Biopsychosocial characteristics as predictors of treatment outcome of TMD patients with symptoms of pain

Acknowledgements I am grateful to my supervisor Associate Professor Adrian Yap who has introduced me to this work. I truly appreciate my co-supervisor, Dr. Chan Yiong Huak for his never-ending patience and understanding in guiding me through this journey. To Professor Jennifer Neo, I appreciate her kindness and understanding. She has accommodated me in many ways. I deeply appreciate Professor Richard Ohrbach, University at Buffalo, New York, USA, for his valuable advice and constructive criticism in the writing up of the proposal for the research. Thanks are also due to Dr. David Tay who has given me a lot of feedback for this study. I thank University Malaya for sponsoring my study leave and to my colleagues at the Dental Faculty, University Malaya who have helped ease the burden at work during my write-up period. I am so indebted to Dr. Awang Bulgiba who has been a great friend and has given me tremendous amounts of help. I express my thanks to the nurses who have helped me organize the subjects for this study. I can never thank enough my husband, Kenneth, who has been my rock over the years, giving me support in all ways he knew; when it was plain sailing and also during my darkest hours. He even took on the role of mother for our children, Justin and Julian, when I was nowhere available for them. I love you all endlessly. I am truly thankful to my dearest friend Pauline; she has been most kind, generous and supportive of me in so many ways. i And to my most precious teacher Gehlek Rimpoche who encouraged, prayed and guided me along this journey. With His wisdom and compassion, He has made this possible. December 2006. ii Dedication To Gelek Rimpoche who has taught and shown me how to live. To your good health, longevity & may your sun of teachings forever shine. To Kenneth, Justin & Julian for your love, support, understanding and tolerance All of you have made this journey possible. iii Table of contents ACKNOWLEDGEMENTS I TABLE OF CONTENTS . IV SUMMARY VIII NOTICE X LIST OF TABLES XII LIST OF FIGURES . XIV LIST OF SYMBOLS . XVI CHAPTER INTRODUCTION . CHAPTER LITERATURE REVIEW 2.1 OROFACIAL PAIN .5 2.1.1 Epidemiology of orofacial pain 2.1.2 Mechanisms of orofacial pain . 2.1.3 Theories of pain 2.1.4 Measurement of orofacial pain 11 2.2 TEMPOROMANDIBULAR DISORDERS (TMD)………………………….16 2.2.1 Definition 16 2.2.2 Epidemiology of TMD 18 2.2.3 Etiology of TMD . 21 2.2.3.1 Neural basis of TMD & masticatory muscle pain 22 2.2.3.2 Pathobiology of TMD . 22 2.2.3.3 Evolution of theories for TMD . 25 2.2.3.4 Risk Factors 26 2.2.4 Diagnosis of TMD 48 2.2.4.1 Diagnostic classifications 48 2.2.4.2 Biopsychosocial model for TMD 49 2.2.4.3 Research Diagnostic Criteria for TMD . 53 iv 2.2.5 Management of TMD . 58 2.2.5.1 Pharmacotherapy . 59 2.2.5.2 Physical therapy 60 2.2.5.3 Occlusal therapy 63 2.2.5.4 Biobehavioral approach to TMD . 65 2.2.5.5 Surgical therapies 68 2.2.6 Treatment outcome of TMD . 68 2.2.6.1 Measuring treatment outcome . 68 2.2.6.2 Outcome measures 70 2.2.6.3 Predictors of TMD treatment outcome . 75 CHAPTER OBJECTIVE OF RESEARCH 85 3.1 RESEARCH QUESTIONS………………………………………………… 85 3.2 GAPS IN KNOWLEDGE…………………………………………………….85 3.3 AIMS OF STUDY…………………………………………………………….88 3.4 SCOPE OF STUDY………………………………………………………… 89 CHAPTER 4.1 PRELIMINARY PREPARATION……………………………………………91 4.1.1 Cross-cultural adaptation of RDC/TMD . 91 4.1.1.1 Translation process . 92 4.1.1.2 Evaluation of Semantic equivalence . 94 4.1.1.3 Evaluation of Internal consistency and validity 95 4.1.2 4.2 METHODOLOGY . 91 Calibration of clinical examiners . 98 BIOPSYCHOSOCIAL DIFFERENCES IN TMD PATIENTS……………….101 4.2.1 Determination of sample size 101 4.2.2 Patient selection 103 4.2.3 Research instrument 105 4.2.4 Procedure 105 4.2.5 Study measures . 106 4.2.6 Data analysis . 107 4.2.6.1 Pre-analysis data preparation 107 v 4.2.6.2 4.3 Statistical tests………………………………………………………. 107 PREDICTIVE MODEL FOR TMD TREATMENT OUTCOME………… 111 4.3.1 Predictive model for non-response to treatment . 111 4.3.1.1 Determination of sample size 111 4.3.1.2 Patient sample . 112 4.3.1.3 Treatment and follow-up . 112 4.3.1.4 Outcome measures 115 4.3.1.5 Data analysis . 118 4.3.2 Changes in psychological distress over time 119 CHAPTER 5.1 RESULTS 121 CROSS-CULTURAL ADAPTATION OF RDC/TMD…………………… 121 5.1.1 Semantic equivalence 121 5.1.2 Internal consistency of translation. 122 5.1.3 Test-retest reliability of translation. . 123 5.1.4 Validity of translation 124 5.1.5 Calibration of examiners . 125 5.2 BIOPSYCHOSOCIAL DIFFERENCES IN TMD PATIENTS…………… 127 5.2.1 Characteristics of TMD patient sample 127 5.2.2 Biopsychosocial differences in TMD patients : association WIPS. . 130 5.2.3 Prediction of pain symptom in TMD patients . 135 5.2.4 CPI and biopsychosocial characteristics of TMD patients WIPS 140 5.3 PREDICTIVE MODEL FOR TREATMENT OUTCOME………………….143 5.3.1 Descriptive statistics . 144 5.3.2 Outcome of conservative TMD treatment 147 5.3.3 Prediction of non-response to TMD treatment . 153 5.3.3 Changes in psychological distress over time 158 5.3.3.1 Psychological changes over time among responders 158 5.3.3.2 Psychological changes over time among non-responders. . 163 5.3.3.3 Comparison of changes in psychological distress. . 166 CHAPTER DISCUSSION 172 vi 6.1 CROSS-CULTURAL ADAPTATION OF THE RDC/TMD……………….172 6.1.1 General discussion 172 6.1.2 Methodological considerations . 172 6.1.3 Cross-cultural adaptation of instrument 173 6.1.4 Reliability and validity of RDC Axis II 175 6.1.5 Reliability and validity of clinical examination 179 6.2 BIOPSYCHOSOCIAL DIFFERENCES IN TMD PATIENTS…………… 185 6.2.1 General discussion 185 6.2.2 Methodological considerations . 185 6.2.3 Biopsychosocial differences in TMD patients. . 191 6.2.3.1 Characteristics of TMD patients . 191 6.2.3.2 Factors associated with TMD patients with pain symptom 201 6.3 PREDICTIVE MODEL FOR TREATMENT OUTCOME…………………207 6.3.1 General discussion 207 6.3.2 Methodological considerations . 207 6.3.3 Predictive model for non-responders to treatment. . 213 6.3.4 Changes in psychological distress over time. . 231 6.4 LIMITATIONS OF STUDY……………………………………………… 236 CHAPTER CONCLUSIONS 238 BIBLIOGRAPHY . 243 APPENDIX A RDC/TMD . 303 APPENDIX B MAL-RDC/TMD . 308 APPENDIX C TRAINING & CALIBRATION PROTOCOL 312 APPENDIX D TREATMENT PACKAGE 317 vii Summary Pain is the most common reason why Temporomandibular Disorder (TMD) patients seek treatment. The biopsychosocial factors associated with pain symptoms (WIPS) in TMD patients and their role in the prediction of TMD treatment response are still largely unknown. AIMS: (I) To develop a cross-culturally-adapted RDC/TMD for Malaysian use, (II) to compare the biopsychosocial characteristics of TMD patients with (WIPS) and without pain symptoms (WoPS) in a multiracial urban Asian population, in order to determine which factors were associated WIPS and (III) to construct a predictive model that would enable identification of non-responders to conservative TMD treatment. METHODOLOGY: The RDC/TMD was cross-culturally adapted using standard procedures and was used to collect data on the biopsychosocial characteristics of 314 TMD patients (age between 18-74 years) from an urban Malaysian setting. A cross-sectional study comparing the biopsychosocial characteristics of TMD patients WIPS and WoPS was carried out in order to determine factors which were likely to be associated WIPS. The selection criteria for patients WIPS was the presence of pain as the main symptom. Patients WIPS (N=100) were given conservative TMD treatment without occlusal splints; their progress documented at six weeks and at six months. The outcome measures were improvement in the Characteristic Pain Intensity and maximal width of pain-free mouth opening (WMNoPmax). All data were subjected to univariate followed by multivariate analysis in SPSS. RESULTS: (I) The cross-culturally-adapted RDC/TMD was valid, reliable and suitable for use in Malaysia. (II) There were differences among TMD patients WIPS and WoPS in their sociodemography (race), physical and self-reported characteristics and psychological status. Of these, factors associated WIPS were limited width viii of pain-free mouth opening (p=0.011), self-report of nocturnal tooth-grinding/jaw-clenching (p=0.036), presence of a Myofascial Pain (MFP) diagnosis (p[...]... confounders of nocturnal toothgrinding/jaw-clenching among TMD patients WIPS Table 19 Linear regression analysis of biopsychosocial factors and CPI in TMD patients WIPS Table 20 Model summary of biopsychosocial factors associated with CPI among TMD patients WIPS Table 21 Distribution of patients with TMD subgroup diagnoses in sample (N=100) Table 22 Biopsychosocial characteristics (continuous data) of responders... Biobehavioral Model of Pain Adapted from Epidemiologic studies of chronic pain: A dynamic-ecologic perspective (Dworkin et al, 1992b) Figure 3 Distribution of TMD subgroups alone and in combination in patient sample Figure 4 ROC curve of full model of biopsychosocial factors associated with pain symptoms Figure 5 ROC curve of full model of biopsychosocial factors associated with nonresponse to TMD treatment Figure... (TMD) is described in greater detail in the next section 10 2.1.4 Measurement of orofacial pain Pain measurement provides clues that assist in the differential diagnosis of the underlying causes of pain (Grossman, 1994) and also helps in the assessment of the degree of disability or impairment of function related to pain (Osterweis et al, 1987; Max et al, 1990) Because pain is essentially made up of. .. disease would have to be defined as one which has undergone a complete resolution of symptoms for a significant amount of time for example for at least six months In this light, there has been no definition of a “recurrent” TMD disorder which implies a complete resolution of of TMD pain contributes to why the Characteristic Pain Intensity, as is used in the Research Diagnostic Criteria for TMD (RDC /TMD) ,... integration of peripheral stimuli with cortical variables such as mood and anxiety, in the perception of pain With the hope for a purely biomedical understanding of pain further discredited, the identification of psychological and social factors associated with pain has taken on even greater urgency It is within this context that Turk and Rudy (1987) proposed a multidimensional, biopsychosocial model of pain. .. non-responders (N=31) to TMD treatment at baseline, six weeks and six months Table 23 Summary of biopsychosocial characteristics of responders and non-responders to conservative TMD treatment (N=100) Table 24 Comparison of two models (full model and most parsimonious model) for non-response to TMD treatment Table 25 Summary of reliability studies on measures of Depression, NSPS and GCPS of the RDC /TMD Axis II Table... obvious Measurement of pain also differs between the types of pain ie acute pain versus chronic pain Gracely & Dubner (1981) proposed five properties of an ideal pain measurement system They should:- (1) be sensitive, bias-free, (2) provide immediate information of accuracy and reliability of subject, (3) be able to separate inherent stimulus painfulness (sensorydiscriminative aspects of pain experience)... patient characteristics TMD Temporomandibular disorders WIPS With pain symptoms WoPS Without pain symptoms WMNoPmax Maximal width of pain- free mouth opening WMPmax Maximal width of mouth opening with pain Statement of originality Unless otherwise noted or referenced in the text, the work described in this thesis is that of the author xvii Chapter 1 INTRODUCTION Temporomandibular disorders (TMD) , as defined... of clinical TMD subgroups in sample (N=314) Table 14 Distribution of physical jaw characteristics of sample (N=314) Table 15 Scores of depression and somatization (with and without pain items) in xii sample (N=314) Table 16 Biopsychosocial characteristics of TMD patients WIPS and WoPS (N=314) Table 17 Comparison of characteristics between full and most parsimonious model Table 18 Precision assessment... of reliability studies on the RDC /TMD Axis I measures Table 27 Distribution of Axis I TMD subgroup diagnoses in studies Table 28 Treatment outcome and sample characteristics of studies xiii List of Figures Figure 1 Model of pathways of vulnerability to Idiopathic Pain Disorders Adapted from Idiopathic pain disorders – pathways to vulnerability (Diatchenko et al, 2006) Figure 2 Biobehavioral Model of . approach to TMD 65 2.2.5.5 Surgical therapies 68 2.2.6 Treatment outcome of TMD 68 2.2.6.1 Measuring treatment outcome 68 2.2.6.2 Outcome measures 70 2.2.6.3 Predictors of TMD treatment outcome. ROC curve of full model of biopsychosocial factors associated with pain symptoms. Figure 5. ROC curve of full model of biopsychosocial factors associated with non- response to TMD treatment. . considerations 185 6.2.3 Biopsychosocial differences in TMD patients. 191 6.2.3.1 Characteristics of TMD patients 191 6.2.3.2 Factors associated with TMD patients with pain symptom 201 6.3 PREDICTIVE