Self-Determination Theory and the Theory of Planned Behaviour Applied to Substance Abuse Treatment in a Therapeutic Community Setting

Transcription

1 180 Substance Abuse Treatment Self-Determination Theory and the Theory of Planned Behaviour Applied to Substance Abuse Treatment in a Therapeutic Community Setting Stefanie Klag B.A. (Japanese), B.Psych (Hons) Submitted in fulfilment of the requirements of Doctor of Philosophy within the School of Psychology, Health, Griffith University January 2006

2 ii ABSTRACT In the 21 st century drug and alcohol abuse presents one of the most serious problems worldwide. Of particular concern is the strong relationship between drug use and crime. While law enforcement strategies, including incarceration, have been revealed to contribute little to break the vicious cycle of drug use and crime, substance abuse treatment has been shown to represent an effective form of intervention. Substantial research on the effectiveness of drug treatment has demonstrated the importance of motivation in predicting treatment retention and success. However, substance users are frequently coerced into therapy by external sources, including the criminal justice system, therefore, typically exhibiting little motivation to enter and remain in treatment long enough to overcome their substance addiction. Although past research investigating the effects of treatment-entry coercion indicates positive treatment results, the vast majority of these studies are seriously impeded by extensive conceptual and methodological problems, questioning the postulated value of coercion in substance abuse treatment. Following the call for a shift in the methodological focus of future studies made by some researchers, the author of the present study tested three models that were based on well-established theories. The first model was based on Self Determination Theory (SDT), a motivational theory, while the second model was based on the Theory of Planned Behaviour (TPB), an expectancy-value theory. The third model consisted of a combination of the two theories, which was argued to provide a more complete and comprehensive model than each theory on its own. The testing of the models allowed the exploration of the dynamic interplay and relationships between a number of variables including perceptions of coercion, motivation, perceived autonomy support, and behavioural intentions in an effort to explain and predict retention and treatment outcomes amongst drug and alcohol abusers. The study was conducted in two phases. Phase 1 involved the development of a 29-item instrument called the Perceived Coercion Questionnaire, which was designed to assess participants perceptions of coercion to enter drug and alcohol treatment originating from six different sources. The scale was shown to be a valid and reliable measure of the coercion construct. Phase 2 involved the testing of the three models longitudinally by using a sample of 350 substance abusers from six therapeutic communities across Australia. Participants were asked to complete a battery of standardised measures within the first two weeks of treatment admission (Time 1), two months into treatment (Time 2), and at completion of the treatment program (Time 3). The models were tested cross-sectionally and longitudinally employing hierarchical multiple regression analysis. In addition, change scores were calculated to test whether changes in predictor variables would

3 iii predict outcomes and changes in outcomes cross-sectionally and longitudinally. Analyses of Time 1 and Time 2 cross-sectional data revealed that the SDT model, compared to the TPB and the combined model, provided a better and more parsimonious account of the factors that influence outcomes in therapeutic community treatment. Given the complexity of the study, it was decided to resume the analysis by focusing on the investigation of the SDT model alone. To highlight some of the most important findings, results demonstrated that motivation was a key factor in the treatment and rehabilitation of substance users. As anticipated, intrinsic motivation was consistently predictive of retention and more positive treatment outcomes, while external motivation and amotivation were associated with more negative outcomes. Results also revealed that clients who entered treatment as the result of a legal mandate experienced substantially higher levels of legal coercion compared to clients who entered treatment voluntarily. Legal coercion, in turn, was found to exert a negative impact on substance users motivation for treatment, thereby indirectly resulting in more negative treatment outcomes. In contrast, self coercion (i.e., feelings of pain and suffering) and health-related pressures seemed to facilitate the development of a more intrinsic motivational attitude towards treatment. Besides, perceptions of competence and control in relation to the therapeutic regime emerged as consistent and important predictors of motivation and treatment outcomes. Finally, findings suggested that treatment staff who employed more autonomous and non-coercive strategies that guided substance users through the change process directly influenced individuals treatment motivation and thereby facilitated more positive treatment outcomes. In sum, findings provided support for the usefulness of the SDT model in predicting dropout as well as processes and outcomes in therapeutic community drug and alcohol treatment. Implications for residential substance abuse treatment were discussed, as well as the strengths and limitations of the study. The discussion concludes with implications for practice and suggestions for future research.

4 iv STATEMENT OF ORIGINALITY This work has not previously been submitted for a degree or diploma in any university. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made in the thesis itself. Signed:

5 v TABLE OF CONTENTS CHAPTER 1:...INTRODUCTION Overview Substance Abuse Treatment Treatment Modalities Diversion Programs The Effectiveness of Legal Coercion in Drug and Alcohol Treatment Motivation in Drug and Alcohol Treatment Retention/Time in Treatment Relationship Between Motivation and Outcomes The Theory of Planned Behaviour Past Behaviour Criticisms of the TPB The Present Study Goal Goal Goal Goal Measures of Treatment Success Drug Court Clients/Legally Mandated Clients Research Outline...45 CHAPTER 2:DEVELOPMENT AND VALIDATION OF THE PERCEIVED COERCION QUEST 2.1 Introduction Critique of Existing Coercion Scales Aims of the Present Study Phase 1 Focus Groups Method Focus Group Results Item Generation Content Validity Phase 2 Reduction and Validation of the Perceived Coercion Questionnaire Method Results Item Reduction Item Analysis Factor Analysis...56

6 vi Dimensionality of the Coercion Construct Convergent Validity Influence of Demographic Variables Phase 3 - Confirmatory Factor Analysis Method Results Confirmatory Factor Analysis Dimensionality of the PCQ Divergent Validity Influence of Demographic Variables Discussion General Discussion...66 CHAPTER 3:MAIN RESULT SECTION PRELIMINARY ANALYSES AND MODEL TESTING 3.1 Method Participants Measures Outcome Measures Demographic Variables Control Variables Validity of Study Measures Procedure Preliminary Results Differences between Participants from the Six Therapeutic Communities Attrition Analysis Past Treatment Behaviour Data Analysis Overview Justification for the Three Approaches Preliminary Analyses Changes between T1 and T2 Predictors Differences between Legally Mandated Clients and Non-Mandated Clients Differences between Drug Court Clients and Non Drug Court Clients Differences between Legally Mandated (N = 35) and Drug Court Clients (N = 33) Hierarchical Multiple Regression Analyses (HMRAs) Issues regarding Categorical Variables Results of The Raw Score Approach - T1 Predictor and T1 Outcome Variables...116

7 3.3.1 Motivation Intention Treatment Engagement Time in Treatment Severity of Dependence Drug Taking Confidence Wellbeing Summary Results of The Raw Score Approach T2 Predictor and T2 Outcome Variables Results of Raw Score Approach T1 Predictors and T2 Outcome Variables T2 Motivation as predicted by T1 SDT T2 Treatment Engagement as predicted by T1 SDT (N = 192) T2 Severity of Dependence as predicted by T1 SDT (N = 192) T2 Drug Taking Confidence as predicted by T1 SDT (N = 192) T2 Wellbeing as predicted by T1 SDT (N = 192) T2 Perceived Training Transfer as predicted by T1 SDT (N = 192) Summary Results of Raw Score Approach T1 Predictors and T3 Outcome Variables T3 Treatment Engagement as predicted by T1 SDT (N = 109) T3 Severity of Dependence as predicted by T1 SDT (N = 109) T3 Drug Taking Confidence as predicted by T1 SDT (N = 109) T3 Wellbeing as predicted by T1 SDT (N = 109) T3 Perceived Training Transfer as predicted by T1 SDT (N = 109) Summary Results of Raw Score Approach T2 Predictors and T3 Outcome Variables T3 Treatment Engagement as predicted by T2 SDT (N = 109) T3 Severity of Dependence as predicted by T2 SDT (N = 109) T3 Drug Taking Confidence as predicted by T2 SDT (N = 109) T3 Wellbeing as predicted by T2 SDT (N = 109) T3 Perceived Training Transfer as predicted by T2 SDT (N = 109) Summary Overall Summary vii CHAPTER 4:SUMMARY OF THE CHANGE IN OUTCOME APPROACH & CHANGE/CHANGE 4.1 Summary of the Hierarchical Multiple Regression Analyses The Change in Outcome Approach...168

8 4.1.1 The Change in Outcome Approach T1 SDT Variables Predict Changes in Outcome Variables from T1 to T The Change in Outcome Approach T1 SDT Variables Predict Changes in Outcome Variables from T1 to T The Change in Outcome Approach T1 SDT Variables Predict Changes in Outcome Variables from T2 to T The Change in Outcome Approach T2 SDT Variables Predict Changes in Outcome Variables from T2 to T Summary of the Hierarchical Multiple Regression Analyses The Change/Change Approach The Change/Change Approach Changes in SDT Variables from T1 to T2 Predict Changes in Outcome Variables from T1 to T The Change/Change Approach Changes in SDT Variables from T1 to T2 Predict Changes in Outcome Variables from T1 to T3 (N = 109) Summary of the Change Score Approaches viii CHAPTER 5:... DISCUSSION Review of Study s Goals Testing of the SDT Model Amotivation External Motivation Introjected Motivation Identified/Integrated Motivation Treatment Engagement Time in Treatment Severity of Dependence Drug Taking Confidence Wellbeing Perceived Training Transfer Predictors of Treatment Dropout Client-Related Factors Overall Summary Implications Strengths & Limitations Future Research REFERENCES...221

9 ix LIST OF FIGURES Figure 1. The self-determination continuum showing types of motivation with their regulatory styles, and loci of causality as defined by SDT (from Deci & Ryan, 2000) Figure 2 The Hierarchical Model of Intrinsic and Extrinsic Motivation (HMIEM) (from Vallerand, 1997) Figure 3 The Theory of Planned Behaviour Figure 4 Hierarchical Model of Intrinsic and Extrinsic Motivation in a drug treatment context Figure 5 The Theory of Planned Behaviour (TPB) applied to a drug treatment context Figure 6 Combined Model (Self-Determination Theory and the Theory of Planned Behavior) LIST OF TABLES Table 1 Table 2 Table 3 Table 4 Table 5 Table 6 Participants statements relating to six sources of perceived pressure to enter drug/alcohol treatment Perceived Coercion Questionnaire 48-item Pilot Measure (Phase 2) Means and standard deviation scores of Phase 2 data for Perceived Coercion Questionnaire (PCQ) subscales, the Total PCQ score, and the MacArthur Perceived Coercion Scale (N = 158) Eigenvalues, percentages of variance explained, Cronbach alphas, and the inter-factor correlations of the six subscales, the Total Perceived Coercion Questionnaire, the Mac Arthur Perceived Coercion Scale, and age for Phase 2 (N = 158) Means and standard deviation scores of Phase 3 data for continuous demographic variables, Perceived Coercion Questionnaire scores, and the Spirituality Transcendence Index (N = 362) Goodness of fit indices for model of the six-factor Perceived Coercion Questionnaire and the one-factor model for Phase 3 data (N = 362) Table 7 29-Item Perceived Coercion Questionnaire (Phase 3) Table 8 Table 9 Means associated with significant group differences between male and female participants (N = 362) in relation to legal, finance and work coercion at T1, using t-tests Means associated with significant group differences between participants who are alcohol dependent, drug dependent and alcohol and drug dependent at T1 in relation to legal and total coercion, using ANOVAs

10 Table 10 Table 11 Table 12 Table 13 Cronbach s alphas, inter-factor correlations, and correlations among the subscales and the total Perceived Coercion Questionnaire with age and spirituality for Phase 3 data (N = 362) Results of Principle Axis Factor Analyses of all predictor and outcome measures (N = 350) Cronbach s alpha scores for the six subscales and total score of the Perceived Coercion Questionnaire at T1, T2, and T Factor loadings of the Principle Axis Factor Analysis of the T1 Client Motivation for Therapy Scale (N = 350) Table 14 Correlations between the four motivational constructs at T1 (N = 350) Table 15 Correlations between the four motivational constructs at T2 (N = 192) Table 16 Table 17 Table 18 Table 19 Table 20 Table 21 Table 22 Table 23 Principle Axis Factor Analysis of the measures of the Self-Determination Theory Model at Time 1 (N = 350) Principle Axis Factor Analysis of the measures of the Theory of Planned Behaviour Model at Time 1 (N = 350) Means, standard deviations, skewness, kurtosis, and Cronbach s alpha values for Time 1 predictor and outcome variables (N = 350) Means, standard deviations, skewness, kurtosis, and Cronbach s alpha values for Time 2 predictor and outcome variables (N = 192) Means, standard deviations, skewness, kurtosis, and Cronbach s alpha values for Time 3 predictor and outcome variables (N = 109) Frequency and percentage values of categorical client-related variables at T1 (N = 350) Frequency and percentage values of categorical client-related variables at T2 (N = 192) Frequency and percentage values of categorical client-related variables at T3 (N = 109) Table 24 Visual presentation of the longitudinal design of the study Table 25 Table 26 Means associated with significant group differences between participants in the six therapeutic communities (TC) at T1 in relation to individuals age and treatment history, using ANOVAs Means associated with significant and marginally significant group differences between treatment graduates and dropouts at T1 in relation to Identified/Integrated Motivation, Perceived Competence/Control, Intention and Attitude, using t-tests x

11 Table 27 Means associated with significant and marginally significant group differences between treatment graduates and dropouts at T2 in relation to Perceived Competence/Control, Work Coercion, Attitude, Intention and Identified/Integrated Motivation, using t- tests Table 28 Correlations between T1 variables (N = 350) Table 29 Correlations between T2 variables (N = 192) Table 30 Table 31 Table 32 Table 33 Table 34 Table 35 Table 36 Table 37 Table 38 Table 39 Table 40 Table 41 Table 42 Table 43 Table 44 Table 45 Bivariate correlations between T1 predictor and T2 outcome variables Bivariate correlations between T1 predictor and T3 outcome variables Bivariate correlations between T2 predictor and T3 outcome variables Bivariate correlations between T1 predictors and T1-T2 outcome variables Bivariate correlations between T1 predictor and T1-T3 outcome variables Bivariate correlations between T1 predictor and T2-T3 outcome variables Bivariate correlations between T2 predictor and T2-T3 outcome variables Bivariate correlations between T1-T2 predictor and T1-T2 outcome variables Bivariate correlations between T1-T2 predictor and T1-T3 outcome variables Means and standard deviations for difference scores (T1-T2) of predictors and outcomes (N = 192) Means and standard deviations for difference scores (T1-T3) of outcome variables (N = 109) Means and standard deviations for difference scores (T2-T3) of outcome variables (N = 109) Reliability of difference scores as a function of the reliability of two variables and their correlations Total sample means, sub-sample means, and difference scores for T1 and T2 dependent measures to explore regression to the mean effects Total sample means, sub-sample means, and difference scores for T1 and T3 dependent measures to explore regression to the mean effects Total sample means, sub-sample means, and difference scores for T2 and T3 dependent measures to explore regression to the mean effects xi

12 xii Table 46 Table 47 Table 48 Table 49 Table 50 Table 51 Table 52 Table 53 Table 54 Table 55 Table 56 Table 57 Means, standard deviations, 95% confidence intervals, and significance levels for paired sample t-tests testing change across T1 and T2 predictor and outcome variables (N = 192) Means, standard deviations, 95% confidence interval, and significance levels for paired sample t-tests testing change across T2 and T3 predictor and outcome variables (N = 109) Means, standard deviations, 95% confidence intervals, and significance levels for paired sample t-tests testing change across T1 and T3 predictor and outcome variables (N = 109) Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for legally mandated clients (N = 68) and matched non-mandated clients (N = 68) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for legally mandated clients (N = 40) and matched non-mandated clients (N = 39) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for legally mandated clients (N = 27) and matched non-mandated clients (N = 17) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for drug court clients (N = 33) and matched non drug court clients (N = 33) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for drug court clients (N = 21) and matched non drug court clients (N = 19) at T Means, standard deviations, and 95% confidence intervals of independent sample t-tests for predictor and outcome variables for drug court clients (N = 14) and matched non drug court clients (N = 11) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for legally mandated clients (N = 35) and drug court clients (N = 33) at T Means, standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables for legally mandated clients (N = 19) and drug court clients (N = 21) at T Means and standard deviations, 95% confidence intervals, and significance levels of independent sample t-tests for predictor and outcome variables of legally mandated clients (N = 13) and drug court clients (N = 14) at T

13 xiii Table 58 Table 59 Table 60 Table 61 Table 62 Table 63 Table 64 Table 65 Table 66 Table 67 Table 68 Table 69 Table 70 Table 71 Table 72 Table 73 Table 74 Table 75 variables predicting T1 Amotivation; N = variables predicting T1 Amotivation; N = variables predicting T1 External Motivation; N = variables predicting T1 Introjected Motivation; N = variables predicting T1 Identified/Integrated Motivation; N = Summary data of standard regression analysis for T1 TPB variables predicting T1 Intention; N = Summary data for the hierarchical regression analysis of the T1 combined model predicting T1 Intention (note that the variable Perceived Competence/Control is part of SDT at Step 2); N = Summary data for the hierarchical regression analysis of the T1 combined model predicting T1 Intention (note that the variable Perceived Competence/Control is part of TPB at Step 4); N = Summary data for the hierarchical regression analysis for T 1 SDT variables predicting T1 Treatment Engagement; N = Summary data for the hierarchical regression analysis for T1 TPB variables predicting T1 Treatment Engagement; N = variables predicting Time in Treatment; N = Summary data for the hierarchical regression analysis for T1 TPB variables predicting Time in Treatment; N = Summary data for the hierarchical regression analysis of the T1 combined model predicting Time in Treatment (note that the variable Perceived Competence/Control is part of SDT at Step2); N = Summary data for the hierarchical regression analysis of the T1 combined model predicting Time in Treatment (note that the variable Perceived Competence/Control is part of TPB at Step3); N = variables predicting T1 Severity of Dependence; N = Summary data for the hierarchical regression analysis for T1 TPB variables predicting T1 Severity of Dependence; N = variables predicting T1 Drug Taking Confidence; N = variables predicting T1 Drug Taking Confidence; N =

14 xiv Table 76 Table 77 Table 78 Table 79 Table 80 Table 81 Table 82 Table 83 Table 84 Table 85 Table 86 Table 87 Table 88 Table 89 Table 90 Table 91 Table 92 Table 93 Summary data for the hierarchical regression analysis for T1 TPB variables predicting T1 Drug Taking Confidence; N = Summary data for the hierarchical regression analysis for T1 TPB variables predicting T1 Drug Taking Confidence; N = variables predicting T1 Wellbeing; N = Summary data for the hierarchical regression analysis for T1 TPB variables predicting T1 Wellbeing; N = Summary data for the hierarchical regression analysis of the T1 combined model predicting T1 Wellbeing (note that Perceived Competence/Control is part of SDT at Step 3); N = Summary data for the hierarchical regression analysis of the T1 combined model predicting T1 Wellbeing (note that Perceived Competence/Control is part of TPB at Step 5); N = variables predicting T2 Amotivation; N = variables predicting T2 External Motivation; N = variables predicting T2 Introjected Motivation; N = variables predicting T2 Identified/Integrated Motivation; N = Summary data of standard regression analysis for T2 TPB variables predicting T2 Intention; N = Summary data for the hierarchical regression analysis of the T2 combined model predicting T2 Intention (note that Perceived Competence/Control is part of SDT at Step 2); N = Summary data for the hierarchical regression analysis of the T2 combined model predicting T2 Intention (note that Perceived Competence/Control is part of TPB at Step 4); N = variables predicting T2 Treatment Engagement; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting T2 Treatment Engagement; N = Summary data for the hierarchical regression analysis of the T2 combined model predicting T2 Treatment Engagement (note that Perceived Competence/Control is part of SDT at Step 2); N = variables predicting Time in Treatment; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting Time in Treatment; N =

15 xv Table 94 Table 95 Table 96 Table 97 Table 98 Table 99 Table 100 Table 101 Table 102 Table 103 Table 104 Table 105 Table 106 Table 107 Table 108 Table 109 Table 110 Table 111 Summary data for the hierarchical regression analysis of the T2 combined model predicting Time in Treatment (note that Perceived Competence/Control is part of SDT at Step 2); N = Summary data for the hierarchical regression analysis of the T2 combined model predicting Time in Treatment (note that Perceived Competence/Control is part of TPB at Step 4); N = variables predicting T2 Severity of Dependence; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting T2 Severity of Dependence; N = Summary data for the hierarchical regression analysis of the T2 combined model predicting T2 Severity of Dependence (note that Perceived Competence/Control is part of SDT at Step 3); N = variables predicting T2 Drug Taking Confidence; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting T2 Drug Taking Confidence; N = variables predicting T2 Wellbeing; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting T2 Wellbeing; N = Summary data for the hierarchical regression analysis of the T2 combined model predicting T2 Wellbeing (note that Perceived Competence/Control is part of SDT at Ste p3); N = variables predicting T2 Perceived Training Transfer; N = Summary data for the hierarchical regression analysis for T2 TPB variables predicting T2 Perceived Training Transfer; N = Summary data for the hierarchical regression analysis of the combined model predicting T2 Perceived Training Transfer (note that Perceived Competence/Control is part of SDT at Step 2); N = Summary of the amount of variance (Adjusted R ) accounted for by SDT, TPB, and the Combined Models at T1 and T variables predicting T2 Amotivation; N = Summary data of standard regression analysis for T1 SDT variables predicting T2 External Motivation; N = Summary data of standard regression analysis for T1 SDT variables predicting T2 Introjected Motivation; N = variables predicting T2 Identified/Integrated Motivation; N =

16 xvi Table 112 Table 113 Table 114 Table 115 Table 116 Table 117 Table 118 Table 119 Table 120 Table 121 Table 122 Table 123 Table 124 Table 125 Table 126 Table 127 Table 128 Table 129 Table 130 Table 131 Table 132 variables predicting T2 Treatment Engagement; N = variables predicting T2 Severity of Dependence; N = variables predicting T2 Drug Taking Confidence; N = variables predicting T2 Wellbeing; N = variables predicting T2 Perceived Training Transfer; N = Variance explained by the SDT model for T1 predictor and T2 outcome variables Summary data for the hierarchical regression analysis for T1 STD variables predicting T3 Treatment Engagement; N = Summary data for the hierarchical regression analysis for T1 STD variables predicting T3 Severity of Dependence; N = Summary data for the hierarchical regression analysis for T1 STD variables predicting T3 Drug Taking Confidence; N = Summary data for the hierarchical regression analysis for T1 STD variables predicting T3 Wellbeing; N = Summary data of standard regression analysis for T1 STD variables predicting T3 Perceived Training Transfer; N = Variance explained by the SDT model for T1 predictor and T3 outcome variables variables predicting T3 Severity of Dependence; N = variables predicting T3 Drug Taking Confidence; N = Summary data of standard regression analysis for T2 STD variables predicting T3 Wellbeing; N = Variance explained by the SDT model for T2 predictor and T3 outcome variables Summary of the variance accounted for by the two cross-sectional and the three longitudinal SDT Raw Score Approaches Summary of hierarchical multiple regression for T1 SDT predictors and T1 outcome variables Summary of hierarchical multiple regression analyses for T2 SDT predictor and T2 outcome variables Summary of hierarchical multiple regreession analyses of T1 SDT variables predicting T2 outcomes (N = 192) Summary of hierarchical multiple regression analysis of T1 SDT variables predicting T3 Outcomes (N = 109)

17 xvii Table 133 Table 134 Table 135 Table 136 Table 137 Table 138 Table 139 Table 140 Table 141 Table 142 Table 143 Table 144 Table 145 Table 146 Summary of hierarchical multiple regression analysis of T2 SDT variables predicting T3 Outcomes (N = 109) Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which T1 SDT variables predicted changes in outcomes measures from T1 to T variables predicting changes in Amotivation from T1 to T2; N = variables predicting changes in External Motivation from T1 to T2; N = Summary data of the standard regression analysis for T1 SDT variables predicting changes in Identified/Integrated Motivation from T1 to T2; N = variables predicting changes in Treatment Engagement from T1 to T2; N = Summary data of the standard regression analysis for T1 SDT variables predicting changes in Drug Taking Confidence from T1 to T2; N = Summary data for the standard regression analysis for T1 SDT variables predicting changes in Wellbeing from T1 to T2; N = variables predicting changes in Treatment Engagement from T1 to T3; N = Summary data of the standard regression analysis for T1 SDT variables predicting changes in Severity of Dependence from T1 to T3; N = Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which T1 SDT variables predicted changes in outcomes measures from T1 to T Summary data of standard regression analysis for T1 SDT variables predicting changes in Treatment Engagement from T2 to T3; N = Summary data of standard regression analysis for T1 SDT variables predicting changes in Perceived Training Transfer from T2 to T3; N = Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which T1 SDT variables predicted changes in outcomes measures from T2 to T

18 xviii Table 147 Table 148 Table 149 Table 150 Table 151 Table 152 Table 153 Table 154 Table 155 Table 156 Table 157 Table 158 Table 159 Table 160 Table 161 variables predicting changes in Treatment Engagement from T2 to T3; N = variables predicting changes in Perceived Training Transfer from T2 to T3; N = Summary data of standard regression analysis for Time 2 SDT variables predicting changes in Wellbeing from T2 to T3; N = variables predicting changes in Severity of Dependence from T2 to T3; N = Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which T2 SDT variables predicted changes in outcomes measures from T2 to T Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which changes in SDT variables from T1 to T2 predict changes in outcomes measures from T1 to T Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Amotivation from T1 to T2; N = Summary data of standard regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in External Motivation from T1 to T2; N = Summary data of standard regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Introjected Motivation from T1 to T2; N = Summary data of standard regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Identified/Integrated Motivation from T1 to T2; N = Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict differences in Treatment Engagement from T1 and T2; N = Summary for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict Time in Treatment; N = Summary data of standard regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Dependence Severity from T1 to T2; N = Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Drug Taking Confidence from T1 to T2; N = Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict differences

19 xix Table 162 Table 163 Table 164 Table 165 Table 166 Table 167 Table 168 Table 169 Table 170 Table 171 Table 172 Degrees of freedom, F-values, significance levels, R-square values, and Adjusted R-square values of HMRAs and standard regression analyses testing the extent to which changes in SDT variables from T1 to T2 predict changes in outcomes measures from T1 to T Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict differences in Treatment Engagement from T1 to T3; N = Summary data for the hierarchical regression analysis assessing changes in SDT predictors from T1 to T2 to predict differences in Severity of Dependence from T1 to T3; N = Summary data of standard regression analysis assessing changes in SDT predictors from T1 to T2 to predict changes in Drug Taking Confidence from T1 to T3; N = Summary of the variance accounted for by the Change in Outcome Approach and the Change/Change Approach Summary of hierarchical multiple regression analysis of T1 SDT variables predicting changes in Outcomes from T1 to T2 (N = 109) Summary of hierarchical multiple regression analysis of T1 SDT variables predicting changes in Outcomes from T1 to T3 (N = 109) Summary of hierarchical multiple regression analysis of T1 SDT variables predicting changes in Outcomes from T2 to T3 (N = 109) Summary of hierarchical multiple regression analysis of T2 SDT variables predicting changes in Outcomes from T2 to T3 (N = 109) Summary of hierarchical multiple regression analysis assessing changes in SDT variables from T1 to T2 to predict changes in Outcomes from T1 to T2 (N = 192) Summary of hierarchical multiple regression analysis assessing changes in SDT variables from T1 to T2 to predict changes in Outcomes from T1 to T3 (N = 109)

20 LIST OF APPENDICES xx

21 xxi ACKNOWLEDGEMENTS Firstly, thank you very much to the residents and staff at the six therapeutic communities who participated in this project. I really appreciate the time you took to do this. A very big thank you to my excellent supervisors, Frances O Callaghan and Peter Creed, for the outstanding and invaluable support, guidance, patience and accessibility throughout this project. It has been a very steep learning curve for me, and I feel very fortunate to have had the opportunity to work with you. Thank you also to Wendy Wynne for the much needed assistance you provided in helping to format my thesis. A big thank you to my wonderful friends Julie and Rex for the loving support, encouragement and the many home-cooked meals that kept me on track through the progression of my study. Finally, I would like to thank my wonderful fiancé Paul for his encouragement and support.

22 xxii LIST OF PUBLICATIONS FROM THIS RESEARCH Klag, S., O Callaghan, F., & Creed, P. (2004). The role and importance of motivation in the treatment of substance abuse. Therapeutic Communities, 25, Klag, S., O Callaghan, F., & Creed, P. (2005). The use of legal coercion in the treatment of substance abusers: An overview and critical analysis of thirty years of research. Substance Use and Misuse, 40, Klag, S., Creed, P., & O Callaghan, F. (in press). Development and initial validation of an instrument to measure perceived coercion to enter treatment for substance abuse. Psychology of Addictive Behaviors. O'Callaghan, F., Sonderegger, N., & Klag, S. (2004). Drug and crime cycle: Evaluating traditional methods versus diversion strategies for drug-related offences. Australian Psychologist, 39,