Spouse and Patient Beliefs and Perceptions About Chronic Pain: Effects on Couple Interactions and Patient Pain Behavior

The Journal of Pain, Vol 20, No 10 (October), 2019: pp 1176−1186 Available online at www.jpain.org and www.sciencedirect.com

Spouse and Patient Beliefs and Perceptions About Chronic Pain: Effects on Couple Interactions and Patient Pain Behavior John W. Burns, PhD* Kristina M. Post,y David A. Smith,z Laura S. Porter,x Asokumar Buvanendran,{ Anne Marie Fras,x and Francis J. Keefex

* Department of Psychiatry and Behavioral Science, Rush University Medical Center, Chicago, Illinois, yDepartment of Psychology, University of La Verne, La Verne, California, zDepartment of Psychology, University of Notre Dame, Notre Dame, Indiana, x Department of Psychiatry and Behavioral Science, Duke University Medical Center, Durham, North Carolina, {Department of Anesthesiology, Rush University Medical Center, Chicago, Illinois

Abstract: Patient beliefs and perceptions about the causes and meaning of their chronic pain are related to their psychosocial functioning. Beliefs and perceptions about chronic pain held by spouses may also be related to patient functioning. We used a laboratory procedure to evaluate whether spouse beliefs about and perceptions of chronic pain were related to spouse negative responses toward patients with chronic low back pain during a conflictual discussion and to their attributions about patient pain behavior during a subsequent pain-induction task. Patients (n = 71) and their spouses (n = 71) participated in a 10-minute discussion followed by the patient undergoing a 10-minute structured pain behavior task. Findings were that a) spouse perceptions that patient’s pain was a mystery were significantly related to greater patient perceived spouse critical/invalidating responses toward the patient during the discussion; and b) spouse perceptions that patient’s pain was a mystery were related to internal and negative attributions spouses made while observing patients display pain behaviors during the structured pain behavior task. Inasmuch as both spouse critical/ invalidating speech toward patients and negative attributions regarding the cause of patient behavior are related to poor patient functioning, spouse uncertainty about the source and potential legitimacy of their partner’s pain may play crucial roles in affecting patient well-being. Perspective: Spouse beliefs about and perceptions of patient chronic pain were related to spouse behavior toward patients during a discussion and to attributions explaining patient pain during physical activity. If spouse confusion and doubt about patient pain is related to negative behavior and attributions, then modifying these perceptions may be a fundamental intervention target. © Published by Elsevier Inc. on behalf of the American Pain Society Key words: Chronic low back pain, couples, critical/invalidating responses, negative attributions, pain behaviors. atient beliefs and perceptions about the causes and meaning of their chronic pain have been shown to correlate with pain levels, mood, and various domains of psychosocial functioning.17,19,20,31,33,42 Although the vast majority of this research has focused on the impact of patient beliefs and perceptions about chronic pain, the beliefs and perceptions about chronic

P

pain held by family and friends, particularly spouses, may also be related to patient well-being. Couples research by Cano et al11 supports this claim. The influence of spouse beliefs and perceptions could be pronounced when the precise cause of the pain is invisible, unclear, or otherwise ambiguous. Many cases of chronic low back pain (CLBP) fit this description.

Received September 19, 2018; Revised February 28, 2019; Accepted April 2, 2019. Supported by Grant # R01 NR010777 from the National Institute of Nursing Research/NIH. The authors report no conflicts of interest. Supplementary data accompanying this article are available online at www.jpain.org and www.sciencedirect.com.

Address reprint requests to John W. Burns, PhD, Department of Behavioral Sciences, Rush University Medical Center, 1645 W. Jackson Blvd., Chicago, IL 60612. E-mail: [email protected] 1526-5900/$36.00 © Published by Elsevier Inc. on behalf of the American Pain Society https://doi.org/10.1016/j.jpain.2019.04.001

1176

Burns et al Unlike some disease-related pain conditions (eg, cancer pain, rheumatoid arthritis pain), the cause of CLBP is often unclear, and may be poorly understood by both patient and spouse.28,29,36,37 Without a clear explanation, a spouse’s preexisting beliefs about the nature of pain, in general, and his or her perceptions and opinions of the cause and source of the patient’s persistent pain, in particular, can fill the knowledge void. In the present study, we conducted secondary analyses of data reported in Burns et al.7 There, patients with CLBP and their spouses participated in a 10-minute discussion about how patients might improve their ability to cope with pain. This discussion was followed by the patient engaging in a 10-minute structured pain behavior task22 while being observed by the spouse. Here, we examined whether spouse beliefs about the nature of pain in general and perceptions of the patient’s persistent pain in particular are related to spouse behavior toward patients during the marital discussion and to spouse attributions regarding the causes of patient pain behaviors they observe during the pain behavior task. We assessed spouse beliefs that pain in general is due to illness and organic factors and that pain may be influenced by psychological factors (eg, anxiety may worsen pain). The degree to which a spouse believes that the patient’s pain is produced by understandable organic causes that are external to the patient (ie, rooted in physical illness) may contribute to spouse empathy for the patient’s plight.25 Such beliefs may therefore be related to low levels of critical or invalidating responses toward the patient during the marital discussion task and may be related to high levels of relatively compassionate attributions regarding patient pain behavior during the pain behavior task. We also assessed the degree to which the patient’s pain was a mystery to the patient and to the spouse. Thus, in addition to assessing convictions about the nature of pain in general, we measured, in contrast, judgments and perceptions about the nature of the patient’s pain in particular. On 1 level, the spouse’s inability to understand why the patient is in pain could contribute to alarming and catastrophic appraisals of the patient’s well-being and potential to improve. To the degree such perceptions elicit distress in the spouse, findings by Cano et al10 suggest that catastrophic appraisals of patient pain may be related to high levels of critical or invalidating responses during the discussion task. On a second level, high levels of spouse uncertainty about the cause of the patient’s persistent pain could give rise to doubt about the legitimacy of the pain. Over time, providing empathic and supportive responses may become difficult when the cause of the pain is unclear and possibly not viewed by the spouse as stemming from an entirely definitive medical source. Thus, a lack of certainty about the source of the patient’s pain may be related to high levels of critical/invalidating responses toward patients during the discussion task1,5−7,9,10,16,27,38 and negative attributions regarding the cause of patient pain behaviors during the pain behavior task. General beliefs about pain and the degree to which the patient’s pain is a mystery were also assessed among patients. Following past work,19 we expected to find

The Journal of Pain 1177 associations between patient beliefs that pain has roots in organic, medically related causes and patient adjustment factors. Namely, to the extent that patients believe that pain signals damage to tissues, we expected these beliefs to be related to high levels of negative affect, pain intensity, and pain behavior during the pain behavior task.

Methods Participants Seventy-one married couples were recruited through referrals from staff at the pain clinics of Rush University Medical Center in Chicago, Illinois; Duke University Medical Center in Durham, North Carolina; Memorial Hospital in South Bend, Indiana; and through advertisements in local newspapers and flyers provided at various other health care agencies. Each participant received $75. The protocol was approved by the Institutional Review Boards at Rush University Medical Center, Duke University Medical Center, and the University of Notre Dame. Patient inclusion criteria were a) pain of the lower back stemming from degenerative disk disease, spinal stenosis, or disk herniation (radiculopathy subcategory), or muscular or ligamentous strain (chronic myofascial pain subcategory); b) a pain duration of ≥6 months, with an average intensity of at least 3 of 10 (on a scale form 0 [no pain] to 10 [the worst pain possible]); and c) age between 18 and 70 years. The inclusion criterion for patients’ spouses was age between 18 and 70 years. Exclusion criteria for both patients and spouses were a) current alcohol or substance abuse problems, or meeting Diagnostic and Statistical Manual of Mental Disorders-IV criteria for alcohol or substance abuse or dependence (within the past 12 months); b) past or current psychotic or bipolar disorders; c) inability to understand English well enough to complete questionnaires without assistance; d) acute suicidality; and e) meeting criteria for obsessive-compulsive disorder or posttraumatic stress disorder within the past 2 years. A further exclusion criterion for patients was if their pain was due to malignant conditions (eg, cancer, rheumatoid arthritis), migraine or tension headache, fibromyalgia, or complex regional pain syndrome. A further exclusion criterion for spouses was if they reported currently suffering from a condition that caused episodes of acute pain (eg, migraine headaches) or reported a history of chronic pain of their own within the past 12 months. These inclusion and exclusion criteria were assessed using a detailed medical and psychosocial history, including administration of the Mood Disorder, Psychotic Screening, and Substance Use Disorders modules of the Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders-IV axis I disorders—NonPatient Edition.15 The demographic characteristics of the couples are shown in Table 1.

Self-Report Measures Patient and Spouse Pain Beliefs The Pain Beliefs Questionnaire (PBQ)14 was used to measure patient and spouse beliefs about pain in

Spouse and Patient Beliefs and Perceptions About Chronic Pain

1178 The Journal of Pain

Table 1.

Demographic Characteristics

Female sex Age, y Hispanic African American Caucasian Employed Disability insurance Length of marriage, y Pain duration, y

PATIENT

SPOUSE

33 (46.5) 43.94 § 9.8 5 (6.0) 18 (25.4) 51 (71.8) 32 (45.0) 19 (26.7) 14.30 § 14.0 9.04 § 7.80

37 (52.1) 44.61 § 11.3 5 (6.0) 18 (25.4) 51 (71.8) 47 (66.2) 8 (11.3) — —

NOTE. Values are mean § standard deviation or number (%).

general. The organic beliefs subscale of the PBQ tapped beliefs that pain is primarily organic in nature (PBQ-O; eg, “pain is the result of damage to body tissue,” “pain is a sign of illness”), whereas the psychological beliefs subscale of the PBQ tapped beliefs that pain is influenced by psychological factors (PBQ-P; eg, “being anxious makes pain worse,” “thinking about pain makes it worse”). The 2 subscales have been shown to have adequate psychometric qualities.14 In the present data, the Cronbach alpha for the PBQ-O for patients and spouses, respectively, were .72 and .68. The Cronbach alpha for the PBQ-P for patients and spouses, respectively, were .67 and .65.

Patient and Spouse Pain Perceptions We also administered the “my pain is a mystery” subscale of the Pain Beliefs and Perceptions Inventory.45 The PBQ (ad described) taps convictions about the nature of pain in general; that is, “beliefs” about what pain is and what may affect it. With the “my pain is a mystery” subscale, in contrast, we assessed judgments and perceptions about the source of the patient’s pain in particular. The 3-item “my pain is a mystery” subscale was used to tap patient and spouse perceptions about the degree to which the source and persistence of the patient’s pain is unknown (eg, “I can’t figure out why I am in pain”). The subscales of this inventory have been shown to have adequate psychometric characteristics.43 The 3 items were modified for the spouse to refer to the patient’s pain (eg, “I can’t figure why my spouse is in pain”). For the spouse, this subscale is hereafter referred to as the “(my spouse’s) pain is a mystery” subscale. The items were not modified for patients, and remain focused on the patient’s perception of his or her pain (eg, “I can’t figure why I am in pain”). In the present data, Cronbach’s alpha for these items for patients and spouses, respectively, were .87 and .86.

Affect Checklists The Emotional Assessment Scale (EAS)12—a standardized measure of emotional state—was used to assess changes in patient anger, anxiety, and sadness aroused during the discussion task and structured pain behavior task (SPBT). Rating scales were used to measure these negative affects rated at baseline and immediately after

the discussion and SPBT (from 0 [none] to 10 [very much]).

Patient Perceived Critical or Invalidating Responses During the Discussion Task Patients reported on the degree of negative responses they perceived from their spouse during the discussion task using 4 items. These were, “My spouse was critical of me,” “My spouse acted and spoke with hostility toward me,” “My spouse blamed me for the problem we discussed,” and “My spouse kept telling me what he/ she thinks I should do.” They rated these items on a scale ranging from 0 [strongly disagree] to 10 [strongly agree]. The Cronbach alpha for these 4 items was .87. Thus, we summed these items to form a composite, hereafter referred to as patient perceived critical or invalidating spouse responses.

Pain Intensity Patients rated their pain intensity at baseline and after the SPBT on a 0 (none) to 10 (extremely) scale.

Attributions for Patient Pain Behaviors During the Structured Pain Behavior Task Spouses reported on the degree to which they attributed patient pain behavior during the pain behavior task to five causes. These items were adapted from the work of Williamson et al,46 who advanced the notion of “external” and “internal” attributions regarding illness behavior. They hypothesized that unpleasant and disturbing patient behavior (eg, grimacing and groaning) may be attributed by caregivers to the patient’s medical condition, and thus may be seen by spouses as beyond the patient’s control. These external attributions would be associated with low levels of caregiver resentment and other negative responses. In contrast, unpleasant and disturbing patient behavior may be attributed by caregivers to the patient’s efforts to manipulate and control the caregiver, to foment guilty feelings, and so forth, and thus may be seen by the spouse as within the patient’s control. These internal attributions would be associated with high levels of resentment and other negative responses. Thus, external attributions for unpleasant patient behavior (ie, ascribing behavior to the illness) are conceptualized to be distinct from internal attributions (ie, ascribing behavior to patient attempts to manipulate and control). In the present study, the items tapping spouse attributions were, “He/ she acts that way because of the pain his/her medical condition causes,” “He/she acts that way because the pain really gets him/her ‘down,’” “He/she acts that way because he/she gives in to the pain too easily,” “He/she acts that way to make me feel bad for him/her,” and “He/she acts that way to get out of doing something.” Spouses indicated their level of agreement from 0 (strongly disagree) to 10 (strongly agree). We wrote these items expressly to tap internal and external attributions. A principal components analysis with varimax

Burns et al rotation revealed 2 factors that accounted for 76.86% of the total variance. The items “He/she acts that way because he/she gives in to the pain too easily,” “He/she acts that way to make me feel bad for him/her,” and “He/she acts that way to get out of doing something” loaded on the first factor. The Cronbach alpha for these items was .82. We summed these 3 items to form the internal attribution scale. The items “He/she acts that way because of the pain his/her medical condition causes,” and “He/she acts that way because the pain really gets him/her ‘down’” loaded on the second factor. The Cronbach alpha for these 2 items was .73, and so we summed them to form the external attribution scale.

Patient and Spouse Distress Symptoms Self-reported depressive symptoms for the patient and spouse were measured with the Beck Depression Inventory-II (BDI-II).2 The psychometric quality of the BDI-II has been extensively reported.3 The Spielberger Trait Anxiety Inventory (TAI; 43) is a well-validated measure of trait anxiety.40 The BDI-II and TAI were used as proxy measures of spouse distress levels to test the hypothesis that general spouse distress may be related to greater criticism/invalidation of the patient. These scales and the Cook-Medley Hostility Scale (CMHS), described elsewhere in Patient and Spouse Trait Hostility, were also administered to allow us to examine the unique relationships between pain beliefs and dependent variables, controlling for the variance that may also be accounted for by depressive symptoms, anxiety, hostility, and marital satisfaction.

Patient and Spouse Trait Hostility The CMHS13 was used to assess the extent to which spouses endorsed cynical and mistrustful beliefs about others and were prone to anger or to getting annoyed by others’ behaviors. The scale has demonstrated adequate psychometric qualities.13

Patient and Spouse Marital Satisfaction The Dyadic Adjustment Scale (DAS)39 was used to measure patient and spouse’s perceived marital adjustment. Scores range from 0 to 151, with higher scores indicating greater marital adjustment. Reports indicate that the scale shows acceptable levels of reliability and validity.21

Marital Discussion Procedure Couples participated in a laboratory discussion task that was adapted from a protocol that has been used successfully in studies of social support, marital discord, and depression.34,35 To identify a topic for this discussion, patients and spouses independently listed 5 things they would like to see their partners change about the way they responded to, or coped with, the patients’ chronic pain. These 5 items were listed by each patient and spouse

The Journal of Pain 1179 in order of perceived importance. Items were further rated by each participant on a scale of 1 to 10 for the degree to which each participant believed the other person in the couple disagreed with their views about the topic. As they generated items, participants were reminded that one of the topics would be chosen for a subsequent video recorded discussion with their spouse. Participants were also told that the target (namely, patient or spouse) for the discussion would be randomly selected by a coin flip. However, spouse-nominated topics pertaining to patient change were always selected so that patients were always the target of the discussion about making changes. From the spouse’s list, the experimenter then identified the item that a) was ranked highest in importance and b) received the highest disagreement rating. That is, couples discussed something about the way the patient handled their chronic pain that was important to the nonpatient spouse but was not listed as something the patient necessarily wanted to change. This unbalanced topic was chosen to provide ample opportunity for hostile and critical spouse comments to occur during the laboratory discussion task. Participants were instructed to discuss the identified topic for 10 minutes, with the spouse receiving specific instructions to initiate the conversation. During discussion, participants were seated such that the cameras captured facial expressions and nonverbal behaviors yet still allowed the participants to face each other. The angle and distance between the chairs was the same for all participants.

Observational Measures of Discussion Task We limited our assessment of marital interaction behavior during the discussion task to criticism and hostility. To measure criticism, we adopted the coding system of Peterson and Smith.35 This system has been used in healthy and depressed samples reliably (intraclass correlation coefficient = .8). Definitions of criticism from existing coding systems24,26,43 are integrated in this system, and various behavioral indicators of criticism identified in prior studies were trained to permit coders to rate reliably. Specifically, we defined spouse criticism as expressions of dislike, disapproval, or resentment of the other person’s behavior. To assess hostility, we followed Vaughn and Leff.43 Briefly, spouse hostility is a global rating defined by a) critical generalizations (ie, critical comments about character or personality rather than about specific behaviors), and/or b) rejecting remarks. Thus, we coded spouse criticism and hostility separately, chiefly because of the distinction between comments regarding the other person’s behavior versus comments regarding the other person’s character. Coders rated criticism and hostility on separate 4-point scales (0 [no criticism or hostility]; 1 [low intensity and duration of criticism or hostility]; 2 [high intensity or high duration criticism or hostility]; and 3 [high intensity and high duration criticism or hostility]).

Observational Coding Coders were trained on the marital interaction coding system at the University of Notre Dame’s Marital

1180 The Journal of Pain

Therapy and Research Clinic. Training of coders included an initial education in the definitions of criticism and hostility as culled from the observational coding literature.24,26,43 Familiarity with the constructs and initial evidence of rater agreement was then established via group coding of sample (nonstudy) tapes, with independent ratings subsequently discussed aloud to help clarify definitions and to resolve discrepancies. Each 10-minute interaction was coded in ten 60-second intervals. Three coders viewed each interaction, rating spouse criticism and hostility following each 60-second interval. Seventy-one interactions were coded in this manner. Coders were blind to specific study hypotheses. Ratings were then aggregated separately across the ten minutes of the interactions. Inter-rater agreement was good, with intraclass correlation coefficients (3, 3) ranging from .78 to .88 across the patient and spouse criticism and hostility variables.

Structured Pain Behavior Task A structured pain behavior task22 was used as a naturalistic pain induction manipulation that allowed for the assessment of both self-reported pain intensity and observable pain behaviors (eg, grimacing). Previous reports indicate that the pain behavior task produces significant increases in pain intensity among people with CLBP.8 This task involved sitting, standing, walking, reclining, bending, and stretching to lift light-weight objects (see Procedure); these everyday activities typically produce mild low back pain in patients with CLBP. The spouse watched as the patient performed the pain behavior task activities, which then provided a basis for their later responding to items tapping their attributions for what they observed. The task was presented to both patient and spouse as providing the investigators with “a sample of certain kinds of everyday activities” to help them “get a sense” of the couple’s daily life.

Procedure The task was conducted in a room with an examination table, a chair without arms, and a shelf with waistand shoulder-high shelves (on which a box with weights sat). The procedure as described in Keefe and Block22 was followed with some exceptions. Participants were asked by the tester to engage in two 15- and 45-second sitting and standing periods, three 30-second reclining periods, and two 60-second walking periods. We added 2 separate bending and lifting sequences, in which subjects lifted a covered box containing either 2.5 kilos of weight (for women) or 5 kilos of weight (for men). In the first sequence, participants lifted the box from the floor, stood erect, placed it on the waist-high shelf, and then replaced the box on the floor. Participants repeated this procedure once more. In the second sequence, participants lifted the box from the waisthigh shelf and placed it on the shoulder-high shelf and then replaced the box on the waist-high shelf. They repeated this procedure once more. The order of positions and activities was varied randomly across participants. The tester communicated with participants only

Spouse and Patient Beliefs and Perceptions About Chronic Pain to request position and activity changes. The 12-minute sessions were video recorded by the tester. The spouse observed the SPBT while seated in a chair about 3 meters from the patient.

Observational Measures of Pain Behaviors Two groups of behaviors were coded from the pain behavior task. The first group consists of mutually exclusive behaviors.22 Thus, whether the participant is sitting, standing, bending, stretching, or walking is coded, and whether he or she is shifting to assume another position is coded. The second group consists of motor pain behaviors, which can occur during the mutually exclusive behaviors. These behaviors are guarding, bracing, rubbing, grimacing, and sighing. Two raters rated all 71 protocols and achieved an inter-rater reliability for the mutually exclusive behaviors of r = .96, and for the 5 motor pain behaviors a range from r = .88 to r = .96.

General Laboratory Session Procedure Patients and spouses were instructed to refrain from consuming alcohol 12 hours before the session, and patients were further instructed to refrain from consuming caffeine and pain medications 3 hours before the session. Upon arrival at the laboratory, patients and spouses read and signed informed consent forms, and the procedures and measures were explained. Patients and spouses completed the PBQ, pain is a mystery subscale, the BDI-II, TAI, CMHS, and the DAS. Patients also completed the EAS. All participants then compiled their lists of discussion topics and ratings. The experimenter then selected topics for discussion, informed participants of the topic, and then instructed them to discuss the topic for 10 minutes. Participants were encouraged to stay on topic for the full 10 minutes. After the discussion, patients completed the EAS, and the items assessing spouse perceived criticism, hostility, and other negative behaviors during the discussion. The SPBT procedure then commenced. The spouse sat in a chair 3 meters from the patient with a full view of him or her and was instructed to remain silent and still during the SPBT. The patient was quickly rebriefed on the movements required during the task. The experimenter operated the video camera and instructed the patient on when to begin each movement. The transitions between movements were timed to ensure that they were captured during the coding intervals. After the SPBT, patients rated pain intensity and completed an EAS. Spouses completed the attribution items. The couples were then thoroughly debriefed at the end of this laboratory session.

Statistical Analyses Between-subjects analyses of variance were used to compare spouses and patients on scores on the pain beliefs subscales and other questionnaires. Zero-order correlations were generated among categories of responses reported by patients and spouses during the discussion task and SBPT, and the PBQ-O, PBQ-P, and

Burns et al

The Journal of Pain 1181 were less likely to hold the belief that pain is influenced by psychological factors.

Table 2. STUDY MEASURES PBQ-O PBQ-P Pain is a mystery Patient perceived negative spouse responses Coded spouse criticism Coded spouse hostility Spouse internal attributions Spouse external attributions BDI-II TAI DAS CMHS

PATIENT

SPOUSE

32.10 § 6.49 16.96 § 3.47 8.31 § 4.08 19.83 § 10.18

31.34 § 5.55 18.27 § 3.05 8.38 § 3.88 —

— — — — 15.95 § 9.42 19.94 § 11.18 99.97 § 22.66 11.81 § 4.90

28.25 § 11.83 6.42 § 8.02 6.71 § 6.55 13.56 § 5.14 10.28 § 9.91 16.06 § 10.20 101.75 § 21.59 10.88 § 4.01

NOTE. Patient perceived negative responses are a composite of patient perceived spouse negative response items. Spouse internal attributions are a composite of spouse internal attribution items and spouse external attributions are a composite of spouse external attributions. Values are mean § standard deviation.

the pain is a mystery subscale. To determine whether pain beliefs and perceptions contributed unique variance to the prediction of the categories of responses reported by patients and spouses during the discussion task and SBPT, regressions were performed to control for spouse or patient BDI-II, CMHS, and/or DAS scores.

Results Patient and Spouse Differences on SelfReport Measures Table 2 provides the means and standard deviations of study measures. Between-subject analyses of variance were used to test whether patients and spouses differed on the pain beliefs measures, the BDI, CMHO, and DAS. Patients had lower scores on the PBQ-P than spouses, F(1,70) = 7.95, P < .006. Patients reported more depressive symptoms on the BDI-II than spouses, F(1,70) = 14.41, P < .001. Patients reported greater trait anxiety than spouses, F(1,70) = 6.00, P < .01. Patients and spouses did not differ significantly on the PBQ-O, the pain is a mystery subscale, the CMHO, and the DAS (F’s ranged from .02 to 1.61). Thus, patients were more likely than spouses to experience depressive symptoms, but they

Relationships Among Spouse Pain Beliefs, Spouse Distress, and Other Spouse, Patient, and Observer Ratings During the Discussion Task Zero-order correlations were computed among spouse beliefs and perceptions, spouse depressive symptoms, trait anxiety, patient-reported spouse critical or invalidating responses, changes from baseline in patient-reported negative affects, and observer-rated spouse criticism or hostility (Table 3). The most consistent findings were that high scores on the subscale (my spouse’s) pain is a mystery were significantly related to greater patient perceived spouse critical or invalidating responses and increases from baseline in patient anger and anxiety during the discussion. High scores by spouses on the PBQ-O were related significantly only to patient increases in sadness. Spouse beliefs and perceptions were not significantly related to observer rated spouse criticism and hostility. Zero-order correlations were generated among spouse beliefs and perceptions and spouse scores on the BDI, TAI, CMHO, and DAS. The PBQ-O and PBQ-P were not related significantly to spouse trait measures (rs = -.05 to .13). However, the (my spouse’s) pain is a mystery subscale was related significantly to spouse BDI (r = .26; P < .03) and CMHS scores (r = .25; P < .04). To determine whether spouse pain perceptions exerted unique effects on patient negative affect and ratings of critical or invalidating responses during the discussion task, we ran a series of hierarchical regressions with spouse BDI and CMHS scores entered first, and spouse scores on the (my spouse’s) pain is a mystery entered second. For patient anger change, the (my spouse’s) pain is a mystery subscale remained a significant predictor (beta = .29; P < .03). For patient anxiety change, the (my spouse’s) pain is a mystery subscale also remained a significant predictor (beta = .26; P < .04). For patient perceived critical or invalidating, the (my spouse’s) pain is a mystery subscale also remained a significant predictor (beta = .32; P < .01). Thus, spouse judgments that the patient’s pain is a mystery to him or her were related to patient increases in negative affect and patient perceived critical/invalidating behavior during a marital

Zero-Order Correlations Among Spouse Pain Beliefs, Depressive Symptoms, Trait Anxiety, and Phenomena During the Discussion Task

Table 3.

PBQ-O PBP-P Pain is a mystery BDI-II TAI

PATIENT ANGER

PATIENT ANXIETY

PATIENT SADNESS

PERCEIVED C/I RESP

CODED CRITICISM

CODED HOSTILITY

.08 .06 .29* .08 .16

.05 .01 .24* .12 .14

.24* .08 .01 .13 .03

.16 .21 .33* .10 .15

¡.11 ¡.07 .10 .08 .09

¡.16 ¡.03 .17 .06 .06

TAI, Spielberger Tait Anxiety Inventory. NOTE. Patient anger is change in anger from baseline. Patient anxiety is the change in anxiety from baseline. Patient sadness is the change in sadness from baseline. Perceived C/I Resp is a composite of patient perceived spouse critical or invalidating response items. *P < .05.

Spouse and Patient Beliefs and Perceptions About Chronic Pain

1182 The Journal of Pain

Zero-Order Correlations Among Spouse Pain Beliefs and Attributions During the SPBT

Table 4.

PBQ-O PBQ-P Pain is a mystery

INTERNAL ATTRIBUTIONS

EXTERNAL ATTRIBUTIONS

.22 .12 .47*

.31* .10 .11

NOTE. Internal attributions are a composite of spouse internal attribution items. External attributions are a composite of spouse external attributions. *P < .05.

discussion task over and above other spouse characteristics that could have accounted for these relationships.

Relationships Among Spouse Pain Beliefs and Perceptions and Attributions during the SPBT Zero-order correlations were generated among spouse beliefs and perceptions, the spouse internal attributions scale, and the spouse external attributions scale (Table 4). Of note, scores on the (my spouse’s) pain is a mystery subscale were related significantly to the spouse internal attributions scale but not to the external attributions scales. In contrast, scores on the PBQ-O were related significantly to the spouse external attributions scale, but not to the internal attributions scale. Spouses with high scores on the (my spouse’s) pain is a mystery subscale tended to have high levels of internal and frankly negative attributions (eg, “He/she acts that way to make me feel bad for him/her”) regarding the causes of patient pain behaviors they observed during the SPBT. Spouses with high scores on the pain is organic subscale tended to have high levels of external attributions (eg, “He/she acts that way because of the pain his/ her medical condition causes”) regarding the causes of patient pain behaviors they observed during the SPBT. To determine whether spouse pain beliefs and perceptions exerted unique effects on spouse attributions during the SPBT, we ran a series of hierarchical regressions with BDI and CMHS scores entered first, and the attribution variable entered second. For the spouse external attribution scale, the spouse PBQ-O remained a significant predictor (beta = .31; P < .01). For the spouse internal attribution scale, scores on the (my spouse’s) pain is a mystery subscale remained a significant predictor (beta = .40; P < .001). Thus, spouse beliefs and perceptions of chronic pain were related to their attributions for patient pain behaviors they witnessed

Table 5.

during the SPBT over and above other spouse characteristics that could have accounted for these relationships.

Relationships Among Patient Pain Beliefs and Phenomena during the SPBT Zero-order correlations were generated among patient pain beliefs and perceptions, anger, anxiety, and sadness changes during the SPBT, and pain intensity changes and pain behaviors (Table 5). The most consistent findings were that patient scores on the PBQ-O were related significantly to patient anger and pain intensity increases during the SPBT and to the frequency of observed pain behaviors. Patient scores on the PBQ-P and my pain is a mystery to me subscales were not related significantly to any of the other measures. Patient scores on the PBQ-O were related significantly to patient scores on the BDI-II (r = .31; P < .01) and CMHS (r = .25; P < .03). To determine whether patient organic pain beliefs exerted unique effects on negative affect and pain during the SPBT, we ran a series of hierarchical regressions with patient BDI and CMHS scores entered first and the target variables entered second. For patient increases in anger, the patient PBQ-O did not remain a significant predictor (beta = .09; P = NS). For pain intensity changes, scores on the patient PBQ-O remained a significant predictor (beta = .27; P < .03). For frequency of pain behaviors, scores on the patient PBQ-O again remained a significant predictor (beta = .28; P < .03). Thus, patient beliefs of the organic nature of pain were related to pain intensity and pain behaviors during the SPBT over and above other patient characteristics that could have accounted for these relationships.

Discussion Patient beliefs and perceptions of the causes and meaning of chronic pain have been shown to correlate with patient pain levels, mood, and function.17,19,20,31,33,42 We expanded this past work to also examine the relationships among spouses’ beliefs and perceptions about chronic pain, their responses to patients as they discussed how the patient coped with pain, and their attributions for patient pain behavior they observed during a naturalistic pain induction task (ie, the pain behavior task). Results suggest that spouse judgments that their partner’s pain is a mystery may underlie not only critical and invalidating

Zero-Order Correlations Among Patient Pain Beliefs and Phenomena During the SPBT

PBQ-O PBQ-P Pain is a mystery

PATIENT ANGER

PATIENT ANXIETY

PATIENT SADNESS

PAIN INTENSITY

PAIN BEHAVIORS

.25* .05 .17

.18 .23 .21

.21 .19 .19

.29* .12 ¡.10

.38* .11 .11

NOTE. Patient anger is change in anger from baseline. Patient anxiety is change in anxiety from baseline. Patient sadness is change in sadness from baseline. Pain intensity is change in pain intensity from baseline. *P < .05.

Burns et al responses toward the patient when speaking about the pain condition, but also internal and negative attributions regarding the causes and purposes of patient pain behavior. Given that spouse criticism and hostility are related to increased pain and decreased function among people with chronic pain,6 and that negative attributions seem to be related to spouse criticism and hostility,7 our findings point strongly to the notion that spouse uncertainty about the source and potential legitimacy of their partner’s pain may play a crucial role in affecting patient well-being. Our key findings highlight the importance of a specific spouse appraisal of their partner’s chronic pain in contrast with spouses’ general beliefs about the nature of chronic pain. Namely, the more the spouse judged the cause and source of patient pain to be a mystery, the more likely the patient was to perceive that the spouse behaved negatively toward them during the discussion task. Recall that the discussion topic was shaped around how the spouse wanted the patient to change the way he or she coped with the chronic pain condition. Spouse judgments that the patient’s pain was confusing and stemming from an unknown source was related to patient ratings of spouse criticism and invalidation. This spouse judgment was also related to increases in patient anger and anxiety during the discussion. On 1 level, as suggested by results of Cano et al,10 spouse distress about the patient’s pain may, paradoxically, give rise to invalidation of patient’s pain disclosures. Although we did not assess spouse pain catastrophizing per se, we did assess spouse depressive symptoms and trait anxiety, which could stand as proxy measures of distress. Spouse depressive symptoms and trait anxiety were not significantly correlated with perceived spouse critical or invalidating remarks during the discussion. On another level, spouse frustration and resentment stemming from having to live with and support someone whose pain they are confused about and may doubt is completely rooted in a medical condition may contribute to spouse efforts to convince the patient to cope more effectively, efforts that may, over time and repetition, come to sound more and more critical and hostile to the patient. However, as we have demonstrated,6,7 spouse criticism and hostility regarding patient coping attempts has negative effects on later patient pain and mood, especially among vulnerable groups (eg, depressed patients). Far from inspiring the patient to change for the better, criticism and hostility may well worsen the patient’s lot. Spouse perceptions that the patient’s pain is a mystery may underlie negative spouse behaviors that may ironically increase patient pain behavior. Spouse appraisals that the patient’s pain is a mystery were also related to the spouse making internal attributions regarding the reasons why the patient displayed pain behaviors during the pain behavior task. Recall that the pain behavior task was intended to elicit discrete pain behaviors (eg, grimacing, groaning, bracing) in response to physical activity (eg, reclining, lifting objects, walking). Ascribing internal attributions for the pain behaviors, per Williamson et al,46 suggests that the spouse interpreted the patient pain behaviors as at least partly under the patient’s control. When the spouse does not understand the source of the patient’s pain, the spouse may think

The Journal of Pain 1183 that it is partly under the patient’s control given that they cannot find or understand other explanations. Moreover, the internal attributions assessed in this study were negative, conveying spouse views that patient grimaces, groans, and bracing were at least partly due to the patient trying to manipulate the spouse or make them feel sorry about the patient’s suffering. We administered similar internal attribution items in a diary study.4 Note that the diary study was a naturalistic study in which we did not supply standardized stimuli designed to stimulate pain behaviors. Still, we found that the relationship between spouse observations of patient pain behaviors at, for example 9 AM and high levels of spouse criticism or hostility at, for example, 12 PM was due to spouse’s making internal and negative attributions regarding the reasons patients showed pain behaviors. Our present results expand these previous findings and imply that 1 basis of negative attributions may be that the spouse does not understand the causes of the patient’s chronic pain condition and is confused by the actual pain behaviors unfolding before them. Or, alternatively, the spouse suspects that the pain behaviors they witness are not actually stemming completely from a medical condition. Taking results of the 2 studies together, we can postulate a vicious cycle wherein spouse appraisals that patient pain is a mystery contribute to spouse negative responses to patients as they attempt to cope with pain, negative attributions regarding patient behavior, more negative responses from spouses about patient efforts to cope, and finally more ambiguous or confusing patient pain behaviors. We also found that spouse beliefs that pain is, in general, an organic, medical condition (eg, “pain is a result of damage to the tissues of the body”) were related to spouse external attributions regarding the causes of patient pain behaviors during the pain behavior task. Spouse organic beliefs were related to attributions that the patient pain behaviors they actually witnessed during the pain induction task were largely outside patients’ control and were due to the pain condition. Although these neutral if not compassionate explanations for the behaviors the spouse saw may avoid detrimental effects linked to negative internal attributions, external attributions rooted in beliefs that pain is a sign of physical illness may not necessarily prove to have beneficial effects. Cognitive, emotional, and behavioral factors clearly affect the pain and function of people with chronic pain—factors that are to some extent under the control of both patients and spouses. Closely held beliefs that pain is nearly exclusively a medical condition could preclude spouses from encouraging patients to use any kind of coping strategy other than seeking medical treatment for a medical condition. Other findings suggest that the patient’s general belief that pain is rooted in organic causes and signals physical injury were related to greater pain intensity and frequency of pain behaviors during the SPBT. Items of the PBQ-O subscale are conceptually similar to items included in subscales of the Survey of Pain Attitudes,18 especially regarding the notions that pain signals

1184 The Journal of Pain

physical harm and is largely a medical condition (ie, harm beliefs subscale). Jensen et al19,20,42 found that this subscale was related to elevated pain intensity and depressive symptoms. Further, Jensen et al19 reported that scores on the harm beliefs subscale were related to observer-rated nonverbal pain behavior during a laboratory activity involving household chores. The present findings are consistent with this past work and extend these findings to include patient reports of increased pain intensity and anger during a pain induction task. Beliefs that pain is a result of tissue damage and that the amount of pain is related to the amount of damage may predispose people to be especially susceptible to painful stimuli, magnifying their negative responses to pain, revealed both through observable behavior and changes in negative affect. Our findings have a number of clinical implications. First, broadly speaking, these and previous findings regarding spouse beliefs and behavior toward their partners with chronic pain emphasize the importance of the social component of a psychosocial model of chronic pain.23,26 Thus, the inclusion of spouses in clinical interventions for chronic pain patients seems to be vital. Second, based on our previous results showing the detrimental effects of spouse criticism and hostility on patient pain and function, we advocated for couple communication training to reduce such negative interactions.6,7 Here, we identify possible antecedents of spouse criticism and hostility and negative attributions, which may require slightly different treatment approaches. To the degree that spouses are mystified by the causes of patients’ chronic pain, providing accurate information about the nature, determinants, and course of chronic pain may allay spouses’ confusion and doubts. Such information may also increase both patient and spouse understanding of the complex interplay of cognitive, emotional, and behavioral factors in ways that expand on otherwise overly simple organic views of pain from which less helpful responses emerge. Pain education interventions to address the uncertainty that patients experience regarding chronic pain already exist and show efficacy.32,41 Other investigators have extended the provision of pain education by including spouses, with the effects exceeding those shown by patient pain education alone.30 If confusion and doubt about the patient’s pain do indeed precede and underlie later criticism, hostility, and negative attributions, then modifying these perceptions would seem to be a fundamental target for intervention. Some limitations should be delineated. First, our assessment of patient and spouse beliefs and perceptions regarding pain was not exhaustive. We did not, for instance administer the Survey of Pain Attitudes,

References 1. Alschuler KN, Otis JD: Significant others’ responses to pain in veterans with chronic pain and clinical levels of post-traumatic stress disorder symptomatology. Eur J Pain 17:245-254, 2013

Spouse and Patient Beliefs and Perceptions About Chronic Pain which would have allowed us insights into a wide variety of both patient and spouse pain attitudes. Second, we modified only the pain is a mystery subscale of the PBPI to assess spouse perceptions of the patient’s pain. The items of the organic and psychological beliefs subscales reflected patient and spouse beliefs about pain, but only from their own perspectives. Given the breadth of our findings for the (my spouse’s) pain is a mystery subscale, future research may need to expand the assessment of spouse beliefs, attitudes, and judgments about the cause and source of patient pain. Third, although the PBQ-O, PBQ-P, and pain is a mystery subscales were administered before the laboratory session, we cannot definitively conclude that pain beliefs and perceptions caused greater spouse levels of criticism, hostility, and negative attributions. Fourth, we chose to study people with CLBP because, unlike disease-related pain conditions (eg, cancer pain, rheumatoid arthritis pain), the cause of CLBP is often unclear to medical specialists and poorly understood by both patients and spouses. These factors may underlie spouse mystification and their lack of clarity about the basis and nature of patient pain complaints and behaviors. Thus, our findings regarding links between spouse judgments that the patient’s pain is a mystery and negative responses and negative attributions may not generalize to chronic conditions in which there is a clearer diagnosis that better accounts for—in the spouse’s mind—pain complaints and reduced function (eg, cancer pain).

Conclusions Our results contribute to understanding why spouses may direct criticism and hostility toward their partners with chronic pain and make negative attributions regarding the causes of patient pain behavior. A vital piece in understanding the genesis of spouse negative responses to patients may be their confusion over, and lack of knowledge and understanding about, the causes and meaning of the patient’s chronic pain. If the pain is perceived as a mystery without readily identifiable causes, spouses may fall prey to attribution biases, and ascribe most of the patients’ undesirable and unpleasant behaviors to motivational, dispositional, or other internal causes, minimizing situational explanations.44 Frustration, resentment, and criticism and hostility may all follow, contributing to poor patient well-being.

Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jpain.2019.04.001. 2. Beck AT, Steer RA, Brown GK: Manual for Beck Depression Inventory − II. San Antonio, TX, Psychological Corporation, 1996 3. Beck AT, Steer RA, Garbin MG: Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clin Psychol Rev 8:77-100, 1988

Burns et al 4. Burns JW, Gerhart J, Post KM, Smith DA, Porter LS, Buvanendran A, Fras AM, Keefe FJ: Spouse criticism/hostility toward partners with chronic pain: The role of spouse attributions for patient control over pain behaviors. J Pain 19:1308-1317, 2018 5. Burns JW, Johnson BJ, Mahoney N, Devine J, Pawl R: Anger management style, hostility, and spouse responses: Gender differences in predictors of adjustment among chronic pain patients. Pain 64:445-453, 1996 6. Burns JW, Peterson KM, Smith DA, Keefe FJ, Porter L, Schuster E, Kinner E: Temporal associations between spouse criticism/hostility and pain among patients with chronic pain: A within-couple daily diary study. Pain 154:2715-2721, 2013 7. Burns JW, Post KM, Smith DA, Porter LS, Buvanendran A, Fras AM, Keefe FJ: Spouse criticism and hostility during marital interaction: Effects on pain intensity and behaviors among individuals with chronic low back pain. Pain 159:25-32, 2018 8. Burns JW, Quartana P, Gilliam W, Gray E, Matsuura J, Nappi C, Wolfe B, Lofland K: Effects of anger suppression on pain severity and pain behaviors among chronic pain patients: Evaluation of an ironic process model. Health Psychol 27:645-652, 2008 9. Cano A, Barterian JA, Heller JB: Empathic and nonempathic interaction in chronic pain couples. Clin J Pain 24:678-684, 2008 10. Cano A, Leong LE, Williams AM, May DK, Lutz JR: Correlates and consequences of the disclosure of pain-related distress to one’s spouse. Pain 153:2441-2447, 2012 11. Cano A, Miller LR, Loree A: Spouse beliefs about partner chronic pain. J Pain 10:486-492, 2009 12. Carlson CR, Collins FL, Stewart JF, Porzelius J, Nitz A, Lind: The assessment of emotional reactivity: A scale development and validation study. J Psychopathol Behav Assess 11:313-325, 1989 13. Cook WW, Medley DM: Proposed hostility and pharisaicvirtue scales for the MMPI. J Appl Psychol 194;38:414-418 14. Edwards LC, Pearcea SA, Turner-Stokes L, Jones A: The Pain Beliefs Questionnaire: An investigation of beliefs in the causes and consequences of pain. Pain 51:267-272, 1992 15. First MB, Spitzer RL, Gibbon M, Williams JBW: Structured clinical interview for DSM-IV Axis I Disorders-Nonpatient edition (SCID-I/NP, Version 2.0). New York, NY, Biometrics Research Department, 1996 16. Grant LD, Long BC, Williams JD: Women’s adaptation to chronic back pain: Daily appraisals and coping strategies, personal characteristics and perceived spousal responses. J Health Psychol 7:545-563, 2000 17. Hanley MA, Raichle K, Jensen M, Cardenas DD: Pain catastrophizing and beliefs predict changes in pain interference and psychological functioning in persons with spinal cord injury. J Pain 9:863-871, 2008 18. Jensen MP, Karoly P, Huger R: The development and preliminary validation of an instrument to assess patients’ attitudes toward pain. J Psychosom Res 31:393-400, 1987 19. Jensen MP, Romano JM, Turner JA, Good AB, Wald LH: Patient beliefs predict patient functioning: Further support for a cognitive-behavioural model of chronic pain. Pain 81:95-104, 1999

The Journal of Pain 1185
-Pires C, de la Vega R, Gala
n S, Sole
E, 20. Jensen MP, Tome
J: What determines whether a pain is rated as mild, Miro moderate, or severe? The importance of pain beliefs and pain interference. Clin J Pain 33:414-421, 2017 21. Kazak AE, Jarmas A, Snitzer L: The assessment of marital satisfaction: An evaluation of the Dyadic Adjustment Scale. J Family Psychol 2:82-91, 1988 22. Keefe FJ, Block AR: Development of an observation method for assessing pain behavior in chronic low back pain patients. Behav Ther 13:363-375, 1982 23. Keefe FJ, Blumenthal J, Baucom D, Affleck G, Waugh R, Caldwell DS, Beaupre P, Kashikar-Zuck S, Wright K, Egert J, Lefebvre J: Effects of spouse-assisted coping skills training and exercise training in patients with osteoarthritic knee pain: A randomized controlled study. Pain 110:539-549, 2004 24. Leff J, Vaughn C: Expressed Emotion in Families. Its Significance for Mental Illness. London, Guilford Press, 1985 25. Leonard MT, Issner JH, Cano A, Williams AM: Correlates of spousal empathic accuracy for pain-related thoughts and feelings. Clin J Pain 29:324-333, 2013 26. Li Q, Loke AY: A systematic review of spousal couplebased intervention studies for couples coping with cancer: Direction for the development of interventions. Psychooncology 23:731-739, 2014 ~ a AB, Goldstein JM, Karno M, Miklowitz DJ, Jen27. Magan kins J, Falloon IR: A brief method for assessing expressed emotion in relatives of psychiatric patients. Psychiatr Res 17:203-212, 1986 28. Manne SL, Zautra AJ: Spouse criticism and support: Their association with coping and psychological adjustment among women with rheumatoid arthritis. J Personality Soc Psychol 56:608-617, 1989 29. Martire LM, Keefe FJ, Schulz R: Older spouses’ perceptions of partners’ chronic arthritis pain: Implications for spousal responses, support provision, and caregiving experiences. Psychol Aging 21:222-230, 2006 30. Martire LM, Schulz R, Keefe FJ, Rudy TE, Starz TW: Couple-oriented education and support intervention for osteoarthritis: Effects on spouses’ support and responses to patient pain. Family Syst Health 26:185-195, 2008
J, Huguet A, Jensen MP: Pain beliefs predict pain 31. Miro intensity and pain status in children: Usefulness of the pediatric version of the survey of pain attitudes. Pain Med 15:887-889, 2014 32. Moseley GL, Nicholas MK, Hodges PW: A randomized controlled trial of intensive neurophysiology education in chronic low back pain. Clin J Pain 20:324-330, 2004
J: Changes in pain33. Nieto R, Raichle KA, Jensen MP, Miro related beliefs, coping, and catastrophizing predict changes in pain intensity, pain interference, and psychological functioning in individuals with myotonic muscular dystrophy and facioscapulohumeral dystrophy. Clin J Pain 28:47-54, 2008 34. Pasch L A, Bradbury TN: Social support, conflict, and the development of marital dysfunction. J Consult Clin Psychol 66:219-230, 1998 35. Peterson KM, Smith DA: An actor-partner interdependence model of spousal criticism and depression. J Abnormal Psychol 119:555-562, 2010

1186 The Journal of Pain

36. Porter LS, Keefe FJ, Hurwitz H: Disclosure between patients with gastrointestinal cancer and their spouses. Psychooncology 14:1030-1042, 2005

Spouse and Patient Beliefs and Perceptions About Chronic Pain T, Smith LJ, Tucker DH: Literacy-adapted cognitive behavioral therapy versus education for chronic pain at lowincome clinics: A randomized controlled trial. Ann Intern Med 168:471-480, 2018

37. Porter LS, Keefe FJ, Lipkus I, Hurwitz H: Ambivalence over emotional expression in patients with gastrointestinal cancer and their caregivers: Associations with patient pain and quality of life. Pain 117:340-348, 2005

42. Turner JA, Jensen MP, Romano JM: Do beliefs, coping, and catastrophizing independently predict functioning in patients with chronic pain? Pain 85:115-125, 2000

38. Schwartz L, Slater MA, Birchler GR: The role of pain behaviors in the modulation of marital conflict among chronic pain couples. Pain 65:227-233, 1996

43. Vaughn C, Leff J: The measurement of expressed emotion in the families of psychiatric patients. Br J Soc Clin Psychol 15:157-165, 1976

39. Spanier GB: The measurement of marital quality. J Sex Marital Ther 5:288-300, 1979

44. Weiner B: An Attributional Theory of Motivation and Emotion. New York, Springer-Verlag, 1986

40. Spielberger CD: Manual for the State-Trait Anxiety Inventory (STAI). Palo Alto, CA, Consulting Psychologists Press, 1983

45. Williams DA, Thorn BE: An empirical assessment of pain beliefs. Pain 36:351-358, 1989

41. Thorn BE, Eyer JC, Van Dyke BP, Torres CA, Burns JW, Kim M, Newman AK, Campbell LC, Anderson B, Block PR, Bobrow BJ, Brooks R, Burton TT, Cheavens JS, DeMonte CM, DeMonte WD, Edwards CS, Jeong M, Mulla MM, Penn

46. Williamson GM, Martin-Cook K, Weiner MF, Svetlik DA, Saine K, Hynan LS, Dooley WK, Schulz R: Caregiver resentment: Explaining why care recipients exhibit problem behavior. Rehabil Psychol 50:215-223, 2005