<%BANNER%>

Patient-Centered Outcomes for Chronic Spine Pain: Multidimensional Success Criteria and Treatment Matching


PAGE 1

PATIENT-CENTERED OUTCOMES FOR CHRONIC SPINE PAIN: MULTIDIMENSIONAL SUCCESS CRITER IA AND TREATMENT MATCHING By JENNIFER LYNN BROWN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2005

PAGE 2

Copyright 2005 by Jennifer Lynn Brown

PAGE 3

ACKNOWLEDGMENTS I would like to thank Dr. Michael E. Robinson for his guidance and mentorship throughout my graduate training. This project could not have been completed without his support. I would also like to recognize the valuable contributions of my committee members, Drs. Samuel F. Sears, Stephen R. Boggs, and Pamela W. Duncan. I would like to acknowledge the following data collection sites: Hampton Oaks SpineCare Center, Ayers Medical Plaza and Surgery Center, Brooks Pain Rehabilitation Program, Malcom Randall VA Medical Center, and the Department of Psychiatrys Pain Clinic. I also thank Dr. John Beauvais, Director of Predoctoral Training at VA Connecticut Healthcare System, for his assistance with statistical analyses. Finally, I thank my family and close friends for their love and support. iii

PAGE 4

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iii LIST OF TABLES .............................................................................................................vi LIST OF FIGURES ..........................................................................................................vii ABSTRACT .....................................................................................................................viii CHAPTER 1 INTRODUCTION........................................................................................................1 Aim 1: Developing Patient-Centered Success Criteria.................................................1 Aims 2 and 3: Identifying Patient Subgroups and Examining Treatment Matching....9 2 METHODS.................................................................................................................14 Participants.................................................................................................................14 Design.........................................................................................................................14 Procedure....................................................................................................................15 Treatments..................................................................................................................15 Palliative Pain Management................................................................................15 Multidisciplinary Pain Management...................................................................17 Measures.....................................................................................................................18 Demographics......................................................................................................18 Pain......................................................................................................................19 Mood/Distress......................................................................................................19 Disability.............................................................................................................20 Patient-Centered Outcomes.................................................................................20 Treatment Satisfaction.........................................................................................22 Analytical Strategy.....................................................................................................23 Aim 1: Developing Patient-Centered Success Criteria.......................................23 Aim 2: Identifying Patient Subgroups.................................................................23 Aim 3: Examining Treatment Matching..............................................................24 3 RESULTS...................................................................................................................25 Descriptive Statistics..................................................................................................25 iv

PAGE 5

Aim 1: Developing Patient-Centered Success Criteria...............................................28 Aim 2: Identifying Patient Subgroups........................................................................29 Aim 3: Examining Treatment Matching.....................................................................31 4 DISCUSSION.............................................................................................................35 APPENDIX A PATIENT-CENTERED OUTCOMES QUESTIONNAIRE (PCOQ).......................43 B FOLLOW-UP PCOQ..................................................................................................45 LIST OF REFERENCES...................................................................................................50 BIOGRAPHICAL SKETCH.............................................................................................55 v

PAGE 6

vi LIST OF TABLES Table page 3-1 Descriptive Statistics from the PCOQ (N = 70).......................................................25 3-2 Descriptive Statistics from the Follow-up PCOQ (N = 55).....................................26 3-3 Pain: Observed Frequencies.....................................................................................28 3-4 Fatigue: Observed Frequencies................................................................................28 3-5 Distress: Observed Frequencies...............................................................................28 3-6 Interference: Observed Frequencies.........................................................................28 3-7 Success Criteria and Associated Cl assification Accuracy Statistics........................29 3-8 Scores on PCOQ and Standardized Measures (Mean SD) by Cluster..................30 3-9 Effect Sizes of Cluster Differences (Cohens d )......................................................31

PAGE 7

vii LIST OF FIGURES Figure page 3-1 Cluster Solution for PCOQ Importance Ratings......................................................30 3-2 Subgroup by Treatment Interaction for Usual Interference.....................................33

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy PATIENT-CENTERED OUTCOMES FOR CHRONIC SPINE PAIN: MULTIDIMENSIONAL SUCCESS CRITERIA AND TREATMENT MATCHING By Jennifer Lynn Brown December 2005 Chair: Michael E. Robinson Major Department: Clinical and Health Psychology Our study aimed to (1) define patient-centered success criteria for chronic pain treatment across multiple relevant domains (pain, fatigue, emotional distress, and interference with daily activities) using our newly developed Patient-Centered Outcomes Questionnaires (PCOQs); (2) derive and validate subgroups of chronic pain patients; and (3) examine the potential for treatment matching using our patient subgroups. We recruited 70 patients with chronic spine pain from five University of Florida or affiliated pain clinics. Participants completed pre-treatment and post-treatment measures, including our PCOQ and Follow-up PCOQ. Post-treatment assessment occurred two months after treatment initiation. Pre-treatment success criteria derived using a direct scaling approach were more stringent than criteria developed using other approaches. At the onset of treatment, our patients considered a 58% reduction in pain to represent treatment success. Success criteria for pain developed using other approaches (e.g., expert opinion) have ranged viii

PAGE 9

from 30 to 50%. Pre-treatment success criteria for fatigue, distress, and interference were 61%, 64%, and 66%, respectively. Results showed that patients adjusted their success criteria over time in the direction of becoming less stringent, and they used their less stringent post-treatment criteria to make global judgments of treatment success. Consequently, future research might investigate the evolution of patients success criteria over the course of different treatments. Success criteria for pain derived using a scale comparison approach were a raw change of 17.5 points (0 to 100 numerical rating scale) and percent change of 25%. Other criteria were 7.5 points (11%) for fatigue, 5.0 points (13%) for distress, and 9.5 points (12%) for interference. Future research should validate these success criteria, particularly for the less studied domains of fatigue, distress, and interference. Once validated, they can be used to calculate effect size estimates, permitting comparisons across studies and ultimately helping to evaluate the efficacy of various treatments. Clinicians can also use these criteria to help them assess the care of individual patients. Finally, our results suggest that pain patients are heterogeneous. Even among patients whose pathology is specific to the spine, two distinct subgroups emerged. Future research might explore whether these and other patient subgroups have different criteria for treatment success. ix

PAGE 10

CHAPTER 1 INTRODUCTION Chronic pain is a major health problem in America. Estimates indicate that up to 18% of the population suffers from persistent pain at any given time (Turk, 2002). Beyond the number of people affected, chronic pain places a substantial economic burden on the United States health care system. Health care expenditures for chronic pain exceed $20 billion annually (Turk, 2002). This figure does not include the cost of treatment at pain rehabilitation programs, which is in the range of $1.4 billion per year (Turk, 2002). Notably, health care expenditures represent only a fraction of the costs associated with chronic pain. Disability compensation, lost productive time, and lost tax revenue also factor into estimates (Turk, 2002). A recent survey found that among working adults with common pain conditions, including back pain and arthritis, lost productivity due to decreased performance costs approximately $61.2 billion per year (Stewart, Ricci, Chee, Morganstein, & Lipton, 2003). Considering all sources of expenditures, chronic pain is projected to cost the United States economy roughly $100 billion each year (Department of Health and Human Services, National Institutes of Health, 1998). Aim 1: Developing Patient-Centered Success Criteria Numerous studies have examined the effectiveness of various treatments for chronic pain (Turk & Melzack, 2001). These studies used standard statistical methods that result in conclusions about statistical significance, or the reliability of treatment changes. Statistical significance, while necessary, is not sufficient to show treatment effectiveness. The statistical significance of a finding largely depends on sample size. A 1

PAGE 11

2 trial involving a large number of patients can detect differences between treatment and control groups or between the same patients preand post-treatment that are statistically reliable, but trivial from a clinical perspective. Conversely, a trial involving a small sample may not achieve statistical significance, but may indicate important differences. Regardless of the specific problem, comparisons of central tendency do not elucidate the proportion of patients who benefit from the treatment, or the extent of improvement in these patients. Statistically significant group differences could reflect major improvement in a few patients, or minor improvement in many patients. They also could reflect improvement in some patients, but no change (or even deterioration) in other patients (Farrar, Portenoy, Berlin, Kinman, & Strom, 2000). Therefore, clinicians might have difficulty interpreting and applying the results of central tendency analyses to the care of individual patients (Farrar et al., 2000). Consequently, studies of chronic pain should also consider whether treatment is clinically significant. A more useful statistic for clinicians is the percentage of patients who attain a clinically important improvement (Farrar et al., 2000). In deciding what constitutes a clinically significant outcome for pain treatments, researchers might consider looking to the patient rather than to health care providers, caregivers, or third-party payers. Pain and its sequelae are intrinsically subjective and influenced greatly by a persons mood, previous pain experience, expectations, and coping resources. Although patient self-report instruments and pain scales are widely used in clinical settings, different health care providers and researchers could still reach different conclusions about treatment success if they apply different interpretive criteria to these instruments. Furthermore, chronic pain patients and their therapists may have

PAGE 12

3 different views of treatment success (Daniel, Long, Murphy, Kores, & Hutcherson, 1983). Given the subjective nature of the pain experience and potential discrepancy between patients and health care providers conclusions about treatment success, decisions about what constitutes treatment success ought to consider the patient perspective. The patient-centered model of health care delivery, which holds that patients views should be recognized, appreciated, and incorporated into treatment decisions and outcome evaluation, has been adopted as a goal of clinical medicine (Lane & Davidoff, 1996), as it promotes greater satisfaction with health care, increased compliance with medical regimens, and maintenance of patient-provider relationships (Fischer, Stewart, Bloch, Lorig, Laurent, & Holman, 1999). Consequently, a growing body of literature addresses patient views of treatment outcome. Investigations have been conducted across medical conditions, such as epilepsy (Gilliam, Kuzniecky, Meador, Martin, Sawrie, Viikinsalo, Morawetz, & Faught, 1999), multiple sclerosis (Schwartz, Cole, & Gelber, 1995), and post-traumatic stress (Zatzick, Kang, Hinton, Kelly, Hilty, Franz, Le, & Kravitz, 2001). To date, there have been few patient-centered investigations of treatment outcome for chronic pain. Furthermore, most investigations focused exclusively on pain reduction, although patients enter pain rehabilitation with multiple presenting problems in addition to pain (e.g., mood disturbance, fatigue, disability) and identify diverse treatment goals. Casarett, Karlawish, Sankar, Hirschman, and Asch (2001) found that 32% of patients in their study identified improved sleep as an important clinical endpoint. Improved activities of daily living and improved mood were identified as important treatment goals

PAGE 13

4 by 30 and 8% of patients, respectively. More recently, the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMPACCT) convened with the purpose of identifying core outcome domains for clinical trials of chronic pain treatment (Turk, Dworkin, Allen, Bellamy, Brandenburg, Carr, Cleeland, Dionne, Farrar, Galer, Hewitt, Jadad, Katz, Kramer, Manning, McCormick, McDermott, McGrath, Quessy, Rappaport, Robinson, Royal, Simon, Stauffer, Stein, Tollett, & Witter, 2003). Included in the core outcome domains were emotional functioning and physical functioning. Therefore, any comprehensive examination of patient-centered outcomes must address a host of treatment endpoints. After identifying relevant endpoints, researchers must establish how much change in a given domain constitutes success. In other words, they must determine the degree of symptom improvement considered important, referred to in the literature as the minimal clinically important difference (MCID). Previous attempts to establish MCIDs used diverse approaches, most of which are not based on patient perceptions. Many approaches are flawed in additional ways, further limiting their utility. The simplest approach was based on the principle of face validity. Moore, McQuay, and Gavaghan (1996) chose 50% pain relief as the MCID because of its clinical intuitiveness. Yet, the significance of this criterion to chronic pain patients has not been determined. The approach taken by Slater, Doctor, Pruitt, and Atkinson (1997) involved comparing patients symptoms to those of healthy individuals. Slater and colleagues defined the MCID in depressive symptoms as change such that the patients score on the Beck Depression Inventory, a self-report measure of depression, following treatment was closer to the mean of healthy individuals than to the mean of chronic pain patients.

PAGE 14

5 Again, the significance of this criterion to the patients it is applied to has not been ascertained. Some investigators relied on expert opinion in establishing MCIDs. For example, Goldsmith, Boers, Bombardier, and Tugwell (1993) conducted eight focus groups with practicing clinicians, epidemiologists, biostatisticians, and physicians working either in various industries (e.g., pharmaceutical) or for regulatory agencies to determine clinically important outcomes for rheumatoid arthritis. The experts independently rated whether profiles containing information about six variables swollen joint count (range = 0 to 66), painful joint count (range = 0 to 68), pain intensity (range = 0 to 10 cm), disability (range = 0 to 10 cm), patient global assessment (range = 0 to 10 cm), and physician global assessment (range = 0 to 10 cm)represented important improvement. When 6 to 8 of the focus groups achieved 70% positive agreement, the profile was considered to reflect important improvement. Measures of central tendency then were calculated on profiles representing important improvement to derive an MCID for each of the six variables. The MCID values were 2 (17%) for swollen joints, 3 ( 27%) for painful joints, 2.2 (36%) for pain, 3.6 (49%) for disability, 2.2 (35%) for patient global assessment, and 2.3 (39%) for physician global assessment. While this approach is more sophisticated than other approaches, it still does not capture the patients view. A common method of deriving MCIDs involves comparing two scales. This approach has two major variations (Redelmeier, Guyatt, & Goldstein, 1996). In the first approach, patients are asked to rate a given domain (e.g., current pain intensity on a 100 mm VAS) and then indicate whether they are doing better or worse in that domain than before. This approach identifies the threshold at which the average patient begins

PAGE 15

6 judging himself as improved. Farrar, Young, LaMoreaux, Werth, and Poole (2001) used this approach with patients with neuropathic and non-neuropathic chronic pain conditions. Their goal was to determine the MCID in pain intensity associated with a much improved global impression of change. Using an 11-point numerical rating scale (NRS), they found that a raw change of 2 (30%) represented clinically important improvement. Salaffi, Stancati, Silvestri, Ciapetti, and Grassi (2004), using the same approach, found that a raw change of 2 (33%) on a 10 cm NRS corresponds to a much better global impression of change among patients with chronic musculoskeletal pain. In a study of patients with osteoarthritis (OA) of the lower extremities, MCIDs associated with a slightly better change of health in general related to the OA joint ranged from .80 to 1.01 points (17 to 22%) on a continuous 0 to 10 NRS (Angst, Aeschlimann, Michel, & Stucki, 2002). The second approach uses interpersonal judgments (Redelmeier et al., 1996). After rating themselves on various domains, patients are asked to converse with another patient with the same condition and then rate themselves on the same domains relative to their conversational partner. The MCID typically is defined as the difference in mean scores between conversations in which partners rate themselves as about the same and those in which they rate themselves as somewhat better. Wells, Tugwell, Kraag, Baker, Groh, and Redelmeier (1993) used this approach to determine the following MCIDs for rheumatoid arthritis patients: .22 units (7%) for physical ability as assessed by the Stanford Health Assessment Questionnaire (HAQ; range = 0 to 3), .62 cm (6%) for pain as measured on a 10 cm VAS, and 9% for overall condition as assessed on a 5-point

PAGE 16

7 scale. In another study of arthritis patients, Redelmeier and Lorig (1993) derived a slightly smaller MCID of .17 units for the HAQ. Although both approaches to scale comparison are patient-centered, important limitations exist. First, much variability exists in researchers a priori definitions of the clinically important differenceWells et al. (1993) chose somewhat better whereas Farrar et al. (2001) chose much improvedmaking comparisons across studies difficult, and raising questions about the ability of any one definition to truly capture the notion of minimal improvement. Most important, these investigator-determined a priori definitions may not reflect the degree of symptom improvement that is important to patients. A final method of determining MCIDs is specific to the domain of pain. Farrar et al. (2000) derived an MCID in analgesia by comparing the change in pain intensity for adequately treated episodes of cancer-related breakthrough pain (i.e., episodes not necessitating rescue medication as an adjunct to oral transmucosal fentanyl citrate) to the change in pain intensity for inadequately treated episodes. They selected as the MCID the change in pain intensity that most accurately distinguished satisfactorily and unsatisfactorily treated episodes in addition to having relatively equal sensitivity and specificity. Analyses yielded an MCID of 2 on an 11-point NRS, which corresponds to 33% change. Farrar and colleagues (2003) then replicated these MCIDs with a second data set. According to Farrar (2000), the use of rescue outcome may offer a more objective and specific measure of the patients perspective on what represents clinically important analgesia and allows the patient to integrate his or her perception of the amount of relief and any potential side effects experienced (p. 292). However, this

PAGE 17

8 approach is most appropriate for trials of fast-acting medications for the treatment of breakthrough or acute pain, and may not be generalizable to chronic pain (Farrar, 2000). Given the limitations associated with previous attempts to establish MCIDs for chronic pain, additional approaches are needed that take into account major areas of concern to patients. Therefore, our first aim was to define MCIDs from the spine pain patients perspective across four domains of concern (pain, fatigue, emotional distress, and interference with daily activities) using newly developed Patient-Centered Outcomes Questionnaires (PCOQs). Preliminary research using the PCOQ found that chronic pain patients might require larger reductions in their pain than has previously been reported in the literature, for them to consider treatment successful (Robinson, Brown, George, Edwards, Atchison, Hirsh, Waxenberg, Wittmer, & Fillingim, in press). In fact, MCIDs derived using a direct scaling approach were 3.4 points (on an 11-point NRS) or 56% for pain, 3.35 points (57%) for fatigue, 3.64 points (65%) for distress, and 4.3 points (68%) for interference (Robinson et al., in press). Our study expands on previous work using the PCOQ by adding a post-treatment measure: the Follow-up PCOQ. Our goal was to derive success criteria for each of the domains by ascertaining the change in scores from pre-treatment to post-treatment that most accurately distinguishes successful treatment from unsuccessful treatment as determined by patient global judgment. Once identified, these MCIDs can be used to calculate effect size estimates, such as the number needed to treat (Cook & Sackett, 1995), permitting comparisons across studies and ultimately helping to establish the efficacy of various treatments (Farrar et al., 2001; Redelmeier et al., 1996).

PAGE 18

9 Aims 2 and 3: Identifying Patient Subgroups and Examining Treatment Matching The etiological mechanisms of many nonmalignant chronic pain conditions, including spine pain, remain poorly understood. Thus, diagnosis is based predominantly on current and past signs and symptoms, rather than on specific pathological findings. Turk and Okifuji (2001) said that because the specificity of these diagnostic criteria has not been established, patients with diverse characteristics may erroneously be given the same diagnosis and treatment. In other words, a heterogeneous group of patients may be treated as if they are homogeneous (Turk & Okifuji, 2001, p. 400). This may partly explain why treatment-outcome studies consistently find that only a subset of patients with a particular diagnosis benefits from the prescribed treatment. To offset the pain patient homogeneity assumption, researchers attempted to identify subgroups of chronic pain patients. For example, Dworkin and LeResche (1992) derived subgroups using physical signs and symptoms. Other investigators (e.g., Jamison, Rock, & Parris, 1988) used traditional measures of personality and psychopathology, such as the Minnesota Multiphasic Personality Inventory or Symptom Checklist-90, to empirically derive pain patient subgroups. Still others (Keefe, Bradley, & Crisson, 1990) created subgroups based on pain behaviors. Although these approaches yielded several classification systems, the question remains as to whether these systems are clinically useful. According to Turk and Okifuji (2001), one way to substantiate the clinical utility of a classification system involves demonstrating the differential effectiveness of therapies matched to the defining characteristics of the subgroups. The few studies that addressed this issue (Guck, Meilman, Skultety, & Poloni, 1988; McGill, Lawlis, Selby, Mooney, & McCoy, 1983; Moore, Armentrout, Parker, & Kivlahan, 1986) failed to

PAGE 19

10 demonstrate the differential effectiveness of treatments customized to subgroup characteristics. Of note, Turk and colleagues had some success with their classification system, which is based on the West Haven-Yale Multidimensional Pain Inventory (WHYMPI). Three subgroups of chronic pain patients emerged from cluster analysis of WHYMPI responses: Dysfunctional, Interpersonally Distressed, and Adaptive Copers. The Dysfunctional subgroup reports high pain severity, life interference, and psychological distress, as well as low activity level (Turk & Rudy, 1988). Interpersonally Distressed patients perceive their significant others as unsupportive of their pain problems. Finally, the Adaptive Coper subgroup reports relatively low levels of pain and interference, and relatively high levels of social support and activity. Preliminary studies examining the differential responsiveness of the WHYMPI subgroups to different treatment programs provide support for the clinical utility of the classification system. Rudy, Turk, Kubinski, and Zaki (1995) evaluated the differential responses of the three WHYMPI subgroups of temporomandibular disorder (TMD) patients to a treatment involving interocclusal appliance, biofeedback-assisted relaxation, and stress management. Results showed that significantly more Dysfunctional patients than Interpersonally Distressed or Adaptive Coper patients achieved clinically significant improvements with respect to pain intensity, depression, and perceived impact of TMD on their lives. Furthermore, a greater number of Interpersonally Distressed patients than Adaptive Coper patients showed clinically significant improvements in these domains. In general, results showed that the treatment was effective, but only for certain subgroups of TMD patients.

PAGE 20

11 Turk and associates (Turk, Rudy, Kubinski, Zaki, & Greco, 1996; Turk, Zaki, & Rudy, 1993) then developed a treatment plan that targets the needs of TMD patients in the Dysfunctional subgroup. They evaluated the efficacy of this treatment program (interocclusal appliance, biofeedback-assisted relaxation, and cognitive therapy for depression) against a more conservative treatment program (interocclusal appliance, biofeedback-assisted relaxation, and supportive counseling). The Dysfunctional patients who received cognitive therapy as part of their treatment showed significantly more improvement, especially in the domains of pain and depression, than did the Dysfunctional patients who received supportive counseling as part of their treatment. Therefore, by incorporating a treatment component tailored to Dysfunctional patients, Turk and colleagues were able to augment treatment outcome. In a study of patients with fibromyalgia syndrome, Turk, Okifuji, Starz, and Sinclair (1998) predicted that the WHYMPI subgroups would show different outcomes to standard rehabilitative treatment involving medical, physical, occupational, and psychological elements. Findings support their hypothesis, with patients in the Dysfunctional subgroup showing improvements in nearly all domains and patients in the Interpersonally Distressed subgroup showing no improvement, presumably because the treatment did not address their particular clinical needs, which reside primarily in the interpersonal domain. Patients in the Adaptive Coper subgroup showed only minor improvements compared to patients in the Dysfunctional subgroup, although they had less room for improvement given their better pre-treatment functioning. Again, these results highlight the importance of matching treatment to the clinical needs of the group.

PAGE 21

12 These studies provide preliminary support for the utility of the treatment-matching approach. However, further research is needed before the treatment-matching approach can reasonably be applied to clinical practice. It is hoped that streamlining treatment based on patient characteristics rather than adopting the one size fits all approach to treatment ultimately will improve outcomes and decrease costs (Turk & Okifuji, 2001). A second aim of our study was to empirically derive subgroups of chronic pain patients using the PCOQ. More specifically, we classified chronic spine pain patients into subgroups based on their ratings of the perceived importance of improvement in pain, fatigue, emotional distress, and interference with daily activities. On the basis of preliminary data from four University of Florida or affiliated pain clinics (Robinson et al., in press), we expected three subgroups to emerge. We predicted that one subgroup (pain-focused subgroup) would show a greater relative interest in pain relief. We expected a second subgroup (multifocused-high subgroup) to report that improvement in all four domains is extremely important. We expected the final subgroup (multifocused-moderate subgroup) to provide moderate importance ratings across domains. In addition to replicating previous findings, our study aimed to establish the validity of the patient subgroups by examining pre-treatment differences among the subgroups on demographic and psychosocial variables. We hypothesized that the two multifocused subgroups would have significantly greater duration of pain than participants in the pain-focused subgroup. We also expected that the multifocused-high subgroup would have the highest scores on self-report measures of pain, distress, and disability. We predicted the pain-focused subgroup to have the lowest scores on measures of pain, distress, and disability.

PAGE 22

13 Our final aim was to examine the potential for treatment matching using the patient subgroups derived from the PCOQ. We predicted that patients who received treatment that was matched to their clinical needs would show superior treatment outcomes, when compared to patients who received treatment that was not matched to their clinical needs. Therefore, we expected patients in the pain-focused subgroup to show better outcomes after treatment that focused exclusively on their pain (i.e., palliative pain management), and we expected patients in the multifocused subgroups to show better outcomes after treatment that addressed not only their pain, but also their fatigue, emotional distress, and interference with daily activities (i.e., multidisciplinary pain management).

PAGE 23

CHAPTER 2 METHODS Participants Participants were 70 patients (27 male, 43 female) with a primary complaint of chronic spine pain of at least 3 months duration. Participants were recruited from five University of Florida or affiliated pain clinics, and they provided informed consent in accordance with Institutional Review Board requirements. The mean age of participants was 50.9 ( SD = 12.5) years. Most participants (88.6%) were Caucasian and 11.4% were African-American. Participants reported a mean of 13.0 ( SD = 2.2) years of formal education. Of participants, 66% percent were married. Participants reported a mean duration of pain of 91.1 ( SD = 95.7) months, with pain duration ranging from 4 to 468 months. Participants underwent one of two types of treatment as part of their routine clinical care: palliative pain management (interventional anesthesia/oral analgesic only) or multidisciplinary pain management (behavioral/rehabilitation intervention component included). The two treatment approaches are described in greater detail below. Design The design of the study was longitudinal, with pre-treatment and post-treatment assessments. Post-treatment assessment occurred approximately 2 months after initiation of treatment. Post-treatment data were collected on 55 of the 70 participants (21 males, 34 females). Nine participants were lost to follow-up for the following reasons: phone number disconnected (3), no longer residing at stated phone number (1), on extended 14

PAGE 24

15 vacation (1), could not reach at home despite multiple attempts (4). Four participants withdrew by choice at follow-up, and two participants were withdrawn by the research team at follow-up because they never received the intended treatment. Procedure Before starting treatment, eligible participants were approached by a co-investigator or other research staff, given a brief description of the study, and then asked if they would consider participating in the study. Interested participants were asked to read and sign the informed consent form. After agreeing to participate in the study, participants completed the Demographics Questionnaire and the following paper-and-pencil measures: McGill Pain Questionnaire, Beck Depression Inventory, Pain Anxiety Symptoms Scale, Roland Disability Questionnaire, Pain Disability Index, and Patient-Centered Outcomes Questionnaire. Participants also rated the intensity and unpleasantness of their pain on visual analog scales (VASs). Two months post-treatment, research staff contacted participants by telephone to complete two measures, the Follow-up PCOQ and Pain Service Satisfaction Test. Treatments Because of ethical considerations, treatment was not randomly assigned. All participants were recruited from existing treatment programs, making our study highly ecologically valid. Withdrawal before completion of the study did not affect the participants treatment. Palliative Pain Management Participants receiving palliative pain management underwent one or more of the following four treatments: injection, radiofrequency denervation, non-narcotic analgesic medication, and opioid medication. Injections consist of epidural administration of

PAGE 25

16 steroid or anti-inflammatory medication, sometimes in combination with local anesthetic. The injection is delivered under fluoroscopic guidance into the epidural space at the spinal level determined by clinical exam to be the area relevant to the clinical pain distribution. Injections serve to decrease pain by reducing nerve root irritation. Radiofrequency denervation is a commonly performed interventional treatment for axial low back pain (Dreyfuss, Halbrook, Pauza, Josi, McLarty, & Bogduk, 2000; North, Han, Zahurak, & Kidd, 1994). The procedure is done by positioning a percutaneous electrode next to the medial branches of the posterior nerve rami for a particular facet joint, and then using a generator to induce heat through the electrode. This causes a lesion of the nerves ability to send messages to the posterior column of the spinal cord. Radiofrequency denervation is completed under fluoroscopic guidance, and it may be done unilaterally or bilaterally and at a single spinal level or at multiple spinal levels. The number of levels to be lesioned is determined by prior diagnostic lidocaine injection of the medial branches. Patients typically receive chronic opioid medication if, and when, there are no other interventions, including non-narcotic analgesic medication (e.g., Ultram), that might manage their pain more effectively. Use of longand short-acting opioids in a pattern to better manage pain is discussed with the patient, as are potential side effects and adverse outcomes. After this discussion, the initial dosing pattern of medication is started. Opioid medications typically prescribed include OxyContin, MS Contin, Methadone, and the Duragesic Patch. In most cases, a second prescription for breakthrough medication also is given. If the patient responds favorably with no intolerable side effects at the initial dose, he or she is maintained on the medication, and the dose is adjusted to achieve as much

PAGE 26

17 pain reduction as possible without intolerable side effects. Should the patient have intolerable side effects from the initial dose, he or she is switched to another opioid medication. Multidisciplinary Pain Management Participants receiving multidisciplinary pain management participated in a pain management group and/or underwent a standard rehabilitation program. Participants sometimes received pharmacological therapy as an adjunct to this behavioral/ rehabilitation approach. The pain management group is a psychoeducational group comprising cognitive-behavioral pain coping skills training, an empirically supported treatment for chronic pain conditions (Keefe, 1996). This 10-session, structured, outpatient group emphasizes instruction, rehearsal, and practice during session and at home. Sessions are held once per week for 90 minutes. Patients initially are taught to conceptualize their pain in terms of the gate control theory of pain (Melzack & Wall, 1965). Then, they are instructed in diaphragmatic breathing, which involves breathing deeply from the diaphragm in a controlled manner. During the next session, patients are taught progressive muscle relaxation, a series of exercises involving tensing and relaxing various muscle groups, and given an opportunity to rehearse the technique in session. The third session is devoted to discussing activity-rest cycling, a method for controlling pain and increasing activity level that consists of planned periods of activity, interspersed with planned regular rest periods. During the fourth session, patients are introduced to cognitive restructuring, which involves systematically identifying and changing negative automatic thoughts to more encouraging, realistic thoughts. They spend the next three sessions participating in cognitive restructuring exercises using examples generated by both group leaders and group members. The eighth session focuses on problem-solving,

PAGE 27

18 and the final sessions focus on evaluating progress and maintaining treatment gains. Other topics covered during group include pleasant activity scheduling, mini-relaxation, and attention diversion techniques (e.g., pleasant imagery, focal point use). The standard rehabilitation program consists of 18 sessions (3 per week), with each session lasting approximately 90 minutes. Each session involves strength, range of motion, flexibility, and aerobic training. Passive treatments (e.g., hot packs, ice packs, electrical stimulation) are not employed during session, but patients are instructed in the proper use of these modalities at home. Patients are prescribed a custom therapeutic exercise program that addresses strength, range of motion, and flexibility deficits detected during their evaluation. In addition, they undergo aerobic exercise, which has been shown in a controlled trial to be an effective treatment for disability reduction in chronic low back pain patients (Mannion, Muntener, Taimela, & Dvorak, 2001). Exercise prescription is done in a graded manner (Lindstrom, Ohlund, Eek, Wallin, Peterson, Fordyce, & Nachemson, 1992). Graded exercise prescription differs from traditional exercise prescription because it focuses on improvement in activity tolerance, not on pain abatement. During the final session, patients are issued a home exercise program for continuation of unsupervised therapeutic exercise. Measures Demographics The Demographics Questionnaire obtains information about a persons age, sex, race, marital status, education, medical diagnosis, pain duration, and previous treatment for the pain condition.

PAGE 28

19 Pain Participants completed visual analog scale (VAS) measures of pain intensity and pain unpleasantness. For pain intensity, the scale ranges from 0 to 100, where 0 represents no pain sensation and 100 represents most intense pain sensation imaginable. For pain unpleasantness, the scale ranges from 0 (not at all unpleasant) to 100 (most unpleasant imaginable). The VAS provides ratio level pain measurement and has been validated in multiple studies (Price, McGrath, Rafii, & Buckingham, 1983). The McGill Pain Questionnaire (MPQ; Melzack & Katz, 2001) consists of 102 words in 20 subclasses. For each of the subclasses, patients are asked to choose the word that best describes their pain at the present moment. Each word in each subclass has a rank value, and rank values are summed to derive scores for the sensory, affective, and evaluative dimensions of pain, as well as a total score. The MPQ is a reliable, valid, and consistent measurement tool (Melzack & Katz, 2001). Mood/Distress The Beck Depression Inventory (BDI; Beck, Steer, & Garbin, 1988), a widely used self-report measure of depression, assesses the extent to which an individual exhibits each of 21 behaviors, thoughts, or affective features commonly associated with depression. The BDI yields a total score (range = 0 to 63), as well as two subscale scores measuring cognitive and somatic symptoms of depression. Geisser, Roth, and Robinson (1997) showed that the BDI is a valid instrument for use with chronic pain patients. The Pain Anxiety Symptoms Scale (PASS) is a 40-item self-report questionnaire that measures four dimensions of pain-related anxiety, namely cognitive anxiety, escape/avoidance, fearful appraisal, and physiological anxiety (McCracken, Zayfert, & Gross, 1992). Participants respond to each item by selecting a frequency rating from

PAGE 29

20 0 (never) to 5 (always). Five items are reverse scored. Frequency ratings are added to derive a total score and four subscale scores. Total scores can range from 40 to 200. The PASS is valid measure, as evidenced by significant correlations with other measures of anxiety (McCracken et al., 1992). Alpha coefficients ranging from .81 to .94 indicate the PASS and its subscales have good internal consistency (McCracken et al., 1992). Disability The Pain Disability Index (PDI) measures the degree to which chronic pain interferes with the following life areas: family/home responsibilities, recreation, social activity, occupation, sexual behavior, self care, and life-support activity (Pollard, 1984). Amount of disability experienced in each of the seven domains is rated on an 11-point scale ranging from 0 (no disability) to 10 (total disability). The ratings are summed to derive a total score (range = 0 to 70). The PDI has acceptable psychometric properties, with an internal consistency coefficient of .86 (Tait, Pollard, Margolis, Duckro, & Krause, 1987). The Roland Disability Questionnaire (RDQ) is a 24-item checklist designed to measure functional impairment due to spine pain (Roland & Morris, 1983). The number of items endorsed is summed to arrive at a total score ranging from 0 (no disability) to 24 (severe disability). The RDQ has adequate psychometric properties, with test-retest reliability values ranging from .72 to .91 (Jensen, Strom, Turner, & Romano, 1992; Roland & Morris, 1983). Patient-Centered Outcomes The Patient-Centered Outcomes Questionnaire (PCOQ; Appendix A) consists of five sections. In the first section, patients indicate their usual levels (during the past week) of pain, fatigue, emotional distress, and interference with daily activities on

PAGE 30

21 101-point NRSs (0 = none, 100 = worst imaginable). In the second section, patients indicate their desired levels of pain, fatigue, emotional distress, and interference with daily activities using the same 101-point NRSs. The third section requires patients to indicate where their pain, fatigue, emotional distress, and interference with daily activities would have to be on 101-point NRSs (0 = none, 100 = worst imaginable) for them to consider treatment successful. In the fourth section, patients indicate the levels of pain, fatigue, emotional distress, and interference with daily activities they expect after treatment on the same 101-point NRSs. Finally, participants are asked to rate on 101-point NRSs (0 = not at all important, 100 = most important) how important it is for them to see improvement in each of the four domains. Pilot data on 21 subjects indicates the PCOQ has acceptable test-retest reliability over a 48-hour period, with reliability values ranging from .84 to .90 ( p s < .001) for usual levels. Reliability values for importance ratings ranged from to .62 to .82 ( p s < .01) for the domains of pain, emotional distress, and interference with daily activities. The correlation for fatigue did not reach statistical significance (r = .33, p = .14). Reliability values for success criteria are rather low in this sample despite achieving statistical significance across all domains (r = .43 to .58, p s < .05) but emotional distress (r = .287, p = .21). Data from our study indicate the PCOQ has good concurrent validity with standardized measures of pain, mood/distress, and disability. For example, the correlation between usual level of pain from the PCOQ and the pain intensity VAS was .52 ( p < .001). Usual emotional distress was highly related to total scores on the BDI (r = .65, p < .001) and PASS (r = .72, p < .001) and usual interference was significantly associated with PDI (r = .75, p < .001) and RDQ (r = .69, p < .001) total scores.

PAGE 31

22 The Follow-up PCOQ (Appendix B) includes six sections. The first section has patients estimate their pre-treatment levels of pain, fatigue, emotional distress, and interference with daily activities on the same 101-point NRS used in the PCOQ (0 = none, 100 = worst imaginable). The next section requires patients to indicate the levels of pain, fatigue, emotional distress, and interference with daily activities associated with treatment success on the same 101-point NRSs. In the third section, patients rate on 101-point NRSs (0 = not at all important, 100 = most important) how important it is for them to see improvement in each of the four domains. In the next section, patients rate on 101-point NRSs (0 = none, 100 = worst imaginable) their usual levels of pain, fatigue, emotional distress, and interference with daily activities after treatment. In the fifth section, patients specify whether they improved, stayed the same, or worsened across each domain since the start of treatment and then, if applicable, rate on 101-point NRSs (0 = none, 100 = complete) the amount of improvement or worsening they experienced in each of the domains. Patients also indicate whether they improved, stayed the same, or worsened overall. In the last section, patients decide whether they would consider their treatment successful. For each of the domains, patients choose yes or no. Patients also evaluate the overall success of their treatment by choosing yes or no. Treatment Satisfaction The 20-item Pain Service Satisfaction Test (PSST) assesses patient satisfaction with pain treatment (McCracken, Klock, Mingay, Asbury, & Sinclair, 1997). The PSST has high internal consistency, and is positively correlated with treatment-related changes in pain, depression, and anxiety (McCracken & Turk, 2002).

PAGE 32

23 Analytical Strategy Aim 1: Developing Patient-Centered Success Criteria Analyses used usual ratings from the PCOQ, and usual ratings and success judgments from the Follow-up PCOQ. First, we calculated change scores for each of the four domains (pain, fatigue, emotional distress, interference with daily activities) by subtracting Follow-up PCOQ usual ratings (i.e., post-treatment usual ratings) from PCOQ usual ratings (i.e., pre-treatment usual ratings). Then, to characterize the association between change scores and patient judgment of treatment success, we derived receiver-operating characteristic (ROC) curves for each of the four domains using logistic regression analyses. For each analysis, judgment of success (yes, no) served as the dependent variable and either the raw change or percent change score served as the independent variable. For these analyses, area under the ROC curve was interpreted as the probability of distinguishing participants who were successfully treated on the domain from participants who were not successfully treated on the domain. The area under the ROC curve can range from .5 (no accuracy) to 1.0 (perfect accuracy). We then computed optimal cut-offs (i.e., MCIDs) for each domain from these ROC curves. We expected optimal cut-offs to have a high hit rate coupled with a good balance of sensitivity and specificity. Aim 2: Identifying Patient Subgroups We conducted a hierarchical agglomerative cluster analysis (Wards method, squared Euclidian distance) on PCOQ importance ratings to derive the patient subgroups. Once we identified the subgroups, we conducted a series of analyses to examine potential subgroup differences on demographic and psychosocial variables. First, we performed one-way analyses of variance (ANOVAs) to examine possible subgroup differences in

PAGE 33

24 age, education, and pain duration. Then, we performed a discriminant function analysis using PCOQ usual ratings of pain, fatigue, emotional distress, and interference with daily activities to discriminate spine pain patients classified into each of the subgroups. We performed a second discriminant function analysis using standardized measures of pain (MPQ), distress (BDI, PASS), and disability (PDI, RDQ) to discriminate spine pain patients classified into each of the subgroups. Lastly, we calculated effect sizes (Cohens d; Cohen, 1988) on PCOQ and standardized measures of pain, fatigue, distress, and disability to examine the magnitude of subgroup differences in standard deviation units. More specifically, we calculated effect sizes by dividing mean subgroup differences by pooled standard deviation for each measure ( M 1 M 2 /SD P ). We interpreted the effect sizes using Cohen's (1988) criteria of .20, .50, and .80 for small, medium, and large effects, respectively. Aim 3: Examining Treatment Matching Analyses used PSST scores, PCOQ usual ratings, Follow-up PCOQ usual ratings, and MCIDs derived in Aim 1. First, we performed a 2 (subgroup) X 2 (treatment) ANOVA on post-treatment PSST scores to examine the effects of subgroup membership and type of treatment on pain treatment satisfaction. Next, we performed 2 (subgroup) X 2 (treatment) X 2 (time) mixed-model ANOVAs on usual ratings of pain, fatigue, emotional distress, and interference with daily activities to investigate the effects of subgroup membership and type of treatment on these ratings across time. Subgroup and treatment served as between-subjects factors and time was the within-subjects factor. Finally, we conducted chi-square analyses for each of the domains to examine the relationship between type of treatment and treatment success as determined by MCID attainment using MCIDs established in Aim 1.

PAGE 34

CHAPTER 3 RESULTS Descriptive Statistics Participants reported moderate usual levels of pain, fatigue, emotional distress, and interference with daily activities before treatment ( M = 51.1 to M = 65.4) (Table 3-1). Furthermore, participants considered a 26.9/100 pain level or a mean reduction in pain of 38.5 points (58%) to represent treatment success. A 23.2/100 fatigue level or a mean reduction in fatigue of 36.9 points (61%), an 18.3/100 distress level or a mean reduction in distress of 32.8 points (64%), and a 21.6/100 interference level or a mean reduction in interference of 43.1 points (66%) also represented successful treatment. As a group, participants viewed improvements in all domains (pain, fatigue, emotional distress, and interference with daily activities) as very important ( M = 83.4 to M = 89.1). Table 3-1. Descriptive Statistics from the PCOQ (N = 70) Level Domain Mean* SD Observed Range Usual Pain 65.4 21.6 20-97 Fatigue 60.1 24.4 0-100 Distress 51.1 30.0 0-100 Interference 64.7 27.4 0-100 Success Pain 26.9 18.3 0-75 Fatigue 23.2 18.4 0-70 Distress 18.3 19.3 0-80 Interference 21.6 18.5 0-80 Importance Pain 89.1 20.3 10-100 Fatigue 83.4 24.4 0-100 Distress 84.0 26.2 0-100 Interference 87.0 23.7 0-100 Numerical Rating Scale = 0 to 100 25

PAGE 35

26 Participants reported moderate usual levels of pain, fatigue, distress, and interference post-treatment ( M = 47.5 to M = 56.4) (Table 3-2). After treatment, they considered a 33.9/100 pain level, 30.2/100 fatigue level, 26.1/100 distress level, and 28.1/100 interference level to represent treatment success. Again, participants viewed improvements in all domains as quite important ( M = 78.8 to M = 88.9). Table 3-2. Descriptive Statistics from the Follow-up PCOQ (N = 55) Level Domain Mean* SD Observed Range Usual Pain 56.4 27.4 0-100 Fatigue 53.5 27.9 0-100 Distress 47.5 30.9 0-100 Interference 53.5 30.6 0-100 Success Pain 33.9 17.6 0-75 Fatigue 30.2 17.9 0-70 Distress 26.1 17.9 0-70 Interference 28.1 16.6 0-75 Importance Pain 88.9 21.2 0-100 Fatigue 78.8 28.1 0-100 Distress 83.0 25.1 0-100 Interference 82.8 26.9 0-100 Numerical Rating Scale = 0 to100 Four paired-samples t-tests were conducted to evaluate whether participants PCOQ (i.e., pre-treatment) success criterion and Follow-up PCOQ (i.e., post-treatment) success criterion significantly differed for each domain. The results showed that participants Follow-up PCOQ (i.e., post-treatment) success criterion for pain was significantly higher than their PCOQ (i.e., pre-treatment) success criterion for pain [ t (52) = 2.125, p = .038], suggesting that participants had higher target values for pain treatment success (i.e., less stringent success criteria for pain) after treatment. Similarly, patients had higher target values (i.e., less stringent success criteria) for successful treatment of fatigue and distress after treatment, t (52) = 2.14, p = .037 and t (52) = 2.16, p = .035, respectively. Results for interference approached statistical significance [ t (52) = 1.77,

PAGE 36

27 p = .083] but did not reach the p .05 standard. These findings indicate that participants adjusted or re-defined their success criteria during the course of treatment in the direction of becoming less strict. A series of chi-square analyses were conducted to determine if participants who met their post-treatment success criterion for a particular PCOQ domain were more likely to judge themselves as successfully treated on that domain (yes/no success judgments) than participants who did not meet their post-treatment success criterion for the domain. The results showed that success criteria and success judgments were significantly related for the domain of pain, Pearson 2 (1, N = 52) = 22.768, p < .001. Compared to patients who did not meet their post-treatment success criterion for pain, patients who met their post-treatment success criterion for pain were 3.7 times (100% v 27%) more likely to judge their pain as having been successfully treated (Table 3-3). Similar results were found for the domains of fatigue [Pearson 2 (1, N = 51) = 11.574, p = .001], distress [Pearson 2 (1, N = 51) = 27.019, p < .001], and interference [Pearson 2 (1, N = 51) = 19.538, p < .001]. Thus, patients who met their post-treatment success criteria for fatigue, distress, and interference were 2.6 (86% v 32%), 5.3 (94% v 18%), and 3.7 (89% v 24%) times more likely, respectively, than patients who did not meet their post-treatment success criteria for these domains to judge treatment of these domains a success (Table 3-4, Table 3-5, and Table 3-6). These findings suggest participants used their success criteria to assist them in making global judgments of treatment success.

PAGE 37

28 Table 3-3. Pain: Observed Frequencies Success Judgment No Yes Achieved No n = 27 n = 10 Post-treatment 73% 27% Success Yes n = 0 n = 15 Criterion? 0% 100% Table 3-4. Fatigue: Observed Frequencies Success Judgment No Yes Achieved No n = 25 n = 12 Post-treatment 68% 32% Success Yes n = 2 n = 12 Criterion? 14% 86% Table 3-5. Distress: Observed Frequencies Success Judgment No Yes Achieved No n = 28 n = 6 Post-treatment 82% 18% Success Yes n = 1 n = 16 Criterion? 6% 94% Table 3-6. Interference: Observed Frequencies Success Judgment No Yes Achieved No n = 25 n = 8 Post-treatment 76% 24% Success Yes n = 2 n = 16 Criterion? 11% 89% Aim 1: Developing Patient-Centered Success Criteria Optimal cut-off points (i.e., MCIDs) for pain generated from the ROC curves are a raw change of 17.5 points and a percent change of 25% (Table 3-7). That is, MCIDs for pain were a reduction in pain greater than or equal to 17.5 points (on a 0 to 100 NRS) or a 25% or greater reduction in pain. Optimal cut-off points for the other domains were 7.5 points or 11% for fatigue, 5.0 points or 13% for emotional distress, and 9.5 points or 12% for interference with daily activities (Table 3-7). Hit rates for these cut-offs ranged from

PAGE 38

29 73 to 85%. In other words, these cut-offs accurately classified 73 to 85% of patients as either successfully or unsuccessfully treated. Table 3-7. Success Criteria and Associated Classification Accuracy Statistics Domain Variable (Cut-off*) ROC (SE) HR SENS SPEC PPV NPV Pain Raw Change (17.5) .86 (.05)*** .85 .72 .96 .95 .79 Percent Change (25%) .85 (.06)*** .85 .72 .96 .95 .79 Fatigue Raw Change (7.5) .80 (.07)*** .73 .70 .75 .70 .75 Percent Change (11%) .78 (.07)*** .73 .71 .75 .71 .75 Distress Raw Change (5) .76 (.07)*** .77 .82 .73 .69 .85 Percent Change (13%) .82 (.07)*** .82 .81 .83 .77 .86 Interference Raw Change (9.5) .76 (.07)*** .73 .67 .79 .73 .73 Percent Change (12%) .77 (.08)*** .75 .70 .79 .73 .76 Classification accuracy statistics are based on identification of patients who were successfully treated on the domain relative to patients who were not successfully treated on the domain. ROC = area under the receiver-operating characteristic curve; HR = hit rate; SENS = sensitivity, SPEC = specificity; PPV = positive predictive value; NPV = negative predictive value; Numerical Rating Scale = 0 to 100; *** p .001. Aim 2: Identifying Patient Subgroups Contrary to the hypothesized three-cluster solution, the hierarchical agglomerative cluster analysis (Wards method, squared Euclidian distance) on PCOQ importance ratings revealed a two-cluster solution (Figure 3-1, Table 3-8). Cluster 1 comprised 7 participants who had low to moderate importance ratings across the domains of pain, fatigue, distress, and interference, whereas Cluster 2 comprised 63 participants who had high importance ratings across domains. The obtained cluster solution limited statistical power in subsequent inferential analyses. A series of one-way ANOVAs did not reveal any differences between the clusters on the demographic variables of age [ F (1, 68) = 2.270, p = .136], education [ F (1, 68) = .310, p = .579], and pain duration [ F (1, 66) = 1.689, p = .198]. A discriminant function analysis was conducted to determine whether four variablesusual pain, usual fatigue, usual distress, and usual interference from the PCOQcould predict cluster membership.

PAGE 39

30 PainFatigueDistressInterferenceDomain 0102030405060708090100Mean Importance Rating Cluster 1 (n = 7) Cluster 2 (n = 63) Figure 3-1. Cluster Solution for PCOQ Importance Ratings. Table 3-8. Scores on PCOQ and Standardized Measures (Mean SD) by Cluster Measure Cluster 1 (n = 7) Cluster 2 (n = 63) PCOQ Usual Pain 53.6 21.0 66.7 21.4 PCOQ Usual Fatigue 55.7 30.5 60.6 23.9 PCOQ Usual Distress 31.4 32.8 53.3 29.1 PCOQ Usual Interference 42.1 36.0 67.2 25.5 MPQ Total 29.6 17.8 32.0 15.3 BDI Total 13.7 11.8 19.3 11.5 PASS Total 78.6 36.0 86.1 37.6 PDI Total 33.2 20.0 41.7 15.0 RDQ Total 11.0 8.1 14.6 5.6 PCOQ = Patient-Centered Outcomes Questionnaire; MPQ = McGill Pain Questionnaire; BDI = Beck Depression Inventory; PASS = Pain Anxiety Symptoms Scale; PDI = Pain Disability Index; RDQ = Roland Disability Questionnaire

PAGE 40

31 The overall Wilks lamda did not quite reach statistical significance, = .884, 2 (4, N = 70) = 8.102, p = .088, indicating that the predictors did not differentiate the two clusters. A second discriminant function analysis using five predictorsMPQ, BDI, PASS, PDI, and RDQ total scoresalso did not predict cluster membership, = .945, 2 (5, N = 68) = 3.577, p = .612. However, examination of effect sizes (Table 3-9), revealed moderate to large effect sizes (range = .48 to .94) for PCOQ usual pain, PCOQ usual distress, PCOQ usual interference, BDI Total, PDI Total, and RDQ Total. Participants in Cluster 1 had lower scores than participants in Cluster 2 across these measures (Table 3-8), suggesting that participants in Cluster 1 had significantly less pain, depression/distress, and disability/interference than participants in Cluster 2. Taken together, these findings partially support hypotheses regarding subgroup differences. Table 3-9. Effect Sizes of Cluster Differences (Cohens d) Measure Cohens d PCOQ Usual Pain .61 PCOQ Usual Fatigue .19 PCOQ Usual Distress .74 PCOQ Usual Interference .94 MPQ Total .15 BDI Total .48 PASS Total .19 PDI Total .55 RDQ Total .61 Aim 3: Examining Treatment Matching Hypotheses regarding treatment matching could not be explored because the previous finding of a pain-focused versus multifocused subgroups was not replicated. However, several analyses were done involving the two clusters obtained from the present sample to examine potential treatment effects. A 2 (subgroup: Cluster 1,

PAGE 41

32 Cluster 2) X 2 (treatment: palliative, multidisciplinary) ANOVA on post-treatment PSST scores revealed only a main effect of subgroup, F (1, 48) = 6.652, p = .013, eta 2 = .12, indicating that participants in Cluster 1 were significantly more satisfied with their treatment ( M = 57.4) than participants in Cluster 2 ( M = 44.1) regardless of the type of treatment received. A series of 2 (subgroup: Cluster 1, Cluster 2) X 2 (treatment: palliative, multidisciplinary) X 2 (time: pre-treatment, post-treatment) mixed-model ANOVAs were performed on usual ratings of pain, fatigue, distress, and interference to investigate the effects of subgroup membership and type of treatment on these ratings across time. Subgroup and treatment served as between-subjects factors and time was the within-subjects factor. The mixed-model ANOVA on usual ratings of pain did not reveal any statistically significant main or interaction effects. Likewise, mixed-model ANOVAs on usual ratings of fatigue and distress did not reveal any statistically significant main or interaction effects. The mixed-model ANOVA on usual ratings of interference revealed significant main effects of subgroup [ F (1, 51) = 6.367, p = .015, eta 2 = 0.11] and treatment [ F (1, 51) = 7.112, p = .010, eta 2 = 0.12], and a significant subgroup by treatment interaction effect [[ F (1, 51) = 6.459, p = .014, eta 2 = 0.11]. The interaction (Figure 3-2) may be misleading, however, due to the fact that only two participants in Cluster 1 received multidisciplinary treatment. Closer examination of the main effect of subgroup revealed that participants in Cluster 1 had a lower mean usual level of interference ( M = 32.8) than participants in Cluster 2 ( M = 59.6). Additionally, participants receiving multidisciplinary treatment had a lower mean usual level of interference ( M = 32.0) than participants receiving palliative treatment ( M = 60.4).

PAGE 42

33 PalliativeMultidisciplinaryTreatment 0102030405060708090100Mean Usual Interference Cluster 1 (n = 7) Cluster 2 (n = 63) Figure 3-2. Subgroup by Treatment Interaction for Usual Interference. Finally, a series of chi-square analyses were conducted to examine the relationship between type of treatment (palliative v multidisciplinary) and treatment success as determined by MCID attainment (yes, no) for each of the domains using MCIDs established in Aim 1. For the domain of pain, results showed that type of treatment was unrelated to MCID attainment, Pearson 2 (1, N = 54) = .616, p = .433. In other words, participants who received palliative pain management and participants who received multidisciplinary pain management were equally likely to attain the MCID for pain. Nonsignificant results also were obtained for the domains of fatigue [Pearson 2 (1, N = 53) = .190, p = .663], distress [Pearson 2 (1, N = 54) = 2.030, p = .154], and

PAGE 43

34 interference [Pearson 2 (1, N = 54) = .533, p = .465], suggesting that type of treatment was unrelated to treatment success in this sample.

PAGE 44

CHAPTER 4 DISCUSSION Our study aimed to establish patient-determined MCIDs for treatment of chronic pain. Consistent with previous research using the PCOQ (Robinson et al., in press), we found that spine pain patients require large reductions in pain for treatment to be deemed successful. At the onset of treatment, patients in our study considered a 26.9/100 pain level or a mean reduction in pain of 38.5 points (58%) to represent treatment success. This patient-defined success criterion derived using a direct scaling approach is more stringent than criteria developed using other methods, including expert opinion (e.g., Goldsmith et al., 1993; Moore et al., 1996) and scale comparisons (e.g., Farrar et al., 2001; Angst et al., 2002). Our study expands on previous research by deriving MCIDs for other domains relevant to pain treatment. Our findings suggest spine pain patients also require large reductions in fatigue (23.2/100 or 61%), emotional distress (18.3/100 or 64%), and interference with daily activities (21.6/100 or 66%) for treatment to be considered successful. These pre-treatment MCIDs are highly consistent with those established in our prior study (Robinson et al., in press), suggesting they are reliable and valid. Our findings also confirm that successful treatment of chronic pain is not defined exclusively in terms of pain reduction, but involves a host of endpoints. If patients adhere to these success criteria throughout treatment, then clinical trials that apply expert-generated MCIDs or use only standard statistical methods might overestimate the efficacy of their treatments. Therefore, it is important to know whether 35

PAGE 45

36 chronic pain patients adjust their criteria for treatment success over time. Results of our study suggest patients do indeed modify their success criteria from pre-treatment to post-treatment in the direction of becoming less stringent, as target values for successful treatment of pain, fatigue, and emotional distress were significantly higher two months after treatment initiation (33.9, 30.2, and 26.1, respectively). Based on our findings, it appears that patients used these post-treatment cut-offs to make decisions about treatment success. In our study, patients who met their post-treatment success criterion for a particular domain were significantly more likely than patients who did not meet their post-treatment success criterion for the domain to judge treatment of the domain a success. These results further support the utility of patient-defined success criteria. It should be noted that 48-hour test-retest reliability values for PCOQ success criteria were somewhat low in another sample of 21 chronic pain patients. One possible explanation for this finding is that patients success criteria are influenced by their current levels of pain, fatigue, emotional distress, and interference with daily activities, such that as their current levels naturally fluctuate from day to day so do their success criteria. Fluctuations in success criteria might be larger than corresponding fluctuations in usual levels. Patients also might be less familiar with and experienced at rating success criteria than they are at rating usual levels, thereby introducing more error variance into ratings of success. These explanations are admittedly speculative and somewhat beyond the scope of our study. Despite the questionable test-retest reliability of PCOQ success criteria, mean success criteria for the four PCOQ domains were highly consistent across our two samples of chronic pain patients. As discussed, success criteria derived in this study were almost identical to success criteria derived in our original study (Robinson et

PAGE 46

37 al., in press) aside from the change from a 0 to 10 NRS to a 0 to 100 NRS. Furthermore, in our study, pre-treatment and post-treatment success criteria still were reliably different even with additional error variance incorporated into the analyses. The aforementioned patient-centered success criteria were derived using a direct scaling approach. Another more commonly used approach to developing MCIDs involves comparing two scales, typically the change on a given domain (e.g., change in pain on a 0 to 10 NRS) and a global rating of improvement (e.g., much improved). This methodology has promise, but use of different anchors (e.g., much improved versus somewhat better) has led to differing results (e.g., Farrar et al., 2001; Wells et al., 1993). In fact, Salaffi and colleagues (2004) calculated MCIDs in pain associated with both a slightly better and much better global impression of change among their sample of chronic musculoskeletal pain patients, and obtained different raw and percent changes for the two anchors. Furthermore, investigator-determined anchors might not reflect the degree of improvement necessary for patients to consider treatment successful. Therefore, our study sought to (1) characterize the association between NRS change scores and treatment success as determined by patient global judgment and (2) select as MCIDs the change scores that best distinguished between successfully and unsuccessfully treated patients. This procedure was done not only for the domain of pain, but also for the domains of fatigue, emotional distress, and interference with daily activities. Using the scale comparison approach, we derived the following MCIDs: 17.5 points (on a 0 to 100 NRS) or 25% for pain, 7.5 points (11%) for fatigue, 5.0 points (13%) for emotional distress, and 9.5 points (12%) for interference with daily activities.

PAGE 47

38 These MCIDs accurately classified 73 to 85% of patients as successfully or unsuccessfully treated. MCIDs for pain were most accurate in classifying patients. Interestingly, our MCIDs for pain are fairly consistent with those derived by Farrar et al. (2001) using a 0 to 10 NRS and a much improved global impression of change (2 points or 30%). Future research should focus on validating these patient-determined MCIDs, particularly for the less studied domains of fatigue, emotional distress, and interference with daily activities. Once validated, they can be used to calculate effect size estimates, such as the number needed to treat (Cook & Sackett, 1995), permitting comparisons across studies and ultimately helping to evaluate the efficacy of various treatments for chronic pain (Farrar et al., 2001; Redelmeier et al., 1996). Additionally, MCIDs can be useful in assessing the care of individual patients. Future research should also explore whether success criteria are absolute or related to pre-treatment levels. A recent study by Salaffi and colleagues (2004) suggests MCIDs may not be standardized along the whole NRS. They found that patients with a higher level of pain before treatment required greater reductions in pain to obtain clinically important improvements than did patients with relatively low pain levels before treatment. Consequently, in applying our MCIDs to individual patients or other pain populations, one should note that our sample had moderate to moderately high mean levels (on a 0 to 100 NRS) of pain ( M = 65.4), fatigue ( M = 60.1), emotional distress ( M = 51.1), and interference with daily activities ( M = 64.7) at the onset of treatment. Finally, future research should investigate the evolution of success criteria throughout the course of different treatments. Our results suggest patients change their success criteria while undergoing treatment. However, it

PAGE 48

39 remains unclear whether patients are more or less likely to adjust their success criteria with different treatments. For example, patients receiving cognitive behavioral therapy (CBT) might be expected to change their success criteria considerably, given that the goals of CBT are to examine potential maladaptive beliefs (e.g., unrealistic expectations regarding treatment) and re-define treatment outcomes. A second aim of our study was to empirically derive and validate patient subgroups using patient ratings of the perceived importance of improvement in the four PCOQ domains. Contrary to our hypothesized three-cluster solution, only two clusters emerged. One subgroup (Cluster 2) rated improvement in all domains as extremely important, whereas the other, smaller subgroup (Cluster 1) reported relatively low importance ratings across domains. Further examination of the cluster analysis revealed that the hypothesized multifocused-moderate subgroup merged with the multifocused-high subgroup to form Cluster 2. Patients in Cluster 2 had higher pre-treatment usual levels of pain than patients in Cluster 1. They also were more distressed and disabled at the start of treatment. Interestingly, the clusters did not differ on chronicity of pain, suggesting that pain duration does not influence subgroup membership. Based on these data, one could speculate that patients in Cluster 1 might be less likely than patients in Cluster 2 to require multidisciplinary treatment to obtain clinically important treatment outcomes. Future research might examine this hypothesis. Our findings provide support for the notion that chronic pain patients are heterogeneous. Even among patients whose pathology is specific to the spine, distinct subgroups emerged. Of note, we did not replicate our previous finding of a truly pain-focused subgroup (Robinson et al., in press). The possibility exists that a subset of

PAGE 49

40 patients with chronic pain in other body regions might believe that treatments addressing nociception are most pertinent to them. Future investigations using larger and more diverse samples of chronic pain patients are needed to examine this hypothesis. Another explanation for our inconsistent findings might lie with the variables on which the subgroups were derived. Importance ratings might not be the best criteria on which to base subgroup membership. Other variables, such as usual ratings or success criteria, might produce more stable and meaningful patient subgroups. Therefore, future investigations should consider using other PCOQ variables to develop patient subgroups. Our final aim was to examine the potential for treatment matching using the patient subgroups derived from the PCOQ. Because a pain-focused subgroup did not emerge, we could not fully explore hypotheses regarding treatment matching. Nonetheless, we did analyses to examine treatment effects across the two subgroups derived in Aim 2. Although patients in Cluster 1 were more satisfied with treatment than patients in Cluster 2, they did not achieve greater reductions in their pain, fatigue, emotional distress, and interference with daily activities from pre-treatment to post-treatment. This pattern of findings occurred irrespective of the type of treatment received (palliative pain management or multidisciplinary pain management). Additional analyses showed, at least in our sample, that type of treatment is unrelated to MCID attainment. That is, patients receiving palliative pain management and patients receiving multidisciplinary pain management were equally likely to obtain MCIDs for pain, fatigue, emotional distress, and interference with daily activities. This finding is somewhat surprising given overwhelming evidence for the superiority of multidisciplinary treatment approaches

PAGE 50

41 (Flor, Fydrich, & Turk, 1992). Nonetheless, it further underscores the need to determine what treatments are effective for which subgroups of chronic pain patients. Potential limitations of our study deserve mention. First, our reduced sample size likely prevented a separate multifocused-moderate patient subgroup from emerging in the cluster analysis. Notably, we conducted our original cluster analysis (Robinson et al., in press) on 110 patients, whereas we performed this cluster analysis on 70 patients. Likewise, we could not conduct certain follow-up chi-square analyses, particularly those examining possible subgroup by treatment interactions, because of small cell sizes. A second limitation of our study pertains to the non-uniformity of the two treatment approaches. Ideally, patients receiving palliative pain management would have been given the same treatment (or at least the same class of treatment, pharmacological or interventional) and patients undergoing multidisciplinary pain management would have received identical treatment components. Another potential limitation of our study involves the use of dichotomous judgments of success as the standard against which MCIDs were developed. The possibility exists, for instance, that patients global judgments of treatment success for the domain of pain were influenced by factors unrelated to pain intensity, such as satisfaction with care. Similarly, patients global judgments of treatment success for the other PCOQ domains (fatigue, distress, and interference) might have been unduly influenced by their global judgments for pain. Nevertheless, enabling patients to incorporate into their global judgments of treatment success all factors perceived as germane to the decision produces success criteria that are ecologically valid. A final concern relates to the generalizability of these findings to other chronic pain populations. Patients with chronic spine pain might view treatment success

PAGE 51

42 differently than patients with other types of chronic pain. Therefore, our MCIDs might not be applicable to other painful conditions. However, we are encouraged that our MCIDs for pain are similar to those developed with other pain populations (e.g,. Farrar et al., 2001). These limitations notwithstanding, our study is the first to develop MCIDs from the patient perspective across multiple domains relevant to treatment of chronic pain. Once validated, these MCIDs can easily be used by both clinicians and researchers to evaluate the effectiveness of various treatment approaches for chronic pain. Our study also further debunks the pain patient homogeneity assumption (Turk & Okifuji, 2001) by identifying at least two distinct subgroups of spine pain patients that differ on important variables, such as pain intensity, distress, and disability. It remains to be seen whether additional subgroups will emerge from larger and more diverse samples of chronic pain patients. Finally, as we were unable to adequately examine the treatment-matching hypothesis (Turk & Okifuji, 2001), future research should explore the potential benefits of tailoring treatment to the behavioral and psychosocial characteristics of these and other subgroups of patients with chronic pain.

PAGE 52

APPENDIX A PATIENT-CENTERED OUTCOMES QUESTIONNAIRE (PCOQ) MANY PEOPLE EXPERIENCE PAIN FATIGUE (I.E., FEELING TIRED), EMOTIONAL DISTRESS (E.G., WORRIES, FEELING SAD), AND INTERFERENCE WITH DAILY ACTIVITIES (E.G., NOT BEING ABLE TO WORK OR DO HOUSEHOLD CHORES) AS A RESULT OF THEIR MEDICAL CONDITION. WE WOULD LIKE TO UNDERSTAND HOW YOU HAVE BEEN IMPACTED IN EACH OF THESE AREAS. WE WOULD ALSO LIKE TO LEARN MORE ABOUT WHAT YOU WANT YOUR TREATMENT TO DO FOR YOU. ________________________________________________________________________ FIRST, WE WOULD LIKE TO KNOW YOUR USUAL LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE. On a scale of 0 (none) to 100 (worst imaginable), please indicate your usual level (during the past week) of pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ NOW, WE WOULD LIKE TO LEARN ABOUT YOUR DESIRED LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE. IN OTHER WORDS, WE WOULD LIKE TO UNDERSTAND WHAT YOUR IDEAL TREATMENT OUTCOME WOULD BE. On a scale of 0 (none) to 100 (worst imaginable), please indicate your desired level of pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ 43

PAGE 53

44 ________________________________________________________________________ PATIENTS UNDERSTANDABLY WANT THEIR TREATMENT TO RESULT IN DESIRED OR IDEAL OUTCOMES LIKE YOU INDICATED ABOVE. UNFORTUNATELY, AVAILABLE TREATMENTS DO NOT ALWAYS PRODUCE DESIRED OUTCOMES. THUS, IT IS IMPORTANT FOR US TO UNDERSTAND WHAT TREATMENT OUTCOMES YOU WOULD CONSIDER SUCCESSFUL. On a scale of 0 (none) to 100 (worst imaginable), please indicate the level each of these areas would have to be at for you to consider treatment successful. pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ NOW, WE WOULD LIKE TO KNOW WHAT YOU EXPECT YOUR TREATMENT TO DO FOR YOU. On a scale of 0 (none) to 100 (worst imaginable), please indicate the levels you expect following treatment. pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ FINALLY, WE WOULD LIKE TO UNDERSTAND HOW IMPORTANT IT IS FOR YOU TO SEE IMPROVEMENT IN YOUR PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE FOLLOWING TREATMENT. On a scale of 0 ( not at all important ) to 100 ( most important ), please indicate how important it is for you to see improvement in your pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____

PAGE 54

APPENDIX B FOLLOW-UP PCOQ ________________________________________________________________________ FIRST, WE WOULD LIKE TO KNOW YOUR PRE-TREATMENT LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE. On a scale of 0 (none) to 100 (worst imaginable), please estimate your pre-treatment level of pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ BASED ON YOUR PRE-TREATMENT LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE, PLEASE INDICATE WHAT TREATMENT OUTCOMES YOU WOULD CONSIDER SUCCESSFUL. On a scale of 0 (none) to 100 (worst imaginable), please indicate the level each of these areas would have to be at for you to consider treatment successful pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ 45

PAGE 55

46 ________________________________________________________________________ NEXT, WE WOULD LIKE TO UNDERSTAND HOW IMPORTANT IT IS FOR YOU TO SEE IMPROVEMENT IN YOUR PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE. On a scale of 0 ( not at all important ) to 100 ( most important ), please indicate how important it is for you to see improvement in your pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ FINALLY, WE WOULD LIKE TO KNOW YOUR USUAL LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE FOLLOWING TREATMENT. IN OTHER WORDS, WHAT ARE YOUR CURRENT LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE? On a scale of 0 (none) to 100 (worst imaginable), please indicate your usual level (during the past week) of pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____

PAGE 56

47 ACCORDING TO OUR RECORDS, YOUR PRE-TREATMENT LEVELS OF PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE [ON A SCALE FROM 0 (NONE) TO 100 (WORST IMAGINABLE)] WERE AS FOLLOWS: pain _____ fatigue (or tiredness) _____ emotional distress _____ interference with daily activities _____ ________________________________________________________________________ NOW WE WOULD LIKE TO KNOW IF YOU HAVE IMPROVED, STAYED THE SAME, OR WORSENED SINCE THE START OF TREATMENT. 1. Has your pain improved, stayed the same, or worsened? Please choose one: _____ Improved (Please complete Part A below.) _____ Stayed the same (Please move on to #2.) _____ Worsened (Please complete Part B below.) Part A: On a scale of 0 (no improvement) to 100 (complete improvement), please indicate how much your pain has improved _____ (Please move on to #2.) Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please indicate how much your pain has worsened _____ (Please move on to #2.) 2. Has your fatigue (or tiredness) improved, stayed the same, or worsened? Please choose one: _____ Improved (Please complete Part A below.) _____ Stayed the same (Please move on to #3.) _____ Worsened (Please complete Part B below.) Part A: On a scale of 0 (no improvement) to 100 (complete improvement), please indicate how much your fatigue has improved _____ (Please move on to #3.) Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please indicate how much your fatigue has worsened _____ (Please move on to #3.)

PAGE 57

48 3. Has your emotional distress improved, stayed the same, or worsened? Please choose one: _____ Improved (Please complete Part A below.) _____ Stayed the same (Please move on to #4.) _____ Worsened (Please complete Part B below.) Part A: On a scale of 0 (no improvement) to 100 (complete improvement), please indicate how much your emotional distress has improved _____ (Please move on to #4.) Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please indicate how much your emotional distress has worsened _____ (Please move on to #4.) 4. Has your interference improved, stayed the same, or worsened? Please choose one: _____ Improved (Please complete Part A below.) _____ Stayed the same (Please move on to #5.) _____ Worsened (Please complete Part B below.) Part A: On a scale of 0 (no improvement) to 100 (complete improvement), please indicate how much your interference has improved _____ (Please move on to #5.) Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please indicate how much your interference has worsened _____ (Please move on to #5.) 5. Overall, have you improved, stayed the same, or worsened? Please choose one: _____ Improved (Please complete Part A below.) _____ Stayed the same (Please move on to the next page.) _____ Worsened (Please complete Part B below.) Part A: On a scale of 0 (no improvement) to 100 (complete improvement), please indicate how much you have improved _____ (Please move on to the next page.) Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please indicate how much you have worsened _____ (Please move on to the next page.)

PAGE 58

49 ________________________________________________________________________ FINALLY, WE WOULD LIKE TO LEARN WHETHER YOU WOULD CONSIDER YOUR TREATMENT SUCCESSFUL. Was treatment of your pain successful? Please choose one: _____ Yes _____ No Was treatment of your fatigue (or tiredness) successful? Please choose one: _____ Yes _____ No Was treatment of your emotional distress successful? Please choose one: _____ Yes _____ No Was treatment of your interference with daily activities successful? Please choose one: _____ Yes _____ No Overall, was your treatment successful? Please choose one: _____ Yes _____ No

PAGE 59

LIST OF REFERENCES Angst, F., Aeschlimann, A., Michel, B. A., & Stucki, G. (2002). Minimal clinically important rehabilitation effects in patients with osteoarthritis of the lower extremities. J Rheumatol, 29 131-138. Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation. Clin Psychol Rev, 8 77-100. Casarett, D., Karlawish, J., Sankar, P., Hirschman, K., Asch, D. A. (2001). Designing pain research from the patients perspective: What trial end points are important to patients with chronic pain? Pain Med, 2 309-316. Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2 nd ed). Hillsdale, NJ: Lawrence Erlbaum Associates. Cook, R. J., & Sackett, D. L. (1995). The number needed to treat: A clinically useful measure of treatment effect. BMJ, 310 452-454. Daniel, M., Long, C., Murphy, W., Kores, R., and Hutcherson, W. (1983). Therapists and chronic pain patients perceptions of treatment outcome. J Nerv Ment Dis, 171 729-733. Department of Health and Human Services, National Institutes of Health. (1998, September 4). New directions in pain research: I Retrieved July 28, 2005, from http://grants2.nih.gov/grants/guide/pa-files/PA-98-102.html Dreyfuss, P., Halbrook, B., Pauza, K., Josi, A., McLarty, J., & Bogduk, N. (2000). Efficacy and validity of radiofrequency neurotomy for chronic lumbar zygapophysial joint pain. Spine, 25 1270-1277. Dworkin, S. F., & LeResche, L. (1992). Research diagnostic criteria for temporomandibular disorders: Review, criteria, examination and specifications, critique. J Craniomandib Disord, 6 301-355. Farrar, J. T. (2000). What is clinically meaningful: Outcome measures in clinical pain trials? Clin J Pain, 16 S106-S112. Farrar, J. T., Berlin, J. A., & Strom, B. L. (2003). Clinical important changes in acute pain outcome measures: A validation study. J Pain Symptom Manage, 25 406-411. 50

PAGE 60

51 Farrar, J. T., Portenoy, R. K., Berlin, J. A., Kinman, J. L., & Strom, B. L. (2000). Defining the clinically important difference in pain outcome measures. Pain, 88 287-294. Farrar, J. T., Young, J. P., LaMoreaux, L., Werth, J. L., & Poole, R. M. (2001). Clinical importance of changes in chronic pain intensity measured on an 11-point numerical pain rating scale. Pain, 94 149-158. Fischer, D., Stewart, A. L., Bloch, D. A., Lorig, K., Laurent, D., & Holman, H. (1999). Capturing the patients view of change as a clinical outcome measure. JAMA, 282 1157-1162. Flor, H., Fydrich, T., & Turk, D. C. (1992). Efficacy of multidisciplinary pain treatment centers: A meta-analytic review. Pain, 49 221-230. Geisser, M.E., Roth, R. S., & Robinson, M. E. (1997). Assessing depression among persons with chronic pain using The Center for Epidemiological Studies-Depression Scale and the Beck Depression Inventory: A comparative analysis. Clin J Pain, 13, 163-170. Gilliam, F., Kuzniecky, R., Meador, K., Martin, R., Sawrie, S., Viikinsalo, M., Morawetz, R., & Faught, E. (1999). Patient-oriented outcome assessment after temporal lobectomy for refractory epilepsy. Neurology, 53 687-694. Goldsmith, C. H., Boers, M., Bombardier, C., & Tugwell, P. (1993). Criteria for clinically important changes in outcomes: Development, scoring, and evaluation of rheumatoid arthritis patient and trial profiles. J Rheumatol, 20 561-565. Guck, T., Meilman, P., Skultety, F., & Poloni, L. (1988). Pain-patient Minnesota Multiphasic Personality Inventory (MMPI) subgroups: Evaluation of long-term treatment outcome. J Behav Med, 11 159-169. Jamison, R. N., Rock, D. L., & Parris, W. C. V. (1988). Empirically derived Symptom Checklist-90 subgroups of chronic pain patients: A cluster analysis. J Behav Med, 11 147-158. Jensen, M. P., Strom, S. E., Turner, J., & Romano, J. M. (1992). Validity of the Sickness Impact Profile Roland scale as a measure of dysfunction in chronic pain patients. Pain, 50, 157-162. Keefe, F. J. (1996). Cognitive behavior therapy for managing pain. Clin Psychol, 49 (3), 4-5. Keefe, F. J., Bradley, L. A., & Crisson, J. E. (1990). Behavioral assessment of low back pain: Identification of pain behavior subgroups. Pain, 40 153-160. Lane, C., & Davidoff, F. (1996). Patient-centered medicine: A professional evolution. JAMA, 275 152-156.

PAGE 61

52 Lindstrom, I., Ohlund, C., Eek, C., Wallin, L., Peterson, L. E., Fordyce, W. E., & Nachemson, A. L. (1992). The effect of graded activity on patients with subacute low back pain: A randomized prospective clinical study with an operant-conditioning behavioral approach. Phys Ther, 72 279-290. Mannion, A. F., Muntener, M., Taimela, S., & Dvorak, J. (2001). Comparison of three active therapies for chronic low back pain: Results of a randomized clinical trial with one-year follow-up. Rheumatology, 40 ,772-778. McCracken, L. M., Klock, A., Mingay, D., Asbury, J. K., & Sinclair, D. M. (1997). Assessment of satisfaction with treatment for chronic pain. J Pain Symptom Manage, 14 (5), 292-299. McCracken, L.M., & Turk, D.C. (2002). Behavioral and cognitive-behavioral treatment for chronic pain. Spine, 27 2564-2573. McCracken, L. M., Zayfert, C., & Gross, R. T. (1992). The Pain Anxiety Symptoms Scale: Development and validation of a scale to measure fear of pain. Pain, 50, 67-73. McGill, J., Lawlis, G., Selby, D., Mooney, V., & McCoy, C. (1983). The relationship of Minnesota Multiphasic Personality Inventory (MMPI) profile clusters to pain behaviors. J Behav Med, 6 77-92. Melzack, R., & Katz, J. (2001). The McGill Pain Questionnaire: Appraisal and current status. In D. C. Turk & R. Melzack (Eds.), Handbook of pain assessment (2 nd ed., pp. 35-52). New York: The Guilford Press. Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150 971-979. Moore, A., McQuay, H., & Gavaghan, D. (1996). Deriving dichotomous outcome measures from continuous data in randomised controlled trials of analgesics. Pain, 66 229-237. Moore, J. E., Armentrout, D. P., Parker, J. C., & Kivlahan, D. R. (1986). Empirically derived pain-patient MMPI subgroups: Prediction of treatment outcome. J Behav Med, 9 51-63. North, R. B., Han, M., Zahurak, M., & Kidd, D. (1994). Radiofrequency lumbar facet denervation: Analysis of prognostic factors. Pain, 57 77-83. Pollard, C. A. (1984). Preliminary validity study of the Pain Disability Index. Percept Mot Skills, 59 (3), 974. Price, D. D., McGrath, P.A., Rafii, A. and Buckingham, B. (1983). The validations of visual analogue scales as ratio scale measures for chronic and experimental pain. Pain, 17 45-56.

PAGE 62

53 Redelmeier, D. A., Guyatt, G. H., & Goldstein, R. S. (1996). Assessing the minimal important difference in symptoms: A comparison of two techniques. J Clin Epidemiol, 49 1215-1219. Redelmeier, D. A., & Lorig, K. (1993). Assessing the clinical importance of symptomatic improvements. Arch Intern Med, 153 1337-1342. Robinson, M. E., Brown, J. L., George, S. Z., Edwards, P. S., Atchison, J. W., Hirsh, A. T., Waxenberg, L. B., Wittmer, V., & Fillingim, R. B. (in press). Multidimensional success criteria and expectations for treatment of chronic pain: The patient perspective. Pain Med Roland, M., & Morris, R. (1983). A study of the natural history of back pain. Part I: Development of a reliable and sensitive measure of disability in low-back pain. Spine, 8, 141-144. Rudy, T. E., Turk, D. C., Kubinski, J. A., & Zaki, H. S. (1995). Differential treatment responses of TMD patients as a function of psychological characteristics. Pain, 61 103-112. Salaffi, F., Stancati, A., Silvestri, C. A., Ciapetti, A., & Grassi, W. (2004). Minimal clinically important changes in chronic musculoskeletal pain intensity measured on a numerical rating scale. Eur J Pain, 8 283-291. Schwartz, C. E., Cole, B. F., & Gelber, R. D. (1995). Measuring patient-centered outcomes in neurologic disease. Arch Neurol, 52 754-762. Slater, M. A., Doctor, J. N., Pruitt, S. D., & Atkinson, J. H. (1997). The clinical significance of behavioral treatment for chronic low back pain: An evaluation of effectiveness. Pain, 71 257-263. Stewart, W., Ricci, J., Chee, E., Morganstein, D., & Lipton, R. (2003). Lost productive time and cost due to common pain conditions in the US workforce. JAMA, 290 2443-2454. Tait, R. C., Pollard, C. A., Margolis, R. B., Duckro, P. N., & Krause, S. J. (1987). The Pain Disability Index: Psychometric and validity data. Arch Phys Med Rehabil, 68 438-441. Turk, D. C. (2002). Clinical effectiveness and cost-effectiveness of treatment of patients with chronic pain. Clin J Pain, 18 355-365.

PAGE 63

54 Turk, D. C., Dworkin, R. H., Allen, R. R., Bellamy, N., Brandenburg, N., Carr, D. B., Cleeland, C., Dionne, R., Farrar, J. T., Galer, B. S., Hewitt, D. J., Jadad, A. R., Katz, N. P., Kramer, L. D. Manning, D. C., McCormick, C. G., McDermott, M. P., McGrath, P., Quessy, S., Rappaport, B. A., Robinson, J. P., Royal, M. A., Simon, L., Stauffer, J. W., Stein, W., Tollett, J., & Witter, J. (2003). Core outcome domains for chronic pain clinical trials: IMMPACT recommendations. Pain, 106 337-345. Turk, D. C., & Melzack, R. (Eds.). (2001). Handbook of pain assessment (2 nd ed.). New York: Guilford Press. Turk, D. C., & Okifuji, A. (2001). Matching treatment to assessment of patients with chronic pain. In D. C. Turk & R. Melzack (Eds.), Handbook of pain assessment (2 nd ed., pp. 400-414). New York: Guilford Press. Turk, D. C., Okifuji, A., Starz, T., & Sinclair, J. (1998). Differential responses by psychosocial subgroups of fibromyalgia syndrome patients to an interdisciplinary treatment. Arthritis Care Res, 11 397-404. Turk, D. C., & Rudy, T. E. (1988). Toward an empirically derived taxonomy of chronic pain patients: Integration of psychological assessment data. J Consult Clin Psychol, 56 233-238. Turk, D. C., Rudy, T. E., Kubinski, J. A., Zaki, H. S., & Greco, C. M. (1996). Dysfunctional patients with temporomandibular disorders: Evaluating the efficacy of a tailored treatment protocol. J Consult Clin Psychol, 64 139-146. Turk, D. C., & Zaki, H. S., & Rudy, T. E. (1993). Effects of intraoral appliance and biofeedback/stress management alone and in combination in treating pain and depression in patients with temporomandibular disorders. J Prosthet Dent, 70 158-164. Wells, G. A., Tugwell, P., Kraag, G. R., Baker, P. R. A., Groh, J., & Redelmeier, D. A. (1993). Minimum important difference between patients with rheumatoid arthritis: The patients perspective. Journal Rheumatol, 20 557-560. Zatzick, D. F., Kang, S., Hinton, W. L., Kelly, R. H., Hilty, D. M., Franz, C. E., Le, L., & Kravitz, R. L. (2001). Posttraumatic concerns: A patient-centered approach to outcome assessment after traumatic physical injury. Med Care, 39 327-339.

PAGE 64

BIOGRAPHICAL SKETCH Jennifer Lynn Brown was born in Alexandria, Virginia on November 24, 1974. She graduated from Potomac Senior High School in 1992, and received her Bachelor of Science degree in psychology from the University of Mary Washington in 1996. After earning a Master of Arts degree in psychology from Wake Forest University in 1998, she worked as a Research Associate for the American Institutes for Research in Washington, DC and then as an Associate for Caliber Associates in Fairfax, Virginia. In 2000, Jennifer entered the Department of Clinical and Health Psychologys doctoral program at the University of Florida (UF) with a concentration in clinical health psychology. While at UF, she worked for the Center for Pain Research and Behavioral Health under the mentorship of Michael E. Robinson, Ph.D. She is completing her predoctoral clinical internship at the VA Connecticut Healthcare System in West Haven, Connecticut. 55


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E20101123_AAAABX INGEST_TIME 2010-11-23T10:50:02Z PACKAGE UFE0011653_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES
FILE SIZE 3899 DFID F20101123_AABAOG ORIGIN DEPOSITOR PATH brown_j_Page_54thm.jpg GLOBAL false PRESERVATION BIT MESSAGE_DIGEST ALGORITHM MD5
82df9351fbadc30d46b7ce03eef49f78
SHA-1
1bf00eb8aab0b31ea8178ad970706a7e30883999
1772 F20101123_AABAJJ brown_j_Page_57.txt
38cfa0b5b421fd50ad4d9b720de5e407
4c129c243d953c7979fe1d6b0f55ee3d2481a2ad
1053954 F20101123_AABAEL brown_j_Page_49.tif
6c5f8c3cc84d379321c59512669bee43
b3a289334fab1b96be4c73929bfc164d74e7daa0
74891 F20101123_AAAZYJ brown_j_Page_51.jp2
ffefc43fc47d2a9226c57f7867bc11ff
6ec3f61f3a12d57f76434bf7726d96db443b83f4
720 F20101123_AABAJK brown_j_Page_58.txt
1fe964983092827a768d064bd920a05c
cf8c26273e068533857ff1fcf9f20422a1b56d0f
F20101123_AABAEM brown_j_Page_50.tif
41462900f1348f5c1fb5ab3cbc74b463
e713a9ca58ee31841ca2a3e4c1c6cfb514766f5b
6172 F20101123_AAAZYK brown_j_Page_63thm.jpg
1a17164485fb552882d6716621f96beb
4e04be98bf75734b89ee5d0fa4ff0f7be7d65c47
2132 F20101123_AABAJL brown_j_Page_59.txt
a75ca5b1a9482904539fb83a334f361e
cb43b60f764929198e9cd69ec22bdca4a5e95295
F20101123_AABAEN brown_j_Page_51.tif
4a824eae67c078ac00a617e50a9c25f7
69887b13f29d39e8fce43dc891803814d68cef82
75556 F20101123_AAAZYL UFE0011653_00001.mets FULL
94b837d8dd0a5414c3c3e2901b317e1e
f45ff20e0aad9543256263b708e84fd09655ef8c
2552 F20101123_AABAJM brown_j_Page_60.txt
7d63fbed0f04be9ba2e31b816cfdd0af
5b7a549ed1572bd96b334427c6e138607665fc45
F20101123_AABAEO brown_j_Page_52.tif
9f45a267e98702fec91634f938e1f7a2
ab4b6344ce0f7f26e97c17352b1c8623fbfc5ad1
2463 F20101123_AABAJN brown_j_Page_61.txt
4f93696ba34880aabf32b9d7ed1dda15
a0d5051a8ceaec78dd2e60e80ba5a898456af3f9
F20101123_AABAEP brown_j_Page_53.tif
c2e6c9b5bcbda7270e1b0adc96fe7cb0
d821a5f9969442bc8471643c4dd11fcbacc796eb
2173 F20101123_AABAJO brown_j_Page_62.txt
0d478423ad619fa577ae92d86ebeceac
2cb99c6ff5a958b24287155b509bac4986597b3a
F20101123_AABAEQ brown_j_Page_54.tif
af308e0ac5d9f2c1f67530e3063791a7
2701c64c52312dcbc396e2b9ac784fabb6f69fb1
25124 F20101123_AAAZYO brown_j_Page_01.jpg
017756b97e51468fa81958f064aab390
ea7fa4375107ceb5a52bec565fc16e3795fb37b2
F20101123_AABAER brown_j_Page_55.tif
e9b5d475ae9aadc66d8b420dfe8e78e8
49347e6e3ea99b7564c509ae1ed705a94ce3bb14
10499 F20101123_AAAZYP brown_j_Page_02.jpg
ed3843840a5685273735f815685d4f2e
fc0aceaa55e6e51ed03dca48995d9d84ff9196ce
2343 F20101123_AABAJP brown_j_Page_63.txt
673a7834fdb8cce3efd07ee2fd3a2566
e9de015a5842a061de17989c52b0f507ccdf3f1d
F20101123_AABAES brown_j_Page_56.tif
b0ab05142730557994c08a3ad567dc5d
f6fc04c4f09197953e7a32bd1d2db1f0a5483802
37239 F20101123_AAAZYQ brown_j_Page_03.jpg
1ec878e2bfe22f64be4cfc7904fc962a
addfdf68200b42be1edcbda2bcfcc7559fb1bfdb
1034 F20101123_AABAJQ brown_j_Page_64.txt
b69618e3970a348622f6832e27dd65ba
1cd86b45e016c43968b0942aeba55670f393806d
F20101123_AABAET brown_j_Page_57.tif
f437781189dc97d4920eeefd8e875a81
211db4bf880076d849a844ee70166482f617dfd4
65816 F20101123_AAAZYR brown_j_Page_04.jpg
e89a536abb47864d5aa5fa12db54d4f0
dc3b27a7ad2df6decb7466f3170909b2af8a6cf2
412300 F20101123_AABAJR brown_j.pdf
ba7472028ee8be53f62148324c551dbf
c02ad9145e66216f71725f818233340ed9b3b1ba
F20101123_AABAEU brown_j_Page_59.tif
894067ce6cc0c035d08d9ee557b02bc3
eb2fa8c21b762bd1f99e8ad54bb15242dfa6cd9e
29637 F20101123_AAAZYS brown_j_Page_05.jpg
4beb94a2eba91ce40dad32b04d0be3a7
7dafb27fe86f418929ae8f46e09f7c03dce58bfb
21995 F20101123_AABAJS brown_j_Page_62.QC.jpg
f5196b73653e2b400b818fc263128d39
a09d3e795ae2e31bd7e9809efa35c376f42451ac
F20101123_AABAEV brown_j_Page_60.tif
949ce57ff74c3ce257b0f6b9224efad9
89da7383fe459075df361679b69cdf38754d930a
15660 F20101123_AAAZYT brown_j_Page_07.jpg
00925808326a9ddc8180a30f06e066fb
929d1d512893553168b83dbaba0d9c9451a62b21
6334 F20101123_AABAJT brown_j_Page_28thm.jpg
f1b4526a4c82bf3ecb319d0de22028e1
228bc944e569c494eb5d26622c4bebec69c0ec78
F20101123_AABAEW brown_j_Page_62.tif
0397b8bd5b7f57e8ba8ffbb9c1b207a0
473f9d434d82a895709116a53f2de2f0fde3fa23
59890 F20101123_AAAZYU brown_j_Page_08.jpg
b28060544476aea1f833c54f1f7d9ada
5eb9ed3b183969a81861f5fdeb0e473ff30e3217
23548 F20101123_AABAJU brown_j_Page_21.QC.jpg
6a7581b908d9ed6f4678ca0ad6602af0
15d497460d839cc7a49c355915eb5509436c5113
F20101123_AABAEX brown_j_Page_63.tif
271eff95668e0794d22648b02bddd30d
22de8df23f88b0dd3e28112dc71d1649d8b9d438
55699 F20101123_AAAZYV brown_j_Page_09.jpg
3420ae17c5a0b610b1c1b3afb71f3370
f40fee31a709e71c4d83b54b8c05ba7ce07da7c4
5013 F20101123_AABAJV brown_j_Page_51thm.jpg
5f323b59ff5cadf6e52cb0ef3788c030
aef8064b9c0db0216e95398f8cd75fcc2c0e66b3
54860 F20101123_AABACA brown_j_Page_39.jp2
ce3461a9faee8e469144bdbcca81afee
f7626027e1a5ba2f592c85bd0d7755a4c67dac62
F20101123_AABAEY brown_j_Page_64.tif
5580fa04edfc854ec69e16554e1645cf
e7c483cf3adcd8b5b579f1574bce303c355ab5a1
67144 F20101123_AAAZYW brown_j_Page_10.jpg
1bdad53115a51375479e52938a82948b
c8682aec5feb7c218f0a60976ed545175538379a
24016 F20101123_AABAJW brown_j_Page_50.QC.jpg
81848b3faa0cf7db772813829dffdbe9
621b7c8a2b1438f413d5eb827191dbc3cc377853
91569 F20101123_AABACB brown_j_Page_40.jp2
8c5f7af5a3b3c34bb8bedbbecee2941f
b0d7feff4570766c4bd7daf1fd26ae15e0b63b45
9111 F20101123_AABAEZ brown_j_Page_01.pro
41865b1c4de009cbd42c99c170d97d88
6eb7644306d1e1c279f69e976d7e13edc93d56f2
72457 F20101123_AAAZYX brown_j_Page_11.jpg
0c7766734172ae6a76fee024fd2119de
e2eff1583c1a28ddc6d9b397aab7de80fd9f8c1f
22965 F20101123_AABAJX brown_j_Page_63.QC.jpg
87792b5a1664ce9d78eed50668a2e8c5
9ad1d9340804697b0d41e5951b3f1698173f35a3
107545 F20101123_AABACC brown_j_Page_41.jp2
2e007cf42ddb098ad762de10c0505bb0
7d3ef441f473174988ede05b9b810f59de3713d3
72374 F20101123_AAAZYY brown_j_Page_13.jpg
e033ee922bbe2900fe3ff1a4e551c000
f5fbd5e27942e096cdef973d7f9b4e4cefe48d90
25252 F20101123_AABAJY brown_j_Page_38.QC.jpg
5c08760422da9d58635b018d437bc764
a307f36ab5b557810c7ef4da6232f5e55bbdb827
58474 F20101123_AABACD brown_j_Page_42.jp2
d920bc3c4df6c3c4f12e5975a8e82386
22d61bab1641eccf62397479faffab3d010affcd
71528 F20101123_AAAZYZ brown_j_Page_14.jpg
4d600fedf1269accc922b22b40083a16
01f761c48a6e5043d152454888fa12eddf3dc778
37035 F20101123_AABAHA brown_j_Page_56.pro
87680c4953f916b04c9efd79bcb758ca
df21cef53db374b374925e36833442cabbdbeecc
5210 F20101123_AABAJZ brown_j_Page_37thm.jpg
2c0df7ef079eded46381d27b532ff0fd
a4074c69c7b9afb0076b7a25e8c47b43ae66a7fc
11717 F20101123_AABACE brown_j_Page_43.jp2
8d35ed897f41d7e259733d0b8db5b888
45a25282d025f3acf5c03dcb6d7125a4a649ba80
41440 F20101123_AABAHB brown_j_Page_57.pro
b26b7b13b6a95b5ee4b9f8a71de17d8e
3b0489e36553d790d9bb11df8822d191f03fe55d
96030 F20101123_AABACF brown_j_Page_44.jp2
be1b94003b949a4da6cdea5525cfdf12
ce6816830da9b314d3dac879c2883d8b1f2d87dc
52343 F20101123_AABAHC brown_j_Page_59.pro
2502264e17245ace03617f6a4c0f6b4b
7a0f120d195f6fb5d1f15fb48cae884383bb0cfc
114220 F20101123_AABACG brown_j_Page_45.jp2
d57361becc3d6f83970f83efeec30ffd
5fe3880c9b935ef279c321ef980ceed9d5dfbd67
19175 F20101123_AABAMA brown_j_Page_23.QC.jpg
b418a56674cc9fe1c4cb3c618f7a20e6
72460b13f133646574f9d600593627e6c6c42cfa
62623 F20101123_AABAHD brown_j_Page_60.pro
247c93e7c82c27c462a08b298f0dde87
5c7646ecfa9e3a42d5d33f0793978ebe82553607
109588 F20101123_AABACH brown_j_Page_47.jp2
4ca8c9b9a813384365ba1d38be1332fa
c1f68a2633a36dba74c3d4eb0580c5b17f79c2ef
6170 F20101123_AABAMB brown_j_Page_62thm.jpg
75d676a222217120da0e925979ef427f
9823b0f494e4a370ddaf2a575630ec99a0730f10
60652 F20101123_AABAHE brown_j_Page_61.pro
846de8c94349e7232c5ebd7291ae67c9
aef13bceabe7b00767d0e85aff3ef51188ce19be
4160 F20101123_AABAMC brown_j_Page_64thm.jpg
5469f05fcc7101732b59f79698affc12
7a8201e146c597d2a88720bce31ecbf21d94fb27
53031 F20101123_AABAHF brown_j_Page_62.pro
2d8dcb577432961e2e15a573871f267c
d23548a29d1927efca0a77a444e1403c36980bad
113065 F20101123_AABACI brown_j_Page_50.jp2
cfea27f0f9d778f3af46258334780654
31babdfb3ac987955e2622dcf5d02947860bc38e
6093 F20101123_AABAMD brown_j_Page_59thm.jpg
5a55ee3d88d141bd8609f2f4f19264f3
361302ba814575a9b874992360fd13110476ae93
56945 F20101123_AABAHG brown_j_Page_63.pro
353820e29b965d1d49da7689c67de1e5
1fd21bff3143b842ae5e62ec61ea6be2ceb42788
77605 F20101123_AABACJ brown_j_Page_52.jp2
f513e08e4b7c53517800c98657f1d8cd
ab3ab64c84fb62005508884992cf3e00c72280ec
6401 F20101123_AABAME brown_j_Page_17thm.jpg
638d3f4af70f27aa14d1c66e562f5d0d
e8f88c7ac8c0c878946dce16a2c559411bb6cece
24890 F20101123_AABAHH brown_j_Page_64.pro
46e5a6f70c79fff2d12dbd81cb6d4420
578ffbebf284cebc10b69aacdc312e3a568e6767
79197 F20101123_AABACK brown_j_Page_53.jp2
61fe00ecea06bbe872c7518b6d2247e1
03e0962517b88254055d50cb3615d6a4eff34c66
6221 F20101123_AABAMF brown_j_Page_24thm.jpg
ec60e5d8e8f80812e3f2b4305e170373
60928bd57167673bccc35af783f3412c8588f982
114 F20101123_AABAHI brown_j_Page_02.txt
025f6112458c1ed4be06763895179d0e
9818ba5fcca28430449b1ecf76a557bcf99044f6
49449 F20101123_AABACL brown_j_Page_54.jp2
37c5dd92c69a7b8e2ffc6da30ab790df
745a4cbcddfa4d56913a577388a20cfd1cd1aa92
6268 F20101123_AABAMG brown_j_Page_32thm.jpg
dcb2d3381f24e9e7a99d12cf4ade1914
e31d3163f61d67f21735466cc205c826ddc89420
902 F20101123_AABAHJ brown_j_Page_03.txt
e1ee4abc4a2d427cb207670d246f30fd
7c579a245feaa94ece12f701bc2af07ba6207f8b
88750 F20101123_AABACM brown_j_Page_56.jp2
09a6412d1629394b28e8c7dccf27e7d9
dab15e9be27a698b32e227e811c95720c7371627
12057 F20101123_AABAMH brown_j_Page_39.QC.jpg
78bc4c4c2c423b8de8be354d831db4fd
887275c0523b401fb2bb1f9b2c79515c8339b6ed
3117 F20101123_AABAHK brown_j_Page_04.txt
b2f23165226b949a0e27cfa1205aad67
215fb684a124bf36b71ba269440c7e3a1a5c2a97
18524 F20101123_AABAMI brown_j_Page_08.QC.jpg
ba1758693f5debc2e7af178ea937b078
05298f862de23e172c0c5508b2b5353210df1e62
93866 F20101123_AABACN brown_j_Page_57.jp2
900673d06fc980278334edb36ee22522
0c15c93b15f0593f20884f6d5606bb5fa0f52755
1028 F20101123_AABAHL brown_j_Page_05.txt
ba1166b1cb0654aec34853f17af1e78c
5975ee351c321143d5dfa53fb1d1af9246f27e53
3380 F20101123_AABAMJ brown_j_Page_22thm.jpg
4c9be6c3d66fb907027d23754cf64136
d80ec9c7b5a04bf72c956e6d7558d73610afea3b
1089 F20101123_AABAHM brown_j_Page_06.txt
878cac3575990f30d025a25dc1342470
7abb4bbc21b68d83d24e0c504a38fd916f5870b7
108423 F20101123_AABACO brown_j_Page_59.jp2
8ca954986adc186e904ff669d66dcfd8
b352aa347fc5a394e016a3ddd7e4f0f5ac436f6a
23636 F20101123_AABAMK brown_j_Page_47.QC.jpg
f5b42f1f1c4ae1f46c29551b68fca6ec
dbb893be606e7b376aa98770822ac9249cd05166
127328 F20101123_AABACP brown_j_Page_60.jp2
db18cc63daf4130560920aa6cb26e88e
bec575b62e0c5bd3bae6e2050b75b6be30503491
11408 F20101123_AABAML brown_j_Page_06.QC.jpg
1385c745d16f23f57934312d644d3ddd
c8b65ac8c4068ef0238e45f5dce19032a83eb09c
373 F20101123_AABAHN brown_j_Page_07.txt
88e5b115c0228bc545c758447e207336
54305653c6210caf62bd42146cc024935920b0db
126498 F20101123_AABACQ brown_j_Page_61.jp2
5ebf103b95b496703158847f128a74fb
9e42189054db8ebc54c85f8b9b520e799973fac1
3978 F20101123_AABAMM brown_j_Page_55thm.jpg
fc3c9db102f7f73e20fa9df5ce0022ed
7fd7502b05fb9a7d72d1b42ada8f32d016201350
1689 F20101123_AABAHO brown_j_Page_08.txt
b9a592fbb482b0e0fc112ffd5405fdb0
9c3e7cb8ba5015fbeaf5390290380b2585afa57d
109061 F20101123_AABACR brown_j_Page_62.jp2
3f1a84cbb4c92aa23834827431e29c8e
c120b382392094bf606c71139b2e66f311c0cae2
24197 F20101123_AABAMN brown_j_Page_45.QC.jpg
71950779fcc5880287366aa8901bf0b0
4a465bc38aed59984e72541eb6f757b0c26e2a77
1529 F20101123_AABAHP brown_j_Page_09.txt
89789324c6f324c80fbec005774a09b9
ed78dff7d483061e83f43c078dd35084300a0ecc
114353 F20101123_AABACS brown_j_Page_63.jp2
4c7fdb5d48f6d2f12a38a10decaf3499
58dd983e595b3241ef7aa6c5bf9a1312f1a21fff
6995 F20101123_AABAMO brown_j_Page_60thm.jpg
ea2fc20c9a5ab91dc46c3780617684c0
ed97bb5f92187d44d55f2f5b5bac1131f12c6989
1891 F20101123_AABAHQ brown_j_Page_10.txt
bb1f340c8b7eb2d217fe6ee03dd1aae8
f003a90d34b730620e321b4dd3736573e972128b
57979 F20101123_AABACT brown_j_Page_64.jp2
44392d7bd6c93f1c75800a870dd8b481
3d17ecc2acd5d910fa7d93d66499f194a1cec843
3318 F20101123_AABAMP brown_j_Page_58thm.jpg
c511d9b2465859fca0c5cd99156ccaf9
5aae6e2a97cc350a9a04f14f2c85a9b234e10d3a
1911 F20101123_AABAHR brown_j_Page_12.txt
18b837804eccb77fd743059eef3ffb0c
8fc7a6030ca821529d8ce60903aedddb0f17ba63
F20101123_AABACU brown_j_Page_01.tif
df589f8374373505136fcf74365b3a1e
47647459cfc2e98ed892f8fe618677231b0fca15
5461 F20101123_AABAMQ brown_j_Page_23thm.jpg
06bf29fa08128d03e72748dccbda10ac
e944e9072675136930996d796fa4b28a5ee26a1c
1963 F20101123_AABAHS brown_j_Page_13.txt
12f5bd827a7ac7dc5eb3b2dc0e654275
d670e7f1c94899406cbc9dda95c6eed5220e7a6c
F20101123_AABACV brown_j_Page_02.tif
997e33e5405be0f74830999ae6820bd3
8af90f829ec9396dc5e3edb43002766984eafae1
3718 F20101123_AABAMR brown_j_Page_39thm.jpg
b748393f5652f82c32bb7c85ed9f033f
efe08a2c64d43029a464eaf4dd7f694d28cfd25b
1976 F20101123_AABAHT brown_j_Page_14.txt
89310bba2fa5c8e0900810b153538e87
537f30e9305d38d997c735a606411e2deeffbe4e
F20101123_AABACW brown_j_Page_03.tif
3a116d2a5d564998b08dd42df42625d4
37b9fa48b8d08889408d42bcaa1d2268873b748f
1890 F20101123_AABAHU brown_j_Page_15.txt
eed5cf79209d2b5cb72ae6c738f26df6
7284e2f67139f03f15534328017a30a2b020a900
25271604 F20101123_AABACX brown_j_Page_04.tif
9220c4796309f5a89a170f491ac9a82c
89c8655841b0e1d6139c1d20893216f86531aa71
22585 F20101123_AABAMS brown_j_Page_28.QC.jpg
edf60529754bd687a00d8965b9e28548
22f69dbafb2feaf9fd56ea02f9c8c232bd0e3f3d
1946 F20101123_AABAHV brown_j_Page_16.txt
ac048739eba941cd60b942c2f76f3a9c
1fed83dc2f83a648ab6ef5ab27d13ebc6c03d445
72111 F20101123_AABAAA brown_j_Page_49.jpg
0516b35526a54763a702f55a365ffcbc
b93e08309f02a8112172c0883f676b236207d9bb
F20101123_AABACY brown_j_Page_06.tif
ebc8efea82be39af301d2a0a744b4c0d
6812f5180e94eacda22f0a3066f5132baed7cfb8
F20101123_AAAZWW brown_j_Page_58.tif
3cd8d2c17bcbf775c15cdef2808e6637
c0b88a2aed76e39ec418729d0ca4632196c460c6
6390 F20101123_AABAMT brown_j_Page_46thm.jpg
ad612001e5a35ebe831d59dbada9e824
02de17332dbf0fbb5b54f87fc69000fba0cbd111
2039 F20101123_AABAHW brown_j_Page_18.txt
1ee7c308aa2cb61f12db1552ae341ea7
2fa67e912e3300906a58f43bf45701508dfe5b5b
73869 F20101123_AABAAB brown_j_Page_50.jpg
32dd7cb1e640c99060281ce938afd0db
1a49a1fe896de48cb9f67161d2bc7c3a60ee6440
F20101123_AABACZ brown_j_Page_07.tif
ed211a47e1328366006e702f64ec6268
0a6cf73f383e1f40ae50b4369a7782d8fa932a35
22311 F20101123_AAAZWX brown_j_Page_31.QC.jpg
2ebf8f67ae42939faef6dc28a79c2820
cafb8b67a634eadc1e1c3f1e707dda2b358268c0
5722 F20101123_AABAMU brown_j_Page_40thm.jpg
b58f91e50c88535c68c06a6fcc705365
bcbe89d5a4fdd7b813837465402aab2901073d82
1889 F20101123_AABAHX brown_j_Page_19.txt
c9fd9e201fddb633ad75e8d52e4d4628
b772c680ac2af79bfebc5caccc20287b80ea1c92
50692 F20101123_AABAAC brown_j_Page_51.jpg
3a79211ee90d2fb0d152c7fc9d741fd3
3d240d2bac5895a46fa64a712063ccd2ee62a082
73335 F20101123_AAAZWY brown_j_Page_25.jpg
6b5d7a64544d428cb2e64b48f385449a
557c463d95dee8cbff1bfd83dc7b61dd0eaf2a06
6531 F20101123_AABAMV brown_j_Page_25thm.jpg
05333afcad654aff0ff9649f0c8402aa
da9f1abfd23d52ea3fdaabe9fed0dd5de27aa2b1
1276 F20101123_AABAFA brown_j_Page_02.pro
c33e7d3248a3fe00f1fcfce5c8fe5013
2ff0d57ac7801fbae8e42456cec65205a594da53
1906 F20101123_AABAHY brown_j_Page_20.txt
05c0194064def2e056a41bc07ccb2adf
b8e8139c1042f12c7de0df1dfd61bf40bbbd2799
60355 F20101123_AABAAD brown_j_Page_52.jpg
0013e18eeaab32f7f4793be5679cbecc
df9ef11fa89ce43e29b2e2b9b4507a9be8ad1abe
58499 F20101123_AAAZWZ brown_j_Page_37.jpg
7dbfcef0eaa420432d4d02ea37946cf1
bd5d36a65c06731f052b8d97497c30a9428fdd69
12023 F20101123_AABAMW brown_j_Page_54.QC.jpg
e69cdacdcaaf7f51eb1a7e1d7b3900c2
3b4a5f087be6ce0f7cadc1f103627cab4a618ae6
21495 F20101123_AABAFB brown_j_Page_03.pro
160d13783e47c540759ce7dfd746fed6
ef9bced76abf1410d452e1c7ac9947d8ca4037ee
2000 F20101123_AABAHZ brown_j_Page_21.txt
fe88d2f212ad2d37a831db5010802f28
780731bb5b3841f2d33d979b6fe7c73a3965ea1c
58541 F20101123_AABAAE brown_j_Page_53.jpg
f81c501556b49cb17167b3171012951d
6249e94bc7547ab50efd86bf21442da76e32361c
13297 F20101123_AABAMX brown_j_Page_55.QC.jpg
3f5db1e9c37ea21c344acd8ea1512764
438721855ae35317b2544800420bc963aa447c2c
75386 F20101123_AABAFC brown_j_Page_04.pro
82b90cfe1cddd7dfcc670a9d922ebc7b
311425090d2901b58fc65e456be940660977f22e
70659 F20101123_AAAZZA brown_j_Page_15.jpg
7a8dc4b052d94198d33a13b50a426b7a
df630b8611a2789ffc03c99f3365586a572c69bd
38866 F20101123_AABAAF brown_j_Page_54.jpg
3f815ef3ffe88ea6e4a700087ebfaf19
b391e1ff76602699b78f7ea19c2dce7050281f34
5815 F20101123_AABAMY brown_j_Page_44thm.jpg
0ad9702cb2b69f9840860642b6309db9
cc92d92701a5ff955f436487c83e9a54af6918d6
22841 F20101123_AABAFD brown_j_Page_05.pro
1a7c965df8aab6bb413e6009c9d88792
f67a0cd580f9224ae9225a9b9fe81f346934a36e
70353 F20101123_AAAZZB brown_j_Page_16.jpg
a8eb7087662c7d9a829feadcca0168bf
5db57a1ea0d649e3b90db02c83f6f23783bdc5fb
2564 F20101123_AABAKA brown_j_Page_01thm.jpg
5e899a60ae02ab08328ae95b2679500a
e7ca9deeaa8f637b83851ce49e931b003194baa5
24172 F20101123_AABAMZ brown_j_Page_26.QC.jpg
b2bbd673c6a95e2a50d2c6797f5938ea
631acf8829066a1ec36de66ff5d640776e5b6bf3
25555 F20101123_AABAFE brown_j_Page_06.pro
5869a532967c013f68ea9b48225aad02
ff82dd2e7d9594ea0e889d605df79dd507429221
70888 F20101123_AAAZZC brown_j_Page_17.jpg
c4854504429d5ff49d1c90405ea1db93
f32082a8a4dd9ef1f4e56345db73638745b00ee6
42475 F20101123_AABAAG brown_j_Page_55.jpg
883e67a8ce3d9984ebb34d6d0c54d9ed
eaee5ce90d64088d45cce5e5a914ef4b39e59515
6399 F20101123_AABAKB brown_j_Page_48thm.jpg
787757458fc1083e84f7c8d7f2d54ba7
5d632e0a1bc51a012b0f0833103603cf760235a4
7561 F20101123_AABAFF brown_j_Page_07.pro
00cd59387bd066387773275e5f8664e8
982ba2e1366da5fd6aa01bb0d0179c22aea28e36
74721 F20101123_AAAZZD brown_j_Page_18.jpg
da16d8e0ec04f435071bc653891e141c
e51d66d6659ec918e161ded840d865b4373948e9
64566 F20101123_AABAAH brown_j_Page_56.jpg
a4fbff37cafdf433daeb4f0acf1712df
4d6843e0154b4ebcbb4c37438ed218a0a62b128c
7774 F20101123_AABAKC brown_j_Page_01.QC.jpg
77c5d48daba55005560be261865ad7e6
ee52c54369c5e665a51dea176db0c858c710077c
38580 F20101123_AABAFG brown_j_Page_08.pro
61e0e681bfa8990fc0c432ee119d4324
e8737f2c7d725b5b2df8fe44e9231dc84f2831af
69925 F20101123_AAAZZE brown_j_Page_19.jpg
e8b215087e6c1bb62c4c218cddf7efc3
71e8b562eb933848c10502f91cac7fa18efdc41b
63242 F20101123_AABAAI brown_j_Page_57.jpg
d7105a533463a23f628021962bc7a106
307a576470644ac3c9bd7bc5de08bc006ea70474
22407 F20101123_AABAKD brown_j_Page_17.QC.jpg
99c9805d744521a68e6f61e71e01562c
3de8bd2ea86872591466b49f8fecff18bfbc3b82
38214 F20101123_AABAFH brown_j_Page_09.pro
a30c7e3ed62a365be5f23ee34d2e6849
7483a6f45a77e7a6c23b36c2fbdb8dfe7fb36a2e
28558 F20101123_AABAAJ brown_j_Page_58.jpg
9e51f569123566fbec909ec6e620f584
1a628b13bf8e4d8d6972028d7149f9e4189fbcf0
23177 F20101123_AABAKE brown_j_Page_49.QC.jpg
1a03bd6766973f8895ba942b32982b0d
f6ceaf7feea5b7c9a293eb737c48524c4fcdcf31
45641 F20101123_AABAFI brown_j_Page_10.pro
ac271e8a440e246e4b545825d65119fa
0c44371d05ef9af73fa68977163cbc88bcdf50c5
71080 F20101123_AAAZZF brown_j_Page_20.jpg
a3c287baccc890a1a319e9ee2e5c16a4
81d4ed863e02d39c3dd02cb3501faeed0456940a
74518 F20101123_AABAAK brown_j_Page_59.jpg
5191e0369eaa9c99fb716cc1ba2a4216
715864525eb8e5db9707c2c80e556d421e61a9d3
23823 F20101123_AABAKF brown_j_Page_25.QC.jpg
c3f3301a74f97571cc5aba51bf613c97
ae4225dfff98dfc0df38acdd32450fc7eec3c18e
52171 F20101123_AABAFJ brown_j_Page_11.pro
033eba40ecbc1c899f479f547d5cdbe6
befbb91f5e7f288b6333906eadf036eb501eeb08
72884 F20101123_AAAZZG brown_j_Page_21.jpg
2239eb501200480d21b9f8ef61dde6d4
714398616729d70b85663a75f97ccb0492bb2a9d
90115 F20101123_AABAAL brown_j_Page_60.jpg
34f0c733eb6bb817e49eddf891c53b40
5d18c10d32a160c310d0b586c9e039f952cecd42
21470 F20101123_AABAKG brown_j_Page_10.QC.jpg
1cf32ae581747edafcbf70ed742d7d33
58fff1f466a13c741714e715da03ac7a805dcb17
48506 F20101123_AABAFK brown_j_Page_12.pro
2a0096345618dfffcc8380453dc6178e
722032b7e8bf630fa2732c58db0c184d1bf8bb91
58742 F20101123_AAAZZH brown_j_Page_23.jpg
c423a0b38d7b230ee9700e7520cffe1a
5b45611a63c8b296a899eef77bc17ea6478c70e5
85057 F20101123_AABAAM brown_j_Page_61.jpg
ca9d0061d91baf166f4e45ac7a352942
a6715ae5a21af2be994b30b0d041a0b3a7fa7afe
17275 F20101123_AABAKH brown_j_Page_51.QC.jpg
54a0f3018a53ddecce4866de39359556
2525708adc7b1a4ada15ba8215e2e9161d2e8115
66914 F20101123_AAAZZI brown_j_Page_24.jpg
b35c0fa02273616c16fa180b2fd6abfe
04b162cf38912bd47ab01fe4b1e00cfa50df114b
74864 F20101123_AABAAN brown_j_Page_62.jpg
3267fc2e6b0bd1ecd290d4c6d2fab150
93bd1a7d09916284350d7a9d557d45bb19e6a68c
19977 F20101123_AABAKI brown_j_Page_40.QC.jpg
e5f8bd57bcf789a73f9c236c482c2adf
aab6b24882423d97d0d34e8b92fcbc73b9f71eb0
63089 F20101123_AAAZZJ brown_j_Page_27.jpg
57ba8476f8429c3805076207df3c502c
795cd03c8b43d8b23651e518423c163d619839be
81372 F20101123_AABAAO brown_j_Page_63.jpg
470a8269e5ea7f3819f5f2970d121981
3bd0430196736e4ac023d7815a4bcddc0718f7fd
17776 F20101123_AABAKJ brown_j_Page_52.QC.jpg
4e1d88c10fef858d1cac2183b28c6b94
353ecd6fa54d1f03987efa55bb4e457b70fe98a6
49850 F20101123_AABAFL brown_j_Page_13.pro
16f3f99f61964c1d3529a1e03b0cdee1
854c843f08e6361277c2c152349b6c62c4618553
70677 F20101123_AAAZZK brown_j_Page_28.jpg
551caa24734d1c28a3f9f173ebd151b3
2a4bdde14bda70c3478e8ba621392ec601feb71b
42323 F20101123_AABAAP brown_j_Page_64.jpg
11d69538ce8196b0aae04d870f757c01
33a300ae37ec8f04b54fa442295a2758ace14ed0
23704 F20101123_AABAKK brown_j_Page_13.QC.jpg
84c1a7c2a63561512e38287d1aac30d8
cde0b57a9bddb874092ecbd0cd62994691619824
50189 F20101123_AABAFM brown_j_Page_14.pro
445f263fcf2bf99a25fd1f759a81c5b3
11313a58636b0a8e707631c206bcab93fe79f8f9
67896 F20101123_AAAZZL brown_j_Page_29.jpg
84e500dfbf8ffb4b042a28980078b576
34ac5752a7bd70be321e891f5ff4e409c73fee6b
5940 F20101123_AABAAQ brown_j_Page_02.jp2
ed62c53c529b6af983a548cf8bdc1acd
2cd9be2d89a2cd830bbb3092f2e079abb2634340
23652 F20101123_AABAKL brown_j_Page_30.QC.jpg
393ddcbe81722c4607c578b84f89d3d1
841c17f2eb7f6a33a4908bf0fc2e22d78be10ce3
47981 F20101123_AABAFN brown_j_Page_15.pro
e713c5a1f778d3df23a9c94f7962e457
6bb1d100ce371b9d982c83cc23dd7980dabab577
72878 F20101123_AAAZZM brown_j_Page_30.jpg
a63782ea558d1b936be96c706899c01f
b096bf6c418a3892d530e78451d5f43463fdb392
50454 F20101123_AABAAR brown_j_Page_03.jp2
da8a02ce993460c0644dab5a46385d86
e826050e0359e44d2c0eaba8e1689085015b5887
2675 F20101123_AABAKM brown_j_Page_05thm.jpg
bc8f8b9914d9dcfca2709c58327380ca
b8d26d1679f520db8735dcfb902f54ef12e0d23c
49414 F20101123_AABAFO brown_j_Page_16.pro
5d5d545de2ced841a5b849429bb8b5ec
ec64f137294b7ed181acbc1472b510548d0bbdf8
71056 F20101123_AAAZZN brown_j_Page_32.jpg
6d4708720d5efd6b21dd068ad63b5e4c
b645d6fa1614be0acbf6ee7ba68cfa262735372a
1051984 F20101123_AABAAS brown_j_Page_04.jp2
1b9681140c13e5549e81657ee2effb5d
4dc5758f169867e37a1dc700bfd7360300d6b647
6208 F20101123_AABAKN brown_j_Page_19thm.jpg
540f0059b29a9919a1e50ab21ee3d965
d494a2f981540bef5572063bf803535661275451
48788 F20101123_AABAFP brown_j_Page_17.pro
a68908c6f03dc92c9338690861e53616
080c9d02345c2dae1e0daa56c237fd12012fb66d
73707 F20101123_AAAZZO brown_j_Page_33.jpg
eedb4ad3a554ae577b1e4b9aa315c69d
f6ab070d588cb214204d7d89dca6a22450170d24
654291 F20101123_AABAAT brown_j_Page_05.jp2
19dabb60aa3eba9e621feba2ef47575f
9227d6e951aff19a97ac6f8993643ff1e361168f
22122 F20101123_AABAKO brown_j_Page_19.QC.jpg
c168f7c306010f7ff920231f25c8c6a9
45ec22abccfa901bf52d981b6cf76898c54f1c21
51693 F20101123_AABAFQ brown_j_Page_18.pro
4213d01a68e47c43901874a66cb49c21
ef160bbe7f2a6d18821b13839b1d963f6298254d
54103 F20101123_AAAZZP brown_j_Page_34.jpg
546fc38aac43fd0e6486f8d057bfdacd
718bcd6579e634fd0995cd3b67e708481f8205f6
852349 F20101123_AABAAU brown_j_Page_06.jp2
35da21996b94c9be6047bf386ca6650b
123648e6309f4fb4d7b5f901094f0ac682afb0c5
17646 F20101123_AABAKP brown_j_Page_04.QC.jpg
59e1c5ab5c5091c92b66e2be4e8f8e24
d21a0508506534655ef8ac3fcb3e6bf4754e8708
47810 F20101123_AABAFR brown_j_Page_19.pro
841e7126decfa7b7fddd191cea7288b8
745b240569c240806fefe271c8f40a1eaf83bc8b
68273 F20101123_AAAZZQ brown_j_Page_35.jpg
179152c7328fdfe33d414ae1006c85cb
358667f4e6dcad8607def35cb9dd9ffffc0ea4da
226039 F20101123_AABAAV brown_j_Page_07.jp2
2083694cef4d730be0458fe563a9685c
78cccf4c85cb669203a6b93b0e74ac7aaa9d7f3b
48264 F20101123_AABAFS brown_j_Page_20.pro
f65981472cf60b1cb0b9b8195dbde93b
8bb4711e125efd4ecab666c7ad7e9f9c524c1109
65450 F20101123_AAAZZR brown_j_Page_36.jpg
6baec735ba51dd32649f4b78e391fa84
44234bf92337df0ecc9504164e5b673c5ce27619
86643 F20101123_AABAAW brown_j_Page_08.jp2
8457558e32fa5f3bdc9920c3cdc8f045
789ffcd83eb5635e1e0080b2df7995de0bd5f1e1
6451 F20101123_AABAKQ brown_j_Page_16thm.jpg
cfcebd42f57d6a13cbfe48e7d46435b3
37c6db7d5d6e78e5e2b702d8defa2b96f8eb80b2
50735 F20101123_AABAFT brown_j_Page_21.pro
0ff5015bfdf2d68600b5187ca87a4863
34b03d5867ba515f8ca2d7bf5193bade2e1b388b
85264 F20101123_AABAAX brown_j_Page_09.jp2
f1c91fea8e2a8ae019937d950b7e2fa6
ce0774f4359da863e98651002ee8b122f3e83802
6381 F20101123_AABAKR brown_j_Page_20thm.jpg
bde8ebba4b913b190375a25d5fa94b9d
05fea26b7b0646a0a7d186bbff325c13cd33970f
20606 F20101123_AABAFU brown_j_Page_22.pro
ee612b5dd4b1af6150808fc1db5f50a5
27bd3c6915fac645fb16ff5476713222d96502cb
41100 F20101123_AAAZZS brown_j_Page_39.jpg
fd2f91ab99883e6586e0d385f466de33
9bc2f73e0cb7ad1a7f36a3195eea5eca018371dd
98801 F20101123_AABAAY brown_j_Page_10.jp2
8959b3ef171247f9eae49fc4577e358c
ebe9e438fba7dedec3157503dbe52c6eb9511659
24121 F20101123_AABAKS brown_j_Page_33.QC.jpg
6f5cfa4f1114902519f2ed969d16d984
83b613a6aa222cddc66b593b622e0860c5859d6e
45260 F20101123_AABAFV brown_j_Page_24.pro
8fa3c8ab34c5b7b24a6f71ee9dd8a9fd
cff62d39643625d48e14e0ea390097fbca14d21a
62802 F20101123_AAAZZT brown_j_Page_40.jpg
e5ed8d16b383470a947f0b6dc67e7a4d
8c53bc45437bc030590a83376a260ae7906f7909
110669 F20101123_AABAAZ brown_j_Page_11.jp2
94106aa28bb9886f2fe41038c37b090d
1d05b361c04dd9b60590cdb7e30b7082e3c91b1d
F20101123_AABAKT brown_j_Page_14thm.jpg
1477301f33bbf0d4459d28e7eaae7c9b
682e93e825df8ee7bc810ca0b4556dac7e9f7b94
51893 F20101123_AABAFW brown_j_Page_25.pro
158163940643083bcb5fe0744ae0c6e7
b90b02e1b94de742eb37dd1b857b74b6fd854f77
42458 F20101123_AAAZZU brown_j_Page_42.jpg
7630396178bdacb3faab3c42297dd732
3467c73d4a7386e931c767cdcce3b3ed16ebb1cc
4781 F20101123_AABAKU brown_j_Page_34thm.jpg
e08d428ee2cc30ef26829cb4b317ca09
93b4ca4c9fc4ecde45038c6f968a24da2bb9596d
51855 F20101123_AABAFX brown_j_Page_26.pro
b2a46fda15293974c1f53b81aa6cd738
41cc3506f65d6b3fbe8dd8655283d1baff60549d
13502 F20101123_AAAZZV brown_j_Page_43.jpg
0783115c6654c5ad81b46978543235ce
423f8e6de0786927a6790d539097cf87fc5fe914
6473 F20101123_AABAKV brown_j_Page_33thm.jpg
02f628ed750045725492278a92cd4721
93dbfde7b2e32c187a011e6ed364d68b4029882a
F20101123_AABADA brown_j_Page_08.tif
cc2d433bdd8c3f0187670182911f3af5
a2d4d4bead833406322f30a906da13cb5faf9e10
43629 F20101123_AABAFY brown_j_Page_27.pro
7871d1e7c57ffb1255d30d1d2b8507e5
37b7b3a23df7cbf717e3747c307ff1b8151e030e
63873 F20101123_AAAZZW brown_j_Page_44.jpg
2d6073234339776fe8c998d059f0f320
bebea59117e63b7afa1409450dc7d6bdaabb10c2
6912 F20101123_AABAKW brown_j_Page_61thm.jpg
c3c135f3ec2df41b0bf0f050994170f2
eefa8f2804bd1ef83c37db97b539864e9b3f62ce
F20101123_AABADB brown_j_Page_09.tif
7539c85e974ca4f356f5b74f38185ae7
96823aa1e797a2894bfc50cc34edb457ae1b7c23
49036 F20101123_AABAFZ brown_j_Page_28.pro
e5bb20ed9543421fee6a457041724dd8
0eb8fee60dbd19cef4a00bf43f0d180625b85a5f
74603 F20101123_AAAZZX brown_j_Page_45.jpg
496539323c67f5ad7e74aa61ec53c7fd
429cbb59fdd4da5f5502ac2a4585814b84cbeae6
23158 F20101123_AABAKX brown_j_Page_32.QC.jpg
7d88a6acee6e568e7df920cb49b23a8e
307e1dee7b5e68bd2f5031cae10f67872d01129d
F20101123_AABADC brown_j_Page_10.tif
27f75b799dbac9858b96fde63c7182ff
2da971bb1bceb9d59493d36f23c496259d14cddf
F20101123_AAAZXA brown_j_Page_12thm.jpg
4c2a0b0e14721865f4ce19f6cec05bbd
7505771309490331dda0caeed72854ca5b851758
72367 F20101123_AAAZZY brown_j_Page_47.jpg
74b435817421e200a6fbc25c06b7a682
83226942ace7bc9964432101ba5ba03c19f3e423
824 F20101123_AABAIA brown_j_Page_22.txt
74ee8e0842d938a72db4567a7d4ef6d4
ffc2343154f501a7bea380f6ca2506bc45da27b9
23435 F20101123_AABAKY brown_j_Page_46.QC.jpg
de93a3e10f2f8a42781cf1be3b7b0dc2
18b008d8c30b002ccbdd7e213f356cc8feccf2e4
F20101123_AABADD brown_j_Page_11.tif
5d8b879c3fee0351a36128adf080721b
c44ae2c1a8a8c1f87ce788cbe2b3dbf8d7cceaa0
26559 F20101123_AAAZXB brown_j_Page_01.jp2
1f47cf51742c970fca76c3483cc52c38
2f777131579907ef3e632d9200f2266f9baa4241
72458 F20101123_AAAZZZ brown_j_Page_48.jpg
1807eea8994384a6107063808e9a36e5
4d30f2372bbcc72f1dfab2b47ef490b068526980
1732 F20101123_AABAIB brown_j_Page_23.txt
8ea4b5e19912e8975f6c3c9b7db3709a
f24e240d526f52c0c7860c128dde98b80ccf0128
20084 F20101123_AABAKZ brown_j_Page_56.QC.jpg
6603f4991407537d31499aa4c5c94800
fbd86e8af182fac6e4a460b6e44d3525c8ad0482
F20101123_AABADE brown_j_Page_12.tif
ca54c9030d07b987705e9e6cef437c8b
4450ad2207fccd29343b5626b263408884ccf48c
F20101123_AAAZXC brown_j_Page_21.tif
e44e8b4be412f04fde4419ee15abf624
4d15f90fb612b47d4d4fb1a394f96edcf0e24584
1873 F20101123_AABAIC brown_j_Page_24.txt
404626ac1e34958d0fa56d8cb4011ada
aa501683997bfc305a50d8d9f0a26bc116a42723
F20101123_AABADF brown_j_Page_13.tif
aa4b72cf7b98a9b9543687a17a59065c
c8dd4826863df5dd0c6550d5c2aa6b674bb2e8d7
17609 F20101123_AABANA brown_j_Page_53.QC.jpg
97eee3cade3ca1c32fe0bda75c8682d7
f16c9fec8a8909b9762bc34ef9035b62f4f202f8
2041 F20101123_AABAID brown_j_Page_25.txt
661b875976c61aae4773ab27aeb28ee1
946db81e60bca1da98fd2f6b8a5de98e4e4f7ae6
F20101123_AABADG brown_j_Page_14.tif
c8539b569ca0cc0ab1f9466d040da37a
9f2657530560c6ad936f5a599c0ee6aa7ac61eec
479 F20101123_AAAZXD brown_j_Page_01.txt
5ec4c0037e498a92386d6c5d5ba61185
3afaae30b0e205d27a81772be9a8e4e45e5555e8
5426 F20101123_AABANB brown_j_Page_08thm.jpg
c9785f476a91eeb5a486cae1585fcadc
763264861c89e7f4b24ba00307d30565fe392241
2030 F20101123_AABAIE brown_j_Page_26.txt
422a05e7f696b9b8db8147de2f67b474
05ece4495efc113b4a25486f7f6627ee9caa97fd
F20101123_AABADH brown_j_Page_15.tif
efac5b830d40cc1fee2ea626f66039fc
fc6c052dd552700f7e3bc55e3f7858bbb0975e6b
36042 F20101123_AAAZXE brown_j_Page_06.jpg
ae37be798771a29db0bb68e9f16ec9d1
6c87a1ba770a48683edca71060998d08c369c8d6
20389 F20101123_AABANC brown_j_Page_44.QC.jpg
7ba15fc4c607457500351e41bbdd30d5
3ed91aa3edd3a41a44d2897c92a744a38aab9b4b
1773 F20101123_AABAIF brown_j_Page_27.txt
af55ac0c5f3ab0599e28354fa7dd2e1c
261469faeab44609d6d2c2c524c9c31a28b4ffe5
F20101123_AABADI brown_j_Page_16.tif
ad1d94f631e54851cd3019cb1c1c3b8e
683390c95baf3dde48042606c70d58dbdbf802dc
47658 F20101123_AAAZXF brown_j_Page_29.pro
bbba1f24bd959afa5fa68642d4f0a468
0d06bae3706acc757a9a45ac17cc622e47f2f6ee
22658 F20101123_AABAND brown_j_Page_16.QC.jpg
4f412ddbf6fe66ae09edc037715a1d72
7b71fb41aa25777a51e054b2be5b2b61c93ddd8b
1942 F20101123_AABAIG brown_j_Page_28.txt
a90a14917357f8d583f08de812286e5c
3da633bd9c5e8568b514b4ca11a258881385a8b7
35661 F20101123_AAAZXG brown_j_Page_58.jp2
14e1eb92743dce61c6c7557970eafe2b
34f93e0e4a4adaaaf511a11f8826be72f1328037
23335 F20101123_AABANE brown_j_Page_41.QC.jpg
5dbe11fe4f3f49362cf0e2705ffd2302
269153f90b885efb44cbfde24eec12a41672d4fa
F20101123_AABADJ brown_j_Page_17.tif
1fb72f7479aff482d5d9149551aab96a
a8692389b644ef8833659212e6403bbf6ce59484
109651 F20101123_AAAZXH brown_j_Page_48.jp2
193c04fa533b611909b87020767a2636
a6721cbf041e8e20492b3f96bb016c1d3edbee81
F20101123_AABAIH brown_j_Page_29.txt
9705ff7459dfc74c940da18cb15349d7
e81af1f9996fa5b944b1d98e20346f478c438131
22942 F20101123_AABANF brown_j_Page_12.QC.jpg
bfc9dad4bcf3fbc5a1f1f18025e00352
186c8953b89f36cb57a84c0cdfb72634b1b440fa
F20101123_AABADK brown_j_Page_18.tif
05811414e92d29e833758434877280ab
544ecab3d3155ba35b41617385b4f185a2c66d14
111791 F20101123_AAAZXI brown_j_Page_33.jp2
e9a00d02f6d229d3ac061f88c2035546
f5aca01d01a29eb9c650d3e1415c2d0c4ee4e372
2059 F20101123_AABAII brown_j_Page_30.txt
5ee132dc8a6a39dea69970678206b391
0bb3e7d2d75ed053b7c5bc2e5c33780e6389dea2
22714 F20101123_AABANG brown_j_Page_20.QC.jpg
91afddb5a81080008ac7e38f4feec33a
f07651073aa223f30c65dbaa1f2be8ce971d2d7d
F20101123_AABADL brown_j_Page_19.tif
a25b1c8a95fa8ab5239e8230e9303f08
85af5bb94810289801ca0f309d14cc1f4364ded8
70857 F20101123_AAAZXJ brown_j_Page_46.jpg
3b6f943f3694083482488a849018ae3c
b0f4694563f783a2e69e97e57b33e7988a298254
F20101123_AABAIJ brown_j_Page_31.txt
717d6a8933b13eeffcf066c88ed0a689
c3aa760cd57725f1e3eb4c0c7160946a0267f44e
5031 F20101123_AABANH brown_j_Page_09thm.jpg
57553e2e7c292b4bb11bbf4e00efa152
b8e778b6bd7c1a7020e6ecd5439a05c53190434e
F20101123_AABADM brown_j_Page_20.tif
8c983b0660e678590f452e2e97e3de06
b771e6c0b16e3367eb6ba88807dc4f8d51cca6cd
20749 F20101123_AAAZXK brown_j_Page_36.QC.jpg
8b010a42028e558c1c52b248f540b58b
1ae00f9c6d985e42484472cc8511646ecff04c06
1953 F20101123_AABAIK brown_j_Page_32.txt
e0823af54487152aa770bba23fe50164
c4835c6a27576e1d723739649e6a082fc83e2775
6178 F20101123_AABANI brown_j_Page_29thm.jpg
48544a28d4fb93dd24916ca562e9d654
fe9e7b7a18cefe511ab806d5fb475c8b72d4cd8d
F20101123_AABADN brown_j_Page_22.tif
d3707ead206955873baef32fae90a71e
eaab02c3ff46acca18db0f0645971885c4c62353
73535 F20101123_AAAZXL brown_j_Page_26.jpg
3bc49bcba586815d0aceca3c8fed771a
f8270add886b950720d77b17a854e2e77a554cc3
1979 F20101123_AABAIL brown_j_Page_33.txt
d4128f2cfcc8f3a2b762330c2859cfaa
6fac32e0c11fdb9883e5e4cc041913fb83d0b174
1406 F20101123_AABANJ brown_j_Page_02thm.jpg
db8e8700b62636af6332eee101f6439b
f7a41a1aee1b42fefc888c3eae469eb7c7d1e7e3
F20101123_AABADO brown_j_Page_23.tif
bb824874deaa09b5141170c9983f46df
7d83e25079922b25018d631ce00861f364e44381
4498 F20101123_AAAZXM brown_j_Page_04thm.jpg
c67126f4fb9e894e2bf6801c5b753a4d
fca7de4c4c4381199e87b9615a45b3af466c4776
2093 F20101123_AABAIM brown_j_Page_34.txt
8221ca1d1c4fdc83f09209554e92c38e
7c5a716687235f689e1d9cee1aee675e0dc6bfb5
23321 F20101123_AABANK brown_j_Page_48.QC.jpg
b7ae9ed72cc1e5a891100f968c20ddba
973384373752792194a182ab15723b0b9bd16c50
F20101123_AABADP brown_j_Page_24.tif
6faf213d3d193e445078cfe018f806c6
f7a59b5683faf507b763e8971dec9b420b006e80
34272 F20101123_AAAZXN brown_j_Page_22.jpg
8826a7cb83392954a9aa63b41c6655e0
3e02630699cc3cb8f539de70048cf390cba96373
2482 F20101123_AABAIN brown_j_Page_35.txt
299de70adff8c18ade17b1252a45ea5a
cd912c8eae2a09f8d219d2d9a0ae52e6e6f638ea
6478 F20101123_AABANL brown_j_Page_49thm.jpg
302c4af19c9b950206431eb605bed737
ced799b7d258d328355bbe348698cb832a605583
F20101123_AABADQ brown_j_Page_25.tif
1f075a0f468b5b8328d04ec02a4d6dc4
04cae6e09dc45be5b6375b4033563094e7105587
17925 F20101123_AAAZXO brown_j_Page_37.QC.jpg
5c56c0f046c731fcdb9cc99e7b0df251
11d4bd00756740cb7d7134efe65ee23012a9cae5
5596 F20101123_AABANM brown_j_Page_56thm.jpg
7f475adc14de5f43e81feedcea612137
64e7ee00c0cc17d60028c76e145ac276c240018c
F20101123_AABADR brown_j_Page_26.tif
6f5d828ff1a623bfa253fd43b86ead4c
a959ac235a8ebb7092e46220d60d5adb4957f2cf
F20101123_AAAZXP brown_j_Page_05.tif
54e5efcb0ff24a7781fd8184acb9cfd9
68bde424223e1551bfde7bc2109af3672b6a9592
1770 F20101123_AABAIO brown_j_Page_36.txt
9e2afd1cbe201ab091d2e5b3ad2a2e61
7dec4a160ef5f497848c4849bacd8c8c4f6c754e
4317 F20101123_AABANN brown_j_Page_42thm.jpg
20f3fadf2f7c9bdf837c4a7c14f49bf4
eb264b0f5cb07a8f54997afbf27e28d0321a7155
F20101123_AABADS brown_j_Page_27.tif
2722e7f5015d2fbf5cc5e33cd63a63cd
9004acc41ecfac264e549e1c8990dcb09ff5391f
71998 F20101123_AAAZXQ brown_j_Page_41.jpg
9cf1cbe392a9043da82069093809aa50
83baeb2ac1874bffad977d1d8330a0f2547a78ba
2009 F20101123_AABAIP brown_j_Page_37.txt
377f9a3b4cd0795c048dfea428181250
82c2d3c0816a0b7eece3fb34cdddb0d7d8d710c2
23392 F20101123_AABANO brown_j_Page_11.QC.jpg
af48b02f8f0b23806a51222e1c838af3
ee4e121e96b5736b2cf2061c6e8b819196acc0fa
F20101123_AABADT brown_j_Page_28.tif
c5689e06d28c2cb2a02f30ce841a1cc6
9338352d53948f4c9ebcecb831a9085a8cafa24d
70447 F20101123_AAAZXR brown_j_Page_12.jpg
63865e6da37e6ef1eed5cb9b7762a363
3373549e1499cd3ff4686523de6b1789ed5536cc
2517 F20101123_AABAIQ brown_j_Page_38.txt
7ad26635fa6b014f94df53fe62eeff20
0136f18b6fe236b824494c5b041034aabcb0e11e
25301 F20101123_AABANP brown_j_Page_60.QC.jpg
d7e79fcd3e631c90885dc4d34cf33c83
7de92afa439a5de6ae5c2eb9c6b397d01db6d241
F20101123_AABADU brown_j_Page_29.tif
26f8b92e4b21267581ff64ceb9c6a4c9
735c5e18b7b14f681c740bbeb9c8934bb5a30e96
2051 F20101123_AAAZXS brown_j_Page_11.txt
fc46934e5f91bdd5e37f9cb009ceac4c
10f30a42a3c2f5288102a52e1f3fe0bb461297ab
1259 F20101123_AABAIR brown_j_Page_39.txt
87c36d6cf7951a053a4a89412b96e607
40d5593f8503ba65e4da69836c3da6a71c469296
9635 F20101123_AABANQ brown_j_Page_58.QC.jpg
d1a340d2ccfefe2a82e673c450839d00
16b5b896f21e478187dab7532c84b713ad21b083
F20101123_AABADV brown_j_Page_30.tif
bbcb018bf565d0af3ef7c95543966ba5
0e73ab0a1c86fc589bc756d36dc7ba106fdb1ebe
F20101123_AAAZXT brown_j_Page_40.tif
fb312dc18c98694e97bd9acf5ebf378d
5737595a5d40b0bc45327093b7d9e7c04ba84235
1936 F20101123_AABAIS brown_j_Page_40.txt
4b1c25d527b8140ba8e3c30d99ea57a1
3b55d4164badc1c707420b022a8f7e011d903bed
6542 F20101123_AABANR brown_j_Page_18thm.jpg
bd5bc053c8f2d519269276a2f41146bf
348872d94f73f8749d6046be5f2dcca742206833
F20101123_AABADW brown_j_Page_31.tif
29dea6904e2a4fc315d743883a5ba69d
386948fe64a5f387698a580204afb2cdc852c425
13779 F20101123_AAAZXU brown_j_Page_64.QC.jpg
f864c144c7af33009fff353592fe069b
c5c98898ccda53d17cc0044327c8e8d5ac4c9dfb
1980 F20101123_AABAIT brown_j_Page_41.txt
673ab9ed6348f78192fb42f0087f57c2
988eedd20635ce754f9038b93d96496013cd24ee
97562 F20101123_AABANS UFE0011653_00001.xml
cc396ed45a66b29874291e3792dea116
4aa0b2c41ad53dcb8cd55d6dfe952c5fb36781b6
F20101123_AABADX brown_j_Page_32.tif
85c847a880035374489f49109bf47649
3ee565033aef11d442e2cecf926c17723fd132ff
F20101123_AAAZXV brown_j_Page_48.tif
99e8cc6692992227e2d74695679c69b9
756194f20d3463c686aedf91779bde4b8d5fc62c
1477 F20101123_AABAIU brown_j_Page_42.txt
202d43793fa732b5162a22367b75e8ad
ba71f1ba775acf82fd043a75046b4d039a726a75
F20101123_AABADY brown_j_Page_33.tif
dce01dba619194b77d8f979de8f55fbb
c4e69a7e5b727e50d508ca6946a110b4f9fe4894
F20101123_AAAZXW brown_j_Page_61.tif
e9e26549f45b20372d0c278c06200a4f
c331861faead90f6b6a7a7d4cf72717bdb038b13
203 F20101123_AABAIV brown_j_Page_43.txt
f3d12dc868bb24fdfccc954b6b8028e1
c7e462c827c55b094a2c416fed52b1e9a0bd0cf9
107125 F20101123_AABABA brown_j_Page_12.jp2
8bb79b2daa0779efd3356269726ec9a7
d770f1a45c3de43e7f64f76532bd63757855a9d2
3722 F20101123_AABANT brown_j_Page_03thm.jpg
f6584b835a16412e9a5d5d62e662bd60
753c36a4415d228d06f5ef7c282e6e1adba5b489
F20101123_AABADZ brown_j_Page_34.tif
b219a14d41324f1f5379e0674ce00cb3
bde8518d9831f1d315ffd4a5f63abc089437c27c
81718 F20101123_AAAZXX brown_j_Page_38.jpg
399ee73a60204b9b87798e4f9814a648
0f2377ab121572002c7b5cb9cf7a5e835cf10f90
1836 F20101123_AABAIW brown_j_Page_44.txt
4cf2828f42f81a9a3016d2eaafbeaf8f
fc9f6ed7b39cc157c88a35e66d862be8c7325b50
109035 F20101123_AABABB brown_j_Page_13.jp2
1f39434a4f08e3839abc1f8ba390ec46
9a9289428364d3615bf9480b3784eb6308db3eaa
3332 F20101123_AABANU brown_j_Page_06thm.jpg
d9a20bf724c2bb6e1ea634e0c0536ca6
45cfe19938684c3c5ed7099568e197e00d42cf7d
6634 F20101123_AAAZXY brown_j_Page_45thm.jpg
6ecf20c44c2b2c366fcb04bc9883cf5e
ffe223488440451d4bd5854424aebd0cb66cc6fd
2126 F20101123_AABAIX brown_j_Page_45.txt
a00d112eac3665c506cd3b9a717ca797
04a625e43cc7a24d603b06c4d083fa4f3b04695a
109708 F20101123_AABABC brown_j_Page_14.jp2
96d31d4c7a7f107ac90036f3d24130c5
02aef14b1a7f411f55ad326cab5728b165df01df
5046 F20101123_AABANV brown_j_Page_07.QC.jpg
cf45d455e0ecc430f91e273eccc48e31
b71083f8d620df411bfe60136629b23c9cff4a50
108580 F20101123_AAAZXZ brown_j_Page_46.jp2
68ea60828f6166fbfca00a7a0913b371
be16f235588a6ef97c09bae3b2567ee716f0316c
52472 F20101123_AABAGA brown_j_Page_30.pro
56e544de7bf2caf3d76602ea20e71484
90764bdbbf43d2001e861dc1f13941482a6844fe
1962 F20101123_AABAIY brown_j_Page_46.txt
8b07e54848662d166e7449eaa0910f0b
ee1d221959effb00bdb8af41947e29ca6ece6380
106133 F20101123_AABABD brown_j_Page_15.jp2
7e396104cab351a35ec8c5143a39bdd0
9cd6ad29b7432eeef3c99a26498d2c8fcd8b549c
1899 F20101123_AABANW brown_j_Page_07thm.jpg
7e34d974505cdd5e4d1ab32467518288
441aea1fda6faeb5f21df0192d28579cdb024d62
47800 F20101123_AABAGB brown_j_Page_31.pro
19dcb1f52c4c5d9d2a3954b2e0894312
aae8480bd60a2672a7cc70716a0f5319e8c94397
1994 F20101123_AABAIZ brown_j_Page_47.txt
954c6bb836555d41d094d7c0f75d4e6e
82922d4e1a79b9df0a2ad1a2e153682b764d810a
106938 F20101123_AABABE brown_j_Page_16.jp2
78278827b9f5cd9b20da8d64220c83bd
f020d271dc131bd7c1e2c136ab7b6e25ac03638b
6486 F20101123_AABANX brown_j_Page_13thm.jpg
c55aa2363130183889b40df86dccd0a0
4ef11641e5768bd98294db12b58506c9699e5414
48609 F20101123_AABAGC brown_j_Page_32.pro
9a7c9a9ceda27ac47a753843fd50dc7d
d9798a81548f30c520d88478730990aa755418c9
107086 F20101123_AABABF brown_j_Page_17.jp2
3ecfd85ce38c9f1db8c1934c8c6a53af
8f199167fe187cc71ff031a2e786bebcd79e2688
6372 F20101123_AABANY brown_j_Page_15thm.jpg
ac1ffc22a4159f876bdc1a20b3763f5e
aa01f7cc413fa7fc34590b16e3d1f085909930d2
24866 F20101123_AABALA brown_j_Page_61.QC.jpg
c796e4627b0707175ddaa7a9bd6124af
162193d4143d0889fc6db9c84bd1fbe188c4591b
50431 F20101123_AABAGD brown_j_Page_33.pro
df23a5f8acf106abdc9fa384fb6867af
a7e46984986e7621ef91f0fc2c896b65d2e8aca0
114523 F20101123_AABABG brown_j_Page_18.jp2
24b69512ae5e45507872ffff3e0dc652
83624992b517f45b0889cd12295d80e2968c6bd9
6597 F20101123_AABANZ brown_j_Page_26thm.jpg
07045202fe5b196a01888bf7936d56e7
b49c0ca5184423233245744b53056ff06876c144
6533 F20101123_AABALB brown_j_Page_11thm.jpg
8706c19f0aed6353d30480a194e0f2e1
9a4f894c97452c39b2a1a3dfaad5fbb9602bc7a0
39713 F20101123_AABAGE brown_j_Page_34.pro
922e4ddd1a6b4eca226d331b49f83a4d
be96fc90bd9fe31bb9e3512a50ba5623e242b7f8
5624 F20101123_AABALC brown_j_Page_57thm.jpg
fd0cec6b63e2208f5e61b3aa9efb5f36
47b80685179a4ebe1b4326eef39e1de810bb5cbb
53125 F20101123_AABAGF brown_j_Page_35.pro
6043c396e7a454d0b3d0c70f51f59ee9
bc5d660b969ba8d7471d7f5cb8393237d8188b93
105254 F20101123_AABABH brown_j_Page_19.jp2
ceba285a916738c627afe960c90a2bdd
4654b62db8e52b854cbb64c57357e6c6eb76bba1
44514 F20101123_AABAGG brown_j_Page_36.pro
38d2156f46fbc5b7b2130476a1d5fc0e
81d47e3657888bb039c505d3457f339b45dd522d
106371 F20101123_AABABI brown_j_Page_20.jp2
2c34ec336008d78917b9b62de4c88a5e
62ce055426912df16fde69c698d754e3ceda1ca9
6477 F20101123_AABALD brown_j_Page_30thm.jpg
c7608ae370afdf310bcb7f023eedfa31
18a7d550f7077fabc0d410d788b309b3f8514fd2
39215 F20101123_AABAGH brown_j_Page_37.pro
97f21357f6162858549dc6b475c15d78
066e137dee79f0cf66f28d8a6d53599ee6a08189
110692 F20101123_AABABJ brown_j_Page_21.jp2
6da9a8198ed8920c3e6fe6dbf592a20e
0b640b41e439690b8cd2c89292b8d52aed7b4248
22180 F20101123_AABALE brown_j_Page_29.QC.jpg
a67d13456c882a8665f4f903730f93fa
8a9b4c421388a805410393992b1b9e8c87d4e155
59774 F20101123_AABAGI brown_j_Page_38.pro
6b49294cc4228d4e0fa4317affba7db6
305044c758eb916a11fbf2df379965be7649257c
47235 F20101123_AABABK brown_j_Page_22.jp2
9109c1448f2777c29973782194151425
3b3b49920e880fa0a9ce6d893679a12d7ebf46f6
4509 F20101123_AABALF brown_j_Page_43.QC.jpg
9f0084cc148065726fde93eceb9ea229
0f3058bca8a6a38b700c6f748157ac42889d6d9d
25719 F20101123_AABAGJ brown_j_Page_39.pro
067eb3a6d82dde3fc1689d2040734854
022e8498d1f39967d26623f293be117d68c64f8f
87950 F20101123_AABABL brown_j_Page_23.jp2
36da7a6f7c714bf8f5cfaed01c3f4ff1
8b85cf648ca43ddedc4ef8ad68ce13bff8b7fdcf
21550 F20101123_AABALG brown_j_Page_35.QC.jpg
8a009b092b91a82b441eab09018cfdfa
bafc0d09badff9f84bc457eaf2da2778a9917222
45107 F20101123_AABAGK brown_j_Page_40.pro
802db7134cefe6add2e00b4c8cdfc0ea
d1e2f9b6d4e435ae2be9c04d15c0730a1184011a
100545 F20101123_AABABM brown_j_Page_24.jp2
103af5085819d32f023249a66d67f31d
b2a6f7979ec951d266fdd7177b0fdc1c0e04d187
23094 F20101123_AABALH brown_j_Page_14.QC.jpg
03b1575f3d0b3a3fc15922edaf8a0436
d074464a254a0ae142a19156cf7e6556ecf3fd65
50420 F20101123_AABAGL brown_j_Page_41.pro
5c020ee25d09a635240728280525f7b9
1df51dd76f1a9a27d2df49a5a24151b90a0cd2d5
110696 F20101123_AABABN brown_j_Page_25.jp2
305025a6cb6e190755fd61db8010b26b
f366faad2ea3a27625ab28ecb74381d657f13c88
6456 F20101123_AABALI brown_j_Page_47thm.jpg
db0da2da0314ebd2ace1dfae308bc2b2
116682a1e860baea1c5e63997c14c588ccd51bd6
112739 F20101123_AABABO brown_j_Page_26.jp2
16b7d275a96a11cc92d64cad9b9212fe
7c1a300df83572092ac9966f27aecd250834ae78
5938 F20101123_AABALJ brown_j_Page_10thm.jpg
ddc8b2cec8895ba011f331c75d838869
2ff7ae44bfe0f94d2df60fdacd52a0504311b470
29527 F20101123_AABAGM brown_j_Page_42.pro
4624bd5c7e60869ea69cf771c781ff93
96fbae07d0f61ca7fc413dc62165ecb185d91d92
96629 F20101123_AABABP brown_j_Page_27.jp2
2aa19c78017f56cc882452ee2737b5cd
589466ab80916b98c2a723a029df4ffaaa02084e
5062 F20101123_AABALK brown_j_Page_52thm.jpg
5a45d44ec32cab82c6a3edfdf0d94f56
aaab8075d507d92357fce987a0f66493934bb25b
4029 F20101123_AABAGN brown_j_Page_43.pro
53a3a2608ad9beabccf0aad40b6d6b75
337d6f0bb03ea7551e4b066956d5e9d71883d569
108625 F20101123_AABABQ brown_j_Page_28.jp2
f82fddab54a2c8cf38f6bcb1e26ed0c3
6387d35b5b56614e7473e93bfdb163b3339f623b
24124 F20101123_AABALL brown_j_Page_18.QC.jpg
4a17625ed3f490e76f0d6bc734152e3b
2d4ee34145fe6bc86c1a07818f0229fa0281c611
44443 F20101123_AABAGO brown_j_Page_44.pro
f9e17880586f26a7248c673f2a89521c
ab7ec4c8745c3e5d659a3bbf6c90fc674b6c7902
104890 F20101123_AABABR brown_j_Page_29.jp2
a8e1336cdf3d547c3e68c22f7921d263
a3ae4ca7a1d62613d2145a4315d9dff4724e4045
8561 F20101123_AABALM brown_j_Page_05.QC.jpg
4997c1580802d4a85a0d94eef12c65ce
c560495e54fa779cb2a38b263a58114a7123e05f
54262 F20101123_AABAGP brown_j_Page_45.pro
17fe5786bdab17e9bdcca7e0c9cc6a42
65489526ec0a45cd1cef532e31cd7741aedb2d17
108886 F20101123_AABABS brown_j_Page_30.jp2
a538c15bb51e6c18231bbb18eb4c07cd
3cd367d69310880b892ebe32fd1b7e1b54fa0593
21885 F20101123_AABALN brown_j_Page_24.QC.jpg
4f8c8cbb4231eff122860e8695eeb103
a9c72cfd7679bc5dc237eaa64c1582087c652730
49679 F20101123_AABAGQ brown_j_Page_46.pro
43659fe73b317516cf33613f8bfd2a49
00f36e80501edfdc2c19b91cf85490973b0da3b5
102353 F20101123_AABABT brown_j_Page_31.jp2
db5ec93f1043864408b831e233c12af7
9e1d1a8cedc168a8be35ed2d1f88753a0ff2236f
6396 F20101123_AABALO brown_j_Page_21thm.jpg
56981459efc17a7c3b24cdddd1515aa0
c9e3dd8a4ae113340e2735a530600f9407dc941c
50803 F20101123_AABAGR brown_j_Page_47.pro
70f6354f3e9e94cfc3d5f97a078c1729
ee3332b8747195ac91b7591dd69c69a36dc177ff
108333 F20101123_AABABU brown_j_Page_32.jp2
622eaefc57f2de6142d55bc0856331a0
779c06e2d933da2061d469f12baf5ee59d6db4b2
10935 F20101123_AABALP brown_j_Page_22.QC.jpg
c29732b050724cb2e19d6d6ddab5892a
e1543295f455c4d1e7b95bbcfeeca1310f92f195
51035 F20101123_AABAGS brown_j_Page_48.pro
c9488865b47ce4b9cb2d4767d79a8528
c0054aeb1abe038175c57bd4e6eaa0916b2cb533
77214 F20101123_AABABV brown_j_Page_34.jp2
8e9b36fe88879cc046de2417734f5113
a235ef51f4e6e93c32517fcba6ec1a6a690035c5
18184 F20101123_AABALQ brown_j_Page_09.QC.jpg
d1278e2431416d40347217b45190b91d
89a38dff396f2b84655fa244e14e5a0050c880c7
50042 F20101123_AABAGT brown_j_Page_49.pro
20e2aa24cf26ef5198c2edf9f90d2fe7
b2e2d5ab849008712ff331b89f8e5328d7be8dde
101903 F20101123_AABABW brown_j_Page_35.jp2
3cb7c682fb6a59cbc13a6644c1347666
52de770f6ca3bdc24c9eb0e7147ab5d4cad3cc3a
52543 F20101123_AABAGU brown_j_Page_50.pro
bf6b8a5aa1e4a0113b8c46ccd5fc9bf2
cdf462741227b7c9ea915e8b949551b98db18886
96886 F20101123_AABABX brown_j_Page_36.jp2
aa5db7689666aa8a8a9ec3227ffa0574
85c2b9b4721f64990bbf1ea5f169e64e94b3c15a
21925 F20101123_AABALR brown_j_Page_59.QC.jpg
323e98d9d7862b002a5d8c07f832648d
f55781178d77abf873d068c69683dcbb89ba07c0
33582 F20101123_AABAGV brown_j_Page_51.pro
17eb2ee8ea0a9cc8007542e7402fa983
2d48b936ba5513108262b21a7c16ce1866547731
772489 F20101123_AABABY brown_j_Page_37.jp2
d77a729c2eb9179b40394718efda81d5
5f7b3932690af956556544ae24cb1877ab435748
1709 F20101123_AABALS brown_j_Page_43thm.jpg
537d3ec6c477a77970401638d0456228
f8f963b16689f19f297b12442232dd52a99d2265
30679 F20101123_AABAGW brown_j_Page_52.pro
6173879d0ee3bc9c4c7afeadd1a4faf0
d65315db582213283af4607cf0c322377966cf11
123962 F20101123_AABABZ brown_j_Page_38.jp2
4995c202fe55b74baa01d4d7db93aa47
135d77aad2186a166e4a624859e2935821a8f783
20345 F20101123_AABALT brown_j_Page_27.QC.jpg
f5d2902dba895ecb7e8b8ec87d138aae
1e90a0893d17dc0fc2cea9ef4c5488a57a56d324
32087 F20101123_AABAGX brown_j_Page_53.pro
712bc946cd7bd832db53fa2542a69e73
b154b6a427a61ce17db53eab432d12b708bb649a
12459 F20101123_AABALU brown_j_Page_03.QC.jpg
1265916b93a0d1c65f3ebc114a1b94a0
0a5dc17933f5147711b34d60a6d1cf05f5a263ac
19788 F20101123_AABAGY brown_j_Page_54.pro
5adfb3f9b22d9f80783fc5c62766b172
d0aa02239f7822054a7ecd995f96d548f484a34a
F20101123_AABALV brown_j_Page_27thm.jpg
70f63c296699ea3e34333870086e657f
d21db058e3a23408ac9b57a7eebace8d19cd196f
F20101123_AABAEA brown_j_Page_35.tif
d8b4395942fe9a36c72c4035b35880e8
3d035241104da33afebf4971d6ccaeaaa0bfe60a
21640 F20101123_AABAGZ brown_j_Page_55.pro
9220a09ac7d215216bcbff822d32d222
4e345de85de386a6854e8ea07c633d9bbf42ef30
5262 F20101123_AABALW brown_j_Page_53thm.jpg
e03654a092e0de559266e47677e3da18
f55609326f4ae9be286f56cdb3412b1c24a7962a
F20101123_AABAEB brown_j_Page_36.tif
16760d53d0489efcfd8d24f263e89d9d
90874df3c2c2c69cfb7be52ff972ad25b3ff91eb
3315 F20101123_AABALX brown_j_Page_02.QC.jpg
524d1e6dd24359925609ea0fc26d03cd
a03c4436f4cc551daecdf0780daa326c082d0de5
F20101123_AABAEC brown_j_Page_37.tif
69ae5122f6ffb327aa4cee4e89794147
23527012758f5673b98a8a1066c27225ffd9e36b
68145 F20101123_AAAZYA brown_j_Page_31.jpg
249025968e2de368d1fab7c31e868c90
4b39945d9d486eee10f1e5ab129f87f92f94b866
F20101123_AABAJA brown_j_Page_48.txt
576953221dcb8efdec5d49dc65e6ce5d
a482710153efa788e02ee8a2f64b193ed2ca1c67
6560 F20101123_AABALY brown_j_Page_38thm.jpg
7156a6a25a755ff7c646fd9937d562b9
7d5124e326f3d010f0ba9bd9b0d6de052831f8b1
F20101123_AABAED brown_j_Page_38.tif
0c36fe4596d6183e5066477584466d0d
543c7a462d4d555c12f5eae61c1d0afa4a7df36d
108614 F20101123_AAAZYB brown_j_Page_49.jp2
131e0eec44fd6f3847b8ee470905189c
fcdfd84ae391eb5be3ffc9a1c4dff18bae94a4c5
1969 F20101123_AABAJB brown_j_Page_49.txt
6bd835a1d4c2173914e0ded18048ba7d
c4342eb90d2fb9d698e7831cda9121294e661038
22575 F20101123_AABALZ brown_j_Page_15.QC.jpg
c832d528de0ea57f6dfea237a5c287c1
1dd4fbc0aa59922565671a4e80d6425809b1d909
F20101123_AABAEE brown_j_Page_39.tif
2cb561910bbe6bbed2ec9be36c4f6f89
1e2e3fe18445ccb365a8e7cf6b33e053ff090848
39863 F20101123_AAAZYC brown_j_Page_23.pro
4ef3093e74f8bcb166765c1cd2179843
891115cfd57d0cc2536ab331a3dbecec92dff111
F20101123_AABAJC brown_j_Page_50.txt
5040970d11d9c12317bcfe84f6e35a59
d2e37bd97f0313f14058e3e6bf98d0a79169f2a5
F20101123_AABAEF brown_j_Page_41.tif
cf83552422afb4a9dbae978a7b37a675
d1bd71f6521ae103ffd60415d54abfc104551abf
13158 F20101123_AAAZYD brown_j_Page_58.pro
ee28aacd6e2c6aa2af40e47ea13967d4
4e1b4458b2c54e45adddabda38a08031062d16b4
6135 F20101123_AABAOA brown_j_Page_31thm.jpg
09632fbf907b509ac384d64639d7d305
cba06c9aef3bde76f0fa4b178eea0ddb6ed01b8e
1340 F20101123_AABAJD brown_j_Page_51.txt
0499e3e9ab130a477c2d60892f162465
ecdb71a0d4e1d8953b0960f60ced65cc41110b85
F20101123_AABAEG brown_j_Page_42.tif
e211c9dba06554f4ea42c4f1192b0a65
a263edc477f24d9954602b809a7151f4bfba20b6
16654 F20101123_AABAOB brown_j_Page_34.QC.jpg
9fd89c7864d31d30f59e11a4e01e3534
67d9ad835e8e81dbb573f8071bd25ea72d33c18f
1282 F20101123_AABAJE brown_j_Page_52.txt
2c1bd1558430e5e4c4762600f078fc79
179a62eaca50dbaa40f6b1ae2577c80bd93c7af2
F20101123_AABAEH brown_j_Page_43.tif
ebe1983372ed48e86098ea97be0885b8
35c1c6c78865011f797638f5f80b140d3e251561
54123 F20101123_AAAZYE brown_j_Page_55.jp2
43f3c27a64ac43ac6e5efbfb957302fe
e18667740e6f429525afc5090aafd805b1a32c5b
6081 F20101123_AABAOC brown_j_Page_35thm.jpg
16a06b94d1238fef19c8e11d2614efc4
af23d1503f3bb5dbca4aa6b61f57dd16467abba0
1311 F20101123_AABAJF brown_j_Page_53.txt
df63e37da65b3e8c049a8965aeab7b6d
e803c821f15cc0c107cc99991678c6ec242a1e85
F20101123_AABAEI brown_j_Page_45.tif
1e5f68be02dc1570f761bb47d1961a45
d71296ae957f30e99414f0db2008253f2ed9a7c4
18585 F20101123_AAAZYF brown_j_Page_57.QC.jpg
d4ea17058751c81c2434bbdfee520c2d
06badbe5711ea4eb8c28ad567cab152dfdbe274c
6035 F20101123_AABAOD brown_j_Page_36thm.jpg
0a0dd929bf78d2d9d0320c4a6efecf68
01e06a8f1936a0f366dc5eabcd04df0c00b8b4a5
891 F20101123_AABAJG brown_j_Page_54.txt
d8900e831e6d6edd17f88b99b89a23e7
feb1615b66734664a5f4f3ecb975ec2df7ecd929
F20101123_AABAEJ brown_j_Page_46.tif
660f16b10bc92315f0b62a886cf50d0c
9b7a54b799ee44620af4cd2be5e6dde6b18b9b1a
1927 F20101123_AAAZYG brown_j_Page_17.txt
5bffe710ba0f0757ef43d414b774fa05
bcca71874716c0aa13a2f93068f099a6a30b2a0f
6505 F20101123_AABAOE brown_j_Page_41thm.jpg
d489e2f6a0075e24ef6def26a20753cc
a28ed24e14b1ddcecc7be36042727182ba5e2f2f
874 F20101123_AABAJH brown_j_Page_55.txt
6e3684d6c06ea7e32090d48630c947f4
83e8d2163ca1961847f1a6c383519f2b37222652
13679 F20101123_AAAZYH brown_j_Page_42.QC.jpg
f46baf549b2a7fd71a3e73412c9e27e7
1c07c8d42a7cafd7459bd6c96e519b68bfe142fd
6548 F20101123_AABAOF brown_j_Page_50thm.jpg
375f6f446256d4368b5748ccdca67311
1d2e0e39fc749c5b2a2c0156694e5be69988f97a
1629 F20101123_AABAJI brown_j_Page_56.txt
667b462c01e45277cd0e5098208f983a
f4115a3218618726ec2ec615c32e68dd4f321175
F20101123_AABAEK brown_j_Page_47.tif
a616a247ecb26c0d39a57f0657958873
1f636219c736f4f449094643d261aa1325f65f03
F20101123_AAAZYI brown_j_Page_44.tif
c51acb9bcd57c4eae68f408e4fdf4de6
4774d2fed44906ee85ed6627e6319850197b7b2a


Permanent Link: http://ufdc.ufl.edu/UFE0011653/00001

Material Information

Title: Patient-Centered Outcomes for Chronic Spine Pain: Multidimensional Success Criteria and Treatment Matching
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0011653:00001

Permanent Link: http://ufdc.ufl.edu/UFE0011653/00001

Material Information

Title: Patient-Centered Outcomes for Chronic Spine Pain: Multidimensional Success Criteria and Treatment Matching
Physical Description: Mixed Material
Copyright Date: 2008

Record Information

Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
System ID: UFE0011653:00001


This item has the following downloads:


Full Text












PATIENT-CENTERED OUTCOMES FOR CHRONIC SPINE PAIN:
MULTIDIMENSIONAL SUCCESS CRITERIA AND TREATMENT MATCHING
















By

JENNIFER LYNN BROWN


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF
DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA


2005































Copyright 2005

by

Jennifer Lynn Brown















ACKNOWLEDGMENTS

I would like to thank Dr. Michael E. Robinson for his guidance and mentorship

throughout my graduate training. This project could not have been completed without his

support. I would also like to recognize the valuable contributions of my committee

members, Drs. Samuel F. Sears, Stephen R. Boggs, and Pamela W. Duncan. I would like

to acknowledge the following data collection sites: Hampton Oaks SpineCare Center,

Ayers Medical Plaza and Surgery Center, Brooks Pain Rehabilitation Program, Malcom

Randall VA Medical Center, and the Department of Psychiatry's Pain Clinic. I also thank

Dr. John Beauvais, Director of Predoctoral Training at VA Connecticut Healthcare

System, for his assistance with statistical analyses. Finally, I thank my family and close

friends for their love and support.
















TABLE OF CONTENTS

page

A C K N O W L E D G M E N T S ......... .................................................................................... iii

LIST OF TABLES ................... ..... ....... ......... ............. vi

LIST OF FIGURE S ......... ..................................... ........... vii

A B STR A C T ..................... ................................... ........... ... .............. viii

CHAPTER

1 IN TR OD U CTION ............................................... .. ......................... ..

Aim 1: Developing Patient-Centered Success Criteria ...............................................1
Aims 2 and 3: Identifying Patient Subgroups and Examining Treatment Matching....9

2 M E T H O D S .......................................................................................................14

P a rtic ip a n ts ........................................................................................................... 1 4
D e s ig n ................................................................................................................... 1 4
P ro c e d u re .......................................................................................................1 5
Treatm ents ...................................................... 15
Palliative Pain M management ................................. ......................... ........ 15
Multidisciplinary Pain Management .................................. ...............17
M e a su re s ............................................................................................................... 1 8
D em graphics ............................................ 18
P a in .........................................................................................1 9
M ood/Distress ......................................................................... ......... ....... ......... 19
D isab ility ....................................................... 2 0
Patient-Centered Outcomes ......... ..............................20
T reatm ent S satisfaction .................................................................................... 22
A n aly tical Strateg y ............................................................................................... 2 3
Aim 1: Developing Patient-Centered Success Criteria .....................................23
Aim 2: Identifying Patient Subgroups ..................... ... ...... ........ 23
Aim 3: Examining Treatment Matching......................... ............. 24

3 R E SU L T S ....................................................... ............. ........ 25

D descriptive Statistics ................... ........................................................ 25










Aim 1: Developing Patient-Centered Success Criteria.............................................28
Aim 2: Identifying Patient Subgroups ............................................. ............... 29
Aim 3: Examining Treatment M watching ................. ...... ..... ................... ............... 31

4 D ISCU SSION ........... .. ....... ........................................................... 35

APPENDIX

A PATIENT-CENTERED OUTCOMES QUESTIONNAIRE (PCOQ).....................43

B F O L L O W -U P P C O Q ....................................................................... .....................45

LIST OF REFEREN CES ............................................................ .................... 50

B IO G R A PH IC A L SK E TCH ..................................................................... ..................55








































v
















LIST OF TABLES

Table pge

3-1 Descriptive Statistics from the PCOQ (N = 70) ..................................................25

3-2 Descriptive Statistics from the Follow-up PCOQ (N = 55) ...................................26

3-3 Pain: Observed Frequencies ......................................................... ............... 28

3-4 Fatigue: Observed Frequencies ........................................ .......................... 28

3-5 D stress: O observed Frequencies ........................................ ......................... 28

3-6 Interference: Observed Frequencies........................................ 28

3-7 Success Criteria and Associated Classification Accuracy Statistics........................29

3-8 Scores on PCOQ and Standardized Measures (Mean + SD) by Cluster ................ 30

3-9 Effect Sizes of Cluster Differences (Cohen's d)............................................. 31
















LIST OF FIGURES

Figure page

3-1 Cluster Solution for PCOQ Importance Ratings ............................................... 30

3-2 Subgroup by Treatment Interaction for Usual Interference. ...................................33















Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

PATIENT-CENTERED OUTCOMES FOR CHRONIC SPINE PAIN:
MULTIDIMENSIONAL SUCCESS CRITERIA AND TREATMENT MATCHING

By

Jennifer Lynn Brown

December 2005

Chair: Michael E. Robinson
Major Department: Clinical and Health Psychology

Our study aimed to (1) define patient-centered success criteria for chronic pain

treatment across multiple relevant domains (pain, fatigue, emotional distress, and

interference with daily activities) using our newly developed Patient-Centered Outcomes

Questionnaires (PCOQs); (2) derive and validate subgroups of chronic pain patients; and

(3) examine the potential for treatment matching using our patient subgroups. We

recruited 70 patients with chronic spine pain from five University of Florida or affiliated

pain clinics. Participants completed pre-treatment and post-treatment measures, including

our PCOQ and Follow-up PCOQ. Post-treatment assessment occurred two months after

treatment initiation.

Pre-treatment success criteria derived using a direct scaling approach were more

stringent than criteria developed using other approaches. At the onset of treatment, our

patients considered a 58% reduction in pain to represent treatment success. Success

criteria for pain developed using other approaches (e.g., expert opinion) have ranged









from 30 to 50%. Pre-treatment success criteria for fatigue, distress, and interference were

61%, 64%, and 66%, respectively. Results showed that patients adjusted their success

criteria over time in the direction of becoming less stringent, and they used their less

stringent post-treatment criteria to make global judgments of treatment success.

Consequently, future research might investigate the evolution of patients' success criteria

over the course of different treatments.

Success criteria for pain derived using a scale comparison approach were a raw

change of 17.5 points (0 to 100 numerical rating scale) and percent change of 25%. Other

criteria were 7.5 points (11%) for fatigue, 5.0 points (13%) for distress, and 9.5 points

(12%) for interference. Future research should validate these success criteria, particularly

for the less studied domains of fatigue, distress, and interference. Once validated, they

can be used to calculate effect size estimates, permitting comparisons across studies and

ultimately helping to evaluate the efficacy of various treatments. Clinicians can also use

these criteria to help them assess the care of individual patients.

Finally, our results suggest that pain patients are heterogeneous. Even among

patients whose pathology is specific to the spine, two distinct subgroups emerged. Future

research might explore whether these and other patient subgroups have different criteria

for treatment success.














CHAPTER 1
INTRODUCTION

Chronic pain is a major health problem in America. Estimates indicate that up to

18% of the population suffers from persistent pain at any given time (Turk, 2002).

Beyond the number of people affected, chronic pain places a substantial economic burden

on the United States health care system. Health care expenditures for chronic pain exceed

$20 billion annually (Turk, 2002). This figure does not include the cost of treatment at

pain rehabilitation programs, which is in the range of $1.4 billion per year (Turk, 2002).

Notably, health care expenditures represent only a fraction of the costs associated with

chronic pain. Disability compensation, lost productive time, and lost tax revenue also

factor into estimates (Turk, 2002). A recent survey found that among working adults with

common pain conditions, including back pain and arthritis, lost productivity due to

decreased performance costs approximately $61.2 billion per year (Stewart, Ricci, Chee,

Morganstein, & Lipton, 2003). Considering all sources of expenditures, chronic pain is

projected to cost the United States economy roughly $100 billion each year (Department

of Health and Human Services, National Institutes of Health, 1998).

Aim 1: Developing Patient-Centered Success Criteria

Numerous studies have examined the effectiveness of various treatments for

chronic pain (Turk & Melzack, 2001). These studies used standard statistical methods

that result in conclusions about statistical significance, or the reliability of treatment

changes. Statistical significance, while necessary, is not sufficient to show treatment

effectiveness. The statistical significance of a finding largely depends on sample size. A









trial involving a large number of patients can detect differences between treatment and

control groups or between the same patients pre- and post-treatment that are statistically

reliable, but trivial from a clinical perspective. Conversely, a trial involving a small

sample may not achieve statistical significance, but may indicate important differences.

Regardless of the specific problem, comparisons of central tendency do not

elucidate the proportion of patients who benefit from the treatment, or the extent of

improvement in these patients. Statistically significant group differences could reflect

major improvement in a few patients, or minor improvement in many patients. They also

could reflect improvement in some patients, but no change (or even deterioration) in

other patients (Farrar, Portenoy, Berlin, Kinman, & Strom, 2000). Therefore, clinicians

might have difficulty interpreting and applying the results of central tendency analyses to

the care of individual patients (Farrar et al., 2000). Consequently, studies of chronic pain

should also consider whether treatment is clinically significant. A more useful statistic for

clinicians is the percentage of patients who attain a clinically important improvement

(Farrar et al., 2000).

In deciding what constitutes a clinically significant outcome for pain treatments,

researchers might consider looking to the patient rather than to health care providers,

caregivers, or third-party payers. Pain and its sequelae are intrinsically subjective and

influenced greatly by a person's mood, previous pain experience, expectations, and

coping resources. Although patient self-report instruments and pain scales are widely

used in clinical settings, different health care providers and researchers could still reach

different conclusions about treatment success if they apply different interpretive criteria

to these instruments. Furthermore, chronic pain patients and their therapists may have









different views of treatment success (Daniel, Long, Murphy, Kores, & Hutcherson,

1983). Given the subjective nature of the pain experience and potential discrepancy

between patients' and health care providers' conclusions about treatment success,

decisions about what constitutes treatment success ought to consider the patient

perspective.

The patient-centered model of health care delivery, which holds that patients' views

should be recognized, appreciated, and incorporated into treatment decisions and

outcome evaluation, has been adopted as a goal of clinical medicine (Lane & Davidoff,

1996), as it promotes greater satisfaction with health care, increased compliance with

medical regimens, and maintenance of patient-provider relationships (Fischer, Stewart,

Bloch, Lorig, Laurent, & Holman, 1999). Consequently, a growing body of literature

addresses patient views of treatment outcome. Investigations have been conducted across

medical conditions, such as epilepsy (Gilliam, Kuzniecky, Meador, Martin, Sawrie,

Viikinsalo, Morawetz, & Faught, 1999), multiple sclerosis (Schwartz, Cole, & Gelber,

1995), and post-traumatic stress (Zatzick, Kang, Hinton, Kelly, Hilty, Franz, Le, &

Kravitz, 2001).

To date, there have been few patient-centered investigations of treatment outcome

for chronic pain. Furthermore, most investigations focused exclusively on pain reduction,

although patients enter pain rehabilitation with multiple presenting problems in addition

to pain (e.g., mood disturbance, fatigue, disability) and identify diverse treatment goals.

Casarett, Karlawish, Sankar, Hirschman, and Asch (2001) found that 32% of patients in

their study identified improved sleep as an important clinical endpoint. Improved

activities of daily living and improved mood were identified as important treatment goals









by 30 and 8% of patients, respectively. More recently, the Initiative on Methods,

Measurement, and Pain Assessment in Clinical Trials (IMPACCT) convened with the

purpose of identifying core outcome domains for clinical trials of chronic pain treatment

(Turk, Dworkin, Allen, Bellamy, Brandenburg, Carr, Cleeland, Dionne, Farrar, Galer,

Hewitt, Jadad, Katz, Kramer, Manning, McCormick, McDermott, McGrath, Quessy,

Rappaport, Robinson, Royal, Simon, Stauffer, Stein, Tollett, & Witter, 2003). Included in

the core outcome domains were emotional functioning and physical functioning.

Therefore, any comprehensive examination of patient-centered outcomes must address a

host of treatment endpoints.

After identifying relevant endpoints, researchers must establish how much change

in a given domain constitutes success. In other words, they must determine the degree of

symptom improvement considered important, referred to in the literature as the minimal

clinically important difference (MCID). Previous attempts to establish MCIDs used

diverse approaches, most of which are not based on patient perceptions. Many

approaches are flawed in additional ways, further limiting their utility. The simplest

approach was based on the principle of face validity. Moore, McQuay, and Gavaghan

(1996) chose 50% pain relief as the MCID because of its clinical intuitiveness. Yet, the

significance of this criterion to chronic pain patients has not been determined.

The approach taken by Slater, Doctor, Pruitt, and Atkinson (1997) involved

comparing patients' symptoms to those of healthy individuals. Slater and colleagues

defined the MCID in depressive symptoms as change such that the patient's score on the

Beck Depression Inventory, a self-report measure of depression, following treatment was

closer to the mean of healthy individuals than to the mean of chronic pain patients.









Again, the significance of this criterion to the patients it is applied to has not been

ascertained.

Some investigators relied on expert opinion in establishing MCIDs. For example,

Goldsmith, Boers, Bombardier, and Tugwell (1993) conducted eight focus groups with

practicing clinicians, epidemiologists, biostatisticians, and physicians working either in

various industries (e.g., pharmaceutical) or for regulatory agencies to determine clinically

important outcomes for rheumatoid arthritis. The experts independently rated whether

profiles containing information about six variables- swollen joint count (range = 0 to

66), painful joint count (range = 0 to 68), pain intensity (range = 0 to 10 cm), disability

(range = 0 to 10 cm), patient global assessment (range = 0 to 10 cm), and physician

global assessment (range = 0 to 10 cm)-represented important improvement. When 6 to

8 of the focus groups achieved 70% positive agreement, the profile was considered to

reflect important improvement. Measures of central tendency then were calculated on

profiles representing important improvement to derive an MCID for each of the six

variables. The MCID values were 2 (17%) for swollen joints, 3 ( 27%) for painful joints,

2.2 (36%) for pain, 3.6 (49%) for disability, 2.2 (35%) for patient global assessment, and

2.3 (39%) for physician global assessment. While this approach is more sophisticated

than other approaches, it still does not capture the patient's view.

A common method of deriving MCIDs involves comparing two scales. This

approach has two major variations (Redelmeier, Guyatt, & Goldstein, 1996). In the first

approach, patients are asked to rate a given domain (e.g., current pain intensity on a

100 mm VAS) and then indicate whether they are doing better or worse in that domain

than before. This approach identifies the threshold at which the average patient begins









judging himself as improved. Farrar, Young, LaMoreaux, Werth, and Poole (2001) used

this approach with patients with neuropathic and non-neuropathic chronic pain

conditions. Their goal was to determine the MCID in pain intensity associated with a

"much improved" global impression of change. Using an 11-point numerical rating scale

(NRS), they found that a raw change of 2 (30%) represented clinically important

improvement. Salaffi, Stancati, Silvestri, Ciapetti, and Grassi (2004), using the same

approach, found that a raw change of 2 (33%) on a 10 cm NRS corresponds to a "much

better" global impression of change among patients with chronic musculoskeletal pain. In

a study of patients with osteoarthritis (OA) of the lower extremities, MCIDs associated

with a "slightly better" change of health in general related to the OA joint ranged from

.80 to 1.01 points (17 to 22%) on a continuous 0 to 10 NRS (Angst, Aeschlimann,

Michel, & Stucki, 2002).

The second approach uses interpersonal judgments (Redelmeier et al., 1996). After

rating themselves on various domains, patients are asked to converse with another patient

with the same condition and then rate themselves on the same domains relative to their

conversational partner. The MCID typically is defined as the difference in mean scores

between conversations in which partners rate themselves as "about the same" and those

in which they rate themselves as "somewhat better." Wells, Tugwell, Kraag, Baker, Groh,

and Redelmeier (1993) used this approach to determine the following MCIDs for

rheumatoid arthritis patients: .22 units (7%) for physical ability as assessed by the

Stanford Health Assessment Questionnaire (HAQ; range = 0 to 3), .62 cm (6%) for pain

as measured on a 10 cm VAS, and 9% for overall condition as assessed on a 5-point









scale. In another study of arthritis patients, Redelmeier and Lorig (1993) derived a

slightly smaller MCID of. 17 units for the HAQ.

Although both approaches to scale comparison are patient-centered, important

limitations exist. First, much variability exists in researchers' a priori definitions of the

clinically important difference-Wells et al. (1993) chose "somewhat better" whereas

Farrar et al. (2001) chose "much improved"-making comparisons across studies

difficult, and raising questions about the ability of any one definition to truly capture the

notion of minimal improvement. Most important, these investigator-determined a priori

definitions may not reflect the degree of symptom improvement that is important to

patients.

A final method of determining MCIDs is specific to the domain of pain. Farrar et

al. (2000) derived an MCID in analgesia by comparing the change in pain intensity for

adequately treated episodes of cancer-related breakthrough pain (i.e., episodes not

necessitating rescue medication as an adjunct to oral transmucosal fentanyl citrate) to the

change in pain intensity for inadequately treated episodes. They selected as the MCID the

change in pain intensity that most accurately distinguished satisfactorily and

unsatisfactorily treated episodes in addition to having relatively equal sensitivity and

specificity. Analyses yielded an MCID of 2 on an 11-point NRS, which corresponds to

33% change. Farrar and colleagues (2003) then replicated these MCIDs with a second

data set. According to Farrar (2000), the "use of rescue" outcome "may offer a more

objective and specific measure of the patient's perspective on what represents clinically

important analgesia" and "allows the patient to integrate his or her perception of the

amount of relief and any potential side effects experienced (p. 292)." However, this









approach is most appropriate for trials of fast-acting medications for the treatment of

breakthrough or acute pain, and may not be generalizable to chronic pain (Farrar, 2000).

Given the limitations associated with previous attempts to establish MCIDs for

chronic pain, additional approaches are needed that take into account major areas of

concern to patients. Therefore, our first aim was to define MCIDs from the spine pain

patient's perspective across four domains of concern (pain, fatigue, emotional distress,

and interference with daily activities) using newly developed Patient-Centered Outcomes

Questionnaires (PCOQs). Preliminary research using the PCOQ found that chronic pain

patients might require larger reductions in their pain than has previously been reported in

the literature, for them to consider treatment successful (Robinson, Brown, George,

Edwards, Atchison, Hirsh, Waxenberg, Wittmer, & Fillingim, in press). In fact, MCIDs

derived using a direct scaling approach were 3.4 points (on an 11-point NRS) or 56% for

pain, 3.35 points (57%) for fatigue, 3.64 points (65%) for distress, and 4.3 points (68%)

for interference (Robinson et al., in press).

Our study expands on previous work using the PCOQ by adding a post-treatment

measure: the Follow-up PCOQ. Our goal was to derive success criteria for each of the

domains by ascertaining the change in scores from pre-treatment to post-treatment that

most accurately distinguishes successful treatment from unsuccessful treatment as

determined by patient global judgment. Once identified, these MCIDs can be used to

calculate effect size estimates, such as the number needed to treat (Cook & Sackett,

1995), permitting comparisons across studies and ultimately helping to establish the

efficacy of various treatments (Farrar et al., 2001; Redelmeier et al., 1996).









Aims 2 and 3: Identifying Patient Subgroups and Examining Treatment Matching

The etiological mechanisms of many nonmalignant chronic pain conditions,

including spine pain, remain poorly understood. Thus, diagnosis is based predominantly

on current and past signs and symptoms, rather than on specific pathological findings.

Turk and Okifuji (2001) said that because the specificity of these diagnostic criteria has

not been established, patients with diverse characteristics may erroneously be given the

same diagnosis and treatment. In other words, "a heterogeneous group of patients may be

treated as if they are homogeneous" (Turk & Okifuji, 2001, p. 400). This may partly

explain why treatment-outcome studies consistently find that only a subset of patients

with a particular diagnosis benefits from the prescribed treatment.

To offset the pain patient homogeneity assumption, researchers attempted to

identify subgroups of chronic pain patients. For example, Dworkin and LeResche (1992)

derived subgroups using physical signs and symptoms. Other investigators (e.g., Jamison,

Rock, & Parris, 1988) used traditional measures of personality and psychopathology,

such as the Minnesota Multiphasic Personality Inventory or Symptom Checklist-90, to

empirically derive pain patient subgroups. Still others (Keefe, Bradley, & Crisson, 1990)

created subgroups based on pain behaviors. Although these approaches yielded several

classification systems, the question remains as to whether these systems are clinically

useful. According to Turk and Okifuji (2001), one way to substantiate the clinical utility

of a classification system involves demonstrating the differential effectiveness of

therapies matched to the defining characteristics of the subgroups. The few studies that

addressed this issue (Guck, Meilman, Skultety, & Poloni, 1988; McGill, Lawlis, Selby,

Mooney, & McCoy, 1983; Moore, Armentrout, Parker, & Kivlahan, 1986) failed to









demonstrate the differential effectiveness of treatments customized to subgroup

characteristics.

Of note, Turk and colleagues had some success with their classification system,

which is based on the West Haven-Yale Multidimensional Pain Inventory (WHYMPI).

Three subgroups of chronic pain patients emerged from cluster analysis of WHYMPI

responses: Dysfunctional, Interpersonally Distressed, and Adaptive Copers. The

Dysfunctional subgroup reports high pain severity, life interference, and psychological

distress, as well as low activity level (Turk & Rudy, 1988). Interpersonally Distressed

patients perceive their significant others as unsupportive of their pain problems. Finally,

the Adaptive Coper subgroup reports relatively low levels of pain and interference, and

relatively high levels of social support and activity.

Preliminary studies examining the differential responsiveness of the WHYMPI

subgroups to different treatment programs provide support for the clinical utility of the

classification system. Rudy, Turk, Kubinski, and Zaki (1995) evaluated the differential

responses of the three WHYMPI subgroups of temporomandibular disorder (TMD)

patients to a treatment involving interocclusal appliance, biofeedback-assisted relaxation,

and stress management. Results showed that significantly more Dysfunctional patients

than Interpersonally Distressed or Adaptive Coper patients achieved clinically significant

improvements with respect to pain intensity, depression, and perceived impact of TMD

on their lives. Furthermore, a greater number of Interpersonally Distressed patients than

Adaptive Coper patients showed clinically significant improvements in these domains. In

general, results showed that the treatment was effective, but only for certain subgroups of

TMD patients.









Turk and associates (Turk, Rudy, Kubinski, Zaki, & Greco, 1996; Turk, Zaki, &

Rudy, 1993) then developed a treatment plan that targets the needs of TMD patients in

the Dysfunctional subgroup. They evaluated the efficacy of this treatment program

(interocclusal appliance, biofeedback-assisted relaxation, and cognitive therapy for

depression) against a more conservative treatment program (interocclusal appliance,

biofeedback-assisted relaxation, and supportive counseling). The Dysfunctional patients

who received cognitive therapy as part of their treatment showed significantly more

improvement, especially in the domains of pain and depression, than did the

Dysfunctional patients who received supportive counseling as part of their treatment.

Therefore, by incorporating a treatment component tailored to Dysfunctional patients,

Turk and colleagues were able to augment treatment outcome.

In a study of patients with fibromyalgia syndrome, Turk, Okifuji, Starz, and

Sinclair (1998) predicted that the WHYMPI subgroups would show different outcomes to

standard rehabilitative treatment involving medical, physical, occupational, and

psychological elements. Findings support their hypothesis, with patients in the

Dysfunctional subgroup showing improvements in nearly all domains and patients in the

Interpersonally Distressed subgroup showing no improvement, presumably because the

treatment did not address their particular clinical needs, which reside primarily in the

interpersonal domain. Patients in the Adaptive Coper subgroup showed only minor

improvements compared to patients in the Dysfunctional subgroup, although they had

less room for improvement given their better pre-treatment functioning. Again, these

results highlight the importance of matching treatment to the clinical needs of the group.









These studies provide preliminary support for the utility of the treatment-matching

approach. However, further research is needed before the treatment-matching approach

can reasonably be applied to clinical practice. It is hoped that streamlining treatment

based on patient characteristics rather than adopting the "one size fits all" approach to

treatment ultimately will improve outcomes and decrease costs (Turk & Okifuji, 2001).

A second aim of our study was to empirically derive subgroups of chronic pain

patients using the PCOQ. More specifically, we classified chronic spine pain patients into

subgroups based on their ratings of the perceived importance of improvement in pain,

fatigue, emotional distress, and interference with daily activities. On the basis of

preliminary data from four University of Florida or affiliated pain clinics (Robinson et

al., in press), we expected three subgroups to emerge. We predicted that one subgroup

("pain-focused" subgroup) would show a greater relative interest in pain relief. We

expected a second subgroup ("multifocused-high" subgroup) to report that improvement

in all four domains is extremely important. We expected the final subgroup

("multifocused-moderate" subgroup) to provide moderate importance ratings across

domains. In addition to replicating previous findings, our study aimed to establish the

validity of the patient subgroups by examining pre-treatment differences among the

subgroups on demographic and psychosocial variables. We hypothesized that the two

multifocused subgroups would have significantly greater duration of pain than

participants in the "pain-focused" subgroup. We also expected that the

"multifocused-high" subgroup would have the highest scores on self-report measures of

pain, distress, and disability. We predicted the "pain-focused" subgroup to have the

lowest scores on measures of pain, distress, and disability.









Our final aim was to examine the potential for treatment matching using the patient

subgroups derived from the PCOQ. We predicted that patients who received treatment

that was matched to their clinical needs would show superior treatment outcomes, when

compared to patients who received treatment that was not matched to their clinical needs.

Therefore, we expected patients in the "pain-focused" subgroup to show better outcomes

after treatment that focused exclusively on their pain (i.e., palliative pain management),

and we expected patients in the multifocused subgroups to show better outcomes after

treatment that addressed not only their pain, but also their fatigue, emotional distress, and

interference with daily activities (i.e., multidisciplinary pain management).














CHAPTER 2
METHODS

Participants

Participants were 70 patients (27 male, 43 female) with a primary complaint of

chronic spine pain of at least 3 months duration. Participants were recruited from five

University of Florida or affiliated pain clinics, and they provided informed consent in

accordance with Institutional Review Board requirements. The mean age of participants

was 50.9 (SD = 12.5) years. Most participants (88.6%) were Caucasian and 11.4% were

African-American. Participants reported a mean of 13.0 (SD = 2.2) years of formal

education. Of participants, 66% percent were married. Participants reported a mean

duration of pain of 91.1 (SD = 95.7) months, with pain duration ranging from 4 to 468

months.

Participants underwent one of two types of treatment as part of their routine clinical

care: palliative pain management (interventional anesthesia/oral analgesic only) or

multidisciplinary pain management (behavioral/rehabilitation intervention component

included). The two treatment approaches are described in greater detail below.

Design

The design of the study was longitudinal, with pre-treatment and post-treatment

assessments. Post-treatment assessment occurred approximately 2 months after initiation

of treatment. Post-treatment data were collected on 55 of the 70 participants (21 males,

34 females). Nine participants were lost to follow-up for the following reasons: phone

number disconnected (3), no longer residing at stated phone number (1), on extended









vacation (1), could not reach at home despite multiple attempts (4). Four participants

withdrew by choice at follow-up, and two participants were withdrawn by the research

team at follow-up because they never received the intended treatment.

Procedure

Before starting treatment, eligible participants were approached by a

co-investigator or other research staff, given a brief description of the study, and then

asked if they would consider participating in the study. Interested participants were asked

to read and sign the informed consent form. After agreeing to participate in the study,

participants completed the Demographics Questionnaire and the following

paper-and-pencil measures: McGill Pain Questionnaire, Beck Depression Inventory, Pain

Anxiety Symptoms Scale, Roland Disability Questionnaire, Pain Disability Index, and

Patient-Centered Outcomes Questionnaire. Participants also rated the intensity and

unpleasantness of their pain on visual analog scales (VASs). Two months post-treatment,

research staff contacted participants by telephone to complete two measures, the

Follow-up PCOQ and Pain Service Satisfaction Test.

Treatments

Because of ethical considerations, treatment was not randomly assigned. All

participants were recruited from existing treatment programs, making our study highly

ecologically valid. Withdrawal before completion of the study did not affect the

participant's treatment.

Palliative Pain Management

Participants receiving palliative pain management underwent one or more of the

following four treatments: injection, radiofrequency denervation, non-narcotic analgesic

medication, and opioid medication. Injections consist of epidural administration of









steroid or anti-inflammatory medication, sometimes in combination with local anesthetic.

The injection is delivered under fluoroscopic guidance into the epidural space at the

spinal level determined by clinical exam to be the area relevant to the clinical pain

distribution. Injections serve to decrease pain by reducing nerve root irritation.

Radiofrequency denervation is a commonly performed interventional treatment for

axial low back pain (Dreyfuss, Halbrook, Pauza, Josi, McLarty, & Bogduk, 2000; North,

Han, Zahurak, & Kidd, 1994). The procedure is done by positioning a percutaneous

electrode next to the medial branches of the posterior nerve rami for a particular facet

joint, and then using a generator to induce heat through the electrode. This causes a lesion

of the nerve's ability to send messages to the posterior column of the spinal cord.

Radiofrequency denervation is completed under fluoroscopic guidance, and it may be

done unilaterally or bilaterally and at a single spinal level or at multiple spinal levels. The

number of levels to be lesioned is determined by prior diagnostic lidocaine injection of

the medial branches.

Patients typically receive chronic opioid medication if, and when, there are no other

interventions, including non-narcotic analgesic medication (e.g., Ultram), that might

manage their pain more effectively. Use of long- and short-acting opioids in a pattern to

better manage pain is discussed with the patient, as are potential side effects and adverse

outcomes. After this discussion, the initial dosing pattern of medication is started. Opioid

medications typically prescribed include OxyContin, MS Contin, Methadone, and the

Duragesic Patch. In most cases, a second prescription for breakthrough medication also is

given. If the patient responds favorably with no intolerable side effects at the initial dose,

he or she is maintained on the medication, and the dose is adjusted to achieve as much









pain reduction as possible without intolerable side effects. Should the patient have

intolerable side effects from the initial dose, he or she is switched to another opioid

medication.

Multidisciplinary Pain Management

Participants receiving multidisciplinary pain management participated in a pain

management group and/or underwent a standard rehabilitation program. Participants

sometimes received pharmacological therapy as an adjunct to this behavioral/

rehabilitation approach. The pain management group is a psychoeducational group

comprising cognitive-behavioral pain coping skills training, an empirically supported

treatment for chronic pain conditions (Keefe, 1996). This 10-session, structured,

outpatient group emphasizes instruction, rehearsal, and practice during session and at

home. Sessions are held once per week for 90 minutes. Patients initially are taught to

conceptualize their pain in terms of the gate control theory of pain (Melzack & Wall,

1965). Then, they are instructed in diaphragmatic breathing, which involves breathing

deeply from the diaphragm in a controlled manner. During the next session, patients are

taught progressive muscle relaxation, a series of exercises involving tensing and relaxing

various muscle groups, and given an opportunity to rehearse the technique in session. The

third session is devoted to discussing activity-rest cycling, a method for controlling pain

and increasing activity level that consists of planned periods of activity, interspersed with

planned regular rest periods. During the fourth session, patients are introduced to

cognitive restructuring, which involves systematically identifying and changing negative

automatic thoughts to more encouraging, realistic thoughts. They spend the next three

sessions participating in cognitive restructuring exercises using examples generated by

both group leaders and group members. The eighth session focuses on problem-solving,









and the final sessions focus on evaluating progress and maintaining treatment gains.

Other topics covered during group include pleasant activity scheduling, mini-relaxation,

and attention diversion techniques (e.g., pleasant imagery, focal point use).

The standard rehabilitation program consists of 18 sessions (3 per week), with each

session lasting approximately 90 minutes. Each session involves strength, range of

motion, flexibility, and aerobic training. Passive treatments (e.g., hot packs, ice packs,

electrical stimulation) are not employed during session, but patients are instructed in the

proper use of these modalities at home. Patients are prescribed a custom therapeutic

exercise program that addresses strength, range of motion, and flexibility deficits

detected during their evaluation. In addition, they undergo aerobic exercise, which has

been shown in a controlled trial to be an effective treatment for disability reduction in

chronic low back pain patients (Mannion, Muntener, Taimela, & Dvorak, 2001). Exercise

prescription is done in a "graded" manner (Lindstrom, Ohlund, Eek, Wallin, Peterson,

Fordyce, & Nachemson, 1992). Graded exercise prescription differs from traditional

exercise prescription because it focuses on improvement in activity tolerance, not on pain

abatement. During the final session, patients are issued a home exercise program for

continuation of unsupervised therapeutic exercise.

Measures

Demographics

The Demographics Questionnaire obtains information about a person's age, sex,

race, marital status, education, medical diagnosis, pain duration, and previous treatment

for the pain condition.









Pain

Participants completed visual analog scale (VAS) measures of pain intensity and

pain unpleasantness. For pain intensity, the scale ranges from 0 to 100, where

0 represents "no pain sensation" and 100 represents "most intense pain sensation

imaginable." For pain unpleasantness, the scale ranges from 0 (not at all unpleasant) to

100 (most unpleasant imaginable). The VAS provides ratio level pain measurement and

has been validated in multiple studies (Price, McGrath, Rafii, & Buckingham, 1983).

The McGill Pain Questionnaire (MPQ; Melzack & Katz, 2001) consists of 102

words in 20 subclasses. For each of the subclasses, patients are asked to choose the word

that best describes their pain at the present moment. Each word in each subclass has a

rank value, and rank values are summed to derive scores for the sensory, affective, and

evaluative dimensions of pain, as well as a total score. The MPQ is a reliable, valid, and

consistent measurement tool (Melzack & Katz, 2001).

Mood/Distress

The Beck Depression Inventory (BDI; Beck, Steer, & Garbin, 1988), a widely used

self-report measure of depression, assesses the extent to which an individual exhibits

each of 21 behaviors, thoughts, or affective features commonly associated with

depression. The BDI yields a total score (range = 0 to 63), as well as two subscale scores

measuring cognitive and somatic symptoms of depression. Geisser, Roth, and Robinson

(1997) showed that the BDI is a valid instrument for use with chronic pain patients.

The Pain Anxiety Symptoms Scale (PASS) is a 40-item self-report questionnaire

that measures four dimensions of pain-related anxiety, namely cognitive anxiety,

escape/avoidance, fearful appraisal, and physiological anxiety (McCracken, Zayfert, &

Gross, 1992). Participants respond to each item by selecting a frequency rating from









0 (never) to 5 (always). Five items are reverse scored. Frequency ratings are added to

derive a total score and four subscale scores. Total scores can range from 40 to 200. The

PASS is valid measure, as evidenced by significant correlations with other measures of

anxiety (McCracken et al., 1992). Alpha coefficients ranging from .81 to .94 indicate the

PASS and its subscales have good internal consistency (McCracken et al., 1992).

Disability

The Pain Disability Index (PDI) measures the degree to which chronic pain

interferes with the following life areas: family/home responsibilities, recreation, social

activity, occupation, sexual behavior, self care, and life-support activity (Pollard, 1984).

Amount of disability experienced in each of the seven domains is rated on an 11-point

scale ranging from 0 (no disability) to 10 (total disability). The ratings are summed to

derive a total score (range = 0 to 70). The PDI has acceptable psychometric properties,

with an internal consistency coefficient of .86 (Tait, Pollard, Margolis, Duckro, &

Krause, 1987).

The Roland Disability Questionnaire (RDQ) is a 24-item checklist designed to

measure functional impairment due to spine pain (Roland & Morris, 1983). The number

of items endorsed is summed to arrive at a total score ranging from 0 (no disability) to

24 (severe disability). The RDQ has adequate psychometric properties, with test-retest

reliability values ranging from .72 to .91 (Jensen, Strom, Turner, & Romano, 1992;

Roland & Morris, 1983).

Patient-Centered Outcomes

The Patient-Centered Outcomes Questionnaire (PCOQ; Appendix A) consists of

five sections. In the first section, patients indicate their usual levels (during the past

week) of pain, fatigue, emotional distress, and interference with daily activities on









101-point NRSs (0 = none, 100 = worst imaginable). In the second section, patients

indicate their desired levels of pain, fatigue, emotional distress, and interference with

daily activities using the same 101-point NRSs. The third section requires patients to

indicate where their pain, fatigue, emotional distress, and interference with daily

activities would have to be on 101-point NRSs (0 = none, 100 = worst imaginable) for

them to consider treatment successful. In the fourth section, patients indicate the levels of

pain, fatigue, emotional distress, and interference with daily activities they expect after

treatment on the same 101-point NRSs. Finally, participants are asked to rate on

101-point NRSs (0 = not at all important, 100 = most important) how important it is for

them to see improvement in each of the four domains.

Pilot data on 21 subjects indicates the PCOQ has acceptable test-retest reliability

over a 48-hour period, with reliability values ranging from .84 to .90 (ps < .001) for usual

levels. Reliability values for importance ratings ranged from to .62 to .82 (ps < .01) for

the domains of pain, emotional distress, and interference with daily activities. The

correlation for fatigue did not reach statistical significance (r = .33, p = .14). Reliability

values for success criteria are rather low in this sample despite achieving statistical

significance across all domains (r = .43 to .58, ps < .05) but emotional distress (r = .287,

p = .21). Data from our study indicate the PCOQ has good concurrent validity with

standardized measures of pain, mood/distress, and disability. For example, the correlation

between usual level of pain from the PCOQ and the pain intensity VAS was .52

(p < .001). Usual emotional distress was highly related to total scores on the BDI (r = .65,

p < .001) and PASS (r = .72, p < .001) and usual interference was significantly associated

with PDI (r = .75, p < .001) and RDQ (r = .69, p < .001) total scores.









The Follow-up PCOQ (Appendix B) includes six sections. The first section has

patients estimate their pre-treatment levels of pain, fatigue, emotional distress, and

interference with daily activities on the same 101-point NRS used in the PCOQ

(0 = none, 100 = worst imaginable). The next section requires patients to indicate the

levels of pain, fatigue, emotional distress, and interference with daily activities associated

with treatment success on the same 101-point NRSs. In the third section, patients rate on

101-point NRSs (0 = not at all important, 100 = most important) how important it is for

them to see improvement in each of the four domains. In the next section, patients rate on

101-point NRSs (0 = none, 100 = worst imaginable) their usual levels of pain, fatigue,

emotional distress, and interference with daily activities after treatment. In the fifth

section, patients specify whether they improved, stayed the same, or worsened across

each domain since the start of treatment and then, if applicable, rate on 101-point NRSs

(0 = none, 100 = complete) the amount of improvement or worsening they experienced in

each of the domains. Patients also indicate whether they improved, stayed the same, or

worsened overall. In the last section, patients decide whether they would consider their

treatment successful. For each of the domains, patients choose yes or no. Patients also

evaluate the overall success of their treatment by choosing yes or no.

Treatment Satisfaction

The 20-item Pain Service Satisfaction Test (PSST) assesses patient satisfaction

with pain treatment (McCracken, Klock, Mingay, Asbury, & Sinclair, 1997). The PSST

has high internal consistency, and is positively correlated with treatment-related changes

in pain, depression, and anxiety (McCracken & Turk, 2002).









Analytical Strategy

Aim 1: Developing Patient-Centered Success Criteria

Analyses used usual ratings from the PCOQ, and usual ratings and success

judgments from the Follow-up PCOQ. First, we calculated change scores for each of the

four domains (pain, fatigue, emotional distress, interference with daily activities) by

subtracting Follow-up PCOQ usual ratings (i.e., post-treatment usual ratings) from PCOQ

usual ratings (i.e., pre-treatment usual ratings). Then, to characterize the association

between change scores and patient judgment of treatment success, we derived

receiver-operating characteristic (ROC) curves for each of the four domains using logistic

regression analyses. For each analysis, judgment of success (yes, no) served as the

dependent variable and either the raw change or percent change score served as the

independent variable.

For these analyses, area under the ROC curve was interpreted as the probability of

distinguishing participants who were successfully treated on the domain from participants

who were not successfully treated on the domain. The area under the ROC curve can

range from .5 (no accuracy) to 1.0 (perfect accuracy). We then computed optimal cut-offs

(i.e., MCIDs) for each domain from these ROC curves. We expected optimal cut-offs to

have a high hit rate coupled with a good balance of sensitivity and specificity.

Aim 2: Identifying Patient Subgroups

We conducted a hierarchical agglomerative cluster analysis (Ward's method,

squared Euclidian distance) on PCOQ importance ratings to derive the patient subgroups.

Once we identified the subgroups, we conducted a series of analyses to examine potential

subgroup differences on demographic and psychosocial variables. First, we performed

one-way analyses of variance (ANOVAs) to examine possible subgroup differences in









age, education, and pain duration. Then, we performed a discriminant function analysis

using PCOQ usual ratings of pain, fatigue, emotional distress, and interference with daily

activities to discriminate spine pain patients classified into each of the subgroups. We

performed a second discriminant function analysis using standardized measures of pain

(MPQ), distress (BDI, PASS), and disability (PDI, RDQ) to discriminate spine pain

patients classified into each of the subgroups. Lastly, we calculated effect sizes

(Cohen's d; Cohen, 1988) on PCOQ and standardized measures of pain, fatigue, distress,

and disability to examine the magnitude of subgroup differences in standard deviation

units. More specifically, we calculated effect sizes by dividing mean subgroup

differences by pooled standard deviation for each measure (Mi M2/SDp). We

interpreted the effect sizes using Cohen's (1988) criteria of .20, .50, and .80 for small,

medium, and large effects, respectively.

Aim 3: Examining Treatment Matching

Analyses used PSST scores, PCOQ usual ratings, Follow-up PCOQ usual ratings,

and MCIDs derived in Aim 1. First, we performed a 2 (subgroup) X 2 (treatment)

ANOVA on post-treatment PSST scores to examine the effects of subgroup membership

and type of treatment on pain treatment satisfaction. Next, we performed 2 (subgroup) X

2 (treatment) X 2 (time) mixed-model ANOVAs on usual ratings of pain, fatigue,

emotional distress, and interference with daily activities to investigate the effects of

subgroup membership and type of treatment on these ratings across time. Subgroup and

treatment served as between-subjects factors and time was the within-subjects factor.

Finally, we conducted chi-square analyses for each of the domains to examine the

relationship between type of treatment and treatment success as determined by MCID

attainment using MCIDs established in Aim 1.















CHAPTER 3
RESULTS

Descriptive Statistics

Participants reported moderate usual levels of pain, fatigue, emotional distress, and

interference with daily activities before treatment (M = 51.1 to M = 65.4) (Table 3-1).

Furthermore, participants considered a 26.9/100 pain level or a mean reduction in pain of

38.5 points (58%) to represent treatment success. A 23.2/100 fatigue level or a mean

reduction in fatigue of 36.9 points (61%), an 18.3/100 distress level or a mean reduction

in distress of 32.8 points (64%), and a 21.6/100 interference level or a mean reduction in

interference of 43.1 points (66%) also represented successful treatment. As a group,

participants viewed improvements in all domains (pain, fatigue, emotional distress, and

interference with daily activities) as very important (M = 83.4 to M = 89.1).

Table 3-1. Descriptive Statistics from the PCOQ (N= 70)
Level Domain Mean* SD Observed Range
Usual Pain 65.4 21.6 20-97
Fatigue 60.1 24.4 0-100
Distress 51.1 30.0 0-100
Interference 64.7 27.4 0-100
Success Pain 26.9 18.3 0-75
Fatigue 23.2 18.4 0-70
Distress 18.3 19.3 0-80
Interference 21.6 18.5 0-80
Importance Pain 89.1 20.3 10-100
Fatigue 83.4 24.4 0-100
Distress 84.0 26.2 0-100
Interference 87.0 23.7 0-100
* Numerical Rating Scale = 0 to 100









Participants reported moderate usual levels of pain, fatigue, distress, and

interference post-treatment (M = 47.5 to M = 56.4) (Table 3-2). After treatment, they

considered a 33.9/100 pain level, 30.2/100 fatigue level, 26.1/100 distress level, and

28.1/100 interference level to represent treatment success. Again, participants viewed

improvements in all domains as quite important (M = 78.8 to M = 88.9).

Table 3-2. Descriptive Statistics from the Follow-up PCOQ (N = 55)
Level Domain Mean* SD Observed Range
Usual Pain 56.4 27.4 0-100
Fatigue 53.5 27.9 0-100
Distress 47.5 30.9 0-100
Interference 53.5 30.6 0-100
Success Pain 33.9 17.6 0-75
Fatigue 30.2 17.9 0-70
Distress 26.1 17.9 0-70
Interference 28.1 16.6 0-75
Importance Pain 88.9 21.2 0-100
Fatigue 78.8 28.1 0-100
Distress 83.0 25.1 0-100
Interference 82.8 26.9 0-100
* Numerical Rating Scale = 0 tol00

Four paired-samples t-tests were conducted to evaluate whether participants'

PCOQ (i.e., pre-treatment) success criterion and Follow-up PCOQ (i.e., post-treatment)

success criterion significantly differed for each domain. The results showed that

participants' Follow-up PCOQ (i.e., post-treatment) success criterion for pain was

significantly higher than their PCOQ (i.e., pre-treatment) success criterion for pain [t(52)

= 2.125, p = .038], suggesting that participants had higher target values for pain treatment

success (i.e., less stringent success criteria for pain) after treatment. Similarly, patients

had higher target values (i.e., less stringent success criteria) for successful treatment of

fatigue and distress after treatment, t(52) = 2.14, p = .037 and t(52) = 2.16, p = .035,

respectively. Results for interference approached statistical significance [t(52) = 1.77,









p = .083] but did not reach the p < .05 standard. These findings indicate that participants

adjusted or re-defined their success criteria during the course of treatment in the direction

of becoming less strict.

A series of chi-square analyses were conducted to determine if participants who

met their post-treatment success criterion for a particular PCOQ domain were more likely

to judge themselves as successfully treated on that domain (yes/no success judgments)

than participants who did not meet their post-treatment success criterion for the domain.

The results showed that success criteria and success judgments were significantly related

for the domain of pain, Pearson X2 (1, N = 52) = 22.768, p < .001. Compared to patients

who did not meet their post-treatment success criterion for pain, patients who met their

post-treatment success criterion for pain were 3.7 times (100% v 27%) more likely to

judge their pain as having been successfully treated (Table 3-3).

Similar results were found for the domains of fatigue [Pearson X2 (1, N = 51) =

11.574, p = .001], distress [Pearson X2 (1, N = 51) = 27.019, p < .001], and interference

[Pearson X2 (1, N = 51) = 19.538, p < .001]. Thus, patients who met their post-treatment

success criteria for fatigue, distress, and interference were 2.6 (86% v 32%), 5.3

(94% v 18%), and 3.7 (89% v 24%) times more likely, respectively, than patients who

did not meet their post-treatment success criteria for these domains to judge treatment of

these domains a success (Table 3-4, Table 3-5, and Table 3-6). These findings suggest

participants used their success criteria to assist them in making global judgments of

treatment success.









Table 3-3. Pain: Observed Frequencies
Success Judgment
No Yes
Achieved No n = 27 n= 10
Post-treatment 73% 27%
Success Yes n = 0 n = 15
Criterion? 0% 100%

Table 3-4. Fatigue: Observed Frequencies
Success Judgment
No Yes
Achieved No n = 25 n = 12
Post-treatment 68% 32%
Success Yes n = 2 n = 12
Criterion? 14% 86%

Table 3-5. Distress: Observed Frequencies
Success Judgment
No Yes
Achieved No n = 28 n = 6
Post-treatment 82% 18%
Success Yes n = 1 n = 16
Criterion? 6% 94%

Table 3-6. Interference: Observed Frequencies
Success Judgment
No Yes
Achieved No n = 25 n = 8
Post-treatment 76% 24%
Success Yes n = 2 n = 16
Criterion? 11% 89%


Aim 1: Developing Patient-Centered Success Criteria

Optimal cut-off points (i.e., MCIDs) for pain generated from the ROC curves are a

raw change of 17.5 points and a percent change of 25% (Table 3-7). That is, MCIDs for

pain were a reduction in pain greater than or equal to 17.5 points (on a 0 to 100 NRS) or a

25% or greater reduction in pain. Optimal cut-off points for the other domains were 7.5

points or 11% for fatigue, 5.0 points or 13% for emotional distress, and 9.5 points or 12%

for interference with daily activities (Table 3-7). Hit rates for these cut-offs ranged from









73 to 85%. In other words, these cut-offs accurately classified 73 to 85% of patients as

either successfully or unsuccessfully treated.

Table 3-7. Success Criteria and Associated Classification Accuracy Statistics
Domain Variable (Cut-off*) ROC (SE) HR SENS SPEC PPV NPV
Pain Raw Change (>17.5) .86 (.05)*** .85 .72 .96 .95 .79
Percent Change (>25%) .85 (.06)*** .85 .72 .96 .95 .79
Fatigue Raw Change (>7.5) .80 (.07)*** .73 .70 .75 .70 .75
Percent Change (>11%) .78 (.07)*** .73 .71 .75 .71 .75
Distress Raw Change (>5) .76 (.07)*** .77 .82 .73 .69 .85
Percent Change (>13%) .82 (.07)*** .82 .81 .83 .77 .86
Interference Raw Change (>9.5) .76 (.07)*** .73 .67 .79 .73 .73
Percent Change (>12%) .77 (.08)*** .75 .70 .79 .73 .76
Classification accuracy statistics are based on identification of patients who were
successfully treated on the domain relative to patients who were not successfully treated
on the domain. ROC = area under the receiver-operating characteristic curve; HR = hit
rate; SENS = sensitivity, SPEC = specificity; PPV = positive predictive value; NPV =
negative predictive value; Numerical Rating Scale = 0 to 100; *** p < .001.

Aim 2: Identifying Patient Subgroups

Contrary to the hypothesized three-cluster solution, the hierarchical agglomerative

cluster analysis (Ward's method, squared Euclidian distance) on PCOQ importance

ratings revealed a two-cluster solution (Figure 3-1, Table 3-8). Cluster 1 comprised 7

participants who had low to moderate importance ratings across the domains of pain,

fatigue, distress, and interference, whereas Cluster 2 comprised 63 participants who had

high importance ratings across domains. The obtained cluster solution limited statistical

power in subsequent inferential analyses.

A series of one-way ANOVAs did not reveal any differences between the clusters

on the demographic variables of age [F(1, 68) = 2.270, p = .136], education [F(1, 68) =

.310, p = .579], and pain duration [F(1, 66) = 1.689, p = .198]. A discriminant function

analysis was conducted to determine whether four variables-usual pain, usual fatigue,

usual distress, and usual interference from the PCOQ-could predict cluster membership.

































Fatigue


Cluster 1 (n 7)

Cluster 2 (n = 63)


Distress Interference


Domain




Figure 3-1. Cluster Solution for PCOQ Importance Ratings.


Table 3-8. Scores on PCOQ and Standardized Measures (Mean SD) by Cluster
Measure Cluster 1 (n = 7) Cluster 2 (n = 63)
PCOQ Usual Pain 53.6 + 21.0 66.7 + 21.4
PCOQ Usual Fatigue 55.7 + 30.5 60.6 + 23.9
PCOQ Usual Distress 31.4 + 32.8 53.3 + 29.1
PCOQ Usual Interference 42.1 36.0 67.2 25.5
MPQ Total 29.6 + 17.8 32.0 + 15.3
BDI Total 13.7 11.8 19.3 11.5
PASS Total 78.6 36.0 86.1 37.6
PDI Total 33.2 20.0 41.7 + 15.0
RDQ Total 11.0 + 8.1 14.6 5.6
PCOQ = Patient-Centered Outcomes Questionnaire; MPQ = McGill Pain Questionnaire;
BDI = Beck Depression Inventory; PASS = Pain Anxiety Symptoms Scale; PDI = Pain
Disability Index; RDQ = Roland Disability Questionnaire









The overall Wilks' lamda did not quite reach statistical significance, A = .884,

X2 (4, N = 70) = 8.102, p = .088, indicating that the predictors did not differentiate the

two clusters.

A second discriminant function analysis using five predictors-MPQ, BDI, PASS,

PDI, and RDQ total scores-also did not predict cluster membership, A = .945,

X2 (5, N = 68) = 3.577, p = .612. However, examination of effect sizes (Table 3-9),

revealed moderate to large effect sizes (range = .48 to .94) for PCOQ usual pain, PCOQ

usual distress, PCOQ usual interference, BDI Total, PDI Total, and RDQ Total.

Participants in Cluster 1 had lower scores than participants in Cluster 2 across these

measures (Table 3-8), suggesting that participants in Cluster 1 had significantly less pain,

depression/distress, and disability/interference than participants in Cluster 2. Taken

together, these findings partially support hypotheses regarding subgroup differences.

Table 3-9. Effect Sizes of Cluster Differences (Cohen's d)
Measure Cohen's d
PCOQ Usual Pain .61
PCOQ Usual Fatigue .19
PCOQ Usual Distress .74
PCOQ Usual Interference .94
MPQ Total .15
BDI Total .48
PASS Total .19
PDI Total .55
RDQ Total .61

Aim 3: Examining Treatment Matching

Hypotheses regarding treatment matching could not be explored because the

previous finding of a "pain-focused" versus multifocused subgroups was not replicated.

However, several analyses were done involving the two clusters obtained from the

present sample to examine potential treatment effects. A 2 (subgroup: Cluster 1,









Cluster 2) X 2 (treatment: palliative, multidisciplinary) ANOVA on post-treatment PSST

scores revealed only a main effect of subgroup, F(1, 48) = 6.652, p= .013, eta2 = .12,

indicating that participants in Cluster 1 were significantly more satisfied with their

treatment (M = 57.4) than participants in Cluster 2 (M = 44.1) regardless of the type of

treatment received.

A series of 2 (subgroup: Cluster 1, Cluster 2) X 2 (treatment: palliative,

multidisciplinary) X 2 (time: pre-treatment, post-treatment) mixed-model ANOVAs were

performed on usual ratings of pain, fatigue, distress, and interference to investigate the

effects of subgroup membership and type of treatment on these ratings across time.

Subgroup and treatment served as between-subjects factors and time was the

within-subjects factor. The mixed-model ANOVA on usual ratings of pain did not reveal

any statistically significant main or interaction effects. Likewise, mixed-model ANOVAs

on usual ratings of fatigue and distress did not reveal any statistically significant main or

interaction effects. The mixed-model ANOVA on usual ratings of interference revealed

significant main effects of subgroup [F(1, 51) = 6.367, p= .015, eta2 = 0.11] and

treatment [F(1, 51) = 7.112, p = .010, eta2 = 0.12], and a significant subgroup by

treatment interaction effect [[F(1, 51) = 6.459, p = .014, eta2 = 0.11]. The interaction

(Figure 3-2) may be misleading, however, due to the fact that only two participants in

Cluster 1 received multidisciplinary treatment. Closer examination of the main effect of

subgroup revealed that participants in Cluster 1 had a lower mean usual level of

interference (M = 32.8) than participants in Cluster 2 (M = 59.6). Additionally,

participants receiving multidisciplinary treatment had a lower mean usual level of

interference (M = 32.0) than participants receiving palliative treatment (M = 60.4).















100

90 -

80 -

70

S60 --------------------- -- Clusterl(n=7)
60 Cluster 1 (n 7)

0 D ----- Cluster 2(n 63)
4 40

30

20 -

10 -

0
0 --------------------
Palliative Multidisciplinary
Treatment



Figure 3-2. Subgroup by Treatment Interaction for Usual Interference.

Finally, a series of chi-square analyses were conducted to examine the relationship

between type of treatment (palliative v multidisciplinary) and treatment success as

determined by MCID attainment (yes, no) for each of the domains using MCIDs

established in Aim 1. For the domain of pain, results showed that type of treatment was

unrelated to MCID attainment, Pearson X2 (1, N = 54) = .616, p = .433. In other words,

participants who received palliative pain management and participants who received

multidisciplinary pain management were equally likely to attain the MCID for pain.

Nonsignificant results also were obtained for the domains of fatigue [Pearson X2

(1, N = 53) = .190, p = .663], distress [Pearson 2 (1, N = 54) = 2.030, = .154], and






34


interference [Pearson X2 (1, N = 54) = .533, p = .465], suggesting that type of treatment

was unrelated to treatment success in this sample.














CHAPTER 4
DISCUSSION

Our study aimed to establish patient-determined MCIDs for treatment of chronic

pain. Consistent with previous research using the PCOQ (Robinson et al., in press), we

found that spine pain patients require large reductions in pain for treatment to be deemed

successful. At the onset of treatment, patients in our study considered a 26.9/100 pain

level or a mean reduction in pain of 38.5 points (58%) to represent treatment success.

This patient-defined success criterion derived using a direct scaling approach is more

stringent than criteria developed using other methods, including expert opinion

(e.g., Goldsmith et al., 1993; Moore et al., 1996) and scale comparisons (e.g., Farrar et

al., 2001; Angst et al., 2002).

Our study expands on previous research by deriving MCIDs for other domains

relevant to pain treatment. Our findings suggest spine pain patients also require large

reductions in fatigue (23.2/100 or 61%), emotional distress (18.3/100 or 64%), and

interference with daily activities (21.6/100 or 66%) for treatment to be considered

successful. These pre-treatment MCIDs are highly consistent with those established in

our prior study (Robinson et al., in press), suggesting they are reliable and valid. Our

findings also confirm that successful treatment of chronic pain is not defined exclusively

in terms of pain reduction, but involves a host of endpoints.

If patients adhere to these success criteria throughout treatment, then clinical trials

that apply expert-generated MCIDs or use only standard statistical methods might

overestimate the efficacy of their treatments. Therefore, it is important to know whether









chronic pain patients adjust their criteria for treatment success over time. Results of our

study suggest patients do indeed modify their success criteria from pre-treatment to

post-treatment in the direction of becoming less stringent, as target values for successful

treatment of pain, fatigue, and emotional distress were significantly higher two months

after treatment initiation (33.9, 30.2, and 26.1, respectively). Based on our findings, it

appears that patients used these post-treatment cut-offs to make decisions about treatment

success. In our study, patients who met their post-treatment success criterion for a

particular domain were significantly more likely than patients who did not meet their

post-treatment success criterion for the domain to judge treatment of the domain a

success. These results further support the utility of patient-defined success criteria.

It should be noted that 48-hour test-retest reliability values for PCOQ success

criteria were somewhat low in another sample of 21 chronic pain patients. One possible

explanation for this finding is that patients' success criteria are influenced by their current

levels of pain, fatigue, emotional distress, and interference with daily activities, such that

as their current levels naturally fluctuate from day to day so do their success criteria.

Fluctuations in success criteria might be larger than corresponding fluctuations in usual

levels. Patients also might be less familiar with and experienced at rating success criteria

than they are at rating usual levels, thereby introducing more error variance into ratings

of success. These explanations are admittedly speculative and somewhat beyond the

scope of our study. Despite the questionable test-retest reliability of PCOQ success

criteria, mean success criteria for the four PCOQ domains were highly consistent across

our two samples of chronic pain patients. As discussed, success criteria derived in this

study were almost identical to success criteria derived in our original study (Robinson et









al., in press) aside from the change from a 0 to 10 NRS to a 0 to 100 NRS. Furthermore,

in our study, pre-treatment and post-treatment success criteria still were reliably different

even with additional error variance incorporated into the analyses.

The aforementioned patient-centered success criteria were derived using a direct

scaling approach. Another more commonly used approach to developing MCIDs involves

comparing two scales, typically the change on a given domain (e.g., change in pain on a

0 to 10 NRS) and a global rating of improvement (e.g., "much improved"). This

methodology has promise, but use of different anchors (e.g., "much improved" versus

"somewhat better") has led to differing results (e.g., Farrar et al., 2001; Wells et al.,

1993). In fact, Salaffi and colleagues (2004) calculated MCIDs in pain associated with

both a "slightly better" and "much better" global impression of change among their

sample of chronic musculoskeletal pain patients, and obtained different raw and percent

changes for the two anchors. Furthermore, investigator-determined anchors might not

reflect the degree of improvement necessary for patients to consider treatment successful.

Therefore, our study sought to (1) characterize the association between NRS change

scores and treatment success as determined by patient global judgment and (2) select as

MCIDs the change scores that best distinguished between successfully and

unsuccessfully treated patients. This procedure was done not only for the domain of pain,

but also for the domains of fatigue, emotional distress, and interference with daily

activities.

Using the scale comparison approach, we derived the following MCIDs: 17.5

points (on a 0 to 100 NRS) or 25% for pain, 7.5 points (11%) for fatigue, 5.0 points

(13%) for emotional distress, and 9.5 points (12%) for interference with daily activities.









These MCIDs accurately classified 73 to 85% of patients as successfully or

unsuccessfully treated. MCIDs for pain were most accurate in classifying patients.

Interestingly, our MCIDs for pain are fairly consistent with those derived by Farrar et al.

(2001) using a 0 to 10 NRS and a "much improved" global impression of change

(2 points or 30%).

Future research should focus on validating these patient-determined MCIDs,

particularly for the less studied domains of fatigue, emotional distress, and interference

with daily activities. Once validated, they can be used to calculate effect size estimates,

such as the number needed to treat (Cook & Sackett, 1995), permitting comparisons

across studies and ultimately helping to evaluate the efficacy of various treatments for

chronic pain (Farrar et al., 2001; Redelmeier et al., 1996). Additionally, MCIDs can be

useful in assessing the care of individual patients. Future research should also explore

whether success criteria are absolute or related to pre-treatment levels. A recent study by

Salaffi and colleagues (2004) suggests MCIDs may not be standardized along the whole

NRS. They found that patients with a higher level of pain before treatment required

greater reductions in pain to obtain clinically important improvements than did patients

with relatively low pain levels before treatment. Consequently, in applying our MCIDs to

individual patients or other pain populations, one should note that our sample had

moderate to moderately high mean levels (on a 0 to 100 NRS) of pain (M = 65.4), fatigue

(M = 60.1), emotional distress (M = 51.1), and interference with daily activities

(M = 64.7) at the onset of treatment. Finally, future research should investigate the

evolution of success criteria throughout the course of different treatments. Our results

suggest patients change their success criteria while undergoing treatment. However, it









remains unclear whether patients are more or less likely to adjust their success criteria

with different treatments. For example, patients receiving cognitive behavioral therapy

(CBT) might be expected to change their success criteria considerably, given that the

goals of CBT are to examine potential maladaptive beliefs (e.g., unrealistic expectations

regarding treatment) and re-define treatment outcomes.

A second aim of our study was to empirically derive and validate patient subgroups

using patient ratings of the perceived importance of improvement in the four PCOQ

domains. Contrary to our hypothesized three-cluster solution, only two clusters emerged.

One subgroup (Cluster 2) rated improvement in all domains as extremely important,

whereas the other, smaller subgroup (Cluster 1) reported relatively low importance

ratings across domains. Further examination of the cluster analysis revealed that the

hypothesized "multifocused-moderate" subgroup merged with the "multifocused-high"

subgroup to form Cluster 2. Patients in Cluster 2 had higher pre-treatment usual levels of

pain than patients in Cluster 1. They also were more distressed and disabled at the start of

treatment. Interestingly, the clusters did not differ on chronicity of pain, suggesting that

pain duration does not influence subgroup membership. Based on these data, one could

speculate that patients in Cluster 1 might be less likely than patients in Cluster 2 to

require multidisciplinary treatment to obtain clinically important treatment outcomes.

Future research might examine this hypothesis.

Our findings provide support for the notion that chronic pain patients are

heterogeneous. Even among patients whose pathology is specific to the spine, distinct

subgroups emerged. Of note, we did not replicate our previous finding of a truly

"pain-focused" subgroup (Robinson et al., in press). The possibility exists that a subset of









patients with chronic pain in other body regions might believe that treatments addressing

nociception are most pertinent to them. Future investigations using larger and more

diverse samples of chronic pain patients are needed to examine this hypothesis. Another

explanation for our inconsistent findings might lie with the variables on which the

subgroups were derived. Importance ratings might not be the best criteria on which to

base subgroup membership. Other variables, such as usual ratings or success criteria,

might produce more stable and meaningful patient subgroups. Therefore, future

investigations should consider using other PCOQ variables to develop patient subgroups.

Our final aim was to examine the potential for treatment matching using the patient

subgroups derived from the PCOQ. Because a "pain-focused" subgroup did not emerge,

we could not fully explore hypotheses regarding treatment matching. Nonetheless, we did

analyses to examine treatment effects across the two subgroups derived in Aim 2.

Although patients in Cluster 1 were more satisfied with treatment than patients in Cluster

2, they did not achieve greater reductions in their pain, fatigue, emotional distress, and

interference with daily activities from pre-treatment to post-treatment. This pattern of

findings occurred irrespective of the type of treatment received (palliative pain

management or multidisciplinary pain management). Additional analyses showed, at least

in our sample, that type of treatment is unrelated to MCID attainment. That is, patients

receiving palliative pain management and patients receiving multidisciplinary pain

management were equally likely to obtain MCIDs for pain, fatigue, emotional distress,

and interference with daily activities. This finding is somewhat surprising given

overwhelming evidence for the superiority of multidisciplinary treatment approaches









(Flor, Fydrich, & Turk, 1992). Nonetheless, it further underscores the need to determine

what treatments are effective for which subgroups of chronic pain patients.

Potential limitations of our study deserve mention. First, our reduced sample size

likely prevented a separate "multifocused-moderate" patient subgroup from emerging in

the cluster analysis. Notably, we conducted our original cluster analysis (Robinson et al.,

in press) on 110 patients, whereas we performed this cluster analysis on 70 patients.

Likewise, we could not conduct certain follow-up chi-square analyses, particularly those

examining possible subgroup by treatment interactions, because of small cell sizes. A

second limitation of our study pertains to the non-uniformity of the two treatment

approaches. Ideally, patients receiving palliative pain management would have been

given the same treatment (or at least the same class of treatment, pharmacological or

interventional) and patients undergoing multidisciplinary pain management would have

received identical treatment components. Another potential limitation of our study

involves the use of dichotomous judgments of success as the standard against which

MCIDs were developed. The possibility exists, for instance, that patients' global

judgments of treatment success for the domain of pain were influenced by factors

unrelated to pain intensity, such as satisfaction with care. Similarly, patients' global

judgments of treatment success for the other PCOQ domains (fatigue, distress, and

interference) might have been unduly influenced by their global judgments for pain.

Nevertheless, enabling patients to incorporate into their global judgments of treatment

success all factors perceived as germane to the decision produces success criteria that are

ecologically valid. A final concern relates to the generalizability of these findings to other

chronic pain populations. Patients with chronic spine pain might view treatment success









differently than patients with other types of chronic pain. Therefore, our MCIDs might

not be applicable to other painful conditions. However, we are encouraged that our

MCIDs for pain are similar to those developed with other pain populations (e.g,. Farrar et

al., 2001).

These limitations notwithstanding, our study is the first to develop MCIDs from the

patient perspective across multiple domains relevant to treatment of chronic pain. Once

validated, these MCIDs can easily be used by both clinicians and researchers to evaluate

the effectiveness of various treatment approaches for chronic pain. Our study also further

debunks the pain patient homogeneity assumption (Turk & Okifuji, 2001) by identifying

at least two distinct subgroups of spine pain patients that differ on important variables,

such as pain intensity, distress, and disability. It remains to be seen whether additional

subgroups will emerge from larger and more diverse samples of chronic pain patients.

Finally, as we were unable to adequately examine the treatment-matching hypothesis

(Turk & Okifuji, 2001), future research should explore the potential benefits of tailoring

treatment to the behavioral and psychosocial characteristics of these and other subgroups

of patients with chronic pain.














APPENDIX A
PATIENT-CENTERED OUTCOMES QUESTIONNAIRE (PCOQ)

MANY PEOPLE EXPERIENCE PAIN, FATIGUE (I.E., FEELING TIRED),
EMOTIONAL DISTRESS (E.G., WORRIES, FEELING SAD), AND INTERFERENCE
WITH DAILY ACTIVITIES (E.G., NOT BEING ABLE TO WORK OR DO
HOUSEHOLD CHORES) AS A RESULT OF THEIR MEDICAL CONDITION. WE
WOULD LIKE TO UNDERSTAND HOW YOU HAVE BEEN IMPACTED IN EACH
OF THESE AREAS. WE WOULD ALSO LIKE TO LEARN MORE ABOUT WHAT
YOU WANT YOUR TREATMENT TO DO FOR YOU.

FIRST, WE WOULD LIKE TO KNOW YOUR USUAL LEVELS OF PAIN,
FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE.

On a scale of 0 (none) to 100 (worst imaginable), please indicate your usual level
(during the past week) of...

pain

fatigue (or tiredness)

emotional distress

interference with daily activities

NOW, WE WOULD LIKE TO LEARN ABOUT YOUR DESIRED LEVELS OF PAIN,
FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE. IN OTHER WORDS,
WE WOULD LIKE TO UNDERSTAND WHAT YOUR IDEAL TREATMENT
OUTCOME WOULD BE.

On a scale of 0 (none) to 100 (worst imaginable), please indicate your desired level of ...

pain

fatigue (or tiredness)

emotional distress

interference with daily activities










PATIENTS UNDERSTANDABLY WANT THEIR TREATMENT TO RESULT IN
DESIRED OR IDEAL OUTCOMES LIKE YOU INDICATED ABOVE.
UNFORTUNATELY, AVAILABLE TREATMENTS DO NOT ALWAYS PRODUCE
DESIRED OUTCOMES. THUS, IT IS IMPORTANT FOR US TO UNDERSTAND
WHAT TREATMENT OUTCOMES YOU WOULD CONSIDER SUCCESSFUL.

On a scale of 0 (none) to 100 (worst imaginable), please indicate the level each of these
areas would have to be at for you to consider treatment successful.

pain

fatigue (or tiredness)

emotional distress

interference with daily activities

NOW, WE WOULD LIKE TO KNOW WHAT YOU EXPECT YOUR TREATMENT
TO DO FOR YOU.

On a scale of 0 (none) to 100 (worst imaginable), please indicate the levels you expect
following treatment.

pain

fatigue (or tiredness)

emotional distress

interference with daily activities

FINALLY, WE WOULD LIKE TO UNDERSTAND HOW IMPORTANT IT IS FOR
YOU TO SEE IMPROVEMENT IN YOUR PAIN, FATIGUE, EMOTIONAL
DISTRESS, AND INTERFERENCE FOLLOWING TREATMENT.

On a scale of 0 (not at all important) to 100 (most important), please indicate how
important it is for you to see improvement in your...

pain

fatigue (or tiredness)

emotional distress


* interference with daily activities















APPENDIX B
FOLLOW-UP PCOQ


FIRST, WE WOULD LIKE TO KNOW YOUR PRE-TREATMENT LEVELS OF
PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE.

On a scale of 0 (none) to 100 (worst imaginable), please estimate your pre-treatment
level of...

pain

fatigue (or tiredness)

emotional distress

interference with daily activities


BASED ON YOUR PRE-TREATMENT LEVELS OF PAIN, FATIGUE, EMOTIONAL
DISTRESS, AND INTERFERENCE, PLEASE INDICATE WHAT TREATMENT
OUTCOMES YOU WOULD CONSIDER SUCCESSFUL.

On a scale of 0 (none) to 100 (worst imaginable), please indicate the level each of these
areas would have to be at for you to consider treatment successful.

pain

fatigue (or tiredness)


* emotional distress

* interference with daily activities










NEXT, WE WOULD LIKE TO UNDERSTAND HOW IMPORTANT IT IS FOR YOU
TO SEE IMPROVEMENT IN YOUR PAIN, FATIGUE, EMOTIONAL DISTRESS,
AND INTERFERENCE.

On a scale of 0 (not at all important) to 100 (most important), please indicate how
important it is for you to see improvement in your...

pain

fatigue (or tiredness)


* emotional distress

* interference with daily activities


FINALLY, WE WOULD LIKE TO KNOW YOUR USUAL LEVELS OF PAIN,
FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE FOLLOWING
TREATMENT. IN OTHER WORDS, WHAT ARE YOUR CURRENT LEVELS OF
PAIN, FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE?

On a scale of 0 (none) to 100 (worst imaginable), please indicate your usual level
(during the past week) of...


* pain

* fatigue (or tiredness)

* emotional distress


* interference with daily activities











ACCORDING TO OUR RECORDS, YOUR PRE-TREATMENT LEVELS OF PAIN,
FATIGUE, EMOTIONAL DISTRESS, AND INTERFERENCE [ON A SCALE FROM
0 (NONE) TO 100 (WORST IMAGINABLE)] WERE AS FOLLOWS:

pain

fatigue (or tiredness)

emotional distress

interference with daily activities




NOW WE WOULD LIKE TO KNOW IF YOU HAVE IMPROVED, STAYED THE
SAME, OR WORSENED SINCE THE START OF TREATMENT.

1. Has your pain improved, stayed the same, or worsened?

Please choose one:
Improved (Please complete Part A below.)
Stayed the same (Please move on to #2.)
Worsened (Please complete Part B below.)

Part A: On a scale of 0 (no improvement) to 100 (complete improvement),
please indicate how much your pain has improved (Please move on to #2.)

Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please
indicate how much your pain has worsened (Please move on to #2.)

2. Has your fatigue (or tiredness) improved, stayed the same, or worsened?

Please choose one:
Improved (Please complete Part A below.)
Stayed the same (Please move on to #3.)
Worsened (Please complete Part B below.)

Part A: On a scale of 0 (no improvement) to 100 (complete improvement),
please indicate how much your fatigue has improved (Please move on to
#3.)

Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please
indicate how much your fatigue has worsened (Please move on to #3.)









3. Has your emotional distress improved, stayed the same, or worsened?

Please choose one:
Improved (Please complete Part A below.)
Stayed the same (Please move on to #4.)
Worsened (Please complete Part B below.)

Part A: On a scale of 0 (no improvement) to 100 (complete improvement),
please indicate how much your emotional distress has improved (Please
move on to #4.)

Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please
indicate how much your emotional distress has worsened (Please move on
to #4.)

4. Has your interference improved, stayed the same, or worsened?

Please choose one:
Improved (Please complete Part A below.)
Stayed the same (Please move on to #5.)
Worsened (Please complete Part B below.)

Part A: On a scale of 0 (no improvement) to 100 (complete improvement),
please indicate how much your interference has improved (Please move on
to #5.)

Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please
indicate how much your interference has worsened (Please move on to
#5.)

5. Overall, have you improved, stayed the same, or worsened?

Please choose one:
Improved (Please complete Part A below.)
Stayed the same (Please move on to the next page.)
Worsened (Please complete Part B below.)

Part A: On a scale of 0 (no improvement) to 100 (complete improvement),
please indicate how much you have improved (Please move on to the next
page.)

Part B: On a scale of 0 (no worsening) to 100 (complete worsening), please
indicate how much you have worsened (Please move on to the next page.)











FINALLY, WE WOULD LIKE TO LEARN WHETHER YOU WOULD CONSIDER
YOUR TREATMENT SUCCESSFUL.

Was treatment of your pain successful?

Please choose one:

Yes
No

Was treatment of your fatigue (or tiredness) successful?

Please choose one:

Yes
No

Was treatment of your emotional distress successful?

Please choose one:

Yes
No

Was treatment of your interference with daily activities successful?

Please choose one:

Yes
No

Overall, was your treatment successful?

Please choose one:

Yes
No















LIST OF REFERENCES


Angst, F., Aeschlimann, A., Michel, B. A., & Stucki, G. (2002). Minimal clinically
important rehabilitation effects in patients with osteoarthritis of the lower
extremities. J Rheumatol, 29, 131-138.

Beck, A. T., Steer, R. A., & Garbin, M. G. (1988). Psychometric properties of the Beck
Depression Inventory: Twenty-five years of evaluation. Clin Psychol Rev, 8, 77-
100.

Casarett, D., Karlawish, J., Sankar, P., Hirschman, K., Asch, D. A. (2001). Designing
pain research from the patient's perspective: What trial end points are important to
patients with chronic pain? Pain Med, 2, 309-316.

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed). Hillsdale,
NJ: Lawrence Erlbaum Associates.

Cook, R. J., & Sackett, D. L. (1995). The number needed to treat: A clinically useful
measure of treatment effect. BMJ, 310, 452-454.

Daniel, M., Long, C., Murphy, W., Kores, R., and Hutcherson, W. (1983). Therapists'
and chronic pain patients' perceptions of treatment outcome. J Nerv Ment Dis, 171,
729-733.

Department of Health and Human Services, National Institutes of Health. (1998,
September 4). New directions in pain research: I. Retrieved July 28, 2005, from
http://grants2.nih.gov/grants/guide/pa-files/PA-98-102.html

Dreyfuss, P., Halbrook, B., Pauza, K., Josi, A., McLarty, J., & Bogduk, N. (2000).
Efficacy and validity of radiofrequency neurotomy for chronic lumbar
zygapophysial joint pain. Spine, 25, 1270-1277.

Dworkin, S. F., & LeResche, L. (1992). Research diagnostic criteria for
temporomandibular disorders: Review, criteria, examination and specifications,
critique. J Craniomandib Disord, 6, 301-355.

Farrar, J. T. (2000). What is clinically meaningful: Outcome measures in clinical pain
trials? Clin J Pain, 16, S106-S112.

Farrar, J. T., Berlin, J. A., & Strom, B. L. (2003). Clinical important changes in acute
pain outcome measures: A validation study. J Pain Symptom Manage, 25, 406-411.









Farrar, J. T., Portenoy, R. K., Berlin, J. A., Kinman, J. L., & Strom, B. L. (2000).
Defining the clinically important difference in pain outcome measures. Pain, 88,
287-294.

Farrar, J. T., Young, J. P., LaMoreaux, L., Werth, J. L., & Poole, R. M. (2001). Clinical
importance of changes in chronic pain intensity measured on an 11-point numerical
pain rating scale. Pain, 94, 149-158.

Fischer, D., Stewart, A. L., Bloch, D. A., Lorig, K., Laurent, D., & Holman, H. (1999).
Capturing the patient's view of change as a clinical outcome measure. JAMA, 282,
1157-1162.

Flor, H., Fydrich, T., & Turk, D. C. (1992). Efficacy of multidisciplinary pain treatment
centers: A meta-analytic review. Pain, 49, 221-230.

Geisser, M.E., Roth, R. S., & Robinson, M. E. (1997). Assessing depression among
persons with chronic pain using The Center for Epidemiological Studies-
Depression Scale and the Beck Depression Inventory: A comparative analysis. Clin
JPain, 13, 163-170.

Gilliam, F., Kuzniecky, R., Meador, K., Martin, R., Sawrie, S., Viikinsalo, M.,
Morawetz, R., & Faught, E. (1999). Patient-oriented outcome assessment after
temporal lobectomy for refractory epilepsy. Neurology, 53, 687-694.

Goldsmith, C. H., Boers, M., Bombardier, C., & Tugwell, P. (1993). Criteria for
clinically important changes in outcomes: Development, scoring, and evaluation of
rheumatoid arthritis patient and trial profiles. J Rheumatol, 20, 561-565.

Guck, T., Meilman, P., Skultety, F., & Poloni, L. (1988). Pain-patient Minnesota
Multiphasic Personality Inventory (MMPI) subgroups: Evaluation of long-term
treatment outcome. J Behav Med, 11, 159-169.

Jamison, R. N., Rock, D. L., & Parris, W. C. V. (1988). Empirically derived Symptom
Checklist-90 subgroups of chronic pain patients: A cluster analysis. J Behav Med,
11, 147-158.

Jensen, M. P., Strom, S. E., Turner, J., & Romano, J. M. (1992). Validity of the Sickness
Impact Profile Roland scale as a measure of dysfunction in chronic pain patients.
Pain, 50, 157-162.

Keefe, F. J. (1996). Cognitive behavior therapy for managing pain. Clin Psychol, 49(3),
4-5.

Keefe, F. J., Bradley, L. A., & Crisson, J. E. (1990). Behavioral assessment of low back
pain: Identification of pain behavior subgroups. Pain, 40, 153-160.

Lane, C., & Davidoff, F. (1996). Patient-centered medicine: A professional evolution.
JAMA, 275, 152-156.









Lindstrom, I., Ohlund, C., Eek, C., Wallin, L., Peterson, L. E., Fordyce, W. E., &
Nachemson, A. L. (1992). The effect of graded activity on patients with subacute
low back pain: A randomized prospective clinical study with an operant-
conditioning behavioral approach. Phys Ther, 72, 279-290.

Mannion, A. F., Muntener, M., Taimela, S., & Dvorak, J. (2001). Comparison of three
active therapies for chronic low back pain: Results of a randomized clinical trial
with one-year follow-up. Rheumatology, 40,772-778.

McCracken, L. M., Klock, A., Mingay, D., Asbury, J. K., & Sinclair, D. M. (1997).
Assessment of satisfaction with treatment for chronic pain. J Pain Symptom
Manage, 14(5), 292-299.

McCracken, L.M., & Turk, D.C. (2002). Behavioral and cognitive-behavioral treatment
for chronic pain. Spine, 27, 2564-2573.

McCracken, L. M., Zayfert, C., & Gross, R. T. (1992). The Pain Anxiety Symptoms
Scale: Development and validation of a scale to measure fear of pain. Pain, 50, 67-
73.

McGill, J., Lawlis, G., Selby, D., Mooney, V., & McCoy, C. (1983). The relationship of
Minnesota Multiphasic Personality Inventory (MMPI) profile clusters to pain
behaviors. J Behav Med, 6, 77-92.

Melzack, R., & Katz, J. (2001). The McGill Pain Questionnaire: Appraisal and current
status. In D. C. Turk & R. Melzack (Eds.), Handbook of pain assessment (2nd ed.,
pp. 35-52). New York: The Guilford Press.

Melzack, R., & Wall, P. D. (1965). Pain mechanisms: A new theory. Science, 150, 971-
979.

Moore, A., McQuay, H., & Gavaghan, D. (1996). Deriving dichotomous outcome
measures from continuous data in randomised controlled trials of analgesics. Pain,
66, 229-237.

Moore, J. E., Armentrout, D. P., Parker, J. C., & Kivlahan, D. R. (1986). Empirically
derived pain-patient MMPI subgroups: Prediction of treatment outcome. J Behav
Med, 9, 51-63.

North, R. B., Han, M., Zahurak, M., & Kidd, D. (1994). Radiofrequency lumbar facet
denervation: Analysis of prognostic factors. Pain, 57, 77-83.

Pollard, C. A. (1984). Preliminary validity study of the Pain Disability Index. Percept
Mot Skills, 59(3), 974.

Price, D. D., McGrath, P.A., Rafii, A. and Buckingham, B. (1983). The validations of
visual analogue scales as ratio scale measures for chronic and experimental pain.
Pain, 17, 45-56.









Redelmeier, D. A., Guyatt, G. H., & Goldstein, R. S. (1996). Assessing the minimal
important difference in symptoms: A comparison of two techniques. J Clin
Epidemiol, 49, 1215-1219.

Redelmeier, D. A., & Lorig, K. (1993). Assessing the clinical importance of symptomatic
improvements. Arch Intern Med, 153, 1337-1342.

Robinson, M. E., Brown, J. L., George, S. Z., Edwards, P. S., Atchison, J. W., Hirsh, A.
T., Waxenberg, L. B., Wittmer, V., & Fillingim, R. B. (in press). Multidimensional
success criteria and expectations for treatment of chronic pain: The patient
perspective. Pain Med.

Roland, M., & Morris, R. (1983). A study of the natural history of back pain. Part I:
Development of a reliable and sensitive measure of disability in low-back pain.
Spine, 8, 141-144.

Rudy, T. E., Turk, D. C., Kubinski, J. A., & Zaki, H. S. (1995). Differential treatment
responses of TMD patients as a function of psychological characteristics. Pain, 61,
103-112.

Salaffi, F., Stancati, A., Silvestri, C. A., Ciapetti, A., & Grassi, W. (2004). Minimal
clinically important changes in chronic musculoskeletal pain intensity measured on
a numerical rating scale. Eur J Pain, 8, 283-291.

Schwartz, C. E., Cole, B. F., & Gelber, R. D. (1995). Measuring patient-centered
outcomes in neurologic disease. Arch Neurol, 52, 754-762.

Slater, M. A., Doctor, J. N., Pruitt, S. D., & Atkinson, J. H. (1997). The clinical
significance of behavioral treatment for chronic low back pain: An evaluation of
effectiveness. Pain, 71, 257-263.

Stewart, W., Ricci, J., Chee, E., Morganstein, D., & Lipton, R. (2003). Lost productive
time and cost due to common pain conditions in the US workforce. JAMA, 290,
2443-2454.

Tait, R. C., Pollard, C. A., Margolis, R. B., Duckro, P. N., & Krause, S. J. (1987). The
Pain Disability Index: Psychometric and validity data. Arch Phys Med Rehabil, 68,
438-441.

Turk, D. C. (2002). Clinical effectiveness and cost-effectiveness of treatment of patients
with chronic pain. Clin J Pain, 18, 355-365.









Turk, D. C., Dworkin, R. H., Allen, R. R., Bellamy, N., Brandenburg, N., Carr, D. B.,
Cleeland, C., Dionne, R., Farrar, J. T., Galer, B. S., Hewitt, D. J., Jadad, A. R.,
Katz, N. P., Kramer, L. D. Manning, D. C., McCormick, C. G., McDermott, M. P.,
McGrath, P., Quessy, S., Rappaport, B. A., Robinson, J. P., Royal, M. A., Simon,
L., Stauffer, J. W., Stein, W., Tollett, J., & Witter, J. (2003). Core outcome
domains for chronic pain clinical trials: IMMPACT recommendations. Pain, 106,
337-345.

Turk, D. C., & Melzack, R. (Eds.). (2001). Handbook of pain assessment (2nd ed.). New
York: Guilford Press.

Turk, D. C., & Okifuji, A. (2001). Matching treatment to assessment of patients with
chronic pain. In D. C. Turk & R. Melzack (Eds.), Handbook of pain assessment
(2nd ed., pp. 400-414). New York: Guilford Press.

Turk, D. C., Okifuji, A., Starz, T., & Sinclair, J. (1998). Differential responses by
psychosocial subgroups of fibromyalgia syndrome patients to an interdisciplinary
treatment. Arthritis Care Res, 11, 397-404.

Turk, D. C., & Rudy, T. E. (1988). Toward an empirically derived taxonomy of chronic
pain patients: Integration of psychological assessment data. J Consult Clin Psychol,
56, 233-238.

Turk, D. C., Rudy, T. E., Kubinski, J. A., Zaki, H. S., & Greco, C. M. (1996).
Dysfunctional patients with temporomandibular disorders: Evaluating the efficacy
of a tailored treatment protocol. J Consult Clin Psychol, 64, 139-146.

Turk, D. C., & Zaki, H. S., & Rudy, T. E. (1993). Effects ofintraoral appliance and
biofeedback/stress management alone and in combination in treating pain and
depression in patients with temporomandibular disorders. J Prosthet Dent, 70, 158-
164.

Wells, G. A., Tugwell, P., Kraag, G. R., Baker, P. R. A., Groh, J., & Redelmeier, D. A.
(1993). Minimum important difference between patients with rheumatoid arthritis:
The patient's perspective. Journal Rheumatol, 20, 557-560.

Zatzick, D. F., Kang, S., Hinton, W. L., Kelly, R. H., Hilty, D. M., Franz, C. E., Le, L., &
Kravitz, R. L. (2001). Posttraumatic concerns: A patient-centered approach to
outcome assessment after traumatic physical injury. Med Care, 39, 327-339.















BIOGRAPHICAL SKETCH

Jennifer Lynn Brown was born in Alexandria, Virginia on November 24, 1974. She

graduated from Potomac Senior High School in 1992, and received her Bachelor of

Science degree in psychology from the University of Mary Washington in 1996. After

earning a Master of Arts degree in psychology from Wake Forest University in 1998, she

worked as a Research Associate for the American Institutes for Research in Washington,

DC and then as an Associate for Caliber Associates in Fairfax, Virginia. In 2000, Jennifer

entered the Department of Clinical and Health Psychology's doctoral program at the

University of Florida (UF) with a concentration in clinical health psychology. While at

UF, she worked for the Center for Pain Research and Behavioral Health under the

mentorship of Michael E. Robinson, Ph.D. She is completing her predoctoral clinical

internship at the VA Connecticut Healthcare System in West Haven, Connecticut.