• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Reliability and validity of a treatment outcome satisfaction questionnaire for patients with low back pain (LBP)

Dissertation
Author: Angela Lis
Abstract:
The purpose of this study is to define what treatment outcome satisfaction (TOS) means for the patient with low back pain (LBP) and to develop a theoretically-based, valid and reliable measure to assess this construct in a physical therapy setting. Two preliminary studies were conducted to provide support for the main study: (1) An explorative study that examined the concept of TOS at the end of a physical therapy (PT) intervention, and (2) A cross-sectional study that evaluated the content, face validity, and internal consistency of a TOS questionnaire (TOSQ). Methods. A longitudinal study established construct validity by evaluating the content, convergent, and discriminate validity of a TOSQ. Data were collected using three methods: written, phone, and proxy. Results. An exploratory and confirmatory factor analysis detected one factor that accounted for 58% to 78% of the total variance in the multidimensional questionnaire depending on the method of data collection (written, phone, or proxy). All of the items highly and significantly correlated with each other and with the global measures of satisfaction, demonstrating good internal consistency and concurrent validity. Cronbach's alpha (α) values were calculated to find the most homogenous set of items. The highest α and average inter-item correlation in all of the cases was found when eight items were included. Criterion validity was supported by the strong correlations between each of the specific items and the two global items, and the significant association between the total score of the TOSQ and these global questions. Convergent validity was established as TOS was associated with the theoretical models of performance, expectation/disconfirmation, and embodiment. Discriminate validity was suggested by the low associations with process and structure of care satisfaction. Results remained similar for all methods of data collection. Conclusions. TOS was found to be a complex attitude reflected in the patient's beliefs, emotional responses, and behavioral reactions resulting from the physical therapy (PT) intervention. Construct validity of the TOSQ was proven through content, face, criterion, convergent, and discriminant validity. This research project was successful in developing a valid TOSQ for patients with LBP in a PT setting.

viii TABLE OF CONTENTS DEDICATIONS .................................................................................................................. iii AKNOWLEDGEMENTS ..................................................................................................... iv ABSTRACT ........................................................................................................................ vi LIST OF FIGURES .............................................................................................................. xii LIST OF TABLES ............................................................................................................... xiv LIST OF ABBREVIATIONS ................................................................................................. xx LIST OF APPENDICES .................................................................................................... xxii INTRODUCTION ............................................................................................................... 1 Satisfaction: Definition ................................................................................... 3 Satisfaction: Underlying Theories ................................................................. 5 Low Back Pain and Outcome Satisfaction ............................................... 15 Outcome Satisfaction: Instruments ............................................................ 16 PRELIMINARY STUDIES ................................................................................................... 23 Study #1: ........................................................................................................ 23 Study Design .................................................................................................. 23 Research Question ........................................................................................ 23

ix Study Population ........................................................................................... 24 Recruitment.................................................................................................... 24 Sampling ......................................................................................................... 25 Inclusion Criteria ............................................................................................ 26 Exclusion Criteria ........................................................................................... 27 Data Collection ............................................................................................. 27 Data analysis .................................................................................................. 28 Results ............................................................................................................. 31 Study #2 ......................................................................................................... 38 Study Design .................................................................................................. 38 Research Question ........................................................................................ 38 Study Population ........................................................................................... 38 Recruitment.................................................................................................... 38 Sampling ......................................................................................................... 39 Inclusion Criteria ............................................................................................ 40 Exclusion Criteria ........................................................................................... 40 Data Collection ............................................................................................. 40

x Data Analysis ................................................................................................. 41 Sample size ..................................................................................................... 42 Results ............................................................................................................. 42 METHODS ....................................................................................................................... 52 Study Design .................................................................................................. 52 Research Questions...................................................................................... 52 Hypotheses .................................................................................................... 53 Variables and Instruments ........................................................................... 55 Study Population .......................................................................................... 60 Recruitment ................................................................................................... 61 Sampling ........................................................................................................ 63 Inclusion Criteria ........................................................................................... 64 Exclusion Criteria ........................................................................................... 64 Sample size .................................................................................................... 65 Data Collection ............................................................................................ 66 Data Analysis ................................................................................................. 67 Factor Analysis ............................................................................................... 69

xi Multi-Trait Multi-Method (MTMM) Matrix Analysis ...................................... 72 RESULTS ........................................................................................................................... 81 Sample Characteristics ................................................................................ 81 Treatment outcome satisfaction questionnaire (TOSQ) Psychometric Analysis ....................................................................................................................... 85 Factor Analysis ............................................................................................... 85 Associations among variables..................................................................... 92 Multi-trait Multi-method (MTMM) Analysis Results .................................. 100 Compliance analysis .................................................................................. 110 DISCUSSION ................................................................................................................. 120 Strengths of the research study ................................................................ 136 For Future Research:................................................................................... 143 CONCLUSIONS ............................................................................................................ 146 APPENDICIES ............................................................................................................... 149 REFERENCES................................................................................................................. 301

xii LIST OF FIGURES

Fig. 1 Outcome satisfaction theories…………………………………………………….8 Fig. 2 PCA, Scree plot……………………………………………………………..…….244 Fig. 3 General schematic representation of a MTMM……………………………...74 Fig. 4 MTMM#1……………………………………………………………………………245 Part 1 MTMM: Written Vs written methods…………………………………...246 Part 2 MTMM: Written Vs phone methods…………………………….……..247 Part 3 MTMM: Written Vs proxy methods……………………………….........248 Part 4 MTMM: Phone Vs written methods…………………………………….249 Part 5 MTMM: Phone Vs phone methods…………………………………….250 Part 6 MTMM: Phone Vs proxy methods……………………………..……….251 Part 7 MTMM: Proxy Vs written methods……………………………………...252 Part 8 MTMM: Proxy Vs phone methods……………………………...………253 Part 9 MTMM: Proxy; Vs proxy methods………………………………………254 Fig. 5 MTMM#2……………………………………………………………………………255 Part 1 TOSQ MTMM: Written Vs written methods……………………...........256

xiii Part 2 TOSQ MTMM: Written Vs phone methods…………………..………..257 Part 3 TOSQ MTMM: Written Vs proxy methods…………………….............258 Part 4 TOSQ MTMM: Phone Vs written methods…………………………….259 Part 5 TOSQ MTMM: Phone Vs phone methods…………………………….260 Part 6 TOSQ MTMM: Phone Vs proxy methods……………………..............261 Part 7 TOSQ MTMM: Proxy Vs written methods……………………..............262 Part 8 TOSQ MTMM: Proxy Vs phone methods………………………………263 Part 9 TOSQ MTMM: Proxy Vs proxy methods………………………………..264

xiv LIST OF TABLES Table 1: Descriptive statistics of the study population…………………………...…266

Table 2: Correlation matrix of the multidimensional items of the TOSQ……….....44

Table 3: Factor analysis principal component analysis without rotation………...45

Table 4: Un-rotated PCA: Commonalities……………………………………………...46

Table 5: Correlation matrix (Correlation of each of the 8 multidimensional items and the global questions)……………………………………………………………..….48

Table 6: Correlation coefficient among the global measurements………….…267

Table 7: PCA un-rotated analysis excluding the item of effort………………...…268

Table 8: Factor loadings for the PCA un-rotated analysis (excluding the item of effort)……………………………………………………………………………………...…..49

Table 9: PCA un-rotated analysis (excluding the item of effort) minimal residuals……………………………………………………………………………………...269

Table 10: Number of days between completion of questionnaires….…………270

Table 11: MTMM correlation matrix: traits and methods……...…………………….77

Table 12: Descriptive statistics of the study population…………………………..…83

xv Table 13: Descriptive statistics of the study population……………………………..84

Table 14a: Descriptive statistics of the TOSQ written method Questionnaire………………………………………………………………………………271

Table 14b: Descriptive statistics of the TOSQ phone method Questionnaire…272

Table 14c: Descriptive statistics of the TOSQ proxy method Questionnaire…...273

Table 15: Correlation matrix - TOSQ multidimensional items, written questionnaire………………………………………………………....…………………..274

Table 16: Correlation matrix - TOSQ multidimensional items, phone questionnaire………………………………………………………………………………275

Table 17: Correlation matrix - TOSQ multidimensional items, proxy questionnaire……………………………………………………………………………....276

Table 18: Principal component analysis - Factor analysis using data from the written questionnaire……………………………………………………………………...............277

Table 19: Principal component analysis - Factor analysis using data from the phone questionnaire….………………………………………………………………….278

Table 20: Principal component analysis - Factor analysis using data from the proxy questionnaire……………………………………………………………………….279

Table 21: Factor analysis results per method of data collection………...……..…86

xvi Table 22: Commonalities, un-rotated PCA factor analysis TOSQ written questionnaire data……………………………………………………...………………..280

Table 23: Commonalities, un-rotated PCA factor analysis TOSQ phone questionnaire data………………………………………………………...……………..281

Table 24: Commonalities, un-rotated PCA factor analysis TOSQ proxy questionnaire data…………………………………………………………...…………..282

Table 25: Residual correlations, un-rotated PCA factor analysis TOSQ written questionnaire…………………………………………………………………...………….283

Table 26: Residual correlations, un-rotated PCA factor analysis TOSQ phone questionnaire……………………………………………………………………...……….284

Table 27: Residual correlations, un-rotated PCA factor analysis TOSQ proxy questionnaire………………………………………………………………………...….…285

Table 28: Correlation of multidimensional items Vs total score (per method of data collection)……………………………………………………………………….....…88

Table 29: Correlation of multidimensional items Vs global items (per method of data collection)………………………………………………………………………….…89

Table 30: Correlation of TOSQ total score (per method) Vs global items (per method of data collection)………………………………………………………………90

Table 31: Cronbach alpha calculations: Multidimensional items TOSQ written questionnaire……………………………………………………………………………....286

xvii Table 32: Cronbach alpha calculations: Multidimensional items TOSQ phone questionnaire……...……………………………………………………………………….287

Table 33: Cronbach alpha calculations: Multidimensional items TOSQ proxy questionnaire…...………………………………………………………………………….288

Table 34: Conclusion table: Cronbach alpha calculations using the eight multidimensional items per data collection method (written, phone, and proxy)………………………………………………………………………………………….91

Table 35: TOSQ Vs baseline characteristics –continuous variables…………........93

Table 36: TOSQ Vs baseline characteristics –categorical variables………………94

Table 37: Total score of TOSQ Vs performance at the end of treatment……...289

Table 38: Satisfaction Vs performance at the end of treatment………………..290

Table 39: Treatment outcome satisfaction Vs clinical improvement (measured by clinically meaningful ODI and pain improvement)………………………….......96

Table 40: Treatment outcome satisfaction Vs hindsight expectations (measured by the question “Considering how your back was before physical therapy, is your back better or worse than you expected it would be?”)……………………97

Table 41: Treatment outcome satisfaction Vs hindsight expectations (measured by the question “Right now, compared to before physical therapy, your back is…?”)……………………………………………………………………………………...….98

Table 42: Treatment outcome satisfaction Vs embodiment category………….99

xviii Table 43: Finished treatment or not Vs baseline characteristics; Categorical variables………………………………………………………………………………...….291

Table 44: Finished treatment or not Vs baseline characteristics; Continuous variables………………………………………………………………………………...….292

Table 45: Finished treatment Vs treatment outcome satisfaction (Satisfied= total score > 32)………………………………………………………………………………….111

Table 46: Finished treatment Vs Oswestry (ODI); Clinically meaningful improvement (= change of 6 points or more)……………………………………....293

Table 47: Finished treatment Vs pain intensity; Clinically meaningful change (Subacute patients with NSLBP> 3.5; Chronic patients with NSLBP > 2.5)……………...............................................................................................……....294

Table 48: Finished treatment Vs satisfaction Vs ODI; Clinically meaningful improvement……………………………………………………………………………...112

Table 49: Finished treatment Vs satisfaction Vs pain intensity; Clinically meaningful improvement………………………………………………………………113

Table 50: Finished Vs hindsight expectations measured by the question ”Considering how your back was before your physical therapy treatment, is your back better or worse than you expected it would be?”…………………...295

Table 51: Finished Vs hindsight expectations measured by the question “Considering how your back was before your physical therapy treatment, is your back better or worse than you expected it would be?”…………………...296

Table 52: Finished Vs hindsight expectations measured by the question “Right now, compared to before physical therapy, your back is…?"…………………..114

xix Table 53: Finished Vs hindsight expectations measured by the question “Right now, compared to before physical therapy, your back is…?”………………….115

Table 54: Finished Vs satisfaction Vs hindsight expectations measured by the question “Considering how your back was before your physical therapy treatment, is your back better or worse than you expected it would be?.......297

Table 55: Finished Vs satisfaction Vs hindsight expectations measured by the question “Right now, compared to before physical therapy, your back is…?”………………………………………………………………………………………....298

Table 56: Finished Vs embodiment category……………………………………....116

Table 57: Treatment outcome satisfaction Vs embodiment category……..…117

Table 58: Finished Vs satisfaction Vs embodiment category…………………….299

Table 59: Finished treatment Vs structure and process of care satisfaction….300

Table 60: Finished treatment Vs treatment outcome satisfaction: Multidimensional items, global questions, and total score……………………….119

xx LIST OF ABBREVIATIONS CMI: Clinical Meaningful Improvement CTC(M-1): Correlated Trait Correlated Method Minus One ICC: Intra-class Correlation Coefficient IRB: Institutional Research Board LBP: Low Back Pain MD: Median MRPS: MedRisk Patient Satisfaction MTMM: Multi Trait Multi Method NASS: North American Spine Society NSLBP: Non Specific Low Back Pain ODI: Oswestry Disability Index OIOC: Occupational and Industrial Orthopaedic Center PCA: Principal Component Analysis PSS: Patient Satisfaction Survey PT: Physical Therapy PTOPS : Physical Therapy Outpatient Satisfaction Survey

xxi PTSS: Pain Treatment Satisfaction Scale SD: Standard Deviation SLBP: Specific Low Back Pain SMOG: Simple Measure of Gobbledygook STV: Subject To Ratio Variable TOS: Treatment Outcome Satisfaction TOSQ: Treatment Outcome Satisfaction Questionnaire VAS: Visual Analog Scale

xxii LIST OF APPENDICES Appendix 1: IRB H#12327-01 Approval………………………………………………..150 Appendix 2: IRB H# 118508B Approval and amendment…………………………161 Appendix 3: IRB H#07-233 and extension…………………………………………….172 Appendix 4: Study #1: Purposeful sampling screening questionnaire……….…195 Appendix 5: Study #1: Semi-structured Interview guidelines……………………..197 Appendix 6: Thinking units exploratory analysis: “Patients’ with LBP attitudes about treatment and satisfaction with treatment outcome after a physical therapy intervention”………………………………………………………………………………..201 Appendix 7: A preliminary treatment outcome satisfaction questionnaire: English version……………………………………………………………………………...204 Appendix 8: A preliminary treatment outcome satisfaction questionnaire: Spanish version……………………………………………………………………………..208 Appendix 9: Final Treatment Outcome Satisfaction questionnaire ………….…213 Appendix 10: Oswestry Disability Index (ODI)………………………………………..217 Appendix 11: Pain diagram and scale………………………………………………..225 Appendix 12: Hudak’s modified embodiment profile……………………………...229 Appendix 13: Hindsight expectations questionnaire……………………………….233 Appendix 14: MedRisk satisfaction questionnaire…………………………………..235 Appendix 15: Socio-demographic entrance questionnaire……………………...238 Appendix 16: Figures……………………………………………………………………...243

xxiii

Appendix 17: Tables……………………………………………………………………..265

1 INTRODUCTION

Efforts to improve the quality of health care have lead to an increasing demand for measurements of treatment effectiveness. Concepts such as quality of life, quality of care, patient preferences, and patient satisfaction have been of great interest due to their correlation with the economic and humanistic components of treatment results 1 . Patient satisfaction is an essential health care indicator since it gives information about the providers’ success at meeting the patient’s needs and preferences. Regulatory agencies such as the Joint Commission on Accreditation of Healthcare Organizations and the National Committee on Quality Assurance have shown a strong interest in developing quality indicators such as patient satisfaction measures to document the quality of care. 2

The concept of patient satisfaction has been studied by different disciplines for the last decade. Sociologist, anthropologists, economists, and health care providers have theorized on this topic by using customer satisfaction theories and by attempting to understand satisfaction with different aspects of life including medical care. 3-5, 5, 6 Although patient satisfaction has been measured frequently and is considered an important health care indicator, little is known about the concept of treatment outcome satisfaction in the patient with low back pain (LBP). Determining the patient’s unique perception of his/her health/illness experience still remains a challenge. In fact,

2 sometimes what the clinician considers a successful outcome is unsatisfactory for the patient. Likewise, it has been observed that a patient can be satisfied with an outcome that the clinician considers poor. 3-5, 7-9 Thus, the health care provider’s perception of satisfaction can vary considerably from the patient’s point of view. Although intuitively it may be expected that treatment outcome satisfaction is fully explained by the clinical outcome, this correlation is not always direct and unambiguous. A great percentage of satisfied patients also report poor clinical outcomes. 10-12 For instance, in chronic pain populations it has been found that patients tend to report high levels of satisfaction with their pain management regardless of the level of pain reduction. 12-16 The discrepancy between the clinical outcome and patient satisfaction suggests some possible explanations: that either treatment outcome satisfaction cannot be fully explained by the outcome of the health care intervention, that the relationship between outcome satisfaction and the clinical outcome may be confounded, or that the satisfaction measure used is not a valid measure of treatment outcome satisfaction. The lack of a valid measure of treatment outcome satisfaction was the impetus for this research study. In order to understand treatment outcome satisfaction and develop a valid and reliable measure, it is imperative then to explore the meaning of this construct and its underlying related theories.

3 This first chapter of the thesis is concerned with the definition of satisfaction and the existing theories of satisfaction with emphasis on those theories related to this work. The second and third chapters concentrate on the development of a Treatment Outcome Satisfaction Questionnaire (TOSQ) for patients with LBP and a validity study of that instrument. The last chapter will discuss the results; what was learned about outcome satisfaction in the LBP population and the threats to validity. The purpose of this study is to define what treatment outcome satisfaction means for the patient with LBP and to develop a unique, valid, and reliable measure that can be used to assess this construct in a physical therapy setting. The long-term effect is to encourage health care providers to consider the patient’s needs and to include them as part of the treatment process to achieve higher outcome satisfaction and quality of life for LBP sufferers. Satisfaction: Definition The development of valid measures of patient satisfaction has been hampered by a lack of thorough understanding of the concept. No uniform definition exists. However, factors believed to influence satisfaction have been identified and include the patient expectations, the patient interaction with the health care provider, the environment, the outcome itself, and intrinsic factors such as the patient age and gender. 3, 4, 7-9, 17

4 There are various theories of patient satisfaction. The literature concurs with the definition of satisfaction as the result of an evaluative process of the patient’s experiences with the health care encounter. 3, 18, 19, 20-23 It represents a rather complex concept that cannot be reduced to a single dimensional item. Several authors have proposed satisfaction as a multidimensional concept whose dimensions include: process, structure, and treatment outcome. 2, 8, 16, 24, 25

Process satisfaction refers to the patient’s assessment of intangible elements of health care such as interpersonal interactions between the patient and health care providers. 24 Structure satisfaction is the patient’s assessment of the tangible elements associated with health care such as the physical environment in which the service occurs as well as billing procedures and other amenities like parking. 24, 26 Outcome satisfaction is the patient’s perception of the result of a particular treatment intervention. This may involve biological, physiological, and psychological outcomes and overall quality of life. 24, 27 Most of the studies found in the field of satisfaction have explored the dimensions of structure and process of care. 28-32 Due perhaps to the complexity of the outcome satisfaction dimensions, its disease-specific nature, and its intrinsic relationship to the individual experience of the patient with the disease, treatment outcome satisfaction has been the least well understood and is the focus of this study.

5 Satisfaction: Underlying Theories Well-developed theories of satisfaction exist in the marketing and social psychology literature, but their application to the health care setting have been limited or not tested. 11 Nevertheless, there is a tendency to define satisfaction based on attitudinal theory. 33,34 According to this theory, satisfaction is an attitudinal response described as a construct that represents an individual's degree of like or dislike for an object or experience. Attitudes are generally positive or negative views of a person, place, thing, or event, which is often referred to as the attitude object. Attitudes are judgments and/or appraisals 8, 22, 18 and they can be ambivalent, so they can simultaneously possess both positive and negative views toward the object or experience in question. All attitudes are the result of either direct experiences or indirect (for instance observational) learning. Attitudes are judgments that can be expressed through cognitive, affective, and behavioral responses. The cognitive element of satisfaction refers to the person’s interpretation (cognitive evaluation) of an object, person, outcome, or experience based on existing information and the individual’s beliefs. 3, 8, 18 Thus, the cognitive evaluation of treatment outcome, although strongly related to the actual clinical outcome, 3, 8, 21 cannot be solely defined by the results of the intervention. The individual’s beliefs and his or her assessment of the experience and health care encounter may in fact add important information to the final assessment of satisfaction. 35 For instance,

6 those patients who at the end of the treatment claim to have a better understanding of their disease/illness and feel better than expected have higher level of outcome satisfaction. 3, 8

The affective response is an emotional response that the individual expresses, the degree of preference in this case for an outcome. Thus, emotions such as happiness or distress 3, 8, 18 can be reflected by positive or negative statements such as: “I am happy with the results…or I was worried with the way that things are going….” These statements represent an emotional reaction, the affective component of satisfaction as an evaluative process. However, a person can have either positive (a feeling of favorableness) or negative (a feeling of un-favorableness) affects towards different dimensions of satisfaction (process or outcome). Each one of those affects then will make an independent contribution to overall satisfaction. 20, 22

The behavioral component of satisfaction refers to a person’s observed behaviors or intentions to perform various behaviors as a result of the satisfaction evaluative process. 3, 4, 9, 16, 36 For example, satisfied patients are more likely to maintain a relationship with a specific provider, continue with their care, and follow instructions to improve their quality of life. 37, 38 Dissatisfied patients, on the other hand, tend to avoid the recommended care and “may make services less effective, either by neglecting to seek care when needed or refusing to comply with the prescribed course of treatment.” 39, 40 Patient dissatisfaction then, may have a strong health-related economic impact 41 :

7 dissatisfied patients may be more likely to be out of the work force, be less productive, consume more medical care services, and therefore, “cost” more to society. 42, 43, 44 We can then infer that high outcome satisfaction has a positive impact on the prevention of disability and the promotion of quality of life. Attitude theories of satisfaction differ based on the variables that are central to the theory and focus on the emotional response, the cognitive evaluation, or the behavioral change associated with satisfaction (Fig. 1). For instance, Bradburn’s work on quality of life relies on an affect-balance model of satisfaction. 45 In this model, positive events are associated with positive emotional experiences while negative events are associated with negative emotional experiences. Thus, positive and negative affects make independent contributions to expressions of life satisfaction. 3, 8, 22, 20 In this conceptualization, satisfaction is an affective reaction of an outcome independent of any standard or reference point. This model is simplistic in assuming that emotions are the only variables that influence the patient’s satisfaction with the intervention. 20, 20, 22

8 Figure 1 Outcome satisfaction theories

Other attitudinal theoretical models describe satisfaction as a cognitive interpretation by the patient. Oliver and Desarbo categorized these theories as equity, attribution, performance, and expectancy-disconfirmation theories. 21

In the health care setting, the theories of performance and expectancy- disconfirmation have been the most frequently used. According to the performance theory , satisfaction is a function of perceived performance. 21, 22

9 In this theory, the patient’s performance is the most important variable; there is no comparison to any type of expectation or pre-generated standard, and therefore outcome satisfaction is based only on the final result of the clinical intervention. 3, 25, 46 A similar theory is the cognitive consistency theory where attitude is defined as the weighted function of all of the evaluative beliefs associated with the outcome. 18 In marketing, this theory refers to satisfaction as a function of either the algebraic sum of all of the beliefs towards the object (or outcome) under study, or an average/mean value of the weighted evaluative belief scores of the different aspects of care under study (additive and averaging linear compensatory models). 3, 19, 33, 34 In both these theories, there is no comparative process with pre-existing standards or beliefs but rather an independent evaluation of the outcome. On the other hand, in the expectancy-disconfirmation theory , expectations and pre-existing standards influence the evaluation of satisfaction. In this case, satisfaction is the result of a comparative process between the actual outcome and a personal reference. Thus, satisfaction is based on the patient’s reaction to the perceived difference between the performance appraisal and expectations. 24, 47 The theory consists of a number of cognitive processes including the formation of expectations, a comparison of performance with expectations, the formation of confirmation- disconfirmation perceptions, the comparison of these perceptions with the initial expectation levels, and finally, the formation of satisfaction judgments. In

10 this theory, the patient compares the results to his/her personal standards. The patient’s experiences with the health service that fall within a range of acceptance relative to the patient’s standards can produce an assimilation effect where differences between the experiences and the standards are minimized, and experiences that are either more positive or negative would produce a contrast effect where the differences are amplified. 24, 48 Thus, as a comparative process, satisfaction is the difference between the actual outcome and some standard or reference point. The outcome may be better than expected (positive disconfirmation), worse than expected (negative disconfirmation), or as expected (simple disconfirmation). Both the direction and size of the discrepancy between what is expected and the experience will affect the interpretation of discrepancy. 3, 8, 48

Patient expectations differ widely among individuals and it has been hypothesized that they are a function of personality, previous experiences, social and cultural values, and the context in which health care is received. 49

Still, there is support in the literature for the contribution of expectations to satisfaction. 49,

3, 11, 33, 34 According to the expectancy-disconfirmation theory, both pre-treatment expectations/foresight expectations and hindsight expectations/as recalled after treatment are independently related to satisfaction. Disconfirmation, however, is determined by asking directly after the treatment about the extent to which the expectations have been met (hindsight expectations) or by an algebraic difference between foresight

11 expectations and the outcome of the treatment. Although expectations have been found to be associated with ratings of satisfaction, there is still a group of patients who feel satisfied even though they do not perceive a significant change from their baseline health status. 35 In fact, results that were rated as ‘meeting expectations’ have explained only a small percentage of the variance of patient satisfaction. 50

Beaton et al. explored the concept of ‘feeling better’ in a group of patients with musculoskeletal disorders to better understand the patients’ perception after treatment. 51 After interviewing 24 workers they concluded that patients mean different things when they say they feel better. They may not actually rate feeling better based on the symptomatology or impairment; rather, the ‘perception of feeling better’ is contextualized by the unique experience of the individual, which may be influenced by factors such as the state of the disorder, the adjustment of life to the disorder, or the adaptations to living with the disorder. 51 Thus, if treatment outcome satisfaction is associated with the unique experience of the patient, it is necessary to investigate the multidimensional nature of satisfaction from the patient’s disease-specific experience to understand what it means for the patient with LBP to feel satisfied with the treatment outcome after a physical therapy intervention. Although satisfaction models have made important contributions to our understanding of outcome satisfaction, they all have serious shortcomings. Bradburn's affective model 45 denies the cognitive nature of outcome

12 satisfaction and the importance of other factors such as expectations. It is a non-comparative model that assumes satisfaction as the affective response to an evaluation of the outcome independent of any standard or reference point. The performance theory model recognizes the importance of cognition; however, it is overly simplistic since it reduces outcome satisfaction to only the appraisal of performance, denying the existence of the patient’s personal standards or the existence of an affective response. 3, 25, 46 Although this model has been frequently used to understand patient satisfaction with treatment outcome, it cannot explain, for example, those patients who are satisfied but do not achieve clinical improvement. The expectancy-disconfirmation theory also ignores the complex nature of outcome satisfaction. In this model, expectations are the only variables used to explain outcome satisfaction. 24

Although there is support in the literature for the contribution of expectations to the expression of satisfaction, expectations do not explain more than 30% of the variance in outcome satisfaction. 3, 8, 52-55 . In conclusion, although all of the above theories may explain some of the concepts of outcome satisfaction, they do not seem to reflect the full spectrum of the concept and they do not explore what outcome satisfaction means for a patient in a health care setting. Hudak et al. explored the concept of treatment outcome satisfaction in hand patients by using Gadow’s phenomenological interpretation of satisfaction (embodiment theory). 7, 8 In this theoretical ’embodiment’ model , the body is a site of existential and social experiences rather than a physical

Full document contains 338 pages
Abstract: The purpose of this study is to define what treatment outcome satisfaction (TOS) means for the patient with low back pain (LBP) and to develop a theoretically-based, valid and reliable measure to assess this construct in a physical therapy setting. Two preliminary studies were conducted to provide support for the main study: (1) An explorative study that examined the concept of TOS at the end of a physical therapy (PT) intervention, and (2) A cross-sectional study that evaluated the content, face validity, and internal consistency of a TOS questionnaire (TOSQ). Methods. A longitudinal study established construct validity by evaluating the content, convergent, and discriminate validity of a TOSQ. Data were collected using three methods: written, phone, and proxy. Results. An exploratory and confirmatory factor analysis detected one factor that accounted for 58% to 78% of the total variance in the multidimensional questionnaire depending on the method of data collection (written, phone, or proxy). All of the items highly and significantly correlated with each other and with the global measures of satisfaction, demonstrating good internal consistency and concurrent validity. Cronbach's alpha (α) values were calculated to find the most homogenous set of items. The highest α and average inter-item correlation in all of the cases was found when eight items were included. Criterion validity was supported by the strong correlations between each of the specific items and the two global items, and the significant association between the total score of the TOSQ and these global questions. Convergent validity was established as TOS was associated with the theoretical models of performance, expectation/disconfirmation, and embodiment. Discriminate validity was suggested by the low associations with process and structure of care satisfaction. Results remained similar for all methods of data collection. Conclusions. TOS was found to be a complex attitude reflected in the patient's beliefs, emotional responses, and behavioral reactions resulting from the physical therapy (PT) intervention. Construct validity of the TOSQ was proven through content, face, criterion, convergent, and discriminant validity. This research project was successful in developing a valid TOSQ for patients with LBP in a PT setting.