Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
MENTAL HEALTH AND CRIMINALITY AMONG PSYCHIATRIC OFFENDERS IN SOUTHEASTERN ONTARIO by Vicky Jahye Bae A thesis submitted to the Centre for Neuroscience Studies In conformity with the requirements for the degree of Master of Science Queen’s University Kingston, Ontario, Canada (May, 2011) Copyright ©Vicky Jahye Bae, 2011 Abstract Mental Health and Criminality Among Psychiatric Offenders in Southeastern Ontario. Master of Sciences Thesis, Queen’s University, Kingston, Ontario, Canada, May 2011. OBJECTIVES: (1) Describe the population of mentally ill offenders over whom Ontario Review Board (ORB) held jurisdiction. (2) Assess the influences of psychopathology and criminal factors on criminal career. METHOD: This study was a retrospective case series design that reviewed all offenders who were court ordered for psychiatric evaluation at Mental Health Services Site of Providence Care in Kingston, Ontario from 1993 to 2007 (N=347). Eighty five subjects were found not criminally responsible on the account of mental disorder and were included in statistical analysis (n=85). Bivariate associations between five key variables and two outcome variables, seriousness of crime and recidivism, were examined. Logistic regressions were conducted to test the role of the predictor variables on the outcome variables. RESULTS: Age and change in principal psychiatric diagnosis over time were shown to be associated with seriousness of crime. Timing of psychiatric onset, early signs of deviance and change in diagnosis were shown to be associated with recidivism. On the whole, study population did not markedly vary in their distribution of variables by the outcome variables. Regression model included timing of psychiatric onset; psychiatric history; existence of criminal associate; child abuse history; and early signs of deviance. Recidivism was shown to be predicted by early signs of deviance (OR=8.154, p<0.05). ii Existence of criminal associates was shown to have substantial values of odds ratio at marginal significance (OR=7.577, p=0.13). CONCLUSION: Seriousness of crime is a complex factor that could not be sufficiently predicted by any one or combinations of study variables. Recidivism is better predicted by criminality factors than psychopathology. In the future, an exploratory analysis that more broadly examines the psychopathology and criminal factors in Canadian forensic population is needed. Findings from this study have important clinical and legal implications. iii Acknowledgements I wish to express my gratitude to my supervisor, Dr. Julio Arboleda-Florez not only for his excellent graduate supervision but also for so much more. You have been a great mentor and an inspiration. I also would like to thank Dr. Duncan Scott for co-supervising my thesis and providing invaluable input along the way and Dr. James Reynolds for guiding me throughout my graduate education. Thank you Dr. Heather Stuart for your feedback on research design and statistical analysis. Finally, thank you Dr. Michael Chan and Dr. Steve Southmayd for your insights into the Ontario Review Board and working with mentally ill offenders. The help from staff of the Providence Care Mental Health Care Services and the ORB have been central to this project. I would like to especially thank Kim Carroll at Providence Care for mining the moldy case files, sometimes 20 of them in a single day, to allow me to go through over 300 case files. The completion of this project would not have been possible without the love and support from all my family and friends, humans and furry ones alike. Thank you. iv Table of Contents Abstract................................................................................................................................ii Acknowledgements.............................................................................................................iv Table of Contents.................................................................................................................v List of Tables....................................................................................................................viii List of Figures ......................................................................................................................x List of Abbreviations..........................................................................................................xi List of Definitions..............................................................................................................xii Chapter 1. Introduction........................................................................................................1 1.1. Background..............................................................................................................1 1.2. Rationale..................................................................................................................4 1.3. Hypotheses and Objectives......................................................................................5 Chapter 2. Literature Review...............................................................................................7 2.1 Mental Health.....................,.....................................................................................7 2.1.1 Diagnosis Of Psychiatric Disorders.............................................................7 2.1.2 Prevalence..................................................................................................14 2.1.3 Impact of Mental Illness on Canadian Society..........................................17 2.2 Mental Illness and Crime.......................................................................................21 2.2.1 Prevalence of Mental Illness in Legal Systems.........................................21 2.2.2 Explanation of Trends in Crime Rate using Psychiatric Factors...............23 2.2.3 Other Explanations of Trends in Crime Rate.............................................30 v 2.2.4 Criminalization of Mentally Ill..................................................................31 2.3 Forensic Psychiatric System in Canada.................................................................35 2.3.1 Forensic Path of Mentally Ill in Canada....................................................35 2.3.2 Provincial Review Boards..........................................................................37 2.4 Challenges in Working with a Vulnerable Population..........................................39 Chapter 3. Methods............................................................................................................41 3.1 Ethical considerations............................................................................................41 3.2 Subjects..................................................................................................................41 3.3 Materials and Procedure........................................................................................41 3.4 Study Variables......................................................................................................43 3.5 Statistical Analyses................................................................................................43 Chapter 4. Results..............................................................................................................45 4.1 Description of Study Population (Objective 1)......................................................45 4.2 Statistical Analysis (Objectives 2 & 3)..................................................................45 4.2.1 Bivariate associations................................................................................ 45 4.2.2 Distribution of variables.............................................................................46 4.2.3 Multivariate associations............................................................................46 4.2.4 Additional analyses....................................................................................48 4.2.5 Assessment of Regression Model.............................................................50 Chapter 5. Discussion........................................................................................................64 vi 5.1 Population Characteristics......................................................................................64 5.1.1 Criminal responsibility...............................................................................65 5.1.2 Nature of crimes........................................................................................ 66 5.2 Findings and Interpretations from Bivariate Analysis...........................................68 5.2.1 Distribution of Sociodemographic factors.................................................68 5.2.2 Changes in Principal Psychiatric Diagnosis...............................................68 5.2.3 Timing of psychiatric onset with reference to the first crime....................70 5.2.4 Signs of early deviance..............................................................................71 5.3 Findings and Interpretations from Multivariate Analysis......................................71 5.3.1 Seriousness of index offence......................................................................71 5.3.2 Recidivism and early deviance...................................................................73 5.4 Other Contributing Factors....................................................................................73 5.5 Relevance of Findings ............................................................................................75 5.6 Limitations of Study...............................................................................................78 5.7 Future Directions....................................................................................................80 5.8 Conclusion.............................................................................................................81 References..........................................................................................................................82 Appendices.......................................................................................................................105 One. Detailed DSM Classification Tables........................................................105 Two. Forensic Path of Mentally Ill...................................................................107 Three. Data Collection Sheet..............................................................................111 vii List of Tables Table 2.1 Brief descriptions and an example of DSM IV-TR Diagnostic method....11 Table 2.2 An illustration of a personality disorder is diagnosed...............................12 Table 4.1 Demographic information of the study population....................................51 Table 4.2 Distribution of variables by seriousness of index offence. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................................................................................53 Table 4.3 Distribution of variables by recidivism. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................54 Table 4.4 Multivariate logistic regression model. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................56 Table 4.5 Bivariate logistic regression model for change in principal psychiatric diagnosis. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively.................................................................................57 Table 4.6a Multivariate logistic regression model with age as the only adjuster. Pvalues less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................................................................................58 Table 4.6b Multivariate logistic regression model with gender as the only adjuster. Pvalues less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................................................................................59 Table 4.7a Multivariate logistic regression model including changes in principal diagnosis as a predictor. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively...................................................60 viii Table 4.7b Multivariate logistic regression model including changes in principal diagnosis as a predictor and with age as the only adjuster. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................................................................................61 Table 4.7c Multivariate logistic regression model including changes in principal diagnosis as a predictor and with gender as the only adjuster. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively................................................................................................62 Table 4.8 Multivariate logistic regression model using psychopathological and criminal predictors with direct association only. Models were adjusted for only age or gender. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively..........................................................,,....63 Appendix One Detailed description of DSM IV-TR Classification...................................85 ix List of Figures Figure 2.1 Crime rate and violent crime rate trends in Canada from 1962 to 2007............................................................................................................24 Figure A.1 Navigating the forensic system..................................................................90 x List of Abbreviations DSM Diagnostic and Statistic Manual of Mental Disorders DSM IV-TR Diagnostic and Statistic Manual of Mental Disorder Fourth Edition TextRevision UST Unfit to Stand Trial NGRI Not Guilty by Reason of Insanity NCRMD Not Criminally Responsible on the account of Mental Disorder (formerly NGRI) ORB Ontario Review Board PCL-R Hare’s Psychopathy Checklist Revised VRAG Violent Risk Appraisal Guide SORAG Sexual Offender Risk Appraisal Guide xi List of Definitions Criminal career: a longitudinal series of criminal acts committed by an individual as the individual ages across the lifespan from childhood through adolescence and adulthood Principal psychiatric diagnosis: psychiatric disorder from which the accused primarily suffered at the time of crime and is therefore said to have the greatest influence on the criminal offending. Victimology: a scientific study of victimization, including the relationships between victims and offenders and the connections between victims and other social groups and institutions Unfit to stand trial: an accused is unfit to stand trial when they are unable to participate in the defense of charges because of a mental disorder or illness prevents the accused from understanding the nature of the trial, comprehending the possible consequences and communicating effectively with legal counsel. Diminished criminal responsibility: an accused who violated the law is not held fully responsible for the said crime as their mental functions were diminished. Mens rea: in English, translates to “guilty mind” and refers to the mental element of the offence that accompanies the actus reus, the “guilty act” xii Chapter 1 Introduction 1.1. Background Mental illness is one of the fastest growing global health concerns. Over 450 million people develop one or more clinical conditions at some point in life and a greater number suffers milder mental health problems1. It is also a major public health problem in Canada, where 10.4% of all above the age of 15 suffers from a major type of psychiatric disorder such as depression and substance abuse2. Society also suffers financially from mental illness. Between 2003 and 2004, 6.6 billion dollars have been spent per year on average on direct mental health care in Canada with significantly more spent as indirect cost of mental illness3,4. Challenges of psychiatric disorders are often compounded by stigmatization. Internal prejudice about mental illness discourages the patient from seeking help, while the fear of being perceived negatively by others further lessens the willingness to acknowledge its existence, which further precludes the patient from receiving necessary care and support5. This fear is not ungrounded according to past research. Stigmatization against mental illness has been repeatedly shown to be highly prevalent among the general public as well as those better educated on the subject of psychiatry, like medical students6. The issue of stigmatization is further complicated in mentally ill offenders because stigma encountered by these individuals is two-fold: one layer arising out of their psychiatric conditions and another out of their criminal behaviours7. This unique population began to attract academic attention in late 20th century as increases were observed in the number of mentally ill persons reaching the criminal 1! justice system and in crime rate following the deinstitutionalization movement8. Starting in the 1960’s, many western nations went through deinstitutionalization, an essential part of mental health care reform, in which formal, inpatient-based psychiatric care system was reduced significantly in size10-12. This resulted in a large number of psychiatric patients being discharged from hospitals to the community9-11. It was hoped to be a more cost-effective alternative and a better means of re-introducing psychiatric patients into communities than inpatient psychiatric care. While the rationale appeared to be legitimate, the net outcome was negative9-14. Many psychiatric patients who were prematurely discharged from hospitals due to deinstitutionalization entered the criminal justice system, creating what has long been termed the revolving-door patient syndrome12,13. Some researchers attempted to explain the underlying mechanism of this phenomenon by treating criminal offending as a symptom or a byproduct of mental illness16,17. This theory assumes that diminished capacity to comprehend and conform to societal standards caused by mental disorder, rather than mens rea, is accountable for the criminal offence of these people. Psychiatrization of crime theory, on the other hand, directly opposes this theory by positing that criminal offending attracts psychiatric attention18. Another possibility is that this is merely a coincidence arising from sharing common contributors, in which similar sets of genetic and developmental factors give rise to one or both of psychiatric disorders and criminality depending on individual circumstances15. Still more social factors thought to contribute to criminalization of mentally ill are: deficiency in long- and short-term mental health care services; difficulty faced by previously hospitalized or imprisoned individuals in being reintegrated to the community; difficulty faced by previously 2! hospitalized or imprisoned individuals in continuing to receive the necessary care in community setting; antagonizing attitudes of legal personnel in their handling of deviant behaviours; and stigma on mental illness or crime or both15. Mentally ill offenders embody a unique yet vulnerable group. In 1992, the Criminal Code of Canada was amended to better cope with increasing number of psychiatric offenders19. At this time, the provincial review board system, which holds jurisdiction over offenders who have been exempted of criminal responsibility given their severe mental conditions, was established into its current form19. Historically, working with the forensic population has been fraught with ethical issues, from in-house abuse at the hand of hospital staff to misuse as subjects of illegal studies20-22. Abuse by the forensic population on personnel also adds to the difficulty of issue. Scientific study at a population level has been comparatively scarce and information on Canadian psychiatric offenders was produced mainly as byproducts of examination of the general offender population, drawn from policy and administrative reports or extracted from individual legal cases23-44. Recently, several researchers have examined the Canadian psychiatric offender population under the jurisdiction of provincial review boards. Findings in these studies have been reliable but were limited in applicability because they used narrowly localized samples47-54. As such, previous research in this area left many factors unexplored. The limitation in the sample selection is perhaps inevitable given the widely varying residential characteristics between Canadian provinces55. It is not feasible to design a single research plan that would adequately account for all individual differences between localities while still retaining the ability to gather relevant data and draw reliable 3! conclusions. Researchers of the present study opined that the most effective way of generating knowledge about Canadian mentally ill offenders would be to conduct multiple regional studies then pool the results together. Thus created database would eventually allow a systematic review of mentally ill offenders at a national level as well as in-depth examination of the review board system. 1.2 Rationale Despite the clinical and social importance, empirical research on mentally ill offenders as a population has been scarce. The primary goal of the present study is to conduct a systematic review of the criminal offenders who were transferred to the Ontario Review Board jurisdiction after being found not criminally responsible on the account of mental disorder (NCRMD). A few studies provided basic description of regional review board populations19,48-54. Also, administrative reports are produced on regular basis by most provincial governments or individual review boards and more sociodemographic information is available through these channels46. Therefore, this study will provide a description of the population in terms of the psychiatric and criminal elements. The aim is to capture the characteristics that have a direct bearing on psychopathology and criminality of this population. This will be followed by an investigation of relationships between criminality and key contributors. It is believed that in the seriously mentally ill forensic population, mental illness would have a greater influence than criminality over criminal behaviours. In order to test this theory, interaction between criminality and criminal factors will be 4! compared to interaction between criminality and psychopathological or developmental factors. Predictive power of these factors will also be assessed. Information from this study will provide a better understanding of psychopathology and criminality of the southeastern Ontarian forensic population. The role of the Ontario Review Board (ORB) will also be better illustrated throughout the study. Findings may be used to facilitate the appraisal process of criminal responsibility and better direct clinical and legal rehabilitation effort to factors that actually affect future criminal outcomes. For example, different rehabilitation methods could be employed for an offender with stronger predisposition for recidivism or more serious crime and another offender with fewer propensities. Finally, there are several global projects geared toward advancement in the academic and clinical practice of forensic psychiatry56. Together with the results of other regional studies, findings of this study will provide a Canadian perspective, thereby contributing to the international knowledge on forensic population as a whole while providing platforms on which relevant regional programs may be adjusted to best serve Canadian society. 1.3 Hypotheses and Objectives This study tested the following hypotheses that, in NCRMD offenders: 1. Criminal career will be better predicted by psychopathology than criminal factors. 2. Criminal career will be best predicted when using both psychopathological and criminal predictors. The following objectives were met in order to test the above hypotheses. 5! 1. Description of the mentally-ill offender population under the jurisdiction of Ontario Review Board at a forensic psychiatric hospital in southeastern Ontario. 2. Investigation of the associations between criminal career and (1) psychopathology and (2) criminal factors. 3. Exploration of prediction models for any possible predictors of criminal career. 6! Chapter 2 Literature Review 2.1 Mental Health A sound mind in a sound body. This old proverb represents the importance placed on mental health by the ancient. Mental health, or the ‘sound mind’ as described by the ancient, refers to a “capacity of each and all of us to feel, think, and act in ways that enhance our ability to enjoy life and deal with the challenges we face” and “a positive sense of emotional and spiritual well being that respects the importance of culture, equity, social justice, interconnections and personal dignity57.” It is apparent that being mentally healthy requires more than a simple lack of mental illnesses: mental health is an interactive component of overall health, which in turn is an integrated outcome of mental, physical and social well-being58. 2.1.1 Diagnosis of Psychiatric Disorders An accurate diagnosis of a medical disorder requires a clear description of the disorder and determination of defining features59. Historically, the subjective nature of mental illness has made the fulfillment of this prerequisite difficult60, but description of mental illness by Diagnostic and Statistical Manual of Mental Disorders (DSM) provides an unambiguous and scientific characterization of psychiatric disorders. According to DSM, a psychiatric disorder is a “clinically significant behavioural or psychological syndrome or pattern that occurs in an individual61.” An important element in DSM’s definition is that the problem “must currently be considered a manifestation of a behavioural, psychological, or biological dysfunction in the individual61.” In other words, 7 not every psychological or behavioural problem constitutes a disorder. It must result from a dysfunction that the individual is currently suffering from and manifest itself to a clinically significant degree. The issue of shared symptoms has been yet another hurdle in psychiatric diagnosis. There is a substantial overlap of symptoms among different disorders and characterization based on distinctive sets of symptoms is difficult. Most psychiatric diagnostic manuals circumvent this problem by using a categorical classification system, in which mental disorders that share a set of symptoms are grouped together61,62. They are then broken down into individual disorders by a few distinguishing factors. An example with a detailed explanation may be found in Table 2.2 in a later section of this chapter. This classification based on shared and distinguishing features is also a logical alternative of classification based on etiology because it is not possible to identify a single, principal cause of a mental disorder60,62. Finally, in many cases, symptoms are not externalized and therefore objective clinical manifestations are absent60,63. Psychiatric diagnosis must rely on subjective data such as self-reports and observations by relevant parties, which makes the diagnostic process more susceptible to confounding compared to diagnosis of other types of medical conditions that use objective data such as biological markers63. The patient may not accurately express their conditions to the attending professional either deliberately or because they lack the narrative power. The diagnosis may also be affected by the professional’s biases or by a selective rather than exhaustive use of available data. Subsequently, the reliability of psychiatric diagnosis has been constantly contested63. 8 Over the last few decades, much effort has been given to overcome the problems of psychiatric diagnosis60-63. Mental health professionals across the globe collaborated in literature review, empirical data analysis and field trials to produce several rigorous and comprehensive diagnostic tools. The most widely used tool in psychiatric diagnosis is the Diagnostic and Statistical Manual of Mental Disorders (DSM) published by the American Psychiatric Association. Rivaling DSM in usage is Chapter V: Mental and Behavioural Disorders of the International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD-10) published by the World Health Organization64 DSM is used more widely in North America and this was the tool with which subjects of this study were diagnosed. A brief overview of DSM is presented here. The most recent edition, the DSM IV-Text Revision (TR), was published in 2000. There are 16 categories of disorders along with an additional list of conditions that warrant clinical attention. The multiaxial system devised for DSM IV edition has been retained in DSM IV-TR. This system explores several axes, or dimensions, of the patient’s life and allows the collection of additional information relevant for diagnosis and treatment. DSM IV-TR presents the standard conditions for each disorder and the patient is diagnosed with the disorder of which prototype his or her symptoms most closely resemble. Table 2.1 briefly illustrates the DSM IV-TR diagnostic method. Table 2.1 also shows that comorbid disorders are diagnosed and presented. Comorbidity is high in mental disorders and symptom overlap is substantial in many cases. Therefore DSM IV-TR emphasizes that its categories of mental disorder are not ‘completely discrete entities with absolute boundaries.’ When multiple diagnoses have been, the condition that occasioned the admission is designated as the principal diagnosis. DSM IV-TR also 9 recommends usage of severity and course specifiers as well as information on the recurrence of any previous psychiatric disorders. Global assessment of functioning (GAF) scores, which shows how the mental illness affects one’s functionality in a scale of 1 to 100, are typically factored in when deciding which severity specifier to use. The issue of classification by shared symptoms is further explained in Table 2.2 using the example from Table 2.1. The patient satisfied the general diagnostic criteria for personality disorder and was then subsequently interviewed for more specific personality traits. This person had met most of the 301.6 diagnostic criteria and was therefore given a final diagnosis of dependent personality disorder. 10 Table 2.1 Brief descriptions and an example of DSM IV-TR Diagnostic method. Axis I Classification Clinical Disorders Other Conditions That May Be A Focus of Clinical Attention Axis II Personality Disorders Mental Retardation Axis III Axis IV Axis V General Medical Conditions Psychosocial and Environmental Problems Global Assessment of Functioning (on a scale of 1-100) 11 Example 292.23 Major Depressive Disorder, Single Episode, Severe Without Psychotic Features 305.00 Alcohol Abuse 301.6 Dependent Personality Disorder, Moderate, In Partial Remission Frequent Use of Denial None Threat of Job Loss GAF = 35 Table 2.2 An illustration of a personality disorder is diagnosed. General diagnostic criteria for personality disorder A. An enduring pattern of inner experience and behavior that deviates markedly from the expectations of the individual's culture. This pattern is manifested in two (or more) of the following areas: (1) cognition (i.e., ways of perceiving and interpreting self, other people, and events) (2) affectivity (i.e., the range, intensity, liability, and appropriateness of emotional response) (3) interpersonal functioning (4) impulse control B. The enduring pattern is inflexible and pervasive across a broad range of personal and social situations. C. The enduring pattern leads to clinically significant distress or impairment in social, occupational, or other important areas of functioning. D. The pattern is stable and of long duration, and its onset can be traced back at least to adolescence or early adulthood. E. The enduring pattern is not better accounted for as a manifestation or consequence of another mental disorder. F. The enduring pattern is not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., head trauma). Diagnostic criteria for 301.6 dependent personality disorder 1. 2. 3. 4. 5. 6. 7. 8. has difficulty making everyday decisions without an excessive amount of advice and reassurance from others needs others to assume responsibility for most major areas of his or her life has difficulty expressing disagreement with others because of fear of loss of support or approval. Note: do not include realistic fears of retribution. has difficulty initiating projects or doing things on his or her own (because of a lack of selfconfidence in judgment or abilities rather than a lack of motivation or energy) goes to excessive lengths to obtain nurturance and support from others, to the point of volunteering to do things that are unpleasant feels uncomfortable or helpless when alone because of exaggerated fears of being unable to care for himself or herself urgently seeks another relationship as a source of care and support when a close relationship ends is unrealistically preoccupied with fears of being left to take care of himself or herself 12 There are still limitations in common to psychiatric diagnosis in general that DSM has yet to overcome. For one, culture is an important component in psychiatric development and cultural difference presents yet another problem that is unmet by other medical disorders63. DSM has been continually revised to reflect evolving cultural elements, but it is a predominantly the North American tool that produces most reliable and valid results when used on North American population61,63. However, most professionals accept that certain omissions such as specific information for a minor cultural group are necessary if a diagnostic tool is to be of practical use. It may be possible to devise a diagnostic bible that encompasses every single patient group, probes all severity ranges of psychiatric condition and includes all conceivable clinical scenarios, but professionals would find a large fraction of the tool inapplicable to their clientele and therefore unnecessary to their practices. It is more pragmatic to devise a diagnostic manual that can be used on the majority then modify it to serve the individual groups as the necessity arises. This is the approach adopted by DSM: there are now 22 languages into which the DSM has been translated with brief guidelines for clinicians of the respective cultures61. DSM is also limited in its purpose. It is a clinical diagnostic tool with the ability to serve research purposes. It is not sensitive enough to accurately grasp conditions that are not clinically significant and it is inappropriate to use DSM in non-clinical settings without any adjustment in its mode of employment61,63. This latter point is directly relevant to the present study because improper application and subsequent misuse of DSM in forensic setting have been problematic due to the “imperfect fit between the questions of ultimate concern to the law and the information contained in a clinical 13 diagnosis.” When DSM is used to diagnose forensic population or used for forensic research, it must be noted that the clinical diagnosis using DSM-IV is insufficient to establish the existence of varying degrees of mental illness for legal purposes, and that additional information beyond what DSM can yield is necessary in order to determine whether an individual meets a specified legal standard, such as competence63,65,66. 2.1.2. Prevalence In September 2010, the World Health Organization (WHO) estimated 450 million people to be afflicted with a psychiatric disorder around the globe1. Psychiatric disorders are now more prevalent than cancer, diabetes or heart diseases and account for five out of ten most frequent causes of disability2. In 2001, the Assessment, Classification, and Epidemiology (ACE) Group at the WHO launched an international mental health epidemiology project called the World Mental Health Survey Initiative which involved 28 nations58. As a part of this on-going project, prevalence rates of mental disorders between 2001 and 2003 from 17 countries were reported. The lowest rate was observed in Nigeria (12%) and the highest rate was observed in the United States (47.4%; 25-75th interquartile range across nations=18.1-36.1%)67. Of the major disorder groups – anxiety, mood, substance and impulse-control disorders – anxiety disorders were the most prevalent (IQR=9.9-16.7%), followed by mood disorders (IQR=9.8-15.8%)67. Personality disorders and psychotic disorders, which are the disorders most visible to the public, typically account for a very small proportion of overall prevalence except in offender population67,2. 14 In Canada, 10.4% of, or 2.6 million, Canadians above the age of 15 suffer from a self-reportable psychiatric disorder2. The most prevalent psychiatric disorders in Canada are mood (4.9%), anxiety (4.7%) and substance disorders (3.0%). Schizophrenia had a relatively low prevalence (<1%). Gender difference was insignificant, with females slightly more likely to suffer from all types of disorders other than substance2. Similarly little differences were noted between provinces, with mental disorders being the most prevalent in British Columbia (11.9%) and the least prevalent in Prince Edward Island (7.7%)2. Mental illness is currently the second leading cause of disability in Canada68 and it is also the second leading cause of premature death, a large fraction of which is due to suicide2,68. Prevalence of mental illness is consistently reported to be lower in developing countries than developed countries69,70. This may be due to under-reporting and flaws in data collection. In less industrialized countries, public awareness on mental health is inadequate and results in greater reluctance to admit existence of psychiatric disorders70,71. Also, history of independent public opinion research is shorter in developing nations and candidates may not see the importance of their participation72. This specific survey also paid interviewers per interview. This may have increased human error as some interviewers possibly rushed through interviews in order to receive a higher wage67. Several researchers have argued that cultural differences may be the reason for the lower prevalence rates and better prognosis of psychiatric disorders in developing nations73-77. According to these researchers, some clinical disorders as defined by western diagnostic manuals would simply be viewed as a part of the individual’s personality or an oddity at most73. For instance, Jenkins and Karno have shown that Mexicans and 15 Mexican- Americans in the United States referred to schizophrenia as nervios (case of nerves), a less stigmatizing term than a psychiatric disorder, and treated people with schizophrenia as going through a difficult phase of life that everyone else experiences in varying degrees73. Diagnostic and assessment tools directly translated from even the globalized ICD or DSM would be imperfect for highly diversified prevalence studies because some crucial cultural elements may not be sufficiently and adequately incorporated. Participants may respond truthfully, but their responses would ultimately be inaccurate for the interviewing material70,73-76. On the other hand, it is equally possible for some psychiatric conditions that are unrecognized in North American or European diagnostic manuals to be present as clinical disorders in other cultures. An example may be found in what is known as hwa-byung in Korean culture77,78. It is widely recognized as a mental disorder and there is a scale to diagnose and measure the severity of the condition as well as a fairly standard process of treatment in Korea, but it does not exist in western culture77,78. Other examples more familiar to North American culture include Koro or Genital retraction syndrome (GRS) of African and Asian cultures, amok of Malay culture, and Piblokto or Arctic hysteria of Eskimos79. The prevalence of overall mental illness in a non-western culture may be higher if it included the cases of such culture-specific disorders. The nature of mental illness makes it difficult to investigate psychiatric prevalence. However, efforts are being made to ensure careful sample selection, adequate interviewer training, an appropriate comparison group and consideration of cultural factors. 16 2.1.3. Impact of Mental Illness on Canadian Society Mental illness has a substantial health and financial impact on society as a whole. Suffering of secondarily affected persons, such as family and partners of psychiatric patients, is immense. This impact is especially severe for caregivers who provide informal, usually residential, care80. Research traditionally focused on caregivers of the elderly population and undifferentiated illnesses, with dementia and schizophrenia being the most frequently studied mental illness80-83. In all reviewed studies, caregivers have been shown to have a higher risk of psychological problems than non-caregivers81-83. In 1990, for instance, 15% of Ontarians above the age of 15 were providing informal care for some kind of disability or illness84. Caregivers were 1.5 times more likely to be diagnosed with a psychiatric disorder and twice as likely to require some type of mental health service (OR=1.5 [1.1-1.9] and OR=1.9 [1.4-2.5] respectively) than noncaregivers84. Recently, a Minnesota study provided fairly conclusive evidence on the negative influence of dementia care on the caregiver85. Clinically significant burden symptoms decreased from 52.9% to 28.3% six months following the patient’s hospitalization to a nursing home (!2(1)=162.30, p<0.001). Prevalence of depression also decreased slightly from 35.8% to 31.1% (!2(1)=315.65, p<0.001) in the same time frame. By the 12 month post-placement mark, burden had been reduced to 16.7% and depression to 27.2% (!2(1)=92.18, p<0.001 and !2(1)=191.66, p<0.001 respectively)85. This study would have been more reliable if they were able to provide information on the psychosocial state of the caregivers before they took on their caregiver role in order to ensure the negative psychiatric conditions were induced by caregiving activities. 17 Mental illness also weighs heavily on the Canadian economy. In 2004, the total health care expenditure was 132 billion dollars, of which 6.6 billion dollars have been used in the mental health care3,86. Of the 6.6 billion dollars, 5.1 billion dollars were spent on the public health care services that the provincial governments subsidized. This figure also actually represents a decrease in spending. Total health care expenditure in Canada increased from 71.5 to 131.4 billion dollars between 1993 and 200486. Spending on mental health care has decreased from 7.3 in 1993 (10.3%) to 6.6 billion dollars in 2004 (5.5%) resulting in a drop of 4.8%87. The final figure of 5.5% placed Canada lower than other nations that were rated similarly as Canada in the degree of economic development87. This seemingly discouraging comparison result, however, should not be taken as a direct indication that Canada is lagging behind other countries in providing the necessary mental health care service. Canadian Community Health Survey (CCHS) found that 9.5% of Canadians above 15 years of age had a contact with services and support for problems concerning emotions, mental health or use of alcohol and drugs, which closely matches the overall mental illness prevalence rate of 10.4%3. A closer examination revealed that 79% of interviewees that identified themselves as suffering from a mental disorder was able to access mental health care resources within a year immediately preceding the interview2. Of the remaining 21%, the most frequent reasons for not seeking professional help had been that they preferred self-care (31%) and the second most frequent response was that they did not consider it their priority (19%). This suggests that at least half of the remaining 21% did not contact mental health services voluntarily. Finally, it was encouraging to note that 82% of those who received the mental health service found it 18 satisfactory, with the majority of this majority group rating it very highly satisfactory (60%)2. Attention must be paid, however, to the finding that mental health care was most frequently given through general medical care services rather than mental health specific channels. A similar pattern had been seen in other countries such as the United States88, as well as Belgium, France, Germany, Italy, the Netherlands and Spain89. This pattern is concerning because it has negative policy and financial implications. It obscures the distinction between expenditures made on mental health care and general, nonpsychiatric care, thereby confounding the mental health care needs assessment. Also, it may be indicative of the public’s stigmatizing and avoidant attitude toward mental illness, in which the affected persons chose to seek help through primary care rather than mental health services5. This hints that the actual mental health care expenditure including that spent through non-psychiatric channels may be much higher. Meanwhile, indirect cost of mental illness is more alarming. It is estimated that 500,000 people per day miss their day of work due to a mental health issue resulting in 30% of disability claims and 70% of the total disability compensations68. A recent study estimated mental illness to cost the Canadian society a total of 46 billion dollars per year in addition to the direct cost of health care services4. The same researchers estimated that 8.4 billion dollars were disbursed in long-term work loss such as disability pension and unemployment compensation. 9.3 billion dollars were given out as short-term work loss compensations such as short-term disability leave and absenteeism, and 28.2 billion dollars were spent in care of losses in health-related quality of life from mental illness4. Mental illness cases account for only a small fraction of total disability claims, but the 19 above figures far surpass cost of physical illness. Short-term and long-term care and compensation for mental illness is almost twice the amount of physical disability68. While these figures are sufficiently striking, the overall financial impact of mental illness which encompasses the loss of productivity in non-psychiatric individuals resulting from mental illness caregiving is estimated to be much greater. 20 2.2 Mental Illness and Crime 2.2.1 Prevalence of Mental Illness in Forensic System Strong associations have been repeatedly observed between mental health and crime. One of the earliest demonstrations was through the high prevalence of mental illness among the incarcerated population. Prevalence of mental illness in the forensic system is difficult to measure on a broad scale because of multiple reasons. Firstly, most levels of judicial system have insufficient human resources to perform their main tasks and hence, research is seldom the priority. Also, variation in available data89 as well as definitions of shared terms between different sectors89 disallows any universal data collection project and impedes accurate comparison of findings from multiple studies. Finally, diagnosis is a time-consuming process which impinges heavily on the professionals, who are already struggling with providing care for the identified patients89,90. Several researchers have been able to collect prevalence data from highly localized studies, focusing on a single sector at a single location. Occasionally, results from various regional studies have been pooled to compile a national profile on basic descriptive factors. A regional study in London, Ontario, showed that severely mentally ill persons were involved in 3% of police interactions even though they represented less than 1% of all persons who came in contact with the police between 2000 and 200591. This study also revealed that they were more likely than non-psychiatric individuals to be suspected as criminal perpetrators by the police or the reporting public; to be charged for an offence; to have criminal history; and to recidivate more quickly91. 21 One of the largest prevalence studies on remandees in Canada was conducted by Arboleda-Flórez et al. in 1995 in Alberta92. Roughly 1200 arrestees who were being detained at a remand centre were interviewed over four months and 60.7% of the sample was diagnosed with a DSM Axis I or II disorder. Diagnosis came most frequently from Axis I disorders and least frequently from Axis II personality disorders with the prevalence of substance disorders falling in between. A higher proportion of men than women suffered from a disorder from all three types92. In 2002, Fazel and Danesh published their results from reviewing 62 surveys, from 12 countries, on the prevalence of mental illness in correctional system. The most prevalent disorders included personality disorders (65% [61-68]), major depressive disorders (10% [9-11]) and psychotic disorders (3.7% [3.3-4.1]) such as schizophrenia in men93. Women had same order of prevalence, with slightly different specific percent values (42% [38-45], 12% [11-14] and 4% [3.2-5.1], respectively). Altogether mentally ill individuals accounted for 26% of violent offenders. Odds ratios for psychosis and major depressive disorder were multifold of the general population, although the specific values varied widely across surveys. Antisocial personality disorder was ten times more prevalent among prisoners than the general population. They concluded that several million prisoners worldwide were afflicted with a clinical disorder. This study displayed substantial heterogeneity, only a small fraction of which was explained by difference between detainees and sentenced offenders, but it is nevertheless a powerful study with wide coverage93. 22 2.2.2 Explanations of Trends in Crime Rate Using Psychiatric Factors Association between mental illness and crime has also been demonstrated in recent attempts at explaining a puzzling social phenomenon. It was forecasted unanimously that the crime rate would increase94, but it has decreased noticeably from the early 1990’s to the late 2000’s8,95-97. Figure 2.1 illustrates the overall crime rates of Canada from 1962 to 20078. Nationally, crime rate in Canada has been steadily increasing from 1962 until it peaked in 1992 then it steadily decreased until 2007 and Figure 2.2 shows a similarly steady decrease in the violent crime incidence rate after 19928. A drop in crime rate was observed in the United States as well as England and Germany during the 1990’s95-97. 23 Figure 2.1 Crime rate and violent crime trends in Canada from 1962 to 20078. 24 There have been two landmark events in Canadian forensic psychiatry, namely deinstitutionalization and the Criminal Code reform, which coincided neatly with key changes in crime rate. Deinstitutionalization was immediately followed by the increase in crime rate in the 1960’s and Criminal Code reform had shortly preceded the drop in crime rate in 1992. Deinstitutionalization refers to the large scaled bed cuts in mental health care that started in the 1960’s and continued well into the 1990’s90. In Canada, this reduction was done mainly through closing long stay psychiatric hospitals and discharging patients into alternative care homes and communities98. Humanitarian and financial challenges of institutionalized care for psychiatric patients, such as the public outrage against conditions of asylums and budget cuts, brought in this model with an aim to involve the broader community in a holistic treatment of mental illness99. However, due to financial constraints and the lack of preparedness on the part of the receiving communities, the model has been generally perceived as having produced negative outcomes. Deinstitutionalization in Canada has not been followed by re-creation of matching available specialized psychiatric care. Psychiatric departments were created in general hospitals as an alternative, but this did not increase in the overall number of beds available for mental health care. Many researchers believe that deinstitutionalization merely unloaded the burden of care onto various parties such as the families of patients, other community services like nursing homes and community housing, and the criminal justice system100. Some research demonstrated positive outcome of deinstitutionalization, but their findings were limited in many ways. A study set in Vermont, USA, tracked patients that 25 were discharged as a result of deinstitutionalization and reported that they still needed clinical and social support, and were receiving care from alternative sources five and ten years after discharge101. They also reported that most of these patients had “lost contact” with mental health care after 25 years, but they did not clarify whether the patients voluntarily gave up care at the recommendation of a clinical professional as a result of recovery or they were simply lost in the system101. This ambiguity casts doubt on their conclusion that deinstitutionalization led to better integration back into the community since the ‘lost’ patients could have died without recovering, become vagrant or ended up in prison. Another study from the UK reported that discharged patients were settled down in alternative care services and did not burden relatives102. However, psychiatric symptoms did not improve as was expected as a positive outcome of deinstitutionalization and researchers did not investigate the burden of the illness on other parties, such as the alternative care services, that were providing the necessary care to these patients. Deinstitutionalization appears to have had a particular impact on the criminal justice system. A rising crime rate immediately followed deinstitutionalization in most countries and a steep increase in psychiatric behaviours such as suicide and self-harming rates in prisons was observed in the United States8,103. In 1939, Penrose proposed the idea of an inverse relationship between mental health care and criminal justice system, in which a decrease in supply for one increased the use of another104. Penrose’s Law accurately predicted the outcome of deinstitutionalization in Canada when the increased crime rate directly followed. Furthermore, prevalence of mental illness in federal inmates has increased following deinstitutionalization. While numbers in Canada have not been 26 clearly reported, by 2004 in the United States, where crime rate trends have been observed, mentally ill offenders incarcerated in federal prisons grew as a population by 60%, or 84% including substance abuse as a mental illness, since 1967105. The overall crime rate, which began to rise immediately after the commencement of deinstitutionalization, peaked in 1992 and has been declining since. The year 1992 is the year that the Criminal Code of Canada was amended, resulting in a major reform in the criminal justice system with respect to mentally ill offenders35. Mounting evidence on the link between crime and psychiatry, the increasing number of psychiatric patients who found their ways into prisons and the precedence-setting case of R vs. Swain led the Parliament to amend the Criminal Code (Mental Disorder) to meet the increasing and changing needs of psychiatric offenders. In 1992, the Parliament introduced Bill C-30 following the Supreme Court’s ruling which enacted several crucial changes. Changes and additions that were made to the Mental Disorder section of the Criminal Code as a result of the amendment and that are relevant to the current study are highlighted below. The amendments were largely of a humanitarian nature, such as replacing the stigmatizing “insanity defence” term with a newer and more current “defence of mental disorder” term, but substantial changes in procedure were also made. A definition of “unfit to stand trial” was codified, which stated the accused is unable on the account of mental disorder to understand the legal proceedings, and procedures for determining fitness were implemented. A new requirement of assessment orders was added, replacing the warrants of remand. In the past, an accused suspected of being mentally unstable could be detained indefinitely on the warrant of remand whereas the assessment orders are now to be made with specified purposes and a strict cap is placed on the duration of 27 remand. Quasi-judicial bodies called review boards were established to manage mentally ill offenders at a provincial-level, replacing the former Lieutenant Governor advisory board system that existed in some, but not all, provinces. Review boards were also empowered to make disposition orders, which determine whether and how the mentally ill accused should be located. They may be discharged absolutely, discharged with some requirements or detained in forensic psychiatric hospitals. Another crucial change concerned treatment orders, which in the past dictated how the accused should be treated. Mandatory treatment orders without the accused’s consent are now only given to those found unfit. Finally, the concept of criminal responsibility replaced the system of deciding between guilty and innocent. For instance, an accused may be found to have diminished responsibility or not be responsible for the crime they committed on the account of mental disorder as opposed to being found not guilty which implies that the person is not the one who committed the said offence. In summary, the process through which an accused is psychiatrically evaluated and arrives at sentencing has become more humane, systemic and decentralized35. Violent crime rate decreased for the first time since deinstitutionalization following the amendment and enactment of the amended Criminal Code. Researchers began to investigate the role of the Criminal Code reform on reduction of crime rate and a substantial body of evidence has been built. It has been suggested that the crime rate has been reduced because the crime fighting programs have been improved over the last decade with heavy emphasis focusing on controlling chronic repeat offenders106. In this view, the criminal code reform could have contributed to the crime rate reduction by detaining psychiatrically affected persons, who repeatedly committed crimes and added 28 considerably to the toll of overall crime, in forensic psychiatric hospitals for longer than if they were sentenced to general prisons, thereby taking away the opportunity to commit more crimes. This view is supported by a recent publication by Justice Canada which showed that over 50% of NCRMD offenders were detained in forensic care for at least ten years and a large proportion of these people in turn have been in the review board system for significantly longer46. Several researchers also found that, after a brief transition period spent on adjusting to the new system, significantly and substantially more accused entered the forensic psychiatric system post-reform across Canada19,25,52,107-111. NCRMD verdict was more frequently given during the first year following amendments, stabilizing into the second and third year in Alberta107. Studies also found that NCRMD verdict was more frequently given in British Columbia, along with an increase in number of diminished criminal responsibility findings19, 25,52,108. Finally, in Ontario, a greater number of psychiatric evaluations were conducted post-reform109. Assuming that the forensic psychiatric system provides adequate care that meets the unique needs of psychiatric offenders, it may be inferred that psychiatrically affected persons directed appropriately to the special system geared toward management and maintenance of mental stability of these people, instead of prisons, prevents them from committing multiple criminal offences. It is also intriguing that the crime rate began to decrease in 1992, because it was around the time that Prozac was first introduced into the Canadian market, followed by many more psychiatric medications. The introduction of Prozac into the American market in 1989 and into Canadian market in 1991 closely preceded the decrease in crime rate in 29 both countries112. Some researchers have argued that psychiatric drugs contributed to the reduction of crime by stabilizing mood problems and prevented their manifestation into criminal offences therefore having at least a partial role in crime reduction in North America. An American study showed that the diffusion of psychiatric pharmaceuticals was associated with reduction in violent crimes across various states, attributing the enhanced mental care using drug therapy to an estimated 5% reduction in crime rate113. Another study on alcoholic perpetrators of domestic violence showed a significant druginduced reduction in physical and non-physical Partner Abuse Scale ratings (ANOVA F(1,11)=10.2, p<0.01 and F(1,11)=24.2, p<0.001) compared to other subjects who received only the cognitive behavioural and alcohol therapy114. Other studies, however, have produced outcomes that negated the proponent claim. A placebo-controlled, randomized Prozac trial did not show drug-induced reduction in OAS-M Aggression scores in men with history of intimate partner abuse (ANCOVA F(1,25)=0.36, p=0.56, effect size=0.17 SD)115. At the present, there is a general belief that psychopharmaceuticals have a slight effect in decreasing violence, but the overall evidence is inconclusive as of yet116. 2.2.3 Other Explanation of Trends in Crime Rate Researchers have also attempted to explain trends in crime rate using sociological and legal factors. For instance, improvement in recording technology and practice in the 1960’s is supposed to have contributed to increased reports of crimes between 1960’s and 1970’s117. The increase in crime rate in 1980’s may also have resulted from the difficult economy. Crime rate has been shown to be inversely proportional to indicators of 30 economic strength such as unemployment rate118. Increases in crime overall and especially violent crime between 1970’s and early 1980’s are also attributed to the baby boomers reaching young adulthood, which has been labeled the most violence-prone age, at this time119-121. The subsequent decline is then deduced to be the positive outcome of a large decrease in the number of young adult119,121. The criminal justice system has gone through a number of significant program developments since 1990’s that could have led to better crime fighting and reduced crime rate119,106. Finally, it has been suggested that strengthening of the impaired driving laws and increased access to higher education by young adults could also have contributed to the drop in crime rate119. 2.2.4. Criminalization of Mentally Ill In 1972, Abramson coined the term criminalization of the mentally ill to describe the increase in the number of mentally ill persons reaching the judicial system14. As the link between mental illness and crime became more pronounced, several mechanisms by which the two spheres might influence each other were proposed. Some scholars assert that psychiatric conditions diminish the capacity for one to understand their own actions and this increases one’s chance of inadvertent violation of the law122,123. These researchers rely on the tendency of severely ill individuals to commit disorganized rather than premeditated offences for evidence18,122,123. Highly reported cases of deliberate and scheming psychiatric offenders comprise only a small fraction of crimes committed by mentally ill offenders and mentally ill offenders have been repeatedly shown to mainly commit opportunistic, minor crimes52. 31 It has also been proposed that mental illness and crime often concur because they share many of the same predispositions. Developmental history, such as child abuse, and sociological factors, such as having criminal associates and coming from a low socioeconomic background, contribute to both psychiatric development and criminal offending124-126. Differing combinations of similar genetic and developmental factors drive which of the two outcomes will be generated. It is not surprising that often it results in both of these outcomes simultaneously occurring; experience in one sphere increase the likelihood to enter the other. Psychiatrization of criminal behaviour is an intriguing approach to explain the criminalization of the mentally ill. According to this theory, criminalization of the mentally ill is a combined result of mentally ill persons entering judicial system as well as criminal offenders being sent to psychiatric inspection18. Proponents of the thesis argue that a psychiatric label is affixed to individuals who commit illegal and violent behaviours, and mental illness is viewed as a result of crime in the same way criminal offending is viewed as a result of mental illness. There are more sociological factors that collectively point to the lack of adequate accommodation for psychiatrically affected persons as a reason for the criminalization of the mentally ill. In addition to the reduction in availability of institutionalized psychiatric care, resulting from deinstitutionalization, there is also a severe lack of integrative community care programs through which the general psychiatric patients and mentally ill offenders are re-introduced and re-integrated into the community127. Studies have shown that mentally ill persons sometimes commit crime intentionally, in order to be brought back to incarceration settings where they are accustomed to and where they can continue 32 to receive systematic care, or unintentionally as a result of failure to adapt to society127. It has also been found that personnel in the crime-fighting sector often assume hostile attitudes when working with mentally ill offenders. Their general belief that deviant behaviours are best dealt with by taking a firm stance, rather than treating them with an accommodating mindset, is detrimental to the prognosis of mental illness in these people127. Stigmatization on mentally ill offenders also appears to be a crucial factor in criminalizing the mentally ill. Studies have shown that persons with mental illness or suspected to have mental illness are more likely to be reported and arrested rather than diverted91. Studies that investigated the causal role of mental illness on violence have mostly produced negative results. These studies have shown associations of moderate strength at best and the direction of causality is yet to be clearly established128-130. For instance, a study reported that the majority of mentally ill offenders were also comorbid with longterm substance abuse, which is a known cause of violence, and they could not attribute the case of violent crimes entirely to the principal diagnosis129. These same researchers also built a predictive model for violence, which predicted violence using severe mental illness, but only when other predictors such as substance abuse and environmental factors were also included in the model129. Link, Stueve and Phelan (1998) reported an association between violence and psychotic disorders, showing that psychotic symptoms, such as hallucinations, appeared to cause violence if the affected persons had deluded the victims to be dangerous to themselves or if the severity of symptoms was great as to override one’s prohibitions against violence130. This study, while particularly useful because they narrowly defined exactly which component of the mental illness was linked 33 to violence, is still insufficient by itself to establish that mental illness is a cause of violence nor is it generalizeable to mental illness as a whole130. In summary, contrary to the popular belief that mental illness is conducive to violence, scientific evidence is absent. 34 2.3. Forensic Psychiatric System in Canada Since the large influx of mentally ill individuals into the criminal sector, the Canadian criminal justice system has been making adjustments to accommodate this unique population. One measure included active involvement of forensic psychiatrists in the management of mentally ill offenders. Forensic psychiatry is a highly specialized field of medicine dedicated to helping the mentally ill with legal troubles in a continuum of social systems, namely mental health care, criminal justice and correction130. Contributions of forensic psychiatry in the management of mentally ill offenders are multi-folds. Forensic psychiatrists perform three types of forensic assessment: fitness to stand trial, criminal responsibility, and dangerousness to community. Results from these assessments are crucial in legal proceedings, as judges and counsels rely heavily on the expert reports in determining the sentences. As crucial as their duty as legal expert in determining the legal outcome for mentally ill offenders is a forensic psychiatrist’s responsibilities as a physician. Forensic psychiatrists are key members of the clinical teams tasked with the mental health care of thus sentenced mentally ill offenders130. Other ways through which the judicial system has endeavoured to meet the demands of a mentally ill offenders include the introduction of various points of diversion in the forensic path of mentally ill accused and the establishment of provincial review boards. 2.3.1 Forensic Path of the Mentally Ill in Canada The legal path of a general offender begins with arrest. The arrestee is booked with or without interrogation by the police and the police lay the charges. The arrestee 35 awaits the court appearance in the remand centre. At the first appearance at the court, one may be allowed to be bailed out until trial, or, given the risk in their management, may be disallowed bail and await trail in the remand centre. A trial occurs with or without preliminary hearing, depending on the seriousness of crime. If convicted, the accused is sentenced and is incarcerated in a provincial prison if the sentence is less than a two-year term and in a federal prison if the sentence is any longer than two years and a day131. The forensic path of the mentally ill is differentiated from this typical path by points of diversion. A diversion refers to a medico-legal alternative to criminal sanctions available to the mentally ill persons and is present at every stage of legal proceedings132. Detailed information on the regular and psychiatric forensic paths and an illustration of forensic path for the mentally ill are given in Appendix Two. Diversions are mainly used in order to more humanely handle the cases of those who committed minor offences and to prevent overloading of criminal courts. A serious offender, on the other hand, may be diverted by the court in the presence of severe mental illness132. Of relevance to the present study is this latter case. The accused is transferred to a quasi-judicial body called the provincial review board if the court determines the accused to be not criminally responsible on the account of mental disorder (NCRMD). This means that the accused was incapable of comprehending their actions and consequences when they committed the crime because of their mental illness. Here they are given the equivalent of sentencing, called a disposition, which determines whether the accused will be detained in forensic psychiatric hospitals or will be released into the community. The issue of fitness may be raised by any party of the court. If the fitness assessment results confirm that the individual is unfit to stand trial, the court issues a make fit order, which orders the 36 accused to receive treatment for up to 60 days to return to a fit state. If the treatment fails to restore them to fitness, the jurisdiction over the accused is transferred to the review board132,133. 2.3.2. Provincial Review Board Prior to the 1992 Criminal Code reform, mentally ill offenders were handled through the main criminal justice system. Accused who were found not guilty by reason of insanity (NGRI) were strictly detained on what was termed the Lieutenant Governor’s Warrant. The detained accused could hope to climb down the security ladder over time, but there was no procedure specified as to how this isolated population should be managed. Lieutenant Governor’s advisory boards had been in existence in some provinces before the reform, but their activities were restricted to reporting outcomes to the Lieutenant Governor23,29,47. Along with amendments to the Criminal Code of Canada, Bill C-30 required each province to strike a review board, an adjudicative body that holds exclusive jurisdiction over persons whom the court have found to be NCRMD or unfit to stand trial and did not regain fitness after the treatment order35. Unlike the Lieutenant Governor’s advisory board, a provincial review board is empowered to autonomously make orders or dispositions for these people. The very first hearing held for an accused by the review board is called the initial hearing. A disposition made here is reviewed every year and may be modified, largely depending on the prognosis of the mental disorder. Another crucial determinant of the disposition, especially in the case of sexual offenders, is the assessment of dangerousness. Using information such as severity of illness, seriousness 37 of index offence and prior criminal history, the psychologist and forensic psychiatrists assess the risk the offender poses on the general public if they were to be released to the community. Counsels on each side are given the opportunity to support or challenge the expert opinion provided by the assessing psychiatrist and the review board ultimately reaches a disposition, which may order the offender to be detained in the forensic hospital if their dangerousness assessment indicates that they pose significant threat to the public35. Review boards have been steadily growing in size across Canada46. This is largely due to the permanence of the review board regulations. The mentally ill accused often take a long time to regain fitness or show sufficient improvements in principal diagnosis as to impose little or no threat to the community. Also, courts have been sending increasingly more offenders to review boards, which created a cycle where there is a constant influx with an incomparably small number of offenders leaving the system46,134. Review boards have a special arrangement with hospitals that meet the Criminal Code definition for forensic psychiatric assessment and care and have been designated by the Ministry of Health. There are 12 hospitals designated for the Ontario Review Boards. Subjects for the current study came exclusively from one of these hospitals, the Providence Care, Mental Health Services in Kingston, Ontario134. 38 2.4. Challenges in Working with a Vulnerable Population Forensic psychiatry, the field of medicine tasked with the care of mentally ill offenders, is a relatively young specialty with conflicting ethical demands on the part of the practitioners130. On the one hand, they have their responsibilities as physicians whose priority is patient care. On the other hand, the nature of the field requires the practitioners to work as an agent for a third party, thereby prioritizing some goal other than patient care130. This has led to many complications in the protection and care of mentally ill offenders. For instance, it is more difficult for forensic psychiatrists to build the necessary trustworthy relationship with their offender patients because the patient, for a justifiable reason on their end, may be reluctant to reveal information crucial for their mental health treatment that may be used against them legally135. For another, it is unclear whether restoration of fitness is truly benefiting the patient because it results in sending the patient away from specialized care to legal proceedings, which has a good chance of sending them to a more psychopathogenic prison environment135. Working with mentally ill offenders is also difficult because they constitute a very unpopular patient group. Shortages of funding and human resources often lead to mental health care services being organized around a few priority patient groups, while largely neglecting other groups such as Aboriginal people and psychiatric offenders130,136. Mentally ill offenders constitute one of the most desperate patient groups as evidenced by psychiatric prevalence unrivalled by any other constituent groups93,136. It should then be more appropriate to accommodate the mentally ill as those with physical disability are accommodated, but therapeutic and rehabilitative care of mentally ill offenders receive far less attention than the correctional and punitive measures130. 39 Moreover, the double stigma of mental illness and crime as well as suppressed human rights, have often led this population to serve as subjects of unethical and dangerous researches. Offenders are especially susceptible because participation is often coerced136. Activists have also addressed the concern that, even in the case of voluntary participation, they may not be able to make truly free and informed consent decisions137. From the Nazi doctors to the Tuskegee syphilis experiment, instances of inhumane and unethical sufferings of prisoners in the name of common good are numberless137. The issue of research participation is especially relevant to mentally ill offenders who often have diminished capacity for comprehending the details of the research project and their risks in participation138. It is evident that mentally ill offenders comprise a difficult group to work with. Yet this population cannot be neglected, not only in light of the size of the population or economic impact, but also from a humanitarian stance. The vulnerability as well as the challenges in providing care and studying this population places clinicians and researchers of forensic psychiatry against higher standards of practice. 40 Chapter 3 Methods 3.1 Ethical considerations This study has been approved by the Research Ethics Board of Queen’s University and the Research Review Committee of Providence Care Hospital. 3.2 Subjects Since the inception of the ORB in 1992, Providence Care Forensic Services has admitted a total of 347 adults for court ordered psychiatric evaluation. All charts were reviewed and basic information was gathered for a descriptive analysis (N=347). Detailed information was collected on patients who received the verdict of NCRMD and who were no longer in-patients of the hospital (n=85). Active in-patients of the hospital were excluded in the study due to ethical considerations. Providence Care has provisions regarding active in-patients which require added measures for protection of confidentiality, such as a consent form. It was difficult to follow this protocol because psychiatric in-patients admitted under treatment orders often had diminished capacity and were not able to comprehend the concept of an informed consent. It was further complicated by the fact that these individuals lacked legal guardians who could give consent on their behalf. Less than 10 patients were excluded on this criterion. 3.3 Materials and Procedure Data were collected by a retroactive chart review. There was not a direct interaction between the student researcher and subjects. Charts included a wealth of 41 information including but not limited to: police records, witness statements, medical records, psychiatric records, reports for the initial and annual hearings of the ORB and results of specialized tests, such as Hare’s Psychopathy Checklist Revised(PCL-R)139, Violent Risk Appraisal Guide (VRAG) and Sexual Offender Risk Appraisal Guide (SORAG)140,141. Physicians at Providence Care began utilizing the Hare’s PCL-R in 1995 and VRAG and SORAG were first introduced in 1998. The gap between the 1993 inception of ORB and the adoption of these tools resulted in the information missing in a significant number of subjects. Consequently, the scores of the aforementioned specialized tests were ultimately excluded from data analysis despite their importance in psycho-criminal research. Most of the sociodemographic and criminal information was extracted directly from the police report. Diagnoses were given by forensic psychiatrists at Providence Care following the guidelines of Diagnostic and Statistical Manual of Mental Disorders IV and later IV-TR61. Psycho-criminal information was collected by the ORB staff and clinical teams at Providence Care. Data collection and electronic data coding were conducted simultaneously (see Appendix Three for data collection sheet). Sociodemographic information collected included: gender, age, place of residency, marital status, living arrangement, level of education, and income source. Psychopathological information collected included: current and life diagnoses, previous psychiatric admission history, known psychiatric associates, and timing of first psychiatric diagnosis. Psycho-social information collected included: child abuse history, adult relationship history, adult violence history, general medical conditions, psychosocial environmental factors, and global assessment of functioning. Criminal 42 information collected included: type of index offence, place and location of offence, victim’s relationship to the offender, childhood criminal history, known criminal associates, criminal history. Psycho-criminal information collected included: legal verdict, psychiatric assessment site, initial ORB hearing site, scores of the aforementioned standardized tests such as PCL-R, VRAG and SORAG, fitness to stand trial, and ORB’s disposition regarding treatment. 3.4 Study Variables Study variables were chosen in three steps: review of literature; consultation with ORB forensic psychiatrists; and consideration of the availability of data. An initial list of over twenty potential study variables was drawn from perusal of earlier literature on ORB and the general Canadian forensic population25-54,142-146. Forensic psychiatrists with a lengthy ORB career were consulted in order to narrow down this initial list. They were asked to select the variables that they had frequently seen in their practice and suspected to have effect on criminal outcomes. Finally, following data collection, some of the variables on the created list were excluded when they had insufficient information. Variables that had a zero cell frequency in one of the categorical level or were missing from more than a third of the study population were excluded. Seriousness of index offence and recidivism were chosen as the outcome measures of criminality. 3.5 Statistical Analyses Chi square tests were performed to explore the distribution of sociodemographic and predictor variables by the two outcome variables. Cross tabulations were also 43 performed between predictor variables and outcome variables in order to explore bivariate associations. Chi square and one-tailed exact p-values were reported, except when there were more than two categorical levels, in which case 2-tailed p-values were reported. When there was a small cell size (<5), Fisher’s exact test values were reported instead of chi square values. A multiple logistic regression analysis was run in order to assess the role of individual variables and combinations of them in predicting the outcomes of seriousness of index offence and recidivism. Because predictors were selected from scarce prior research on a similar topic, none of which focused on Canadian forensic population, it was not feasible to determine the predictors’ order of importance in predicting the outcome. Also, after the initial exploratory analysis using chi square, it was apparent that, although all predictors met the significance cut-off to be included in the model, three predictors had significance values well over 0.15. The remaining two predictors were significantly associated (p<0.05) only to one and not the other outcome variable. Determining the order of importance among thus statistically insignificant associations would have yielded little benefit. As such, forced entry method was selected. The models were adjusted for age and gender to control for confounding. 44 Chapter 4 Results 4.1 Description of Study Population (Objective 1) Out of 347 psychiatrically evaluated offenders, 85 received the verdict of NCRMD and were included in data analysis. This represents a success rate of 25.5% for NCRMD among those who raised the NCRMD defense. The study population was predominantly male (88.2%), had a mean age of 33 years and was most frequently diagnosed with schizophrenia (70.6%). Nearly all subjects came from low socioeconomic status: the majority received no post-secondary education (70.7%) and had no income or received some kind of assistance or pension (94.1%). The most common index offence was assault (48.2%) and the disposition most frequently given by the ORB was full-time detention at the hospital (92.9%). Subjects more often began their psychiatric life before criminal career as shown by first official psychiatric diagnosis predating the first criminal charge (78.8%); more often had psychiatric history than criminal history (84.7% vs. 67.1%); and were more likely to have psychiatric associates than criminal associates (45.9% vs. 17.6%). In some of the subjects, a different primary psychiatric diagnosis was given during the ORB assessment than their first official diagnosis (27%). Detailed demographic information of the study population is presented in Table 4.1. 4.2 Statistical Analysis 4.2.1 Bivariate associations Bivariate associations were explored using chi-square tests between the seriousness of index offence or recidivism. This also allowed for detection of ! 45! dissimilarities in distribution of variables by the outcome variables. Summary of cross tabulations and one-tailed chi-square or Fisher’s exact test p-values are reported in Table 4.2 and 4.3. For some variables that were initially divided into groups of narrow ranges, category levels were collapsed into larger ranges and therefore a fewer number of groups. This practice increased cell sizes for individual groups and allowed chi-square significance testing. 4.2.2 Distribution of variables Variables that did differ by the outcome variable were more closely examined to learn where the differences came from. Both sociodemographic factors and predictor variables were similarly distributed between the subjects who committed petty crimes and the subjects who committed serious crimes with the exception of age. Subjects tended to have committed serious crimes when they were younger (!2(5)=13.624, p<0.05). First-time offenders and repeat offenders showed more marked differences. Psychiatric onset had predated the first crime in all first-time offenders while one third of repeat offenders reported to have developed a psychiatric disorder after they committed their first crime (!2(1)=15.468, p<0.001). Repeat offenders also far surpassed first timers in early deviance history (73.1% and 37.5%, respectively; !2(1)=15.468, p<0.005). 4.2.3 Multivariate associations (Objective 2&3) A cut-off significance value of p=0.2 was applied when selecting variables to be used as predictors in the regression analysis. Predictors that met this cut-off value for ! 46! either of the two outcome variables were included in the model for both. Sociodemographic variables were not included in the main regression model even if they met the cut-off value in order to avoid overloading of the model and also because the primary goal of this study was investigation of association between criminality and psychiatric and criminal factors, not basic sociodemography. Also, the variable that represented changes in principal diagnosis was not included in the model even though it met the significance cut-off for both of the outcome variables. The concern here was that this variable was not applicable to subjects who did not previously have a psychiatric diagnosis (n=14). Also, this variable was suspected to be prone to human error because it was not always indicated in the charts when and where the diagnoses were given. Furthermore, it was not clear how the change was patterned because it did not always change from an initial diagnosis of a type of disorder to another, like going from dysthymic disorder to major depressive disorder. The risk for confounding would have been severe and therefore it was excluded from the regression model and was examined individually. Psychiatric history and timing of psychiatric onset were chosen as psychopathological predictors. Known criminal associates and early signs of deviance were chosen as criminal predictors. The term ‘associate’ was limited to family members and romantic partners and excluded more ambiguous relationships like friends or colleagues. Also, child abuse history was chosen as a developmental predictor that was attributed to both psychopathology and criminality. Both crude and adjusted results from the multiple logistic regression analysis are reported in Table 4.4 to compare confounding. Seriousness of index offence was not ! 47! significantly predicted by any of the examined predictor variables and any substantial values of odds ratio were accompanied by high p-values (p=0.581, OR=1.812 for existence of criminal associate and p=0.396, OR=2.292 for child abuse history) and were consequently not considered further. Subject’s risk of recidivism increased by eight times with a history of early deviance (OR=8.154 [1.401-47.461], p<0.05). Other predictor variables did not significantly affect the likelihood to recidivate, but existence of criminal associates was shown to increase the likelihood to recidivate by seven times at a marginal significance (OR=7.758 [0.541-106.2], p=0.133). Results from individual examination of change in principal diagnosis variable are presented Table 4.5. A meaningful odds ratio for committing more serious crime was observed initially (crude OR=2.914 [0.754-11.271], p=0.121), but the effect disappeared with adjustment (OR=1.986 [0.473-8.332], p=0.348). The variable did not significantly predict recidivism (crude OR=0.488 [0.164-1.452], p=0.197 and adjusted OR=0.637 [0.194-2.093], p=0.458). 4.2.4 Additional analyses Comparison of regression models showed that values adjusted using age and gender were substantially different from the crude values. To test the role of age and gender as possible confounders, models were run again with age or gender as the sole adjusters and results are shown in Table 4.6. Odds ratio of recidivism predicted by early deviance was improved when age was controlled, but remained the same when gender was controlled. Significance values did not change substantially. It may be indicative that ! 48! age is a greater confounder than gender in predicting the criminality of the subjects. Gender should still be monitored as a possible confounder since adjusting for gender as well as age produced a greater change in the odds ratio than when age alone was used. Also, existence of criminal associate substantially predicted recidivism at a marginal significance when adjusted for age and child abuse history substantially predicted recidivism at a marginal significance when adjusted for gender (OR=6.628 [0.56078.501], p=0.134, and OR=5.041 [0.509-49.955], p=0.167, respectively). Changes in principal diagnosis had not been included as a predictor in the main regression analysis given the high chance of human error in information collection process. An additional analysis was done with this variable as one of the predictors to examine its role in predicting criminality. Findings are presented in Table 4.7a. Regression models were also run including changes in principal diagnosis as a predictor and adjusted for age or gender only. Findings may be found in Tables 4.7b and 4.7c. Significant results were not produced in this analysis. Early deviance, however, substantially predicted recidivism at a marginal significance when adjusted for both age and gender as well as when adjusted for age alone. Child abuse history substantially predicted recidivism at a marginal significance when adjusted for gender (OR=4.274 [0.638-28.623], p=0.134, OR=2.886 [0.634-13.137], p=0.171 and OR=3.645 [0.60821.864], p=0.157, respectively). Each of the outcome variables was then tested using the predictor variables that were shown to be associated with only that outcome variable, as opposed to either of the two outcome variables, as was done in the main analysis. They were also adjusted using only the sociodemographic factor that was shown to be associated with only that outcome ! 49! variable. Regression analysis on the outcome of recidivism was ultimately not run because the list of predictor variables was exactly the same as results produced for this outcome variable in Table 4.6b. Therefore, only the seriousness of index offence variable was tested using age as the sole adjuster. Findings are presented in Tables 4.8. 4.2.5 Assessment of Regression Model Goodness of fit statistics (R2L, R2CS, R2N and -2 Log Likelihood), residuals and collinearity diagnostics were generated for the final logistic regression model. Only main effects were tested without assessment of interactions. R2 values in logistic regression should be interpreted with caution since they only provide hypothetical trends in model fitness here. Maximum R2 values indicate that 35.7% and 56.8% of the variation in the final models were explained by the predictors. Both of these figures were significant and substantial improvement from the basic model which only included the intercepts. Residuals appeared mostly normal. There were two cases in the seriousness of crime model whose Cook’s Distance values were slightly above five and two cases in the recidivism model with values over one. There were 12 cases and 18 cases respectively in each of the models that exceeded the predicted leverage value, which again may indicate a problem. Doubtful leverage values, however, do not indicate undue influence on regression coefficient but rather on the outcome variable189. Also, given that quite a few predictors were used in the model, influence may have been overly presented. Collinearity diagnostics did not indicate any multicollinearity. No tolerance value was less than 0.1, no VIF higher than 10 was observed and eigenvalues were not correlated. ! 50! Table 4.1. Demographic information of the study population. Variable (n=85) Gender Male Female Age (years) Mean (SEM) 18-25 26-30 31-35 36-40 41-45 46-50 51-55 56-60 61-65 Highest level of education reached Grade 8 or less High School - some High School - complete Community College - some Community College - complete University - some University - complete Income source Unemployed Paid employment Employment Insurance Social Assistance Disability Pension Retirement/Old Age Pension Other (specify when possible) Primary psychiatric diagnosis Schizophrenia (excluding schizoaffective & schizophreniform) Delusional or other psychotic disorder Schizoaffective disorder Bipolar disorder - manic Personality disorder - Antisocial Other Index offence Homicide Assault Sexual assault Offence against property Others (e.g. prostitution) Treatment disposition by the ORB Detain - inpatient care Conditional discharge - halfway housing Conditional discharge - with supervision only Timing of psychiatric onset Predates first crime Follows first crime Other Psychiatric Characteristics Has psychiatric history Has psychiatric associates ! 51! n(%) 75(88.2) 10(11.8) 33.59(1.254) 22(25.6) 22(25.6) 11(12.7) 11(12.8) 7(8.2) 3(3.5) 3(3.6) 3(3.6) 3(3.5) 10(11.8) 31(36.5) 19(22.4) 7(8.2) 5(5.9) 4(4.7) 7(8.2) 19(22.4) 5(5.9) 2(2.4) 7(8.2) 35(41.2) 1(1.2) 7(8.2) 60(70.6) 5(5.9) 4(4.7) 4(4.7) 3(3.5) 9(10.6) 14(16.5) 41(48.2) 6(7.1) 9(10.6) 14(16.5) 79(92.9) 1(1.2) 1(1.2) 67(78.8) 18(21.2) 72(84.7) 39(45.9) Primary psychiatric diagnosis changed between first official diagnosis and the ORB assessment Other Criminal Characteristics Has criminal history Has criminal associates ! 52! 20 (27.0) 57(67.1) 15(17.6) Table 4.2. Distribution of variables by seriousness of index offence. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Variable (n=84) Seriousness of Index Offence Petty Grave n=23(27.4%) n=61(72.6%) Gender (84) Male (75) Female (9) Age (84) 18-25 (22) 26-30 (22) 31-35 (11) 36-40 (11) 41-45 (7) 46 or above (11) Highest level of education reached (83) Grade 8 or less (10) High School - some (31) High School - complete (19) Community College - some (7) Community College - complete (5) University or above (11) Income source (76) Unemployed (19) Paid employment & income resulting from previous paid employment (retirement plan, employment insurance) (8) Social Assistance (e.g. Disability Pension, Ont. Works, Family Benefits Act) (41) Other (specify when possible) (8) Primary psychiatric diagnosis (84) Schizophrenia (59) Other psychotic disorder (8) Mood Disorder (6) Personality disorder (5) Other (6) Treatment disposition by the ORB (80) Detain - inpatient care (78) Conditional discharge (2) Timing of psychiatric onset (84) Predates first crime (66) Follows first crime (18) Other Psychiatric Characteristics Has psychiatric history (71) Has psychiatric associates (38) Primary psychiatric diagnosis changed between first official diagnosis and the ORB assessment (20) Other Criminal Characteristics Has criminal history (56) Has criminal associates (15) Signs of deviance in childhood (48) Developmental Characteristics Has child abuse history (18) ! 53! !2 p-value 21(91.3%) 2(8.7%) 54(88.5%) 7(11.5%) 0.531 4(17.4%) 5(21.7%) 3(13%) 2(8.7%) 6(26.1%) 3(13%) 18(29.5%) 17(27.9%) 8(13.1%) 9(14.8%) 1(1.6%) 8(13.1%) 0.032* (Fisher’s exact; 2tailed) 3(13%) 8(34.8%) 7(30.4%) 1(4.3%) 2(8.7%) 2(8.7%) 7(11.9%) 22(37.3%) 12(20.3%) 6(10.2%) 3(5.1%) 9(15.3%) 0.825 (Fisher’s exact; 2 tailed) 3(15.8%) 0(0%) 16(28.1%) 8(14%) 0.153 (Fisher’s exact; 2 tailed) 14(73.7%) 27(47.4%) 2(10.5%) 6(10.5%) 15(65.2%) 2(8.7%) 2(8.7%) 2(8.7%) 2(8.7%) 44(72.1%) 6(9.8%) 4(6.6%) 3(4.9%) 4(6.6%) 0.88 (Fisher’s exact; 2 tailed) 22(100%) 0 56(96.6%) 2(3.4%) 0.523 (Fisher’s exact) 16(69.6%) 7(30.4%) 50(82%) 11(18%) 0.173 21(91.3%) 12(54.5%) 3(14.3) 50(82%) 26(44.8%) 17(32.7%) 0.244 0.299 0.093 16(69.6%) 4(19%) 13(68.4%) 40(66.7%) 11(19.3%) 35(60.3%) 0.509 0.628 0.364 6(33.3%) 12(32.1%) 0.570 Table 4.3. Distribution of variables by recidivism. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Variable (n=84) Recidivism First timer Repeat offender n=27(32.1%) n=57(67.9%) Gender (84) Male (74) Female (10) Age (84) 18-25 (22) 26-30 (21) 31-35 (11) 36-40 (11) 41-45 (7) 46 or above (12) Highest level of education reached (82) Grade 8 or less (10) High School - some (30) High School - complete (19) Community College - some (7) Community College - complete (5) University or above (11) Income source (76) Unemployed (19) Paid employment & income resulting from previous paid employment (retirement plan, employment insurance) (8) Social Assistance (e.g. Disability Pension, Ont. Works, Family Benefits Act) (42) Other (specify when possible) (7) Primary psychiatric diagnosis (84) Schizophrenia (59) Other psychotic disorder (8) Mood Disorder (6) Personality disorder (5) Other (6) Index offence (83) Homicide (14) Assault (40) Sexual assault (6) Property and other Criminal Code offences (23) Treatment disposition by the ORB (80) Detain - inpatient care (78) Conditional discharge (2) Timing of psychiatric onset (84) Predates first crime (67) Follows first crime (18) Other Psychiatric Characteristics Has psychiatric history (71) Has psychiatric associates (38) Primary psychiatric diagnosis changed between first official diagnosis and the ! 54! !2 p-value 22(81.5%) 5(18.5%) 52(91.2) 5(8.7) 0.176 8(29.6%) 9(33.3%) 3(11.1%) 2(7.4%) 1(3.7%) 4(14.8%) 14(24.6%) 12(21.1%) 8(14%) 9(15.8%) 6(10.5%) 8(14%) 0.693 (Fisher’s exact; 2 tailed) 2(7.4%) 7(25.9%) 8(29.6%) 3(11.1%) 1(3.17%) 6(22.1%) 8(14.5%) 23(41.8%) 11(20%) 4(7.3%) 4(7.3%) 5(9.1%) 0.339 (Fisher’s exact; 2 tailed) 6(23.1%) 5(19.2%) 13(26%) 3(6%) 0.389 (Fisher’s exact; 2 tailed) 13(50%) 29(58%) 2(7.7%) 5(10%) 17(63%) 3(11.1%) 4(14.8%) 1(3.7%) 2(7.4%) 42(73.7%) 5(8.8%) 2(3.5%) 4(7%) 4(7%) 5(18.5%) 14(51.9%) 1(3.7%) 7(25.9%) 9(16.1%) 26(46.4%) 5(8.9%) 16(28.6%) 24(96%) 1(54%) 54(98.2%) 1(1.8%) 0.530 (Fisher’s exact) 27(100%) 0 39(68.4%) 18(31.6%) 0.000*** 21(77.8%) 12(46.2%) 8(38.1%) 50(87.7%) 26(48.1%) 12(23.1%) 0.195 0.529 0.156 0.435 (Fisher’s exact; 2 tailed) 0.906 (Fisher’s exact; 2 tailed) ORB assessment (20) Other Criminal Characteristics Has criminal associates (15) Signs of deviance in childhood (47) Developmental Characteristics Has child abuse history (23) ! 55! 2(8%) 9(37.5%) 12(22.6%) 38(73.1) 0.101 0.003*** 5(20.8%) 18(36.7%) 0.134 Table 4.4. Multivariate logistic regression model. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Predictors Constant Psychiatric onset Seriousness of Index Offence Crude Adjusted B (SE) OR B (SE) OR [95% [95% CI] CI] 1.849 6.351* 1.691 5.426 (0.873)* (1.007) -0.873 0.418 -0.299 0.742 (0.635) [0.120(0.770)! [0.1641.450] 3.352] Psychiatric history -0.556 (0.858) 0.573 [0.1073.080] -0.050 (0.925)! 0.951 [0.1555.831]! 0.279 (0.806)! 0.321 [0.2726.410]! 0.352 (0.948)! 1.422 [0.2229.125]! Existence of criminal associate(s) Child abuse history 0.111 (0.681) 0.595 (1.077)! 5.239 [0.52752.121] 2.197 [0.5199.304] 3.642 [1.02212.972]* 67.2% 2.025 (1.347)! -0.271 (0.646) 1.812 [0.21914.972] 2.292 [0.33715.600] 0.827 [0.1773.862] 82.4% 1.656 (1.172)! Signs of early deviance Accuracy of prediction -2LL R 2N 1.713 [0.3079.576] 1.117 [0.2944.247] 0.762 [0.2152.705] 75% 7.577 [0.541106.2] 1.958 [0.31712.103] 8.154 [1.40147.461]* 83.6% 0.538 (0.878) 76.5*** 0.067 0.830 (0.978)! -0.190 (0.786) ! 57.705 0.357 ! ! Recidivism Crude Adjusted B (SE) OR B (SE) OR [95% [95% CI] CI] -1.051 0.350 -1.652 0.192 (0.801) (1.079) 21.100 1.457E9 21.389 1.946E9 (9676.72 [0-] (9417.67 [0-] 1)! 0)! 0.787 (0.736)! 1.292 (0.648)* ! ! 56! 84.8*** 0.438 0.672 (0.929)! 2.099 (0.899)* ! ! 49.7*** 0.568 Table 4.5. Bivariate logistic regression model for change in principal psychiatric diagnosis. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Predictors Constant Change in principal diagnosis Accuracy of prediction -2LL R 2N ! Seriousness of Index Offence Crude Adjusted B (SE) OR B (SE) OR [95% [95% CI] CI] 0.665 1.944* 0.948 2.580 (0.290)* (0.602) 1.070 2.914 0.686 1.986 (0.690) [0.754(0.732) [0.47311.271] 8.332] 71.2% 78.1% 87.6*** 0.053 74.207 0.240 57! Recidivism Crude Adjusted B (SE) OR B (SE) OR [95% [95% CI] CI] 1.124 3.077* -0.450 2.307 (0.319)* (0.607) -0.718 0.488 0.836 0.637 (0.557) [0.164(0.553) [0.1941.452] 2.093] 71.2% 72.6% 87.6*** 0.032 78.993 0.159 Table 4.6a. Multivariate logistic regression model with age as the only adjuster. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Adjusted OR B (SE) [95% CI] OR Crude B (SE) [95% CI] Adjusted OR B (SE) [95% CI] OR [95% CI] Constant 1.849 (0.873)* 6.351* 2.585 (1.025)* 13.267* -1.051 (0.801) 0.350 -1.324 (1.023) 0.266 Psychiatric onset -0.873 (0.635) 0.418 [0.1201.450] -0.861 (0.641) 0.423 [0.1201.487] 21.100 (9676.7 21) 1.457E9 [0-] 21.493 (9634.0 86) 2.160E9 [0-] Psychiatric history -0.556 (0.858) 0.573 -0.186 (.909) 0.831 [0.1404.932] 0.279 (0.806) 0.321 [0.2726.410] 0.169 (0.921) 1.185 [0.1957.209] Existence of criminal associate(s) 0.111 (0.681) 0.586 (0.896) 1.797 [0.31010.412] 1.656 (1.172) 5.239 [0.52752.121] 1.891 (1.261) 6.628 [0.56078.501] Child abuse history 0.538 (0.878) 1.117 0.439 [0.2944.247] (-0.726) 1.551 [0.3746.435] 0.787 (0.736) 2.197 [0.5199.304] 0.380 (0.843) 1.462 [0.2807.636] Signs of early deviance -0.271 (0.646) 0.762 0.698 (0.735) 0.497 [0.1182.101] 1.292 (0.648)* 3.642 [1.02212.972] * 1.913 (0.852)* 6.770 [1.27335.991] * Accuracy of prediction -2LL R 2N ! Recidivism [0.1073.080] 1.713 [0.3079.576] [0.2152.705] 75.0% 75.0% 67.2% 80.6% 76.5*** 0.067 71.016 0.114 84.82** 0.438 53.0*** 0.527 58! Table 4.6b. Multivariate logistic regression model with gender as the only adjuster. Pvalues less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Adjusted OR B (SE) [95% CI] OR Crude B (SE) [95% CI] Adjusted OR B (SE) [95% CI] OR [95% CI] Constant 1.849 (0.873)* 6.351* 1.874 (0.884)* 6.517* -1.051 (0.801) 0.350 -1.145 (0.822) 0.318 Psychiatric onset -0.873 (0.635) 0.418 [0.1201.450] -0.891 (0.637) 0.410 [0.1181.430] 21.100 (9676.7 21) 1.457E9 [0-] 21.076 (9634.2 00) 1.423E9 [0-] Psychiatric history -0.556 (0.858) 0.573 -0.554 (0.863) 0.575 [0.1063.120] 0.279 (0.806) 0.321 [0.2726.410] 0.457 (0.837) 1.579 [0.3068.145] Existence of criminal associate(s) 0.111 (0.681) 0.512 (0.881) 1.669 [0.2979.388] 1.656 (1.172) 5.239 [0.52752.121] 1.618 (1.170) 5.041 [0.50949.955] Child abuse history 0.538 (0.878) 0.194 (0.724) 1.214 [0.2945.015] 0.787 (0.736) 2.197 [0.5199.304] 1.050 (0.801) 2.857 [0.59513.725] Signs of early deviance -0.271 (0.646) 1.292 (0.648)* 3.642 [1.02212.972] * 1.297 (0.659)* 3.660 [1.00713.304] Accuracy of prediction -2LL R 2N ! Recidivism [0.1073.080] 1.713 [0.3079.576] 1.117 [0.2944.247] 0.762 [0.2152.705] -0.280 (0.650) 0.756 [0.2122.700] 75.0% 75.0% 67.2% 76.1% 76.5*** 0.067 74.207 0.070 84.82** 0.438 58.4*** 0.454 59! Table 4.7a. Multivariate logistic regression model including changes in principal diagnosis as a predictor. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Adjusted OR B (SE) [95% CI] OR Crude B (SE) [95% CI] Adjusted OR B (SE) [95% CI] OR [95% CI] Constant 1.728 (1.270) 5.631 1.498 (1.420) 4.475 -0.627 (1.205) 0.534 -1.753 (1.492) 0.173 Psychiatric onset -0.671 (0.669) 0.511 [0.1381.899] -0.366 (0.786) 0.694 [0.1493.234] 20.939 (9939.3 04) 1.240E9 [0-] 21.256 (9407.5 93) 1.703E9 [0-] Psychiatric history -0.779 (1.195) 0.459 [0.0444.772] -0.291 (1.251) 0.747 [0.0648.668] 0.223 (1.126) 1.250 0.503 (1.233) 1.653 [0.14818.515] Existence of criminal associate(s) 0.500 (0.900) 1.648 [0.2829.626] 0.356 (1.110) 1.428 [0.16212.582] 1.410 (1.173) 4.096 [0.41140.847] 1.770 (1.392) 5.869 [0.38389.905] Child abuse history -0.005 (0.698) 0.995 [0.2533.909] 0.667 (1.001) 1.948 [0.27413.858] 0.896 (0.797) 2.450 [0.51411.679] 1.453 (1.056) 3.262 [0.41225.827] Signs of early deviance -0.133 (0.702) 0.875 [0.2213.462] -0.107 (0.865) 0.899 [0.1654.895] 0.725 (0.712) 2.064 [0.5128.327] 1.182 (0.970) 4.274 Change in principal diagnosis 0.951 (0.869) 2.589 [0.47114.229] 0.327 (0.974) 1.386 [0.2059.360] 0.103 (0.736) 1.109 [0.2624.695] 0.604 (0.951) Accuracy of prediction -2LL R 2N ! Recidivism 73.3% 69.6*** 0.098 73.3% 65.171 0.103 60! [0.13811.355] [0.63828.623] 1.830 [0.28411.796] 71.2% 83.1% n/a 0.376 42.59** 0.544 Table 4.7b. Multivariate logistic regression model including changes in principal diagnosis as a predictor and with age as the only adjuster. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Adjusted OR B (SE) [95% CI] OR Crude B (SE) [95% CI] Adjusted OR B (SE) [95% CI] OR [95% CI] Constant 1.728 (1.270) 5.631 2.367 (1.418) 10.670 -0.627 (1.205) 0.534 -1.334 (1.309) 0.263 Psychiatric onset -0.671 (0.669) 0.511 -0.689 (0.678) 0.502 [0.1331.897] 20.939 (9939.3 04) 1.240E9 [0-] 20.902 (9935.8 76) 1.196E9 [0-] Psychiatric history -0.779 (1.195) -0.531 (1.222) 0.588 0.223 (1.126) 1.250 -0.074 (1.136) 0.929 Existence of criminal associate(s) 0.500 (0.900) Child abuse history -0.005 (0.698) Signs of early deviance -0.133 (-0.702) Change in principal diagnosis 0.951 (0.869) Accuracy of prediction -2LL R 2N ! Recidivism [0.1381.899] 0.459 [0.0444.772] 1.648 [0.2829.626] 0.995 [0.2533.909] 0.875 [0.2213.462] 2.589 [0.47114.229] 73.3% 69.6*** 0.098 [0.0546.445] [0.13811.355] [0.1008.605] 0.543 (0.912) 1.721 [0.28810.274] 1.410 (1.173) 4.096 [0.41140.847] 1.491 (1.201) 4.443 [0.42246.741] 0.240 (0.746) 1.271 [0.2955.489] 0.896 (0.797) 2.450 [0.51411.679] 0.607 (0.836) 1.835 [0.3569.449] -0.485 (0.800) 0.616 0.725 (0.712) 2.064 [0.5128.327] 1.060 (0.773) 2.886 0.103 (0.736) 1.109 [0.2624.695] 0.152 (0.743) 0.862 (0.885) [0.1282.952] 2.369 [0.41813.418] [0.63413.137] 1.165 [-.2724.993] 70.0% 71.2% 72.9% 64.284 0.123 70.86** 0.376 51.351* 0.403 61! Table 4.7c. Multivariate logistic regression model including changes in principal diagnosis as a predictor and with gender as the only adjuster. P-values less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Adjusted OR B (SE) [95% CI] OR Crude B (SE) [95% CI] Adjusted OR B (SE) [95% CI] OR [95% CI] Constant 1.728 (1.270) 5.631 1.811 (1.309) 6.116 -0.627 (1.205) 0.534 -0.846 (1.256) 0.429 Psychiatric onset -0.671 (0.669) 0.511 -0.695 (0.674) 0.499 [0.1331.869] 20.939 (9939.3 04) 1.240E9 [0-] 20.965 (9861.5 83) 1.274E9 [0-] Psychiatric history -0.779 (1.195) -0.859 (1.229) 0.424 [0.0384.708] 0.223 (1.126) 1.250 0.467 (1.170) 1.594 Existence of criminal associate(s) 0.500 (0.900) 0.475 (0.904) 1.608 [0.2739.461] 1.410 (1.173) 4.096 [0.41140.847] 1.353 (1.173) 3.870 [0.38838.561] Child abuse history -0.005 (0.698) 0.124 (0.752) 1.132 [0.2594.948] 0.896 (0.797) 2.450 [0.51411.679] 1.293 (0.914) 3.645 [0.60821.864] Signs of early deviance -0.133 (-0.702) 0.875 [0.2213.462] -0.119 (0.708) 0.888 [0.2223.556] 0.725 (0.712) 2.064 [0.5128.327] 0.736 (0.731) 2.087 Change in principal diagnosis 0.951 (0.869) 2.589 1.034 (0.897) 2.811 [0.48516.301] 0.103 (0.736) 1.109 [0.2624.695] 0.342 (0.788) Accuracy of prediction -2LL R 2N ! Recidivism [0.1381.899] 0.459 [0.0444.772] 1.648 [0.2829.626] 0.995 [0.2533.909] [0.47114.229] [0.13811.355] [0.16115.792] [0.4988.752] 1.408 [0.3016.597] 73.3% 78.1% 71.2% 72.9% 69.6*** 0.098 52.88** 0.225 70.86** 0.376 51.35** 0.403 62! Table 4.8. Multivariate logistic regression model using psychopathological and criminal predictors with direct association only. Models were adjusted for only age or gender. Pvalues less than 0.05, 0.01 and 0.005 were denoted with *, ** and ***, respectively. Seriousness of Index Offence Crude Predictors B (SE) Age adjusted OR B (SE) [95% CI] [95% CI] Constant 0.745 (0.336)* 2.107* 1.014 (0.616) 2.755 Psychiatric onset -0.294 (0.602) 0.745 [0.2292.423] -0.129 (0.712) 0.879 [0.2173.550] Change in principal diagnosis 1.037 (0.694) 2.822 [0.72510.990] 0.759 (0.725) 2.135 [0.5158.846] Accuracy of prediction -2LL R 2N ! OR 71.2% 78.1% 87.6*** 0.053 74.207 0.240 63! ! Chapter 5 Discussions 5.1 Population Characteristics The study population shared many characteristics with the general offender population as well as the NCRMD population examined in earlier studies25-28,32-46,48-54,141146 . Subjects were primarily young males who came from low socioeconomic background and were most commonly diagnosed with schizophrenia. In most cases, first official psychiatric diagnosis predated the first criminal charge, indicating that psychiatric development preceded criminal career. Also, patients more often had a psychiatric history than a criminal history. Finally, patients more often had psychiatric associates than criminal associates. As expected, almost all subjects were ordered to full-time detention at forensic psychiatric hospitals for treatment (93%). Only 2.4% of the subjects were treated in a community setting and none received an absolute discharge. This finding appears appropriate given that offenders do not successfully reach ORB unless they are severely ill, and the severity of illness warrants hospitalization. The rate at which full-time detention versus conditional discharge is given had varied from study to study, with findings of this study on the conservative side. Two studies in British Columbia found 25% and 37.5% of NCRMD offenders conditionally discharged, respectively146,25. However, most studies generally agree that a NCRMD verdict most frequently results in detention at a hospital, followed by a conditional discharge and almost never results in absolute discharge. ! 64! ! 5.1.1 Criminal Responsibility Out of 347 offenders who were psychiatrically evaluated, 85 were found NCRMD and were included as the subjects of this study (24.5%). This rate appears higher than observations from other countries. In the United States, the national success rate was less than 20% among those who attempted to plead not guilty by reason of insanity and less than 1% of all criminal cases within the same time period147. In a recent study on a Chilean offender population, only 4.3% of all psychiatrically evaluated offenders in two large forensic psychiatric hospitals were found not criminally responsible148. Currently there is no data available to estimate the national success rate of NCRMD in Canada. Justice Canada recently presented that 0.12% of all criminal cases was referred to a review board jurisdiction across Canada, but this report did not break down the success rates of NCRMD among the overall offender population, among psychiatrically evaluated offenders or among those that specifically raised NCRMD defense46. More insight was provided in regionally reported rates of attempt and success of NCRMD defense. A study in British Columbia reported 53 successful NCRMD defenses among 185 offenders who raised the question of criminal responsibility and among all 620 psychiatrically evaluated offenders over a two-year period (28.6% and 8.5%)19. Based on these numbers, the success rate of NCRMD seen in this study population appears higher than the reported average. This seemingly high success rate, however, must be taken with caution. The charts used to gather this study’s data did not clarify the purpose of the court ordered psychiatric evaluations. It was not feasible to track this down because it required reviewing restricted material such as court transcripts. The student researcher therefore ! 65! ! collectively treated them as psychiatric evaluations without further differentiating them by the purpose of evaluation. It is possible that all charts were of those who were specifically sent for criminal responsibility evaluations, but it is equally possible that some of them were sent for fitness to stand trial evaluations. In the latter case, the success rate would be reduced substantially as the denominator would surely have been higher in actuality. Also, the aforementioned rate of 24.5% would be the success rate of NCRMD among those who specifically tried this defense, in which case it would be more in line with findings of other Canadian studies. 5.1.2 Nature of Crimes Most frequently committed crimes were assault (48.2%) followed by homicide (16.5%). Sexual assault, offence against property and other minor crimes such as possession of drugs were less common (7.1%, 10.6% and 16.5%, respectively). Altogether violent crimes comprised 71.8% of all index offences. This observation is dissimilar to the trend seen in the general Canadian offender population. Despite the constant increase in the violent crime rate in Canada from 1960’s until 1992, violent crimes have been a smaller proportion of crimes than the other two categories, namely property and other criminal code crimes. For instance, violent crimes accounted for 11% and 20.5% of total crimes committed in Canada in 1996 and 2009 respectively8, neither of which was comparable to 71.8% observed in the study population. This outcome, while unusually high compared to the general offender population, is in keeping with findings of previous studies of psychiatric offenders where a greater tendency to commit ! 66! ! violent crimes was observed19,149,150. Nationally, violent crimes accounted for 80.8% of all crimes committed by NCRMD offenders46. There are several possible explanations for this irregularity. Firstly, because all ORB patients were referred from the court, an offender with a less serious charge, who was less likely to reach the sentencing stage, would have been less likely to be transferred from court to an ORB jurisdiction. Also, while violent crimes were the most common index offence among the study population, they were not the most commonly committed crimes by these people on a broader scale. Typically, severely ill psychiatric offenders have a long and complex criminal history consisting of mostly petty crimes. It has been postulated that psychiatric offenders are not any more violent in nature than the general offender population but are simply less able to act according to social standards due to the mental illness121,122. Severe psychiatric conditions often render those affected disoriented and disorganized, predisposing them to a wide array of violations18,121,122. A large number of such violations, such as breaking and entering, go unreported because there are many points of diversion throughout the forensic system for the mentally ill133,151. However, violent crimes such as attacking a stranger during a psychotic episode, which comprise a smaller fraction of such violations, would bring the offender into the criminal justice system. In this light, it is owing to the selective nature of diversion and less to the violent intent of the participants that there is a higher than normal incidence of violent crime in this population. ! 67! ! 5.2 Findings and Interpretations from Bivariate Analysis 5.2.1 Distribution of sociodemographic factors Chi square tests showed that most sociodemographic factors and all predictor variables were similarly distributed between the subgroups by seriousness of index offence. A difference was found in the age, with the minor crime group being generally older than the serious crime group. The majority of serious crimes were committed by subjects between the ages of 18 and 25 years old, and then 26 and 30 years old, while less than one fourth of the violent crimes was committed by subjects who were 46 years old or older. This is in agreement with the previous discovery that violent crimes are committed most frequently by young adults around the age of 18 and this is therefore the violence-prone age120. First-time offenders and repeat offenders did not significantly differ in sociodemographic description. 5.2.2 Changes in Principal Psychiatric Diagnosis In a significant number of subjects, the primary diagnosis was seen to have changed between the initial psychiatric assessment and the ORB assessment (27%). This variable met the significance cut-off for the multivariate regression model when tested for bivariate association with recidivism, but it had been excluded due to high chance of human error. When individually examined, it did not significantly increase the chance of committing more serious crime or recidivating. Possibly, this may simply represent misdiagnosis at an earlier stage when clinical manifestations were less visible. The charts used for data capture did not always include detailed information on when, where and by whom the diagnoses were made, which ! 68! ! made it impossible to follow up on the discrepancies. Also, there was no discernible pattern in changes, which led to more doubt on the quality of information. For instance, if it were shown that diagnosis usually changed from a less severe or more general disorder to a more severe or specific one, like changing from general delusional disorder to schizophrenia, it would have increased the perceived integrity of the data as this is often the course of prognosis of psychiatric disorders152. Also, with a visible pattern, it might have been possible to see if a diagnosis changed between otherwise linked disorders. For instance, it was shown recently that a personality disorder quite accurately predicts remission of major depressive disorder, while another study showed that dysthymic disorder affects the prognosis of personality disorder153,154. However, the changes were random and the researchers were led to mainly attribute changes to the result of human error like misdiagnosis or mistake during information transfer. Another explanation may be found in the change of the diagnostic tools. It is likely that a vast majority of subjects had already been diagnosed using an earlier edition of the DSM before they were assessed by the ORB staff. Two new versions of DSM, DSM IV and DSM IV-TR, had been published since the 1992 inception of ORB, reflecting substantial changes in diagnostic criteria and disorder classifications. Different versions may have led psychiatrists to different diagnoses under the same circumstances. Alternatively, a change in the main diagnostic tool may have induced a substantial change in diagnostic trends and led psychiatrists to move toward or away from diagnosing some disorders155,156. Primary diagnoses may have changed also because more information was factored in during the diagnostic process. Results from VRAG, SORAG and PCL-R tests add ! 69! ! valuable information to psychological assessment of violent offenders, but these tools were unavailable until several years after the ORB was established139-141. Perhaps introduction of these tools added a new set of information that was previously unaccounted for. Finally, it has been suggested that criminal lifestyle influences psychiatric disorder development157,158. In light of this theory, it is possible that the diagnosis for some subjects changed accurately reflecting the change in their mental illnesses, which in turn had been caused by their criminal lifestyle. There has not been substantiating evidence for this theory, but it may partially account for the changed diagnoses in the current study population. 5.2.3 Timing of psychiatric onset with reference to the first crime Psychiatric onset was shown to have followed criminal career in some of the repeat offenders while it always predated criminal career in first-time offenders. Overall, it more frequently predated crime as it was observed in other studies of the psychiatric offender population145,147. This finding may support the abovementioned criminogenicity of mental illness, in which crime is viewed as a byproduct of psychiatric condition18,121. Also, some researchers have suggested mentally ill people often commit crime in order to enter the forensic system, where they could receive systemic psychiatric care that they cannot access in the community126. While this explanation nicely fits into the above criminogenicity of mental illness theory, it likely does not apply to this study population. Multiple studies have shown that crimes committed for these reasons tend to be fairly minor ones126 whereas the average crime committed by study subjects was serious. ! 70! ! Instead, these same researchers argue that serious crime offenders could have been driven to commit such crimes because their psychiatric conditions were not adequately cared for126. Timely mental health care may prevent advancement of illness and subsequently prevent advancement in criminal career. Alternatively, seriousness of initial crime may be attributed to the finding that psychiatric onset predated crime in most subjects. One may have committed one or more petty crimes before receiving a diagnosis, but were diverted without criminal charges. In this case, a later, more serious crime, which resulted in the first official criminal charge following the first psychiatric diagnosis, would falsely indicate that psychiatric onset before their criminal career. 5.2.4 Signs of early deviance Most repeat offenders had displayed deviant traits in childhood (73.1%) while most first-time offenders had not (37.5%). This finding is easily understandable given the role played by the early deviance variable in criminalization in adulthood157,159,160 and in prediction of recidivism, as shown in the results of this study. This issue will be revisited in a later section. 5.3 Findings and Interpretations from Multivariate Regression Analysis 5.3.1 Seriousness of Index Offence The regression model showed that the seriousness of the index offence was not predicted by any one or groups of predictor variables. There are several possibilities that may account for this lack of association. A probable reason is that this variable was ! 71! ! defined too loosely. In the current study, the category level ‘grave’ had been used to include all crimes that directly harmed persons from murder to sexual assault, while the level ‘petty’ had been used to include all other crimes, such as prostitution and theft. Perhaps grouping the subjects first by the crime type and then subdividing them according to seriousness would have yielded better results. When seriousness of crime was investigated only within the same crime type (violent assault), the predictive power of study variables was substantial161. A better categorization may also have been achieved by using more than two category levels or coding it into a continuous instead of categorical variable. Significant results were produced from a predictive model in an earlier study that tested seriousness of crime as an ordinal variable with multiple levels162. A vast number of factors were examined and a considerable number of them were included in the model. It is nevertheless possible that the selected predictors were insufficient to fully account for the complexity of crime severity. Seriousness of crime interacts with virtually all aspects of life, from psychopathology to developmental history163-164. There have been numerous attempts at building a model for prediction of seriousness of crime, but such efforts had been limited success at best166. The lack of predictive power was sometimes due to small sample size167 and due to insufficient predictor selection at other times168. Both of these explanations may be applied in explaining the present study’s results. The reduction of sample size after applying all exclusion criteria likely reduced the power, while statistical strengths were lessened from exclusion of some key factors like PCL-R scores. In summary, it is postulated that some associations existed but were not detected in the study. There is a clear indication that a wider range of information must be ! 72! ! collected from a greater number of subjects in order to more accurately investigate this variable. 5.3.2 Recidivism and Early Deviance Recidivism was significantly predicted by deviant behaviours in childhood. Strong correlations have been reported between deviant traits displayed early in life and criminality in adulthood159,160,169-171. Consequently, it was anticipated that it would also be associated with recidivism and analysis results confirmed this theory. Recidivism is an important criminological research topic with academic and practical implications. The detection of early deviance as a significant and substantial predictor of recidivism, therefore, is an important finding. While there is an abundance of explanations for how early conduct problems lead to criminality in adulthood, the exact role it plays in recidivism is less clear because the specific association between deviance in childhood and recidivism in adulthood has been less frequently studied172-174. It may be theorized that childhood truancy, which hinders the opportunity to learn and conform to rules, naturally progresses into psychiatric disorders as well as criminality in adulthood170,171,174-176, and this long history of deviance in turn develops into recidivism. 5.4 Other Contributing Factors Some predictors were analyzed but did not yield statistically significant results. Given the evidence found within past literature on their importance in the understanding ! 73! ! of psychiatric offenders, three variables – existence of criminal associates, existence of psychiatric associates and a history of child abuse – are discussed. Prior research found that people with criminal family members or romantic partners are more likely to commit crime. For instance, a child growing up with exposures to unlawful conducts of a criminal family member may learn to mimic the observed criminal behaviours177. Alternatively, antisocial behaviours of a female romantic partner have been shown to increase the likelihood to commit a criminal offence in young adult males178. Finally, people who are surrounded by criminal individuals may get accustomed to the dissipated environment, accept an antisocial lifestyle as the norm and continue down the transgressive path176. The present study partially supported this last view with a substantial odds ratio with some significance: subjects with criminal associates were seven times more likely to recidivate (OR=7.577 [0.541-106.2]; p=0.13). Many psychiatric disorders are hereditary and hence a family history increases the risk of mental illness179,180. Also, a psychiatric family member or a romantic partner is a significant stressor that may trigger psychiatric development in a previously undiagnosed individual6. In a study on burdens of schizophrenia on the caregiving family, half of the subjects were diagnosed with a minimum of one and a maximum of four clinical mental disorders181. Close associates of bipolar disorder patients were more likely to suffer from similarly varying moods and high level of distress180. Mental illness also fosters intrarelationship conflicts and thus, increased relationship stress from psychiatric partner often deteriorates a patient’s condition, predisposing the patient to an array of violations and thereby increasing the likelihood to commit criminal violation181,182,121,122,. Thus increased chance of criminal offending qualifies the legal inclusion criterion while ! 74! ! increased chance of severe mental illness qualifies the psychiatric inclusion criterion of ORB. Child abuse is another crucial developmental factor in the study of mentally ill offenders. It is often viewed as an outcome of having a criminal or psychiatric family member123,124. This pattern was replicated in this study population as well, where 75% of subjects with a history of child abuse had psychiatric associates or criminal associates or both. In turn, criminal or psychiatric associations as well as child abuse could simply be the symptoms, rather than the causes, of unconventional family environment. Difficult family situations were repeatedly shown as contributing factors of criminality and mental illness123,124. They are also indicative of low socioeconomic status, which again has negative influence on mental health125. As such, insight into developmental and current domestic environment through examination of these variables were a vital element of this study. 5.5 Relevance of Findings This is one of the earliest attempts to systematically examine the review board offender population in Canada and the first to describe the southeastern ORB population in detail studies25-28,32-46,48-54,141-146. Traditionally, psychiatric offenders have been studied as individuals in legal cases25-28 or with reference to law or policy32-46, rather than as subjects of a scientific study. Such scarcity in population-based studies has led to a lack of evidence on mentally ill offenders. Subsequently the standards of evidence-based medicine, in which development and reform of medical care are based on scientific evidence183, have not been rigorously adhered to in management of these people. This ! 75! ! study presented a wealth of psychopathological, criminal and developmental information on mentally ill offenders, which will add to the growing body of information on Canadian forensic offenders. It is hoped that this study will stimulate further research into the forensic offender population. The growth of scientific evidence will also be the platform on which care for forensic offenders will become a better practice of evidence-based medicine. In most review board population studies, data collection only captured the population characteristics within five years of the new system. This includes the adjustment period necessary for the courts to adapt to a new legislation and accuracy of information may be influenced by this adjustment period. The current study captures 15 years of data and findings would be relatively free of confounding influence of adjustment period. It may also be broken down into time blocks to investigate periodical trends. This study also provides a current understanding of the ORB system. There has not been a systematic data collection conducted on provincial review board systems since their inception in 1992. There are two published papers that examined the Lieutenant Governor’s advisory board system in detail, but these were published over 20 years ago, set in a single province and before the review boards were in their current format29,47. Furthermore, one of these studies placed main focus on administrative aspects of the system with little reference to the patient population itself29. As such, information presented in these papers are outdated and do not accurately reflect the elements of the current system, such as specific legislations and patient demographics. Because the ORB was never systematically reviewed since its inception in 1992, understanding of the ! 76! ! effectiveness and efficiency of the system has been largely lacking. Current characterization and identification of risk factors of a key participant group as presented in this study will be helpful not only to service providers of ORB but also in the evaluative efforts of the review board system. Successful evaluation of the review board system may be followed by outcome evaluation of the mental health care reform. Information generated during this review process would also provide the context in which results of this study are better interpreted. On a related note, currently there is no single appraisal method used by physicians across the board to reach clinical decision about criminal responsibility and dangerousness184. It is possible that this resulted in the psychiatrists and subsequently the review boards of different regions to adopt different assessment styles. Such differences range from using different clinical and judicial tools to differential weights given to components of the assessment184. While it is reasonable for the physicians and board to account for the changes that they see through time or to reflect local offender characteristics, it may also lead to different dispositions given by the board based on jurisdictions rather than objective information, as was pointed out in the American criminal courts147. The previously illustrated discrepancies between provinces in legal outcomes, like the rates at which offenders are detained or discharged and NCRMD success rate, may be indicative of this problem19,25,52,106,107,110,145. Clearly, there is a need for a uniformly practiced appraisal method. This study has captured much of the psychiatrist’s decision-making process and proceedings of the mental health courts. It is hoped that this information will contribute to a greater uniformity in management of the population. ! 77! ! Finally, data source was of high reliability since they were collected, examined and re-verified by multiple staff. Also, a new report is drafted every year for each patient for the annual ORB hearing, providing an additional opportunity to supply insufficiencies and correct misinformation. Moreover, the case files included a wealth of information. Much of this information, such as the relationship of the offender to the victim, was not included in this study. All available information was still captured and electronically converted. It is anticipated that these information will be useful in future studies on this population. 5.6 Limitations of Study This study focused on a localized population and hence the applicability of the results is limited. Findings should not be directly applied to the general Canadian population nor is it appropriate for the general offender population. The highly selective nature of the study sample had limited the applicability of findings to similarly unique population, which is a predominantly male mentally ill offender population in North America whose charges are grave enough to bring them to the court sentencing and are suffering from severe psychiatric conditions as to render them not criminally responsible on the account of the condition or unfit to stand trial. After applying all exclusion criteria, the sample size had been reduced from 347 to 84. This led to reduction in power and statistical significance in analysis. As a result, the data collected in this study had been insufficient to appropriately test the first outcome variable. While adjusting for age and primary psychiatric diagnosis contributed ! 78! ! to the significance and power of the associations seen in the recidivism variable, it did not aid to the results from the seriousness of index offence variable. The deficiency in population-based knowledge on the Canadian forensic population added to novelty of this study, but it also presented the researchers with many challenges, one of which was encountered during variable selection. Literature was reviewed and veteran ORB psychiatrists were consulted, but there is a chance that the selected variables did not sufficiently cover the key study areas. Previous studies were dissimilar in scope, target population or primary aims from this study25-28,32-46,48-54,141-146, but they were still referenced for variable selection because no other resources existed. Several psychological, criminal, developmental or legal factors that were initially selected as potential predictors were excluded in the main analysis due to confidentiality issues (e.g. victimology) or missing data (e.g. scores from PCL-R, VRAG and SORAG tests). Inclusion of this information would have expanded the scope of this study and strengthen the results. For instance, seriousness of crime, as well as recidivism, have been shown to be closely related to PCL-R scores in offender population161,185 and to severity of illness in psychiatric offenders186,187. The regression model may have shown a greater predictive power for seriousness of index offence if these variables were included. Also, victim selection as well as the mode of crime, such as time of the day and crime site, could provide valuable insight into the study of psychology of the offender188. It was communicated to the student investigator that it might be premature to posit hypotheses in an area where there exists so little prior research. However, hypothesis testing was only a part of this project. The thesis was designed to first describe the study population and explore data before testing specific hypotheses. Additional ! 79! ! safeguarding against false theory was employed by formulating hypotheses based on sound and related, if not directly transferable, previous research. 5.7 Future Direction A crude exploratory analysis that encompasses variables that were excluded in this study would be useful in detecting unexpected associations. For instance, victimology was uninvestigated due to confidentiality issues, but this is an important psycho-legal factor1 that must be examined in detail in the future. Scores of specialized tests, such as the PCL-R and the VRAG, should also be included for the study of the next generation of the ORB population. These analyses may be extended beyond the study population to all psychiatric offenders in order to provide valuable insight on mentally ill offenders as a whole. Following an exploratory study, predictive models may be run for the same outcome variables, seriousness of crime and recidivism, using newly identified associations as predictor variables. This would be bettered by redefining the seriousness of crime variable, either by assigning it into specific crime types or dividing it into multiple categories. Results of these analyses may serve as measures with which the predictive model of this study is evaluated. Finally, findings will be useful together with and in comparison to results of similar studies that cover other regional samples of the ORB offender population. These data could provide current consensus of Canadian forensic population while highlighting specific attributes of regional samples. ! 80! ! 5.8 Conclusion This thesis systematically reviewed Ontario forensic offenders who were exempted from criminal responsibility on the account of mental disorder. Population characteristics were described and key variables were studied with regard to their effects on seriousness of index offence and recidivism. It was found that early signs of deviance positively influenced the likelihood to recidivate. Findings of this study may have clinical, legal and policy implications in the management of the forensic population in Canada. ! 81! References 1. World Health Organization. (2010). Mental health: strengthening our response. Retrieved October, 2010 from World Health Organization Media Centre website: http://www.who.int/mediacentre/factsheets/fs220/en/index.html 2. Statistics Canada. (2003). Canadian Community Health Survey: Mental health and well-being. Statistics Canada – Catalogue no. 82-617-XIE, 28(7). Retrieved March, 2009 from Statistics Canada website: http://www.statcan.gc.ca/pub/82-617x/4067678-eng.htm 3. Jacobs, P., Yim, R., Ohinmaa, A., Eng, K., Dewa, C. S., Bland, R., et al. (2008). Expenditures on Mental Health and Addictions for Canadian Provinces in 2003/04. Canadian journal of Psychiatry, 53(5), 306-313. 4. Lim, K. L., Jacobs, P., Ohinmaa, A., Schopflocher, D. & Dewa, C. S. (2008). A New Population-Based Measure of the Economic Burden of Mental Illness in Canada. Chronic Diseases in Canada, 28(3), 92-98. 5. Link, B. & Phelan, J. C. (2001). Conceptualizing stigma. Annual Review of Sociology, 27, 363-385. 6. Sartorius, N., Gaebel, W., Cleveland, H., Stuart, H., Akiyama, T., Arboleda-Flórez, J., Baumann A. E., Gureje, O. et al. (2010). World Psychiatric Association guidance on how to combat stigmatization of psychiatry and psychiatrists. World Psychiatry, 9(3), 131–144. 7. Arboleda-Flórez, J. (2003). Integration initiatives for forensic services. World Psychiatry, 2(3), 179-183. 82 8. Dauvergne, M. (2007). Crime Statistics in Canada. Statistics Canada – Catalogue no. 85-002-X, 28(7). 9. Thornicroft, G. & Bebbington, P. (1989). Deinstitutionalisation – from hospital closure to service development. British Journal of Psychiatry, 155, 739-753. 10. Jones, K. & Poletti, A. (1986). The Italian experience in mental health care. Hospital Community Psychiatry, 37(8), 795-802. 11. Lamb, H. R. (1993). Lessons learned from deinstitutionalisation in the US. British Journal of Psychiatry, 162, 587-92. 12. Lamb, H. R. & Weinberger, L. E. (2005). The shift of psychiatric inpatient care from hospitals to jails and prisons. Journal of American Academy of Psychiatry and Law, 33(4), 529-534. 13. Birmingham, L. (1999). Between prison and the community. The 'revolving door psychiatric patient' of the nineties. British Journal of Psychiatry, 174, 378-379. 14. Abramson, M. F. (1972). The criminalization of mentally disordered behavior: possible side-effect of a new mental health law. Hospital Community Psychiatry, 23(4), 101-105. 15. Lamb, H. R., Weinberger, L. E. & Gross, B, H. (2004). Mentally ill persons in the criminal justice system: some perspectives. Psychiatric Quarterly, 75(2), 107-26. 16. Buchanan A, Zonana H. (2009). Mental disorder as the cause of a crime. International Journal of Law and Psychiatry, 32(3), 142-146. 17. Anckarsäter, H., Radovic, S., Svennerlind, C., Höglund, P. & Radovic, F. (2009). Mental disorder is a cause of crime: the cornerstone of forensic psychiatry. International Journal of Law and Psychiatry, 32(6), 342-347. 83 18. Link, B. G. & Stueve, A. (1995). Evidence bearing on mental illness as a possible cause of violent behavior. Epidemiological Reviews, 17(1), 172-181. 19. Roesch, R., Ogloff, J. R. P., Hart, S. D., Dempster, R. J., Zapf, P. A. & Whittemore, K. E. (1997). The impact of Canadian Criminal Code changes on remands and assessments of fitness to stand trial and criminal responsibility in British Columbia. Canadian Journal of Psychiatry, 42, 509–514. 20. Arboleda-Flórez, J. (1991). Ethical issues regarding research on prisoners. International Journal of Offender Therapy and Comparative Criminology, 35(1), 1-5. 21. Young, M. H., Justice, J. V. & Erdberg, P. (2004). Assault in prison and assault in prison psychiatric treatment. Journal of Forensic Sciences, 49(1), 141-149. 22. National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1978). The Belmont report: ethical principles and guidelines for the protection of human subjects of research (DHEW Publication). Washington, DC: U.S. Government Printing Office. 23. Cashman, F. E. (1988). A paradoxical outcome of a regional review board decision: a case report. Canadian Journal of Psychiatry, 33(8), 778. 24. Canada Ontario Court of Appeal. (1991). Fleming v. Reid. Dominion Law Reports, 82, 298-320. 25. Grant, I. (1997). Canada's new mental disorder disposition provisions: a case study of the British Columbia Criminal Code Review Board. International Journal of Law and Psychiatry, 20(4), 419-443. 84 26. Glancy, G. D. & Bradford, J. M. (1999). Canadian landmark case: Regina v. Swain: translating M'Naughton into twentieth century Canadian. Journal of American Academy of Psychiatry and Law, 27(2), 301-307. 27. Balachandra, K., Swaminath, S. & Litman, L. C. (2004). Impact of Winko on absolute discharges. Journal of American Academy of Psychiatry and Law, 32(2), 173-177. 28. Gray, J. E. & O'Reilly, R. L. (2009). Supreme court of Canada's "Beautiful Mind" case. International Journal of Law and Psychiatry, 32(5), 315-322. 29. Kaiser, H. A. (1986). Involuntary psychiatry in Nova Scotia: The Review Board proposals (1979-1983) and recent proposals for legislative change. Dalhousie Law Review, 10, 211-224. 30. Manchanda, R., Fernando, M. L. & Galbraith, D. A. (1989). Consent to treat: clinical and legal implications. Canadian Journal of Psychiatry, 34(1), 60-62. 31. Draper, R. J. & Dawson, D. (1990). Competence to consent to treatment: a guide for the psychiatrist. Canadian Journal of Psychiatry, 35(4), 285-289. 32. Bay, M. (1991). Treatment and the Mental Health Act: a review board perspective. Health Law in Canada, 12(1), 11-14. 33. Bay, M. (1993). Implementing competency legislation for health care. Health Law in Canada, 14(2), 35-39. 34. Davis, S. (1993). Changes to the Criminal Code provisions for mentally disordered offenders and their implications for Canadian psychiatry. Canadian Journal of Psychiatry, 38(2), 122-126. 85 35. Swaminath, R. S., Norris, P. D., Komer, W. J. & Sidhu, G. (1993). A review of the amendments to the Criminal Code of Canada (Mental Disorder). Canadian Journal of Psychiatry, 38(8), 567-570. 36. Davis, S. (1994). Fitness to stand trial in Canada in light of the recent Criminal Code amendments. International Journal of Law and Psychiatry, 17(3), 319-329. 37. Greenberg, D. M. (1994). Capacity to stand trial: pitfalls of legal interpretation. Canadian Journal of Psychiatry, 39(2), 113-115. 38. Laberge, D., Morin, D. & Robert, M. (1995). Penal aspects of psychiatric intervention: management modes and impact of new criminal legislation. Canadian Journal of Community Mental Health, 14(1), 103-122. 39. Hewak, N. (1995). The ethical, medical, and legal implications of the forcible treatment provisions of the Criminal Code. Health Law in Canada, 15(4), 107-116. 40. Arboleda-Flórez, J., Crisanti, A. & Holley, H. L. (1995). The effects of changes in the law concerning mentally disordered offenders: the Alberta experience with Bill C-30. Canadian Journal of Psychiatry, 40(5), 225-233. 41. Zapf, P. A. & Roesch, R. (1992). Fitness to stand trial: characteristics of remands since the 1992 Criminal Code amendments. Canadian Journal of Psychiatry, 43(3), 287-293. 42. O'Reilly, R. L., Komer, W. & Dunbar, S. (1999). Why are patients discharged by review boards? Canadian Journal of Psychiatry, 44(3), 259-263. 43. Crisanti, A. S., Arboleda-Flórez, J. & Stuart, H. (2000). The Canadian Criminal Code provisions for mentally disordered offenders: a survey of experiences, attitudes, and knowledge. Canadian Journal of Psychiatry, 45(9), 816-821. 86 44. Thérien, E. (2003). Criminal code sanctions. Canadian Journal of Public Health, 94(2), 154. 45. Livingston, J. D. (2006). A statistical survey of Canadian forensic mental health inpatient programs. Healthcare Quarterly, 9(2), 56-61. 46. Latimer, J. & Lawrence, A. (2006). The Review Board Systems in Canada: Overview of Results from the Mentally Disordered Accused Data Collection Study. Ottawa, ON: Research and Statistics Division, Department of Justice Canada. 47. Phillips, M. S., Landau, T., Sepejak, D. & Osborne, C. (1985). The Advisory Review Board and characteristics of patients on warrants of the Lieutenant Governor in Ontario. Bulletin of American Academy of Psychiatry and Law, 13(4), 373-384. 48. Adams, S. J., Pitre, N. L. & Cieszkowski, R. (1997). Who applies to regional review boards and what are the outcomes? Canadian Journal of Psychiatry, 42(1), 70-76. 49. Robertson, R. G., Gupton, T., McCabe, S. B. & Bankier, R. G. (1997). Clinical and demographic variables related to "fitness to stand trial" assessments in Manitoba. Canadian Journal of Psychiatry, 42(2), 191-195. 50. Komer, W. J., O'Reilly, R. L., Cernovsky, Z. & Dunbar, S. (1999). Review board outcomes for involuntary patients in provincial psychiatric hospitals. Canadian Journal of Psychiatry, 44(5), 495-498. 51. Kelly, M., Dunbar, S., Gray, J. E., O'Reilly, R. L. (2002). Treatment delays for involuntary psychiatric patients associated with reviews of treatment capacity. Canadian Journal of Psychiatry, 47(2), 181-185. 87 52. Livingston, J. D., Wilson, D., Tien, G. & Bond, L. (2003). A follow-up study of persons found not criminally responsible on account of mental disorder in British Columbia. Canadian Journal of Psychiatry, 48(6), 408-415. 53. Wang, X., Livingston, J. D., Brink, J. & Murphy, E. (2006). Persons found 'not criminally responsible on account of mental disorder': a comparison of British Columbia, Canada and Hunan, China. Forensic Science International, 164(2-3), 9397. 54. Crocker, A. G. & Côté, G. (2009). Evolving systems of care: Individuals found not criminally responsible on account of mental disorder in custody of civil and forensic psychiatric services. European Psychiatry: Journal of Association of European Psychiatrists, 24(6), 356-364. 55. Statistics Canada. (2011). Quarterly Demographic Estimates October to December 2010. Statistics Canada – Catalogue no. 91-002-X. Retrieved March, 2011 from the Statistics Canada website http://www.statcan.gc.ca/pub/91-002-x/91-002-x2010004eng.htm. 56. Assessment, Classification, and Epidemiology Group of World Health Organization. (2009). The World Mental Health Survey Initiative. Retrieved November, 2010 from World Mental Health website: http://www.hcp.med.harvard.edu/wmh/index.php. 57. Government of Canada. (2006). The Human Face of Mental Health and Mental Illness in Canada, 2006. Ottawa, ON: Minister of Public Works and Government Services Canada. 88 58. World Health Organization. (n.d). Mental Health. Retrieved June, 2010 from World Health Organization Media Centre website: http://www.who.int/topics/mental_health/en/ 59. Stengel, E. (1959). Classification of mental disorders. Bulletin of the World Health Organization, 21, 601-663. 60. Kessler, R. C. (2000). Psychiatric epidemiology: selected recent advances and future directions. Bulletin of the World Health Organization, 78(4), 464-474. 61. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author. 62. Lempérière, T. (1995). The importance of classifications in psychiatry. L’Encéphale, 21(Special 5), 3-7. 63. Boyle, M. (2007). The problem with diagnosis. The Psychologists, 20(5), 290-292. 64. World Health Organization. (1992). ICD-10 Classifications of mental and behavioural disorder: clinical descriptions and diagnostic guidelines. Geneva, Swiss: World Health Organization. 65. Frances, A. & First, M. B. (2011). Hebephilia Is Not a Mental Disorder in DSM-IVTR and Should Not Become One in DSM-5. Journal of the American Academy of Psychiatry and the Law, 39(1), 78-85. 66. Duggan, C. (2008). Why are programmes for offenders with personality disorder not informed by the relevant scientific findings? Philosophical transactions of the Royal Society of London. Series B, Biological sciences, 363(1503), 2599-2612. 67. Kessler, R., Angermeyer, M., Anthony, J. C., de Graaf, R., Demyttenaere, K., Gasquet, I., et al. (2007). Lifetime prevalence and age-of-onset distributions of 89 mental disorders in the World Health Organization's World Mental Health Survey Initiative. World Psychiatry, 6(3): 168–176. 68. Sroujian C. (2003). Mental health is the number one cause of disability in Canada. Insurance Journal, August, 8. 69. Murray, R. M., Jones, P. B., Susser, E., van Os, J. & Cannon, M. (2002). The epidemiology of schizophrenia. Cambridge: Cambridge University Press. 70. Saha, S., Chant, D., Welham, J. & McGrath, J. (2005). A systematic review of the prevalence of schizophrenia. Public Library of Science Medicine, 2(5), e141. 71. Allgulander, C. (1989). Psychoactive drug use in a general population sample, Sweden: correlates with perceived health, psychiatric diagnoses, and mortality in an automated record-linkage study. American Journal of Public Health and the Nation’s Health, 79, 1006–1010. 72. Cannell, C. F., Marquis, K. H. & Laurent, A. (1977). A summary of studies of interviewing methodology: 1959-1970. Vital and Health Statistics, 2, 69. 73. Jenkins, J. H. & Karno, M. (1992). The meaning of expressed emotion: theoretical issues raised by cross-cultural research. American Journal of Psychiatry, 149, 9-21. 74. Bolton, P. (2001). Cross-cultural validity and reliability testing of a standard psychiatric assessment instrument without a gold standard. Journal of Nervous and Mental Disease, 189(4), 238–242. 75. Flaskerud, J. H. (2000). Ethnicity, culture, and neuropsychiatry. Issues in Mental Health Nursing, 21(1), 5-29. 76. Alarcón, R. D. (2009). Culture, cultural factors and psychiatric diagnosis: review and projections. World Psychiatry, 8(3), 131-139. 90 77. Ketterer, H., Han, K. & Weed, N. C. (2010). Validation of a Korean MMPI-2 HwaByung scale using a Korean normative sample. Cultural Diversity and Ethnic Minority Psychology, 16(3), 379-385. 78. Min, S. K., Suh, S. Y. & Song, K. J. Symptoms to use for diagnostic criteria of hwabyung, an anger syndrome. Psychiatry Investigation, 6(1), 7-12. 79. Bartelsman, M. & Eckhardt, P. P. (2007). Mental illness in the former Dutch Indies-four psychiatric syndromes: amok, latah, koro and neurasthenia. Netherlands Journal of Medicine, 151(51), 2845-2851. 80. Baronet, A. M. (1999). Factors associated with caregiver burden in mental illness: a critical review of the research literature. Clinical Psychology Review, 19(7), 819-841. 81. Papastavrou, E., Charalambous, A., Tsangari, H. & Karayiannis, G. (2010). The cost of caring: the relative with schizophrenia. Scandinavian Journal of Caring Sciences, 24(4), 817-823. 82. Faison, K. J., Faria, S. H. & Frank, D. (1999). Caregivers of chronically ill elderly: perceived burden. Journal of Community Health Nursing, 16(4), 243-253. 83. Reamy, A. M., Kim, K., Zarit, S. H. & Whitlatch, C. J. (2011). Understanding Discrepancy in Perceptions of Values: Individuals With Mild to Moderate Dementia and Their Family Caregivers. Gerontologist, epub ahead of print. 84. Cochrane, J. J., Goering, P. N. & Rogers, J. M. (1997). The mental health of informal caregivers in Ontario: an epidemiological survey. American Journal of Public Health and the Nation’s Health, 87(12), 2002-2007. 91 85. Gaugler, J. E., Mittelman, M. S., Hepburn, K. & Newcomer, R. (2010). Clinically significant changes in burden and depression among dementia caregivers following nursing home admission. BioMed Central Medicine, 8, 85. 86. Canadian Institute for Health Information. (2006). National Health Expenditure Trends 1975-2006. Ottawa, ON: Canadian Institute for Health Information. 87. Institute of Health Economics. (2007). Mental health economic statistics. Edmonton, AB: Institute of Health Economics. 88. Regier, D. A., Narrow, W. E., Rae, D. S., Manderscheid, R. W., Locke, B. Z. & Goodwin, F. (1993). The de Facto US mental health and addictive disorders service system. Epidemiologic Catchment Area prospective 1-year prevalence rates of disorders and services. Archives of General Psychiatry, 50, 85–94. 89. Arboleda-Flórez J. (2006). Forensic psychiatry: contemporary scope, challenges and controversies. World Psychiatry, 5(2), 87-91. 90. Statistics Canada. (2009). Overview of issues – mental health and the criminal justice system. Crime and Justice Research Paper Series, 2009(16). Ottawa, ON: Statistics Canada. 91. Livingston, M. (1998). Update on health care in Canada: what's right, what's wrong, what's left. Journal of Public Health Policy, 19(3), 267-88. 92. Crocker, A.G., K. Hartford & Heslop, L. (2009). Gender differences in police encounters among persons with and without serious mental illness. Psychiatric Services, 60(1), 86-93. 92 93. Arboleda-Flórez, J.E., Love, E.J., Fick, G., O'Brien, K., Hashman, K. & Aderibigbe, Y. (1995). An epidemiological study of mental illness in a remanded population. International Medical Journal, 2(2), 113-126. 94. Fazel, S. & Danesh, J. (2002). Serious mental disorder in 23000 prisoners: a systematic review of 62 surveys. Lancet, 359(9306), 545-550. 95. Chaiken, J. M. (2000). Crunching numbers: crime and incarceration at the end of the millenium. National Institute of Justice Journal, 242, 10-17. 96. Rennison, C. M. (2000). Criminal victimization 1999. National Crime Victimization Survey. U.S.A: Bureau of Justice Statistics. 97. Povey, D. (2001). Recorded crime, England and Wales – 12 months to March 2001. London, UK: Home Office Statistical Bulletins. 98. Berichtsjahr, P. K. S (2001). Police Crime Statistics – 2000. Wiesbaden, Germany: Bundeskriminalamt (Federal Criminal Police Office). 99. Lesage AD, Morissette R, Fortier L, Reinharz D, Contandriopoulos AP. (2000).Downsizing psychiatric hospitals: needs for care and services of current and discharged long-stay inpatients. Canadian Journal of Psychiatry, 45(6), 526-532. 100. Harding, C. M., Brooks, G. W., Ashikaga, T., Strauss, J. S. & Breier, A. (1987). The Vermont longitudinal study of persons with severe mental illness, II: Long-term outcome of subjects who retrospectively met DSM-III criteria for schizophrenia. American Journal of Psychiatry, 144(6), 727-35. 101. Leff, J. (1993). Evaluating the transfer of care from psychiatric hospitals to district-based services. British Journal of Psychiatry Supplement, 19, 6. 93 102. Lurigio, A. J., Rollins, A., & Fallon, J. (2004). The effects of serious mental illness on offender reentry. Federal Probation, 68(2), 45-52. 103. Penrose, L. S. Mental disease and crime: Outline of a comparative study of European statistics. British Journal of Medical Psychology, 18, 1. 104. Beck, A. J. (2000). Prison and jail inmates at midyear 1999. Bureau of Justice Statistics Bulletin, NCJ 181643. Washington, DC: U.S. Department of Justice. 105. Justice Review Task force. (2005). Beyond the revolving door: a new response to chronic offenders. Report of the street crime working group to the Justice review task force. Vancouver, BC: Justice Review Task force. 106. Arboleda-Flórez, J. (2000). Research on the impact of the Criminal Code amendments in Alberta. In: Eaves, D., Ogloff, J. R. P. & Roesch, R., editors, Mental disorders and the Criminal Code: legal background and contemporary perspectives (129–139). Burnaby, BC: Mental Health, Law, and Policy Institute. 107. MacKay, R. (1995). Insanity and fitness to stand trial in Canada and England: a comparative study of recent developments. Journal of Forensic Psychiatry, 6, 121138. 108. Nussbaum, D., Malcolmson, S. & Dosis, O. Summary of research on Bill C-30 implementation in Ontario during 1993 and 1994. In: Eaves, D., Ogloff, J. R. P. & Roesch, R., editors, Mental disorders and the Criminal Code: legal background and contemporary perspectives (104-118). Burnaby, BC: Mental Health, Law, and Policy Institute. 109. Smith, J. & Grant, F. (1993). Amendments to the Criminal Code of Canada: impacts on inpatient psychiatric remands. Journal of Forensic Psychiatry, 4, 355-360. 94 110. Ohayon, M. M., Crocker, A., St-Onge, B. & Caulet, M. (1998). Fitness, responsibility, and judicially ordered assessment. Canadian Journal of Psychiatry, 43, 491–495. 111. Hemels, M., Koren, G. & Einarson, T. (2002). Increased use of antidepressants in Canada: 1981-2000. The Annals of Pharmacotherapy, 36, 1375-1379. 112. Marcotte, D. E. & Markowitz, S. (2011). A cure for crime? Psycho- pharmaceuticals and crime trends. Journal of Policy Analysis and Management, 30(1), 29–56. 113. George, D. T., Phillips, M. J., Lifshitz, M., Lionetti, T. A., Spero, D. E., Ghassemzedeh, N. et al. (2011). Fluoxetine treatment of alcoholic perpetrators of domestic violence: a 12-week, double-blind, randomized, placebo-controlled intervention study. Journal of Clinical Psychiatry, 72(1), 60-65. 114. Lee, R., Kavoussi, R. J. & Coccaro, E. F. (2008). Placebo-controlled, randomized trial of fluoxetine in the treatment of aggression in male intimate partner abusers. International Clinical Psychopharmacology, 23(6), 337-341. 115. Walsh, M. T. & Dinan, T. G. (2001). Selective serotonin reuptake inhibitors and violence: a review of the available evidence. Acta Psychiatrica Scandinavica, 104(2), 84-91. 116. Hagan, J. (1991) The disreputables pleasures: crime and deviance in Canada (third edition). Toronto, ON: McGraw-Hill. 117. Freeman, R. B .(2000). Does the Booming Economy Help Explain the Drop in Crime. In: Perspectives on Crime and Justice: 1999-2000 Lectures Series. U.S.A: Department of Justice. 95 118. Ouimet, M. (2002). Explaining the American and Canadian Crime Drop in the 1990’s. Canadian Journal of Criminology, 44 (1) 33-50 119. Foote, D (2000). Boom, Bust and Echo (second edition). Toronto, ON: MacFarlane, Walter & Ross. 120. Farrington, D. (1986). Age and Crime. In: Michael Tonry and Norval Morris (Eds). Crime and Justice volume 7. Chicago, IL: University of Chicago Press. 121. Arboleda-Flórez, J. & Holley, H. L. (1988). Criminalization of the mentally ill: Part II. Initial detention. Canadian Journal of Psychiatry, 33(2), 87-95. 122. Watson, A., Hanrahan, P., Luchins, D. & Lurigio, A. (2001). Mental health courts and the complex issue of mentally ill offenders. Psychiatric Services, 52(4), 477-481. 123. Capaldi, D. M., & Patterson, G. R. (1991). Relation of parental transitions to boys' adjustment problems: I. A linear hypothesis. II. Mothers at risk for transitions and unskilled parenting. Developmental Psychology, 27, 489-504. 124. Cutajar, M. C., Mullen, P. E., Ogloff, J. R., Thomas, S. D., Wells, D. L. & Spataro, J. (2010). Schizophrenia and other psychotic disorders in a cohort of sexually abused children. Archives of General Psychiatry, 67(11), 1114-9. 125. Blane D. (1995). Social determinants of health – socioeconomic status, social class, and ethnicity. American Journal of Public Health, 85(7), 903-904. 126. Lamb, R. & Weinberger, L. (1998). Persons with severe mental illness in jails and prisons: a review. Psychiatric Services, 49(4), 483-492. 9. 127. Hiday V A. (1995). The social context of mental illness and violence. Journal of Health and Social Behaviour, 36(2), 122-137. 96 128. Swanson, J., Borum, R., Marvin, S. & Monahan, J. (1996). Psychotic symptoms and disorders and the risk of violent behaviour in the community. Criminal Behaviour and Mental Health, 6(4), 309-329. 129. Link, B. G., Stueve, A. & Phelan, J. (1998). Psychotic symptoms and violent behaviors: probing the components of “threat/control-override” symptoms. Social Psychiatry & Psychiatric Epidemiology, 33(13), S55-S60. 130. Arboleda-Flórez, J. (2006). Forensic psychiatry: contemporary scope, challenges and controversies. World Psychiatry, 5(2), 87-91. 131. Communications Branch. (2007). Civil and criminal cases. Canada's system of justice. Retrieved on March, 2011 from the Canadian Department of Justice website: http://www.justice.gc.ca/eng/dept-min/pub/just/08.html 132. Zgodzinski, Rose. (2009). Forensic mental health: pathways to justice for the mentally disordered accused. Canadian Mental Health Association Network Magazine, 24 (2), 14-15. 133. Hartford, K. (2004). Evidence-based practices in diversion programs for persons with serious mental illness who are in conflict with the law: literature review and synthesis. Retrieved November, 2010 from Canadian Mental Health Association website: http://www.ontario.cmha.ca/justice.asp?cID=5445 134. Authors. (2010). Retrieved January, 2011 from the ORB website: http://www.orb.on.ca/scripts/en/. 135. Arboleda-Flórez, J. (2005). Forensic psychiatry Two masters, one ethics. Die Psychiatrie, 2, 153-157. 97 136. McNeill, P. M. (1993). The ethics and politics of human experimentation. Cambridge, UK: Cambridge University Press. 137. Beecher, H, K. (1966). Ethics and clinical research. New England Journal of Medicine, 274, 1354-1360. 138. Roberts, L. W. & Roberts, B. (1999). Psychiatric research ethics: an overview of evolving guidelines and current ethical dilemmas in the study of mental illness. Biological Psychiatry, 46, 1025-1038. 139. Hare, R. D. (1998). The Hare PCL-R: Some issues concerning its use and misuse. Legal and Criminological Psychology, 3, 101-122. 140. Harris, G.T., Rice, M.E., & Quinsey, V.L. (1993). Violent recidivism of mentally disordered offenders: The development of a statistical prediction instrument. Criminal Justice and Behavior, 20, 315-335. 141. Quinsey, V.L., Rice, M.E. & Harris, G.T. (1995). Actuarial Prediction of Sexual Recidivism. Journal of Interpersonal Violence, 10, 85-105. 142. Stuart, H. L. & Arboleda-Flórez, J. (2001). A Public Health Perspective on Violent Offenses Among Persons With Mental Illness. Psychiatric Services, 52, 654659. 143. Brink, J. H., Doherty, D. & Boer, A. (2001). Mental disorder in federal offenders: A Canadian prevalence study. International Journal of Law and Psychiatry, 24, 339356. 144. Hodgins , S. & Côté, G. (1993). The Criminality of Mentally Disordered Offenders. Criminal Justice and Behavior, 20, 115. 98 145. Bouchard, F. (2004). A Health Care Needs Assessment of Federal Inmates in Canada. Canadian Journal of Public Health, 95(S), S1-S63. 146. Davis, S. (1994). Examining the impact of Bill C-30 in British Columbia. International Bulletin of Law and Mental Health, 5, 5-9. 147. Melton, G. B., Petrila, J., Pothress, N. G. & Slobogin, C. (2007). Psychological evaluations for the courts: a handbook for mental health professionals and lawyers (third edition). New York, NY: Guilford Press. 148. St. Denis, E. (2008). Forensic psychiatry and criminal responsibility in Santiago, Chile. Master of Science dissertation. Retrieved from QSpace: http://hdl.handle.net/1974/1467. Queen’s University, Kingston, ON. 149. Hilton, N. Z. & Simmons, J. L. (2001). The Influence of actuarial risk assessment in clinical judgments and tribunal decisions about mentally disordered offenders in maximum security. Law and Human Behaviour, 25(4), 393-408. 150. Holley, H., Arboleda-Flórez, J. & Crisanti, A. (1998). Do forensic offenders receive harsher sentences? International Journal of Law and Psychiatry, 21(1), 43-57. 151. Criminal Code, R.S.C. 1985, c. C-46, s. 318(1), online: Department of Justice Canada < http://laws.justwice.gc.ca>. Criminal Code, Revised Statutes of Canada, 1985, chapter C-46, section 318, subsection 1, cited to the online version on the Canadian government website. 152. Opjordsmoen, S. & Retterstøl, N. (2007). Dimensions of delusional experiences and their value as predictors of long-term outcome. Psychopathology, 40(5), 278-281. 153. Grilo, C. M., Stout, R. L., Markowitz, J. C., Sanislow, C. A., Ansell, E. B., Skodol, A. E. et al. (2010). Personality disorders predict relapse after remission from 99 an episode of major depressive disorder: a 6-year prospective study. Journal of Clinical Psychiatry, 71(12), 1629-1635. 154. Hellerstein, D. J., Skodol, A. E., Petkova, E., Xie, H., Markowitz, J. C., Yen, S. et al. (2010). The impact of comorbid dysthymic disorder on outcome in personality disorders. Comprehensive Psychiatry, 51(5), 449-457. 155. Flanagan, E. H., Davidson, L. & Strauss, J. S. (2010). The Need for Patient- Subjective Data in the DSM and the ICD. Psychiatry, 73(4), 297-307. 156. Aboraya, A. (2010). Scientific forum on the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-V) – an invitation. Psychiatry (Edgemont), 7(11), 32-36 157. Walters. G. D. (1990). The criminal lifestyle: patterns of serious criminal conduct. Newbury Park, CA: Sage Publications. 158. Walters, G. D. (2007). Predicting institutional adjustment with the Lifestyle Criminality Screening Form and the Antisocial Features and Aggression scales of the PAI. Journal of Personality Assessment, 88(1), 99-105. 159. White, J. L., Moffitt, T. E., Earls, F., Robins, L., & Silva, P. A. (1990). How early can we tell? Predictors of childhood conduct disorder and adolescent delinquency. Criminology, 28, 507-533. 160. Pulkkinen, L. & Tremblay, R. E. (1992). Patterns of boys’ social adjustment in two cultures and at different ages: A longitudinal perspective. International Journal of Behavioral Development, 15, 527-553. 161. Gray, N. S., Hill, C., McGleish, A., Timmons, D., MacCulloch, M. J. & Snowden, R. J. (2003). Prediction of violence and self-harm in mentally disordered offenders: a 100 prospective study of the efficacy of HCR-20, PCL-R, and psychiatric symptomatology. Journal of Consulting and Clinical Psychology, 71(3), 443-51. 162. Gardner, W., Lidz, C. W., Mulvey, E. P. & Shaw, E. C. (1996). Clinical versus actuarial predictions of violence of patients with mental illnesses. Journal of Consulting and Clinical Psychology, 64(3), 602-609. 163. Zoghes, D., Liakos, A. & Potamianos, G. (1998). An investigation into the relationship between hostility and criminality in schizophrenia. International Journal of Social Psychiatry, 34(3), 207-211. 164. Yates, P. M. & Kingston, D. A. (2006). The self-regulation model of sexual offending: the relationship between offence pathways and static and dynamic sexual offence risk. Sexual Abuse: A Journal for Research and Treatment, 18(3), 259-270. 165. Schanda, H. (2006). Investigating the association between psychosis and criminality/violence. Fortschritte der Neurologie-Psychiatrie, 74(2), 85-100. 166. Buchanan, A. (2008). Risk of violence by psychiatric patients: beyond the "actuarial versus clinical" assessment debate. Psychiatric Services, 59(2), 184-190. 167. Langdon, P. E., Maxted, H. & Murphy, G. H. (2007). An exploratory evaluation of the Ward and Hudson Offending Pathways model with sex offenders who have intellectual disability. Journal of Intellectual & Developmental Disability, 32(2), 94105. 168. Kennealy, P. J., Skeem, J. L., Walters, G. D. & Camp, J. (2010). Do core interpersonal and affective traits of PCL-R psychopathy interact with antisocial behavior and disinhibition to predict violence? Psychological Assessment, 22(3), 569580. 101 169. Robins, L. N. (1970). The adult development of the antisocial child. Seminars in Psychiatry, 2(4), 420-434. 170. Huesmann, L. R., Eron, L. D., Lefkowitz, M. M.,&Walder, L. O. (1984). Stability of aggression over time and generation. Developmental Psychology, 20, 1120-1134. 171. Stattin, H., & Magnusson, D. (1989). The role of early aggressive behavior in the frequency, seriousness, and types of later crime. Journal of Consulting and Clinical Psychology, 57, 710-718. 172. Colins, O., Vermeiren, R., Vahl, P., Markus, M., Broekaert, E. & Doreleijers, T. (2011). Parent-reported attention-deficit hyperactivity disorder and subtypes of conduct disorder as risk factor of recidivism in detained male adolescents. European Psychiatry: the Journal of the Association of European Psychiatrists, epub ahead of print. 173. Frick, P. J. (2009). Extending the construct of psychopathy to youth: implications for understanding, diagnosing, and treating antisocial children and adolescents. Canadian Journal of Psychiatry, 54(12), 803-812. 174. Mulder, E., Brand, E., Bullens, R., Van Marle, H. (2010). A classification of risk factors in serious juvenile offenders and the relation between patterns of risk factors and recidivism. Criminal Behaviour and Mental Health, 20(1), 23-38. 175. Farrington, D. P. (1994). Childhood, adolescent, and adult features of violent males. In L. R. Huesmann (Ed.), Aggressive behavior: Current perspectives (215240). New York, NY: Plenum. 176. Moffitt, T. E. (1993). Adolescence-limited and life-course-persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674-701. 102 177. Johnston, D. (1995a). Effects of parental incarceration. In K. Gabel & D. Johnston (Eds.), Children of incarcerated parents (59-88). New York, NY: Lexington Books. 178. Capaldi, D. M., Kim, H. K. & Owen, L. D. (2008). Romantic partners' influence on men's likelihood of arrest in early adulthood. Criminology, 46(2), 267-299. 179. Bourget, D. & Bradford, J. M. (1987). Affective disorder and homicide: a case of familial filicide theoretical and clinical considerations. Canadian Journal of Psychiatry, 32(3), 222-225. 180. Hobus, P. P., Schmidt, H. G., Boshuizen, H. P, Patel, V. L. (1987). Contextual factors in the activation of first diagnostic hypotheses: expert-novice differences. Medical Education, 21(6), 471-476. 181. Rascón, M. L., Caraveo, J. & Valencia, M. (2010). Emotional, physical and psychiatric disorders in relatives of schizophrenia patients in Mexico. Revista de Investigación Clínica, 62(6), 509-515. 182. Maji, K. R., Sood, M., Sagar, R. & Khandelwal, S. K. (2011). A follow-up study of family burden in patients with bipolar affective disorder. International Journal of Social Psychiatry, epub ahead of print. 183. Drake, R., Goldman, H., Leff, H., Lehman, A., Dixon, L., Mueser, K. et al. (2001). Implementing evidence-based practices in routine mental health service settings. Psychiatric Services, 52(2), 179-182. 184. Scott, D. Personal communication. October, 2010. 185. Looman, J., Abracen, J., Serin, R. & Marquis, P. (2005). Psychopathy, treatment change, and recidivism in high-risk, high-need sexual offenders. Journal of Interpersonal Violence, 20(5), 549-568. 103 186. Stokman, C. L. (1984). Dangerousness and violence in hospitalized mentally ill offenders. The Psychiatric Quarterly, 56(2), 138-143. 187. Phillips, H. K., Gray, N. S., MacCulloch, S. I., Taylor, J., Moore, S. C., Huckle, P. & MacCulloch, M. J. (2005). Risk assessment in offenders with mental disorders: relative efficacy of personal demographic, criminal history, and clinical variables. Journal of Interpersonal Violence, 20(7), 833-847. 188. Hamilton, J. R. (1987). Violence and victims. The contribution of victimology to forensic psychiatry. Lancet, 1(8525), 147-150. 189. Field, A. (2002). Discovering Statistics Using SPSS. London, UK: Sage. 104 Appendices Appendix One. Detailed description of DSM IV-TR Classification. Axis I Axis II Axis III Clinical Disorders Other Conditions That May Be A Focus of Clinical Attention Disorders usually first diagnosed in infancy, childhood, or adolescence Delirium, dementia, and amnestic and other cognitive disorders Mental disorders due to a general medical condition not elsewhere classified Substance-related disorders Schizophrenia and other psychotic disorders Mood disorders Anxiety disorders Somatoform disorders Factitious disorders Dissociative disorders Sexual and gender identity disorders Eating disorders Sleep disorders Impulse-control disorders not elsewhere classified Adjustment disorders Other Conditions That May Be A Focus of Clinical Attention Personality Disorders Mental Retardation Paranoid personality disorder Schizoid personality disorder Schizotypal personality disorder Antisocial personality disorder Borderline personality disorder Histrionic personality disorder Narcissistic personality disorder Avoidant personality disorder Dependent personality disorder Obsessive-compulsive personality disorder Personality Disorder Not Otherwise Specified Mental Retardation General Medical Conditions Infectious and Parasitic Diseases Neoplasms Endocrine, Nutritional, and Metabolic Diseases. Immunity Disorders Diseases of the Blood and Blood-Forming Organs Diseases of the Nervous System and Sense Organs Diseases of the Circulatory System Diseases of the Respiratory System Diseases of the Digestive System Diseases of the Genitourinary System Complications of Pregnancy, Childbirth, and the Puerperium Diseases of the Skin and Subcutaneous Tissue Diseases of the Musculoskeletal System and Connective Tissue Congenital Anomalies Certain Conditions Originating in the Perinatal Period Systems, Signs, and Ill-Defined Conditions Injury and Poisoning 105 Axis IV Axis V Psychosocial and Environmental Problems Problems with primary support group Problems related to the social environment Educational problems Occupational problems Housing problems Economic problems Problems with access to health care services Problems related to interaction with the legal system/crime Other psychosocial and environmental problems Global Assessment of Functioning 91-100: Superior functioning in a wide range of activities; No Symptoms 81-90: Absent or minimal symptoms 71-80: If symptoms are present 61-70: Some mild symptoms 51-60: Moderate symptoms 41-50: Severe symptoms 31-40: Some impairment in reality testing or communication 21-30: Behavior is considerably influenced by delusions or hallucinations OR serious impairment in communication or judgment 20-11: Some danger of hurting self or others 10-1: Persistent danger of severely hurting self or others 106 Appendix Two. Forensic Path Forensic path for psychiatric offenders is different from the path taken by general offenders. In the case of a general offender who committed a minor or a summary crime, arrest follows booking and interrogation by the police. The police then charge the individual with a summary conviction offence. A person accused of a crime is not always arrested. The accused may simply receive a “summons,” or an order to appear in court at a certain time to answer to the charge, after the police lay a charge. The accused then appears before a provincial court judge, who determines if the accused should be remanded until the trial or if the accused may be released on bail and return to the court for the trial. This step occurs within 24 hours of arrest unless the accused is released by the police. The trial following remand in a jail or return to the court on a bail determines if the accused is guilty. An accused found guilty is sentenced to a term of imprisonment, a fine or a combination of the two. For an imprisonment sentence under two years, the convict is admitted to a provincial prison. For an imprisonment sentence over two years, the convict is admitted to a federal prison. For more serious crimes than summary conviction offense, arrest is followed by indictment. There may be a preliminary hearing for a judge to examine the case to decide if there is enough evidence to proceed with the trial. This sometimes results in dismissal of the case if the court finds there to be insufficient evidence that crime has been committed or the accused is the perpetrator of the crime. The accused may be tried by a provincial court judge, by a superior court judge or by a judge of a superior court with a jury. 107 When the police encounter a mentally ill person upon arriving at the crime scene, they may use discretion to decide between arresting the perpetrator of the crime or calling in a local crisis team if there is a prior agreement between two services. Arrest diversion occurs when the police arrested a person but decide not to lay charges because the crime is minor and the arrested individual’s psychiatric condition is severe. The police may also take the arrested person to a local hospital if he is suspected of suffering from a mental illness. The attending physician may issue a Form 1, the Application for Psychiatric Assessment. This is a legal form that gives the hospital the power to hold custody of the arrestee in order to conduct a full psychiatric examination. Post-charge diversion, pretrial diversion or court diversion all refer to the diversion before trial by the court. Crown attorneys may divert the person into mental health treatment instead of persecuting, usually when the charges are minor. A judge may also divert the accused to mental health institution instead of trying them at the court depending on the severity of the mental illness. Both of these diversions may happen with or without an application by the defense lawyer and ultimately result in the accused entering the provincial review board system. If an accused is found to be NCRMD, they remain under the review board jurisdiction. If an accused is found unfit to stand trial but later regains the fitness, they are returned to the court for the trial. If fitness is not restored after the timeline set out in the Criminal Code of Canada, they remain under the review board jurisdiction. The review board determines if the person may be released back to the community or if they pose a significant threat to the general public and self and must be detained in the forensic psychiatric facility. This decision, called disposition, is reviewed annually and the accused with the NCRMD verdict may eventually be released if their 108 conditions improve. The clinical team at the facility works with community mental health service providers to create a discharge plan. Forensic path undertaken by mentally ill offenders is illustrated in Figure A.1. 109 Figure A.1. Navigating the forensic system132. 110 Appendix Three. Data Collection Sheet The following list of information was extracted from each patient chart and was compiled into a Microsoft Excel sheet. 1. Hospital location admitted to – current facility (Providence Care), other forensic psychiatric institution, psychiatric hospital, jail/detention centres, or court house 2. Attending physician 3. Record number 4. Date of Birth 5. Gender 6. Clinical status – court referral, recurring patient, outpatient and inpatient 7. Is the subject a current inpatient? 8. Area of residence 9. Marital status 10. Living arrangement 11. Highest education reached 12. Income source 13. Date of Index offence 14. Date of ORB finding 15. Index Offence –homicide, assault, sexual assault, offence against property or other (e.g. prostitution) 16. Geographical location of index offence 17. Crime site – on the street, in the victim’s house – bedroom, kitchen, living room, front door, restroom; in the patient’s house – bedroom, kitchen, living room, front 111 door, restroom; in the victim and patient’s shared house – bedroom, kitchen, living room, front door, restroom; business property in a public setting; business property in a private setting; public property in a public setting; or public property in a private setting 18. Relationship to victim – parent of the victim, child of the victim, spouse, romantic partner, friend, colleague, acquaintance, stranger, or crime did not target/involve a specific person 19. Child Abuse History – No abuse in the family, witness of physical/sexual/physical and sexual abuse, victim of physical/sexual/physical and sexual abuse, victim of passive abuse (negligence) 20. Early signs of deviation – no known childhood problems, early behavior problems, suspension or expulsion from school or incompletion of grade 10 education, nonviolent criminal history, did not live with biological parents until the age of 16, an arrest before age 16 21. Adult romantic relationship history – never been engaged in 1-3 serious relationships ending in a marriage or common law, 1-3 relationships or many unsuccessful marriages/common law 22. Current status of adult romantic relationship – currently in or not in a mutually agreed significant relationship 23. Adult violence history – no previous violent conduct, violence against romantic partner, violence against offspring, violence against family or violence against others 24. Psychiatric history –no previous psychiatric episode, previous psychiatric inpatient admission/outpatient treatment at the current facility, previous psychiatric inpatient 112 admission/outpatient treatment at other facility, previous alcohol and/or substance abuse, previous serious suicide attempt, previous legal problems, history of serious physical violence towards others, currently active medical/surgical condition or previous community mental health agency contact 25. Psychiatric admission history – 1-2, 3-5 or more than 5 previous admissions 26. Criminal history – no previous record; 1-3, 4-5 or more than 5 offences against human (homicide); 1-3, 4-5 or more than 5 offences against human (assault); 1-3, 4-5 or more than 5 offences against human (sexual assault); 1-3, 4-5 or more than 5 offences against property; or 1-3, 4-5 or more than 5 other offences 27. Presence of criminal associates – no known criminal associates, criminal parent(s), criminal sibling(s), criminal relative(s), criminal romantic partner 28. Presence of psychiatric associates – no known psychiatric associates, psychiatric parent(s), psychiatric sibling(s), psychiatric relative(s), psychiatric romantic partner 29. Timing of first official psychiatric – diagnosis before first criminal offence (juvenile crime), at first criminal offence (juvenile crime), after first criminal offence (juvenile crime); diagnosis before first criminal offence (adult crime), at first criminal offence (adult crime), after first criminal offence (adult crime) 30. Has the principal psychiatric diagnosis changed since the first official diagnosis? 31. Current psychopathology 32. Medical history as per DSM IV-TR Axis III A-D 33. Psychosocial history as per DSM IV-TR Axis IV A-H 34. Global assessment of functioning – on a scale of 1-100 35. Life psychopathology 113 36. Fitness to stand trial 37. Hare’s Psychopathy Checklist-Revised 38. Recidivism score 39. ORB disposition – fulltime detention, conditional discharge or absolute discharge 40. Has the patient or legal guardian of the patient given consent to the treatment? 114