치료감호 가종료자의 재범요인에 관한 연구

다국어 초록 (Multilingual Abstract)

The recent increase in the number of crimes committed by people with mental disorder has been making the public concerned and worried. After the long-standing debates whether people with mental disorder are more dangerous than those without mental disorder, it has been consistently recognized in recent research that mental disorder has a correlation with violent crime.
In many countries, as the serious crimes by people with mental disorder increased, psychiatric hospitals were introduced for the purpose of special treatment and community protection. Likewise in 1987, Korea government enacted「Psychiatric Custody Act」whereby the Forensic Hospital were established. However, although twenty three years has elapsed since its inception in 1987, the recidivism rates after discharge from the Forensic Hospital was reported to be higher than any other general offenders. Therefore, there is a need to consider fundamental system changes in managing psychiatric criminals.
It has been demonstrated that the discharged psychiatric criminals leads to recidivism due to mental disorder factors, circumstantial factors or a combination of the two factors.
In many advanced countries, as an attempt to identify the causes of recidivism, the predictive tools have been developed and widely used in practice. However, there have been no development of predictive tools and empirical research on the causes of recidivism in Korea. Thus, the aim of this study is to examine the predictive factors for criminals with mental disorder and suggest policy reform for reducing recidivism. In order to investigate predictors of recidivism among a total of 145 probationers discharged from the Psychiatric Hospital, this study compared the recidivism factors between recidivism group and non-recidivism group and the following research questions was set.
Research Question 1: Demographic variables have effects on recidivism.
- Age, years of education, employment, monthly income, experience of marriage and military service have effects on recidivism.
Research Question 2: Family environment variables have effects on recidivism.
- Cohabitation status, family visit during custody, frequency of housing movement after discharged from custody and family relationship have effects on recidivism.
Research Question 3: Offence-related variables have effects on recidivism.
- Index offence, age at first offence, total criminal history, duration of custody and psychiatric custody have effects on recidivism.
Research Question 4: Probation variables have effects on recidivism.
- Frequency of contact with probation officers, compliance with the condition and alcohol usage have effects on recidivism.
Research Question 5: Mental disorder variables have effects on recidivism.
- Diagnosis, age of onset, prior psychiatric treatment, frequency of outpatient treatment after discharge, and compliance with medication have effects on recidivism.
For addressing the research questions, frequency analysis was employed to examine demographic characteristics, and descriptive analysis was conducted to produce means and standard deviations of the variables. In addition, t-tests were carried out to examine the difference between recidivism group and non-recidivism group. Futhermore, Logistic regression analysis was employed to investigate predictors for recidivism. After the process, this study could find the research results as follows.
First, in the results of Question 1, there was a significant difference on age between recidivists and non-recidivists, whereas t-test failed to find differences on such variables as employment, monthly income, marriage, years of education and history of illness. Futhermore, with regard to the results of the logistic regression analysis, it could not find the relationship with recidivism. These findings may result from relatively small sample size, and thus increasing sample size could strengthen the power of a statistical test and reduce Type II error accordingly.
Second, a series of t-tests on family environment variables revealed significant differences on frequency of housing movement and family relationship. Futhermore, the logistic regression analysis found cohabitation status, frequency housing movement and family relationship was predictive factor for recidivism.
Third, the results of t-tests in Question 3 revealed that there were differences on such variables as age at first offence, total criminal history and history of custody. With regard to the results of the logistic regression analysis, only history of custody entered the model of the logistic regression.
Forth, in the results of Question 4, all three variables in probation variable (i.e. frequency of contact with probation officers, compliance with the condition and alcohol usage) were found to be significantly different between recidivists and non-recidivists, and furthermore entered the model of the logistic regression.
Finally, t-test revealed that age of onset, frequency of outpatient treatment and compliance with medication of recidivists were significantly different from those of non-recidivists, whereas logistic regression analysis found that age of onset, prior psychiatric treatment and compliance with medication were associated with recidivism.
The main policy implication of this study is that the ultimate aim of criminal management agencies in Korea should be to reduce recidivism rate and push for changes of policy in criminal management. To this end, risk assessment tools should be developed and extensively used across the criminal management agencies, and termination of psychiatric custody should be evaluated based on the future risk of recidivism, as opposed to the nature of past offences. In addition, criminal policy should be developed for enhancing compliance with probation conditions, strengthening the liaison between the Probation Service and Psychiatric Hospital, and reinforcing personnel in both Probation Offices and the Psychiatric Hospital for effective treatment. Furthermore, the Probation practice system also should be changed in improving the compliance with outpatient treatment and medication for mentally disordered offenders.
Even though it could be found useful results from this research, it still contains some limitations of the study. Its sample size was relatively not enough for empirical study, and neither affective factors nor treatment status of participants were considered. Thus, it is suggested that sufficient sample should be recruited and supplementary survey on factors associated with recidivism should be added in the future research.

목차 (Table of Contents)

• 제 1 장 서론 1
• 제 1 절 연구의 목적 1
• 제 2 절 연구의 범위 및 방법 7
• 1. 연구의 범위 7
• 2. 연구의 방법 10
• 제 2 장 연구의 이론적 배경 12
• 제 1 절 연구의 이론적인 틀 12
• 제 2 절 정신장애와 범죄 17
• 1. 정신장애의 개념 17
• 2. 정신장애 유형과 범죄 19
• 3. 정신장애자 치료 34
• 제 3 절 사회환경적 요인과 범죄 40
• 1. 인구사회학적 요인과 범죄 40
• 2. 가정환경 및 주거요인과 범죄 43
• 3. 범죄경력요인과 범죄 48
• 4. 보호관찰요인과 범죄 50
• 제 4 절 정신장애인의 처우실태 53
• 1. 외국의 정신장애 범죄인 처우 53
• 2. 우리나라 정신장애 범죄인 처우 60
• 제 5 절 선행연구 검토 75
• 1. 재범위험성 판단문제 75
• 2. 선행연구 검토 81
• 제 3 장 연구의 설계와 결과분석 94
• 제 1 절 연구모형 및 연구문제 94
• 1. 연구모형 94
• 2. 연구문제 97
• 제 2 절 자료수집 및 분석방법 99
• 1. 자료수집 99
• 2. 분석방법 99
• 3. 변수의 정의 및 측정 101
• 제 3 절 연구대상자의 일반적 특성 분석 103
• 1. 재범률 103
• 2. 인구사회학적 특성 104
• 3. 가정환경 및 주거요인 특성 106
• 4. 범죄유형 및 경력별 특성 108
• 5. 보호관찰요인 특성 110
• 6. 정신장애요인 특성 111
• 제 4 절 재범·비재범자집단간의 요인별 차이분석 113
• 1. 인구사회학적 요인 차이 113
• 2. 가정환경 및 주거요인 차이 114
• 3. 범죄경력요인 차이 115
• 4. 보호관찰요인 차이 117
• 5. 정신장애요인 차이 118
• 제 5 절 재범에 영향을 미치는 요인분석 120
• 1. 인구사회적요인이 재범에 미치는 영향 120
• 2. 가정환경 및 주거요인이 재범에 미치는 영향 121
• 3. 범죄경력요인이 재범에 미치는 영향 123
• 4. 보호관찰요인이 재범에 미치는 영향 124
• 5. 정신장애요인이 재범에 미치는 영향 125
• 제 6 절 연구결과 논의 129
• 1. 재범결정요인 논의 129
• 2. 분석결과 논의 131
• 제 4 장 정책적 제언 134
• 제 1 절 치료감호와 보호관찰의 처우환경 개선 134
• 1. 치료감호소 치료환경 개선 134
• 2. 보호관찰순응도 향상방안 138
• 3. 치료감호와 보호관찰 연계 강화 142
• 제 2 절 사회 내 치료 강화방안 144
• 1. 투약순응도 향상방안 144
• 2. 외래진료 강화방안 148
• 제 3 절 재범위험성 평가도구 개발 및 활용 152
• 1. 재범위험성 평가도구 개발 152
• 2. 가종료 심사방식 개선 154
• 제 5 장 결론 158
• 참고문헌 165
• ABSTRACT 183
• 부록 188