957-965. Weiner, B. (1995a) . Inferences of responsibility and social motivation. In Zanna, M. P., Advances in experimental social psychology, Vol. 27. San Diego, CA: Academic Press, pp. 1-47. Weiner, B. (1995b) . Judgments of responsibility: A foundation for a theory of social conduct. New York, NY: Guilford Press. Weiner, B., Perry, R. P., & Magnusson, J. (1988). An attributional analysis of reactions to stigmas. Journal of Personality and Social Psychology, 55, 738-748. Williams, C. W. (1993). The effects of priming causal dimensional categories on social judgments. Social Cognition. 11, 223-242. 53 Appendix A Reasons for Misfortunes Instructions: The items on the following pages present specific misfortunes or problems that might happen to anyone. For each item , think about how such a thing could likely happen to someone (other than yourself) and then write down one plausible (likely) reason that comes to mind. That is, for each item, think over what you know about the world to answer the question, "How does a problem like this happen to someone (excluding myself)?" Then , try to express a piau sible reason for the misfortune in a single sentence. After writing down a likely cause for a misfortune, then rate that cause on each of the twelve scales provided by circling one number on each scale. When doing the ratings, be sure to focus on the cause (that is, the reason for the onset) of the problem, NOT on the problem. This may be difficult at times. In other words, make sure you are rating the cause you write down for the misfortune, and NOT the misfortune itself. "The person" referred to in the rating questions means the person who has the problem; the term "Other people" referred to in the ratings means anyone else (that is, anyone other than the person with the problem) . Please take your time when doing the ratings - make sure you read the question carefully. You may find that there is more than one way of interpreting some of the rating questions. Please interpret these questions in the way that is most meaningful to you . There are no right or wrong answers to these questions. To summarize, for each of the 6 misfortunes, you should : 1) think over what you know about how such a misfortune could likely happen to someone (other than yourself). 2) write down one likely cause of that misfortune - try to express the reason in one sentence. 3) then, rate that cause by circling one number on each of the 12 scales provided each time you do the ratings, be sure to focus on the cause you wrote down (i.e., the reason for the problem) , NOT on the problem. 4) if you find there is more than one way of interpreting a question , interpret it in a way that is most meaningful to you. 5) please read the questions carefully. Please answer all the questions. It should take approximately 10-15 minutes to finish this questionnaire. You are, of course, free to stop participating at any time. 54 PLEASE ANSWER ALL OF THE QUESTIONS. Keep in mind that there are no right or wrong answers. Please refer back to the instructions if you are unsure about what to do. 1. Cancer. One likely cause: Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 1 About others Over which the person has power 9 8 7 6 5 4 3 2 1 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate 2. Divorce. One likely cause : Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 About others Over which the person has power 9 8 7 6 5 4 3 2 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate 55 3. Bankruptcy. One likely cause: Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something_: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 About others Over which the person has power 9 8 7 6 5 4 3 2 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate 4. Facial Disfigurement. One likely cau Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something_: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 About others Over which the person has power 9 8 7 6 5 4 3 2 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate 56 5. Has no friends. One likely cause: Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. Is the cause something_: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 About others Over which the person has power 9 8 7 6 5 4 3 2 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate 6. Loss of all possessions. One likely cause: Think about the reason you have written above. The items below concern your impressions or opinions of this cause of this situation. Circle one number for each of the following questions. /s the cause something_: That reflects an aspect of the person 9 8 7 6 5 4 3 2 Reflects an aspect of the situation Manageable by the person 9 8 7 6 5 4 3 2 Not manageable by the person Permanent 9 8 7 6 5 4 3 2 Temporary The person can regulate 9 8 7 6 5 4 3 2 The person cannot regulate Over which others have control 9 8 7 6 5 4 3 2 Over which others have no control Inside the person 9 8 7 6 5 4 3 2 Outside the person Stable over time 9 8 7 6 5 4 3 2 Variable Over Time Under the power of other people 9 8 7 6 5 4 3 2 Not under the power of other people About the person 9 8 7 6 5 4 3 2 About others Over which the person has power 9 8 7 6 5 4 3 2 Over which the person has no power Unchangeable 9 8 7 6 5 4 3 2 Changeable Other people can regulate 9 8 7 6 5 4 3 2 Other people cannot regulate Thank you for participating! 57 Appendix B We are interested in studying how people who have been injured communicate their experience to others. I will be showing you a brief videotape of a patient who has been experiencing pain in his shoulder. He has been referred for assessment to a clinic that specializes in assessing and treating such problems. The scenes that you will see on the videotape are taken from the patient's first visit with a physiotherapist. You will see the patient's reactions to a series of tests that the therapist uses to diagnose the nature and severity of the problem . Imagine that you are this patient's physician. While viewing the tests, we would like you to rate the highest amount of pain that the patient appears to be experiencing , and after the completion of the tests, we would like you to give us your judgements about several aspects of the patient's experience. To help you get used to the procedure, we will provide you with some practice before you view the target patient. Please feel free to ask the researcher any questions that you may have as they arise. It may be helpful for you to know a little bit about the patient's pain problem . Patient A Approximately 18 months ago, patient A received an injury to his shoulder, and has had shoulder pain since the event. A neurologist has examined patient A, and a nerve conduction study and CT scan were ordered . These tests show clear evidence (no evidence) of nerve damage or any well recognized pain syndrome. After being injured, this patient was advised by his physician to take anti-inflammatory drugs on a prescribed schedule, to apply a cold pack daily, and to engage in activity to distract himself from the pain. It is clear from assessment of the patient that he (COPING - has followed this advice quite closely, but his pain persists. NOT COPING- has not followed this advice very closely, and his pain persists). Please circle the number which best represents the maximum amount of pain that the patient appears to be experiencing for each test. TEST 1 0 1 2 3 4 no pain unbearable pain 5 6 7 8 9 10 6 7 8 9 10 moderate pain TEST2 0 TEST3 1 2 3 4 no pain 0 no pain 5 moderate pain 1 2 3 4 5 moderate pain unbearable pain 6 7 8 9 10 unbearable pain 58 Emotional distress includes negative feelings, emotional upset, and depression. Please rate how much emotional distress you believe that the patient you just viewed experiences: 0 I 1 I 2 I 3 I no distress 4 5 6 I I moderate distress I 7 I 8 I 9 I 10 I extreme distress How much do you think the pain of Patient A interferes with his: Family/home responsibilities? This category refers to activities related to the home or family. It includes chores or duties performed around the house (e.g., yard work) and errands or favours for other family members (e.g., driving the children to school). 0 1 I I no disability 2 I 3 I 4 I 5 I 6 I 7 I 8 I moderate disability 9 I 10 I total disability Recreation? This category includes hobbies, sports, and other similar leisure time activities. 0 I 1 I 2 I 3 I no disability 4 5 6 I moderate disability I I 7 I 8 I 9 I 10 I total disability Social activity? This category refers to activities which involve participation with friends and acquaintances other than family members. It includes parties, theater, concerts, dining out, and other social functions. 0 1 I I no disability 2 I 3 I 4 5 6 I I I moderate disability 7 I 8 I 9 10 I I total disability Occupation? This category refers to activities that are a part or are directly related to one's job. this includes non-paying jobs as well, such as that of a homemaker or volunteer worker. 0 1 I I no disability 2 I 3 I 4 5 6 I I I moderate disability 7 I 8 I 9 I 10 I total disability 59 Sexual behaviour? This category refers to the frequency and quality of one's sex life. 0 1 2 3 4 5 6 7 8 9 total disability moderate disability no disability 10 Self-care? This category includes activities which involve personal maintenance and independent daily living (e.g ., taking a shower, driving, getting dressed, etc.). 0 1 2 3 4 5 6 7 8 10 total disability moderate disability no disability 9 Life support activity? This category refers to basic life-supporting behaviours such as eating, sleeping, and breathing . 0 1 2 3 4 5 6 7 8 10 total disability moderate disability no disability 9 If you were the physician looking after this patient, how much of the responsibility for the pain do you believe is the patient's? 0 1 2 3 4 5 6 7 8 moderately responsible not responsible 9 10 completely responsible If you were the physician looking after this patient, how much anger would you feel toward him? 0 1 2 3 4 5 6 7 8 9 moderate anger no anger 10 alot of anger If you were the physician looking after this patient, how much sympathy would you feel toward him? 0 1 no sympathy 2 3 4 5 6 moderate sympathy 7 8 9 10 extreme sympathy 60 If you were this patient's physician , how much would you like to extend support (such as spending time talking and listening) to him? 0 not willing 1 2 3 4 5 6 moderately willing 7 8 9 10 completely will ing There are several options available to physicians to treat the kinds of shoulder pain experienced by this patient. Two of the most common are described below. If you were this patient's physician, which treatment would you be likely to prescribe? __ Treatment A: Activation therapy. This therapy addresses the shoulder pain through repetitive exercise. The therapy is delivered by a physiotherapist. The patient is prescribed a 30 minute exercise routine, which is performed twice a week for four weeks. The purpose of the exercises is to promote healing by strengthening the muscles that support the shoulder joint. The exercises themselves are uncomfortable, but the overall effectiveness of the therapy is excellent in most cases . __ Treatment 8: Ultrasound therapy. This therapy addresses the shoulder pain through exposure to ultra-sonic stimulation. An ultrasound device is placed on the affected shoulder and the beam is aimed at the underlying muscular tissue. The therapy is delivered by a physiotherapist in twice-weekly sessions that last a half hour. The purpose of the treatment is to promote healing by reducing inflammation in the shoulder joint. Although the treatment itself is not uncomfortable, it does leave the patient with a sensation of warmth in the shoulder for some time after the sessions. The overall effectiveness of the therapy is excellent in most cases . 61 Appendix C Personal Information Sheet Do not put your name on this sheet. Please answer the questions below. 1. What is your gender? male I female (circle one) 2. What is your age? - 3. Student ld. #(for Subject pool participants only) . Chronic pain experience. Chronic pain is defined as pain which persists beyond the expected period of healing (at least six months) , or is associated with long term, ongoing disease (again , at least six months). For the purpose of this study, the term "having lived with" means within the past five years , and having spent significant amounts of time with this individual. 4. Have you lived with a family member who has experienced (or is experiencing) chronic pain? yes I no (circle one) If you responded "yes" to question 4, please complete the following questions. If you responded with a "no" to question 4, please move on to question 10. 5. Please write down your relationship to the chronic pain sufferer (e.g., father, sister, etc.). 6. Where on the sufferers body are they afflicted with pain (e.g. , back, hands, etc.)? 7. Approximately how long has this family member been afflicted with this pain? _ _ _ _ _ _ 6 months to a year 1-2 years 2-3 years 3-4 years 4-5 years more than 5 years 8. How frequently does this family member experience pain? 1-2 times a month 2-4 times a month 1-2 times a week 2-4 times a week 5-6 times a week _daily _ 2-3 times a day 9. How severe are this family members usual pain experiences? 0 no pain 1 2 3 4 5 6 moderate pain 7 8 9 10 unbearable pain 62 10. What did you think this experiment was examining? 11. Was there anything that influenced your judgements about the pain patient?