The Effects of Mental Health and Substance Abuse/Dependence Disorders on Prison Misconduct Among Male Inmates in Taiwan

Abstract

This study examined the relative effects of mental illness, substance abuse/dependence, and co-occurring mental disorders and substance abuse/dependence (CODs) on prison misconduct among male inmates (N = 2,065) incarcerated in Taiwan’s nine correctional facilities. Both bivariate and multivariate analyses revealed that COD-affected inmates have the highest risk of prison misconduct compared to those with singular drug abuse/dependence disorders or no disorders, similar to the findings of previous studies conducted in the United States. These results highlighted the importance of clinical screenings and assessments for inmates who might have CODs. Integrated treatments may be more appropriate for inmates with CODs, rather than providing separate treatments for mental and drug abuse/dependence disorders.

Keywords

co-occurring mental illness and substance abuse/dependence disorders (CODs)importation deprivation Taiwan prison misconduct

Mental health problems are one of the most significant factors associated with institutional misconduct (James & Glaze, 2006; Matejkowski, 2017; O’Keefe & Schnell, 2007; Prince & Wald, 2018; Steiner & Wooldredge, 2009; Wright et al., 2007). Literature evidence suggests that inmates with drug abuse/dependence are more likely to commit misconduct than counterparts without these disorders (Connor & Tewksbury, 2016; Jiang, 2005). Presumably, inmates with both mental health and substance abuse/dependence disorders (hereinafter referred to as CODs) are at increased risk of rule-breaking. Several prior investigations of female inmates (Houser et al., 2012; Houser & Welsh, 2014) corroborate this assumption, but there are very few studies that explore the effect of CODs on prison misconduct among male inmates incarcerated in American prison systems (Wood, 2018) or in other countries/areas. This study complements the existing literature by using a sample of male inmates (N = 2,065) housed in nine prisons across Taiwan. Two questions were addressed: (a) Do singular mental illnesses, drug abuse/dependence disorders, and CODs have independent effects on prison misconduct? and (b) Does the additive effect of CODs increase the likelihood of prison misconduct over that of singular disorders? Unlike prior research that examined mental disorders together, this study breaks down the variable “mental disorders” into specific disorders (e.g., major depression, mania, and psychotic disorders), while also examining combinations of disorders. Research that combines types of mental disorders and treats them as a single disorder may mask the nuanced effects of specific mental health problems on prison misconduct, thereby failing to detect relationships that are indeed present.

The Prevalence of Mental Illness and Substance Abuse/Dependence Disorders Among Inmates

Several recent studies have suggested that there are a high number of inmates with mental illness disorders. Such findings have been made in Brazil (Andreoli et al., 2014), Norway (Cramer, 2016), the United Kingdom (Fazel et al., 2016; Morse, 2017), Australia (Butler & Allnutt, 2003), and the United States (Prins, 2014; Torrey et al., 2014). In the United States, for example, approximately 383,000 individuals in jails and prisons were reported to have serious mental illness (e.g., schizophrenia), 10 times more than those in state hospitals (Torrey et al., 2014). In Brazil, an estimated 31.1% of male inmates were reported to suffer moderate to severe depression. In New South Wales, Australia, Butler and Allnutt (2003) found that 42% of male inmates admitted to the correctional system reported a type of affective (e.g., depression) or anxiety disorder (e.g., social phobia). A systematic review of empirical studies published around the world between 2003 and 2015 revealed that around 13.8% of male inmates had a psychotic illness or major depression (Fazel et al., 2016). The prevalence of mental health disorders among male inmates varies widely across research studies because of the different data collection methods, instruments, and types of mental disorders investigated (Fazel et al., 2016). Nevertheless, the literature data all indicate that mental illnesses are prevalent among male inmates (Fazel et al., 2016).

Mental health problems are often concomitant with substance abuse/dependence (Constantino et al., 2010; James & Glaze, 2006). Among U.S. inmates with mental illness, up to 70% to 75% reported having substance abuse/dependence problems (James & Glaze, 2006; Mumola & Karberg, 2006; Peters et al., 2015). Moreover, incarcerated offenders who had previously used drugs generally reported a history of mental disorders (Cossar et al., 2018; Mumola & Karberg, 2006; Peters et al., 2015). In their study validating two short forms of instruments against three standardized instruments for screening mental disorders, Sacks et al. (2007) identified a high rate of mental illness disorders (78.3%) among inmates who were newly admitted to substance abuse/dependence programs across the United States, with half of them reporting having suffered from severe depression and anxiety at some point in their lives.

Given the prevalence of mental disorders and the fact that mental disorders often accompany drug abuse, it is not surprising to find striking rates of CODs among inmates. Approximately 41% of state male inmates (James & Glaze, 2006) and 32.1% of jail male inmates (Sung et al., 2010) in the United States reported dual disorders. Some studies estimate that 18% to 56% (Chiles et al., 1990; Senior et al., 2013) of inmates suffer from such comorbidities. Given that mental health disorders tend to be underreported and underdetected, the above figures on CODs may be higher (Houser et al., 2012; Peters et al., 2004, 2008, 2015, p. 1) once remarked that inmates with CODs in correctional facilities were “the rule rather than the exception in justice settings,” and that prisons today are commonly referred to as “a new asylum” because of the large number of inmates with CODs (Treatment Advocacy Center, 2016, p. 1). Mental health disorders can cause serious problems, such as suicide, self-harm, victimization, misconduct, and future recidivism (Fazel et al., 2016). A large number of inmates with CODs can further complicate this already problematic situation for correctional administrators.

Mental Illness and Substance Abuse/Dependence Disorders Among Inmates in Taiwan

Empirical evidence shows that many incarcerated people in Taiwan suffer mental illness disorders. Drawing on nationwide population-based data containing information on 82,650 inmates, Tung et al. (2019) found that 11.30% of inmates reported mental disorders, with a significantly higher proportion in female inmates (17.82%) than in male inmates (10.56%). Two other studies found even higher proportions of psychiatric disorders (46%–70%) among male inmates with HIV in Taiwan’s prisons (n = 535; Peng, Lee, et al., 2010; Peng, Yeh, et al., 2010). Three special prisons in Taiwan are designed to house inmates with diagnosed mental illnesses, but their threshold for inmate assignment and qualification for treatment is relatively high. Unfortunately, the limited resources allocated to the professional treatment of mental illness also limit the number of beds in these facilities. Drug-related problems have received more attention, relative to mental health disorders, from Taiwanese scholars and the government. In May 1998, the government enacted the Statute for Narcotics Hazard Control, which mandated that first-time drug offenders receive medical treatment in four drug treatment centers in Taiwan in lieu of being sent to prison. By the end of 2015, there were 29,492 inmates convicted of drug-related offences, accounting for approximately half of the prison population (52,173 male and 4,775 female inmates; Ministry of Justice [MOJ], 2017). Research on drug-related offenders in Taiwan covers a variety of topics, such as drug abuse in relation to HIV infection (Yu et al., 2013), drug policy evaluation (J. Chen, 2016), drug abuse services (Chang, 2013), and recidivism of drug offenders (Chiang et al., 2006). However, no previous research has examined the effects of co-occurring mental and drug abuse disorders on prison misconduct in Taiwan’s institutional context.

CODs and Prison Misconduct

Few studies that have examined the role of CODs in relation to prison misconduct or charges of misconduct in prison show a consistent result: inmates with CODs ranked the highest for engagement in serious misconduct, or tended to be sanctioned, or were sanctioned in the most severe way (Houser & Belenko, 2015; Houser et al., 2012; Houser & Welsh, 2014; Wood, 2018). Houser et al. (2012), the first researchers to study the effects of CODs on prison misconduct, found that female inmates with CODs displayed the highest risk of serious misconduct and were also the most likely to be officially punished for their misbehavior, compared with those reporting singular disorders or no disorder. Similarly, Houser and Welsh (2014) found that inmates with reported symptoms of CODs and mental disorders were more likely to have a record of rule-breaking than those without disorders. Another study of female inmates in the United States conducted by Houser and Belenko (2015) also reported that those of their sample with CODs received more severe institutional sanctions than those without CODs. Wood (2018) recently published a study on American male inmates, revealing that male inmates with CODs reported being charged more often for assaulting staff than their counterparts without dual disorders. In contrast, there was no clear pattern regarding the risk of serious misbehavior or of being charged with misbehavior among inmates with singular mental illness or singular drug abuse/dependence relative to those with no disorders. Previous studies are informative, contributing new knowledge to this important but rarely studied problem; however, the discovery of significant gaps in the research led to the creation of the current study. First, all previous studies used official archival records to document inmates’ prison misbehavior. The official data were rightly criticized for its systematic underreporting of prison misbehavior (Steiner & Wooldredge, 2014). Furthermore, it is well known that administrative records do not capture undetected violations by correctional authorities and do not reflect the offences that prison and jail staff decide not to report. Second, these studies did not distinguish between the various types of mental disorders when examining the effect of mental disorders on misconduct. For these reasons, the subtle effect of specific types of mental disorders on prison misconduct may have been overlooked. Finally, in the Houser et al. (2012) study, the length of incarceration was not controlled in their statistical models, which could have led to a misinterpretation of multivariate results.

Importation and Deprivation Theories and Their Derived Predictors

Although there are predictors discussed in the literature derived from different theories to account for prison misconduct (McGuire, 2018), the data set used in this study makes it possible to apply importation and deprivation theories as guides to generate predictors associated with prison misconduct. The author briefly delineates these two theories and their derived predictors in relation to prison misconduct.

The importation model argues that social interactions in the free world are imported to spaces of incarceration and greatly affect prison behavior (Irwin & Cressey, 1962). In addition, inmates’ subcultures, beliefs, and norms outside the prison are used to cope with stressful prison environments. The effect of subculture has yet to be tested on prison misconduct, however, because subculture is difficult to define and empirically quantify. Therefore, researchers commonly examine inmate characteristics or experiences before incarceration to measure the concepts of importation (Steiner et al., 2014). Examples of variables include inmates’ age, education, marital status, mental health problems, pre-prison victimization, and violence experiences and previous incarceration (Blevins et al., 2010; Jiang & Fisher-Giorlando, 2002). Age, for instance, is an important factor related to prison misbehavior in all studies, with younger inmates being more likely to engage in misconduct than their older counterparts (McGuire, 2018; Steiner et al., 2014). Contrary to importation theory, which argues that inmate behavior in the context of incarceration reflects social interactions prior to incarceration, deprivation theory posits that inmates’ maladaptive behavior, such as the use of violence, anxiety, and attempted suicide, may be due to the highly restrictive prison environment (Sykes, 1958). Indeed, prison is a setting in which the rights and privileges enjoyed in the free world, such as freedom of movement and engagement in sexual activities, are severely curtailed (Sykes, 1958). The “pains” that inmates suffer in prison are mainly due to the ubiquitous constraints of prison life. “Tensions” are released by acting out through various forms of misconduct. The common measures of deprivation are the physical conditions of the prison, its level of security, and the stringency of its management (Jiang & Fisher-Giorlando, 2002; McGuire, 2018; Steiner et al., 2014). Previous studies found that better prison conditions, less hardship, and lower levels of prison security are less likely to result in inmate violence than a poor prison environment and prison management (Steiner et al., 2014). Although commonly used, importation and deprivation theories are often criticized because the concepts are too broad to operationalize (Paterline & Petersen, 1999). Three categories have been discussed in the literature as representing the concepts embedded in the two theories: inmates’ background characteristics, and pre-prison and in-prison conditions and experiences (Jiang & Fisher-Giorlando, 2002; Steiner et al., 2014).

In accordance with the requirements of deprivation and importation theories and previous research on CODs, it was concluded in one study that male inmates with a higher risk of prison misconduct were young, less educated, and single (see a review essay by Steiner et al., 2014). Male inmates’ pre-prison experiences that were assumed to affect their behavior in prison included being convicted of violent offences, having a criminal record, and prior victimization (Houser & Belenko, 2015; Houser et al., 2012; Houser & Welsh, 2014; Steiner et al., 2014). In terms of inmates’ in-prison experiences, inmates serving longer sentences, those housed in high-security prisons, and those who negatively rate prison conditions of confinement and facility management practices are more likely to engage in prison misbehavior than those without such traits and perceptions (Jiang & Fisher-Giorlando, 2002). The presence of CODs was the major variable tested in relation to its effect on prison misconduct, while other variables noted above were used for control purposes in the statistical analyses.

Method

Sites Studied

The Taiwanese prison system is centralized and the Correctional Agency, MOJ, oversees 49 correctional institutions in Taiwan, including 24 prisons housing adult inmates. Among these 24 adult prisons, three are exclusively for female inmates and 11 are for male inmates. The remaining 10 prisons house both male and female inmates in separate areas. In Taiwan, the recruitment, training, salary, promotion, and uniforms of correctional officers are highly regulated and identical for all correctional institutions. Inmate management, the nature of daily operations, the variety of activities (e.g., entertainment), doctor visits, mandated work assignments, and the treatment of inmates are all uniformly stipulated by laws and derived institutional regulations. Unlike American prisons, which are classified with different levels of security, prisons in Taiwan have no clear security classifications and Taiwanese inmates are generally housed in a prison close to where they live (Y. Chen et al., 2014). Therefore, both male and female prisons in Taiwan incarcerate those convicted of serious, minor, violent, nonviolent, or drug-related offences.

Data Collection Procedures and Sampling

The University’s Institutional Review Board for Human Subject Research and the Corrections Agency in Taiwan approved this study prior to implementation. The study used a self-reported survey to capture a wide range of misconduct data based on official disciplinary reports. The survey questionnaire contained 12 sections of questions pertaining to the demographic background characteristics of inmates, their pre-prison experiences, their in-prison experiences, incidents of prison misconduct and prison victimization, their attitude toward prison staff, prison programs, and their mental health-related conditions before and over the course of their incarceration. Some researchers question the validity of self-reported surveys, as it is counterintuitive to ask people to candidly report their misdeeds, even though the surveys are anonymous. Thornberry and Krohn (2000) assessed the reliability and validity of the self-report method as a measurement of delinquency and criminal behavior in adults by comprehensively reviewing the literature. They concluded that crime data collected using self-reported surveys is “acceptably valid and reliable,” and that there is no fundamental difference in reliability and validity between self-reported evidence used to collect data from juvenile offenders or from adult criminals. In Taiwan, the use of a self-reported survey to collect data on deviance and crime among youth and inmates is common practice (Ma, 2001; Meng, 2017). However, no research has been performed on the reliability and validity of self-reported surveys in the specific area of crime in Taiwan. As no official data on rule-breaking among male inmates is accessible, this study uses the self-reported survey method to gather sensitive information from inmates. To increase the validity and reliability of the survey questionnaire, the principal investigator conducted a pilot study with 150 male inmates housed in three prisons not included in the current study. Based on the data collected from the pilot research, the research team examined all of the variables measured with multiple items, and the results showed that the items were internally consistent and represented the constructs of the variables. In addition, the preliminary results revealed the strong effects of the key variables on the dependent variable. The research team made several modifications in line with the results of the pilot study to improve the clarity and suitability of the survey questions. During the formal implementation of the survey, the research team strictly adhered to rigorous protocols to enhance the accuracy of the information collected.

Taiwan’s 21 prisons housing male adult inmates are dispersed inland. To maximize the representativeness of the sample with limited resources, the principal investigator diversified the geographical areas by purposively selecting two prisons located in northern, southern, and eastern Taiwan, respectively, and three in the west, resulting in nine male prisons (N = 26,270) in the study. The principal investigator asked these nine prisons to help arrange sessions in appropriate venues with the inmates for an on-site administration of the survey. On the scheduled dates, the principal investigator and six trained postgraduate students visited the selected prisons and administered the survey. Although the study was voluntary, the inmate participants were required to be able to read and write to fill in the survey questionnaire on their own. In addition, they had to have been incarcerated for more than 3 months, because those in prison for less than 3 months were still in the classification process and had not yet been assigned to a permanent cell (Y. Chen et al., 2014).

In each selected prison, the research team was escorted to a workroom or classroom where the available and willing inmates were waiting for instructions. The principal investigator explained the purpose of the study to the groups of inmate volunteers, reconfirmed their consent to participate in the study, stated the research team’s commitment to anonymity, and clarified their right to terminate their participation at any time. Meanwhile, correctional officers were politely asked to leave the survey rooms and to refrain from walking around the rooms to avoid distracting the survey participants. The research team was present to answer the questions of the inmate participants as they filled in the questionnaire. The research team also asked the participants to double-check for missing information when submitting their survey questionnaire. The survey lasted about 1 hr. Data collection began in June 2015 and ended in September 2015. A total of 2,065 survey questionnaires were collected and the data for the variables used in the study were extracted from the large-scale data set collected, using the procedures described above.

Measurements

The dependent variable was prison misconduct and the independent variable is types of disorders. Eleven variables based upon importation and deprivation theories, and the availability of the current data set, were used to control for confounding impacts on the dependent variable.

Dependent variable

The dependent variable was measured by asking the respondents about their involvement in 41 types of misconduct (see Supplemental Appendix A) since their incarceration in the prison in which they were housed. The list included a variety of inmate-on-inmate rule violations, inmate-on-staff rule violations, and general prison rule violations. Examples of misconduct items include using prohibited drugs, drinking alcohol, using objects as weapons to attack staff or other inmates, and taking other inmates’ property without permission. Due to the low frequency of some types of misdeeds (e.g., 58 inmates reported violating substance-related rules once, twice, or more than 3 times), the study grouped the inmate participants into two groups: those who reported at least one misconduct were coded as 1 and those who reported never having engaged in any type of prison misconduct were coded as 0.

Independent variable

The survey examined mental health disorders and substance abuse/dependence of inmates in the 12 months prior to their most recent imprisonment. Four categories containing 25 question items were used to collect information on the mental state of the participants, including a recent history of mental disorders diagnosed by professionals and symptoms of major depression, mania, and psychotic disorders (see detailed questions and criteria in Supplemental Appendix B). These questions were borrowed from Houser et al.’s (2012) study on CODs and institutional misconduct in U.S. prisons. They recreated questions on mental disorders based on two sources: a modified version of the Structured Clinical Interview for the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) and the Survey of Inmates in State Correctional Facilities (SISCF) of the Bureau of Justice Statistics (Bureau of Justice Statistics, n.d; James & Glaze, 2006). The survey included another set of 11 items pertaining to the survey participants’ drug use habits prior to their incarceration (see detailed question items and criteria in Supplemental Appendix C). These items were adopted from the SISCF (2004), which was based on DSM-IV (Mumola & Karberg, 2006). “Yes” and “No” response options were provided for these two sets of questions. Although the indicators of mental disorders and substance abuse adopted in this study were based on U.S. studies, they have been commonly used in medicine and psychology studies among the Taiwanese population (e.g., Y. L. Chen et al., 2019; Gau et al., 2005; Liao et al., 2012; Yang et al., 2012).

Following Houser et al. (2012), this study also divided the survey participants into four groups: no disorder, substance abuse/dependence only, mental disorders only, and CODs. Other than the types of disorder categorized by Houser et al., this study delineated three specific types of disorders—major depression, mania, and psychotic disorders—and included them in separate statistical models. Each individual model contained the same dependent and control variables but different independent variables with different types of disorders. The coding for the independent variable in the first model is no disorders (0), singular drug abuse/dependence (1), general mental health disorders (inmates reported at least one of the four categories of mental illness in the survey questionnaire) (2), and CODs (co-occurring general mental and drug abuse disorders) (3). This coding method was the same for the independent variable disorder type in Models II, III, and IV that contained inmates with major depression, mania, and psychotic disorders, respectively.

Control variables

Measurements for inmate demographic variables were age, educational attainment, and marital status. Age was measured using respondents’ actual age at the time of the survey. Education consisted of six options, ranging from elementary school to postgraduate studies, and was divided into two groups: junior high or below (0) and high school or above (1). The respondents were asked about their marital status at the time of incarceration, the options being single, divorced, cohabiting, married, widowed, and remarried. Ultimately, three categories were obtained: single (including single, widowed, and divorced) (0), cohabiting (1), and married or remarried (2). The pre-prison measures were the type of crime committed, prior criminal record, and childhood victimization. The respondents’ offences for their current conviction were classified into three broad categories: drug-related offences (e.g., illegal drug use and drug trafficking and manufacturing) (0), nonviolent offences (e.g., gambling, fraud, theft, forgery, and drunk driving) (1), and violent offences (e.g., murder, injury, robbery, extortion, and rape) (2). Asking the following question assessed the criminal history of the inmate participants: “Excluding this conviction, how many times have you been convicted before?” A continuous variable was created for prior criminal records according to times of prior convictions. Previous research has suggested that traumatic exposure early in life can have a negative effect on behavior later in life (Petersen et al., 2014). In this study, eight items were used to capture inmate experiences of violent victimization before the age of 18 (Cronbach’s α = .893; Eigenvalue = 4.597), seven items for sexual victimization (Cronbach’s α = .595; Eigenvalue = 3.121), and eight items for witnessing violence (Cronbach’s α = .802; Eigenvalue = 3.358). All these items were derived from the Juvenile Victimization Questionnaire (JVQ). The questions were reviewed and their reliability and validity confirmed with U.S. subjects (Finkelhora et al., 2005). Originally designed for juvenile participants, the JVQ was adjusted to modify the time frame (“before 18 years old”) to accommodate the current inmate adult participants. Examples of questions are as follows: “Before 18 years old, did anyone ever take your valuables from you?” “Did anyone force you to touch their private parts?” and “Did you see your parents (legal guardians) being physically beaten by each other, or by their boyfriends or girlfriends . . . trying to hurt each other?” Two response options, “Yes” and “No,” were provided for each individual item. The inmates who checked “Yes” to any item listed in the violent victimization section were coded as 1 and all “No” cases were coded as 0. The same coding method was applied to the respondents who experienced sexual victimization and witnessed violence when they were young.

The length of detention in the current prison was measured by asking the question: “How long have you been residing in this specific prison?” This measurement was a continuous variable expressed in months. Two other measurements for in-prison experiences and conditions were derived from the inmates’ perceptions of prison conditions (22 items; Cronbach’s α = .910; Eigenvalue = 7.798) and staff management practices (nine items; Cronbach’s α = .919; Eigenvalue = 5.486). Examples of statements related to prison conditions include: “The prison cell is too small” and “The food is not good.” The four optional responses were strongly agree (4), agree (3), disagree (2), and strongly disagree (1). All items were added and the higher the score, the more negative the perceptions of prison conditions. Two examples of questions used to capture inmate perceptions of staff management practices were “Correctional officers help me correct my bad habits to become a better person” and “Correctional officers’ attitude towards me is friendly.” The same four categorical responses were provided: strongly agree (4), agree (3), disagree (2), and strongly disagree (1). A nine-item additive index of staff management practices was created: the higher the score, the better the perceptions of prison correctional officers.

Data analysis

Preliminary descriptive statistics were run to examine the distribution of the dependent, independent, and control variables of the study. Bivariate analyses using the chi-square statistic were performed to test the strength of the relationship between the respondents’ disorder profile, which included no disorder, drug dependence/abuse without mental illness, four categories of mental disorders without drug abuse problems (general mental disorders only, major depression only, mania only, psychotic disorders only, and CODs) and prison misconduct (misconduct and no misconduct). Multivariate analyses with logistic regression models were then used to estimate the effects of disorder types on prison misconduct, with appropriate controls for potential intervening effects. As a result, four logistic regression models were established.

Results

Descriptive Analysis

The majority (62.5%) of the sample of 2,065 male inmates reported having been engaged in no misconduct since their present imprisonment; more than one in three (37.5%) inmates reported breaking the prison rules at least once (see Table 1). The distribution of disorder type among male inmates indicated that, except for disorder type IV, those with both mental illness (i.e., general mental illness problems, major depression, mania, and psychotic disorders) and drug abuse/dependence accounted for the highest percentage (51.4%, 37.0%, and 48.9%, respectively) (see Table 1). The average age of the inmates was approximately 38 years old and nearly 67% reported being single at the time of incarceration. The survey participants’ educational attainment was almost evenly distributed between those whose levels of education was junior high or lower (48.5%) and those above junior high (51.5%). Not surprisingly, the largest number of inmates were convicted of drug-related offences (43.7%), followed by those convicted of nonviolent (35.6%) and violent offences (20.7%).

Table 1.

The Descriptive Statistics for Male Inmates in Taiwan (N = 2,065).

Variables	M (Range)	SD	%
Dependent variables
Misconduct type
None			62.5
One misconduct			16.7
Multiple misconducts			21.8
Independent variable
Disorder type I
No disorders			12.7
Drug dependence only			9.2
General mental illness only			26.7
CODs			51.4
Disorder type II
No disorders			25
Drug dependence only			23.5
Major depression only			14.5
CODs			37
Disorder type III
No disorders			15.2
Drug dependence only			11.6
Manic disorders only			24.3
CODs			48.9
Disorder type IV
No disorders			32.2
Drug dependence only			43.4
Psychotic disorders only			7.3
CODs			17.1
Control variables
Age	38.3 (18.17–72.92)	9.50
Marital status
Single/widowed/divorced			66.2
Cohabited			14
Married/remarried			19.8
Education
Junior high or lower			48.5
High school of above			51.5
Having prior criminal records	1.99 (0–15)	1.93
Incarceration length (months)	43.6 (3–276)	39.05
Current offences
Drug-related			43.7
Nonviolent			35.6
Violent			20.7
Childhood victimization
Violent			38.3
Sexual			36.4
Witnessing violence			66.5
Perceptions for staff management	42 (15–60)	7.5
Perceptions for prison conditions	52.9 (22–88)	12.8

Note. CODs = co-occurring mental disorders and substance abuse/dependence.

Bivariate Analyses

The results of the bivariate analyses on the relationship between disorder type and prison behavior showed a great similarity across analyses (see Tables 2 –5). First, CODs-affected inmates were most likely to report having been engaged in prison misconduct (47%, 51.1%, 47.8%, and 54.4%, respectively) as opposed to those with no disorders, singular drug abuse/dependence, and singular mental illness disorders. Inmates with general mental illness disorders (35.3%), major depression (38.9%), manic disorders (36.5%), and psychotic disorders (41.8%) reported the second highest level of engagement in misconduct compared with inmates reporting no disorders. Inmates with singular drug abuse/dependence did not significantly report more misconduct-engagement than those with no disorders in the first three types of disorders shown in Tables 2 to 4 (18.7% vs. 17.6%, 29.6% vs. 24.2%, and 21.2% vs. 18.4%, respectively). Table 5 indicates that inmates with only drug abuse/dependence problems (38.1%), however, were significantly (p < .05) more likely to report being involved in misconduct than inmates with no disorders reported (26.8%).

Table 2.

Reported Misconduct by Disorder Type I.

Misconduct/No misconduct	ND (%)	DD only (%)	GMID only (%)	CODs (%)
No misconduct	82.4	81.3	64.7	53
Reported misconduct	17.6	18.7	35.3	47

Note. χ²(3, N = 1,998) = 119.327, p < .001. ND = no disorders; DD = drug abuse/dependence; GMID = general mental illness disorders; CODs = co-occurring mental disorders and substance abuse/dependence.

Table 3.

Reported Misconduct by Disorder Type II.

Misconduct/No misconduct	ND (%)	DD only (%)	MD only (%)	CODs (%)
No misconduct	75.5	70.4	61.1	48.9
Reported misconduct	24.2	29.6	38.9	51.1

Note. χ²(3, N = 1,998) = 108.558, p < .001. ND = no disorders; DD = drug abuse/dependence; MD = major depression; CODs = co-occurring mental disorders and substance abuse/dependence.

Table 4.

Reported Misconduct by Disorder Type III.

Misconduct/No misconduct	ND (%)	DD only (%)	Mania only (%)	CODs (%)
No misconduct	81.6	78.8	63.5	52.2
Reported misconduct	18.4	21.2	36.5	47.8

Note. χ²(3, N = 1,998) = 118.566, p < .001. ND = no disorders; DD = drug abuse/dependence; CODs = co-occurring mental disorders and substance abuse/dependence.

Table 5.

Reported Misconduct by Disorder Type IV.

Misconduct/No misconduct	ND (%)	DD only (%)	PD only (%)	CODs (%)
No misconduct	73.2	61.9	58.2	45.6
Reported misconduct	26.8	38.1	41.8	54.4

Note. χ²(3, N = 1,998) = 74.163, p <. 001. ND = no disorders; DD = drug abuse/dependence; PD = psychotic disorders; CODs = co-occurring mental disorders and substance abuse/dependence.

A pattern was evident from the bivariate analyses: CODs-affected male inmates in this Taiwanese sample tended to commit the most violations of prison rules, followed by inmates who reported mental illness alone. On the contrary, the significant likelihood of misconduct among inmates with singular drug abuse/dependence disorders was not consistent across types of disorders. In one case, inmates reported significantly more involvement in misconduct than no-disorders-reported inmates (Table 5), but not the other three cases (Tables 2 –4).

Logistic Regression

The effect of CODs on prison misconduct remained when control variables were introduced into multivariate statistical analyses, but there were no consistent results for singular disorders in relation to prison misconduct. Tables 6 to 9 show that relative to male inmates with no reported disorders, the odds of prison misconduct were approximately 2 times greater for inmates with CODs, Exp(B) = 2.447, 1.828, 2.591, and 1.728, respectively, than those without disorders, as expected. Male inmates with general mental illness disorders—Exp(B) = 1.968—or mania—Exp(B) = 2.276—also presented a higher risk of misconduct than their counterparts with no reported disorders. In contrast, there was no significant difference in misconduct reported for those with major depression or psychotic disorders as opposed to no disorders. The odds for inmates with drug abuse/dependence only showed no difference in reporting misconduct from those with no disorders across the four regression models. Although the odds ratio from the four models of regression analyses revealed that male inmates with CODs had a slightly higher rate of misconduct than those with only mental illnesses, the difference was not significant (p < .05).

Table 6.

The Results of Logistic Regression Predicting the Likelihood of Male Inmates Self-Reporting Prison Misconduct (N = 2,065; Disorder Type I).

Variables	B	SE	Wald	Odds ratio
Disorder type I
No disorders (reference)
Drug abuse/dependence only	−0.047	.292	0.025	0.954
General Mental illness only	0.677	.228	8.858**	1.968
CODs	0.895	.217	17.023***	2.447
Age	−0.001	.007	0.013	1.001
Marital status
Single (reference)
Cohabited	−0.203	.160	1.619	0.816
Married	−0.331	.152	4.756*	0.718
Education
Junior high or lower (reference)
Junior high or above	0.070	.114	0.380	1.073
Prior criminal records	−0.157	.136	1.333	0.854
Incarceration length	0.006	.002	7.481**	1.006
Current offences
Drug-related (reference)
Nonviolent	0.336	.134	6.266*	1.40
Violent	0.508	.145	12.257***	1.662
Childhood victimization
Violent victimization	0.768	.125	37.550***	2.156
Sexual victimization	0.271	.128	4.467*	1.311
Witness violence victimization	0.520	.146	12.712***	1.683
Perceptions for prison conditions	0.018	.005	13.340***	1.018
Perceptions for staff management	−0.021	.008	6.754**	0.963
−2 log likelihood	1,896.666
Cox and Snell R²	.166
Negelkerke R²	.226

Note. CODs = co-occurring mental disorders and substance abuse/dependence.

p < .05. **p < .01. ***p < .001.

Table 7.

The Results of Logistic Regression Predicting the Likelihood of Male Inmates Self-Reporting Prison Misconduct (N = 2,065; Disorder Type II).

Variables	B	SE	Wald	Odds ratio
Disorder type II
No disorders (reference)
Drug abuse only	0.095	.172	0.307	1.100
Major depression only	0.352	.197	3.177	1.422
CODs	0.603	.161	14.036***	1.828
Age	−0.003	.007	0.188	0.997
Marital status
Single (reference)
Cohabited	−0.217	.157	1.895	0.805
Married	−0.323	.149	4.709*	0.724
Education
Junior high or lower (reference)
Junior high or above	0.096	.111	0.756	1.101
Prior criminal records	−0.006	.031	0.039	0.994
Incarceration length	0.006	.002	8.127**	1.006
Current offences
Drug-related (reference)
Nonviolent	0.350	.131	7.116**	1.42
Violent	0.461	.142	10.522**	1.586
Childhood victimization
Violent victimization	0.761	.123	38.372***	2.140
Sexual victimization	0.258	.125	4.263*	1.295
Witness violence victimization	0.556	.142	15.296***	1.743
Perceptions for prison conditions	0.016	.005	10.333**	1.016
Perceptions for staff management	−0.024	.008	8.665*	.976
−2 log likelihood	1,979.782
Cox and Snell R²	.157
Negelkerke R²	.213

Note. CODs = co-occurring mental disorders and substance abuse/dependence.

p < .05. **p < .01. ***p < .001.

Table 8.

The Results of Logistic Regression Predicting the Likelihood of Male Inmates Self-Reporting Prison Misconduct (N = 2,065; Disorder Type III).

Variables	B	SE	Wald	Odds ratio
Disorder type III
No disorders (reference)
Drug abuse only	0.232	.253	0.842	1.262
Mania only	0.822	.214	14.762***	2.276
CODs	0.952	.201	22.405***	2.591
Age	−0.003	.007	1.92	0.997
Marital status
Single (reference)
Cohabited	−0.226	.157	2.057	0.977
Married	−0.326	.150	4.737*	0.722
Education
Junior high or lower (reference)
Junior high or above	0.084	.111	0.574	1.088
Prior criminal records	0.003	.031	0.007	1.003
Incarceration length	0.006	.002	7.188**	1.006
Current offences
Drug-related (reference)
Nonviolent	0.350	.132	7.067*	1.419
Violent	0.480	.143	11.309*	1.616
Childhood victimization
Violent victimization	0.748	.123	36.901***	2.113
Sexual victimization	0.267	.126	4.5333*	1.307
Witness violence victimization	0.524	.143	13.459***	1.690
Perceptions for prison conditions	0.017	.005	11.825**	1.017
Perceptions for staff management	−0.024	.008	8.406**	0.977
−2 log likelihood	1,965.190
Cox and Snell R²	.164
Negelkerke R²	.223

Note. CODs = co-occurring mental disorders and substance abuse/dependence.

p < .05. **p < .01. ***p < .001.

Table 9.

The Results of Logistic Regression Predicting the Likelihood of Male Inmates Self-Reporting Prison Misconduct (N = 2,065; Disorder Type IV).

Variables	B	SE	Wald	Odds ratio
Disorder type IV
No disorders (reference)
Drug abuse only	0.193	.138	1.903	1.213
Psychotic disorders only	0.187	.240	0.603	1.205
CODs	0.547	.171	10.268**	1.728
Age	−0.004	.007	0.407	0.996
Marital status
Single (reference)
Cohabited	−0.223	.157	2.011	0.800
Married	−0.320	.149	4.630*	0.726
Education
Junior high or lower (reference)
Junior high or above	0.091	.111	0.676	1.095
Prior criminal records	−0.001	.031	0.002	0.999
Incarceration length	0.005	.002	6.832**	1.005
Current offences
Drug-related (reference)
Nonviolent	0.366	.131	7.822*	1.442
Violent	0.478	.142	11.374**	1.613
Childhood victimization
Violent victimization	0.799	.122	42.879***	2.223
Sexual victimization	0.287	.125	5.307*	1.333
Witness violence victimization	0.570	.142	16.157***	1.768
Perceptions for prison conditions	0.017	.005	12.133***	1.017
Perceptions for staff management	−0.024	.008	8.721**	0.976
−2 log likelihood	1,988.883
Cox and Snell R²	.153
Negelkerke R²	.207

Note. CODs = co-occurring mental disorders and substance abuse/dependence.

p < .05. **p < .01. ***p < .001.

The results of the control variables were extremely similar across the four models of logistic regression analysis. The demographic background of male inmates (age and education) and their prior criminal record had no significant effect on their likelihood to engage in prison misconduct in these four models. The remaining control variables in the models all showed a significant effect on prison behavior. The male inmate participants who were married at the time of incarceration had a lower risk of misconduct in prison than their single counterparts, Exp(B) = 0.718, 0.724, 0.722, 0.726, p < .05, respectively, as expected. The participants who stayed in prison longer were more likely to break the rules, Exp(B) = 1.006 or 1.005, p < .001, respectively, than those in prison for a shorter period of time. In addition, the offences for their current conviction had a significant effect on how they acted in prison. Inmates convicted of violent or nonviolent offences for their present incarceration were about twice as likely—for violent offences: Exp(B) = 1.662, 1.586, 1.616, 1.613, respectively; for nonviolent offences: Exp(B) = 1.40, 1.42, 1.419, and 1.442, respectively—to engage in prison misconduct compared with inmates convicted of drug-related offences. Consistent with the predictions of proponents of the importation theory, victimization experiences of male inmates at a young age affected their future behavior in prison. Those who were previously victims of physical or sexual violence, or had witnessed violence in their youth, reported committing approximately 1.307 to 2.223 times more misconduct in prison than those who did not report childhood victimization experiences. The more negative the inmates’ perceptions of prison conditions, the more likely they were to become involved in misconduct (Tables 6 –9). Conversely, the better their perceptions of staff management, the less likely they were to report engagement in misconduct in prison, providing further evidence to support the deprivation theory.

Discussion and Conclusion

Although the proportion of prisoners with co-occurring mental illness and drug abuse/dependence disorders has been relatively high in both American and Taiwan’s prisons, this particular type of inmates is rarely studied (Steiner et al., 2014). Thus, this study made the first attempt to examine the effect of CODs on prison misconduct among Taiwanese male inmates. In contrast to prior studies, this study uniquely used a measure of the specific mental illness disorders (i.e., major depression, mania, and psychotic disorders), in addition to an analysis of combined mental illness disorders, to examine their effect on prison misconduct. The study had two purposes. First, it sought to find the independent effect of singular mental illness and substance disorders CODs on prison misconduct. Second, it analyzed whether CODs constituted the highest risk of prison misconduct. The results were consistent with prior research. The following section discusses the significant results, beginning with the descriptive analyses and followed by bivariate and multivariate analyses.

The descriptive analyses indicated that some of the characteristics among the 2,065 male inmates in the study were, to some degree, similar to those in the male inmate population, although the sample was not selected in a random manner. For example, 20.7% were convicted of violent offences (e.g., robbery, murder, abduction, sexual assault, and extortion), which was approximately the same proportion as the male prison population of all prisons in Taiwan (20.9%) at the end of August 2015 (MOJ, 2017) when the study was conducted. Drug-related prison convicts accounted for 43.7% of prisoners in the survey population, slightly less than Taiwan’s national prison population (45%) (MOJ, 2017). The 30- to 49-year-old age group was slightly higher among drug-related inmates in the survey population (72.8%) than in the national male prison population (69.55%). Similarly, 81.9% of the surveyed inmates had a prior criminal record, a figure slightly higher than the national male prison population (80.8%) (MOJ, 2017).

As for the types of disorders, 17.1% to 51.4% of the present sample reported CODs and 7.3% to 26.7% of the inmates surveyed reported mental health problems. The high rate of CODs and mental illness disorders among male inmates the author found in this study echoes Houser et al.’s (2012) study on CODs. Drawing data from 2,930 female inmates in the United States, Houser et al. (2012) found that mental disorders accounted for 21.5% and CODs for 42.7% among inmates sampled. They suspected that the large number of inmates with mental illness disorders might reflect the limited resources for or accessibility to mental health treatments outside of the prison, in the community. The prevalence of CODs-affected inmates has profound implications for correctional institutions in Taiwan.

Bivariate analyses showed that the group of CODs inmates had the highest risk of misconduct. Similar results were found in two studies on American female inmates, concluding that CODs-affected inmates tended to commit serious misconduct and be subject to severe discipline relative to inmates with no disorders or singular disorders (Houser et al., 2012; Houser & Welsh, 2014). Inmates with singular mental health problems were generally likely to commit misconduct compared with those with no disorders, a conclusion that supports findings of prior studies. Although prior studies indicated that female prison inmates with singular drug abuse/dependence did not commit more serious misconduct or face more severe censure than female inmates with no known disorders (Houser et al., 2012; Houser & Welsh, 2014), the current study on male inmates also found similar results, which suggests no significant difference existed in self-reported misconduct between inmates with substance abuse/dependence and with no disorders, except for the inmates in disorder type IV (Table 5).

Multivariate analyses indicated that male inmates with CODs had the highest likelihood of violating prison rules, for all types of disorders. Previous studies in the United States also found that CODs-affected inmates were likely to report engaging in serious misconduct, to be disciplined, and to be disciplined with serious sanctions (Houser et al., 2012; Houser & Welsh, 2014). Inmates with singular mental illness disorders, however, did not show consistent patterns in the four multivariate analytical models in the present study. Interestingly, the findings from previous studies in the United States and this study both showed that inmates with drug abuse/dependence were not more likely to report engagement in more serious misdeeds than their counterparts with no reported disorder (Houser et al., 2012; Houser & Welsh, 2014). As prior studies did not offer explanations for this result, the author speculates that problems of drug abuse/dependence prior to incarceration among inmates might actually be better managed in the prison setting than the other two types of disorders. It is not difficult to understand why prison inmates are prohibited from using illicit drugs in prison. Thus, it is likely that the high degree of control in the prison setting helps inmates correct their drug abuse behavior.

In terms of the control variables, the same nine variables across the four multivariate analysis models showed a significant relationship with misbehaviors, supporting predictions by importation and deprivation theories. Male inmates who were married, were incarcerated longer, were non-drug-related convicts, had not faced childhood victimization, and had positive attitudes toward prison conditions and staff management reported less engagement in misconduct than those with opposite characteristics and experiences, consistent with expectations.

Like prior research, this study clearly showed that although mental disorders have a negative effect on inmates’ behaviors to some degree, CODs exacerbated the likelihood of engaging in risky behavior, suggesting that COD-affected individuals in the prison population should be given priority for monitoring, counseling, and participation in prison programs. Traditional approaches to treating people with dual and co-occurring disorders are based on the notion that a clear division of labor must be pursued (Osher, 2013), which explains why resources and care for mental illness and drug abuse/dependence problems are offered separately. Two common adverse consequences of this approach are that COD-affected inmates are excluded from treatments or are randomly “shuffled” between these two systems (Osher, 2013, p. 1). The literature on co-occurring disorders suggests that integrated treatments and programs with a multidisciplinary team of staff in a single setting are more effective for treating COD-affected people. These treatment programs have been widely accepted (Houser et al., 2012; Osher, 2013). Intake screening and assessment of inmates with CODs is essential for good practice in corrections. Clinical determination of mental and/or drug abuse disorder problems in the early stages of incarceration can reduce prison misconduct, and lead to safer prison environments and a greater likelihood of “correcting” inmate behavior (Sacks et al., 2007).

Few studies have investigated the risk of misconduct among prison inmates with CODs, and this study was the first to examine the effects of individual mental illnesses on prison misconduct. Overall, the results corroborated those of previous studies conducted in the United States, providing a clear message that inmates with CODs pose challenges to correctional management and safety of both staff and inmates. Although this study produced new knowledge to better understand prison misconduct by Taiwanese male inmates, it is not devoid of limitations.

First, the method used to identify disorders among inmate participants consisted of self-report survey items adapted from previous studies. The classification of the cases into mental disorders, drug abuse/dependence, and CODs was also based on self-classification, as opposed to professional diagnoses. In addition, questions about prison misconduct and inmates’ previous victimization experiences were relatively sensitive. Therefore, underreporting may have occurred among the survey participants. Some of the childhood victimization question items for the male adult participants were borrowed from the JVQ for juvenile victimization. The JVQ was slightly modified to suit the adult participants, but these items were primarily based on U.S. juveniles. It is suggested that future studies develop childhood victimization indicators tailored to Taiwanese adults. Furthermore, the questions on mental disorders and drug abuse/dependence did not inquire about severity levels. Similarly, items pertaining to prison misconduct did not exhaustively list all types of prison misconduct, although the study listed more possible misconduct types than previous research. Finally, caution should be exercised when interpreting the results because the sample of inmates in the study was not randomly selected. Nevertheless, several characteristics of the inmates in the sample matched those of the total male prison population in Taiwan.

The prison population has a large number of inmates with CODs, and the few studies on the topic have consistently reported that misconduct is common among COD inmates. Given the scale of this problem for correctional facilities, scholars should contribute to the literature on co-occurring disorders in the criminal justice system. More correctional administrators and prison staff in Taiwan should understand the potential threat to prison safety posed by COD inmates. This threat has been largely ignored in Taiwan. In other countries, various types of comprehensive and integrated treatment are shown to be effective for high-risk offenders and for reducing recidivism (see Osher, 2013; Van Dorn et al., 2017). Taiwanese policymakers should thus pilot evidence-based treatment and programs to help Taiwanese inmates with CODs.

Supplemental Material

Appendix_A – Supplemental material for The Effects of Mental Health and Substance Abuse/Dependence Disorders on Prison Misconduct Among Male Inmates in Taiwan

Supplemental material, Appendix_A for The Effects of Mental Health and Substance Abuse/Dependence Disorders on Prison Misconduct Among Male Inmates in Taiwan by Shih-Ya Kuo in International Journal of Offender Therapy and Comparative Criminology

Footnotes

Acknowledgements

I greatly appreciate the two anonymous reviewers and the associate editor, Carlo Garofalo, for providing me with constructive comments and opinions and for attempting to help me refine the original manuscript.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the University of Macau under grant MYRG2014-00049-FSS.

ORCID iD

Shih-Ya Kuo

Supplemental Material

Supplemental material for this article is available online.

References

American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.).

Andreoli

S. B.

dos Santos

M. M.

Quintana

M. I.

Ribeiro

W. S.

Blay

S. L.

Jose Geraldo

V. T.

de Jesus Mari

(2014). Prevalence of mental disorders among prisoners in the State of Sao Paulo, Brazil. PLOS ONE, 9(2), Article e88836. https://doi.org/10.1371/journal.pone.0088836

Blevins

Listwan

Cullen

T. F.

Jonson

(2010). A general strain theory of prison violence and misconduct: An integrated model of inmate behavior. Journal of Contemporary Criminal Justice, 26, 148–166. https://doi.org/10.1177/1043986209359369

Bureau of Justice. (n.d.). Data collection: Survey of inmates in state correctional facilities (SISCF). Retrieved from https://www.bjs.gov/index.cfm/content/dataonline/content/acf/index.cfm?ty=dcdetail&iid=275

Butler

Allnutt

(2003). Mental illness among New South Wales’ prisoners. New South Wales Corrections Health Service.

Chang

L. F. K.

(2013). The reduction of drug abuse in Taiwan and its information system application—An IT enabled service innovation in E-Government. Journal of Service Science and Management, 6, 250–255.

Chen

(2016). Harm reduction policy in Taiwan: Toward a comprehensive understanding of its making and effects. Harm Reduction Journal, 13(11), 1–8.

Chen

Lai

Lin

(2014). The impact of prison adjustment among women offenders: A Taiwanese perspective. The Prison Journal, 94(1), 7–29.

Chen

Y. L.

Chen

W. J.

Lin

K. C.

Shen

L. J.

Gau

S. S.

(2019). Prevalence of DSM-5 mental disorders in a nationally representative sample of children in Taiwan: Methodology and main findings. Epidemiology and Psychiatric Sciences, 2019, 1–9.

10.

Chiang

Chan

Chen

Sun

Chang

Chen

W. J.

Lin

Chen

(2006). Recidivism among male subjects incarcerated for illicit drug use in Taiwan. Psychiatry and Clinical Neuroscience, 60(4), 444–451.

11.

Chiles

J. A.

Von Cleve

Jemelka

R. P.

Trupin

E. W.

(1990). Substance abuse and psychiatric disorders in prison inmates. Hospital and Community Psychiatry, 41, 1132–1134.

12.

Connor

D. P.

Tewksbury

(2016). Inmates and prison involvement with drugs: Examining drug-related misconduct during incarceration. Journal of Contemporary Criminal Justice, 32(4), 426–445.

13.

Constantino

de Assis

S. G.

Pinto

L. W.

(2016). The impact of prisons on the mental health of prisoners in the state of Rio de Janeiro, Brazil. Ciencia & Saúde Coletiva, 21(7), 2089–2100.

14.

Cossar

Stoové

Kinner

Dietze

Aitken

Curtis

Kirwan

Ogloff

J. R. P.

(2018). The associations of poor psychiatric well-being among incarcerated men with injecting drug use histories in Victoria, Australia. Health & Justice, 6(1), 1–8. https://doi.org/10.1186/s40352-018-0059

15.

Cramer

(2016). The prevalence of mental disorders among convicted inmates in Norwegian prisons. The Centre for Research and Education in Forensic Psychiatry. http://sifer.no/files/Hel_oppdatert_Victoria_Cramer_rapport_engelsk.pdf

16.

Fazel

Hayes

A. J.

Bartellas

Clerici

Trestman

(2016). The mental health of prisoners: A review of prevalence, adverse outcomes and interventions. Lancet Psychiatry, 3(9), 871–881.

17.

Finkelhora

Hambyb

S. L.

Ormroda

Turnera

(2005). The juvenile victimization questionnaire: Reliability, validity, and national norms. Child Abuse & Neglect, 29, 383–412.

18.

Gau

S. S.

Chong

M. Y.

Chen

T. H. H.

Cheng

A. T. A.

(2005). A 3-year panel study of mental disorders among adolescents in Taiwan. The American Journal of Psychiatry, 162, 1344–1350.

19.

Houser

Belenko

(2015). Disciplinary responses to misconduct among female prison inmates with mental illness, substance use disorders, and co-occurring disorders. Psychiatric Rehabilitation Journal, 38(1), 24–34.

20.

Houser

K. A.

Belenko

Brennan

P. K.

(2012). The effects of mental health and substance abuse disorders on institutional misconduct among female inmates. Justice Quarterly, 29(6), 799–828.

21.

Houser

K. A.

Welsh

(2014). Examining the association between co-occurring disorders and seriousness of misconduct by female prison inmates. Criminal Justice and Behavior, 41(5), 650–666.

22.

Irwin

Cressey

D. R.

(1962). Thieves, convicts, and the inmate culture. Social Problems, 10(2), 142–155.

23.

James

D. J.

Glaze

L. E.

(2006). Mental health problems of prison and jail inmates (NCJ 213600). Bureau of Justice Statistics, U.S. Department of Justice.

24.

Jiang

(2005). Impact of drug use on inmate misconduct: A multilevel analysis. Journal of Criminal Justice, 33, 153–163.

25.

Jiang

Fisher-Giorlando

(2002). Inmate misconduct: A test of the deprivation, importation, and situational models. The Prison Journal, 82(3), 335–358.

26.

Liao

Lai

Liu

(2012). Low prevalence of major depressive disorder in Taiwanese adults: Possible explanations and implications. Psychological Medicine, 42, 1227–1237.

27.

(2001). Meta- analysis of criminal theories: In terms of social evolutionism and behavior reinforcement. Educational and Social Studies, 2, 35–66.

28.

Matejkowski

(2017). The moderating effects of antisocial personality disorder on the relationship between serious mental illness and types of prison infractions. The Prison Journal, 97, 202–223.

29.

McGuire

(2018). Understanding prison violence: A rapid evidence assessment. HM Prison and Probation Service, Ministry of Justice. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/737956/understanding-prison-violence.pdf

30.

Meng

(2017). Across crime. Wu Nan Publisher.

31.

Ministry of Justice. (2017). Statistics of ministry of justice. http://www.rjsd.moj.gov.tw/rjsdweb/common/WebList3.aspx?menu=INF_COMMON_C

32.

Morse

K. C. B.

(2017). Mental health in prisons. National Audit Office. https://www.nao.org.uk/wp-content/uploads/2017/06/Mental-health-in-prisons.pdf

33.

Mumola

C. J.

Karberg

J. C.

(2006). Drug use and dependence, state and federal prisoners, 2004. U.S. Department of Justice.

34.

O’Keefe

M. L.

Schnell

M. J.

(2007). Offenders with mental illness in the correctional system. Journal of Offender Rehabilitation, 45(1/2), 81–104.

35.

Osher

F. C.

(2013). Integrating mental health and substance abuse service for justice-involved persons with co-occurring disorders. SAMHSA’S Gains Center for Behavioral Health and Justice Transformation. https://www.samhsa.gov/gains-center

36.

Paterline

B. A.

Petersen

D. M.

(1999). Structural and social psychological determinants of medium security prison. The Social Science Journal, 35, 253–263.

37.

Peng

E. Y.

Lee

Morisky

D. E.

Yeh

Farabee

Lan

Chen

Y. A.

Lyu

(2010). Psychiatric morbidity in HIV-infected male prisoners. Journal of the Formosan Medical Association, 109(3), 177–184.

38.

Peng

E. Y.

Yeh

C. Y.

Lyu

S. Y.

Morisky

D. E.

Chen

Y. M.

Lee

M. B.

Farabee

Malow

R. M.

(2010). Prevalence and correlates of lifetime suicidal ideation among HIV-infected male inmates in Taiwan. AIDS Care, 22(10), 1212–1220.

39.

Peters

R. H.

Bartoi

M. G.

Sherman

P. B.

(2008). Screening and assessment of co-occurring disorders in the justice system. CMHS National GAINS Center.

40.

Peters

R. H.

LeVasseur

M. E.

Chandler

R. K.

(2004). Correctional treatment for co-occurring disorders: Results of a National Survey. Behavioral Science & the Law, 22, 563–584.

41.

Peters

R. H.

Wexler

H. K.

Lurigio

A. J.

(2015). Co-occurring substance use and mental disorders in the criminal justice system: A new frontier of clinical practice and research. Psychiatric Rehabilitation Journal, 38(1), 1–6.

42.

Petersen

A. C.

Joseph

Feit

(2014). Consequences of child abuse and neglect. In A. C. Petersen, J. Joseph & M. Feit (Eds.), New directions in child abuse and neglect research (pp. 111–155). Washington, DC: National Academies Press.

43.

Prince

J. D.

Wald

(2018). Risk of criminal justice system involvement among people with co-occurring severe mental illness and substance use disorder. International Journal of Law and Psychiatry, 58, 1–8.

44.

Prins

S. J.

(2014). Prevalence of mental illnesses in U.S. State prisons: A systematic review. Psychiatric Services, 65(7), 862–872.

45.

Sacks

Melnick

Coen

Banks

Friendman

P. D.

Grella

Knight

(2007). CJDATS Co-occurring Disorders Screening Instrument (CODSI) for Mental Disorders (MD): A validation study. Criminal Justice and Behavior, 34(9), 1198–1215.

46.

Senior

Birmingham

Harty

M. A.

Hassan

Hayes

A. J.

Kendall

. . . Webb

(2013). Identification and management of prisoners with severe psychiatric illness by specialist mental health services. Psychological Medicine, 43, 1511–1520.

47.

Steiner

Butler

H. D.

Ellison

J. M.

(2014). Causes and correlates of prison inmates misconduct: A systematic review of the evidence. Journal of Criminal Justice, 42, 462–470.

48.

Steiner

Wooldredge

(2009). Individual and environmental effects on assaults and nonviolent rule breaking by women in prison. Journal of Research in Crime and Delinquency, 46(4), 437–467.

49.

Steiner

Wooldredge

(2014). Comparing self-report to official measures of inmate misconduct. Justice Quarterly, 31(6), 1074–1101.

50.

Sung

Mellow

Mahoney

A. M.

(2010). Jail inmates with co-occurring mental health and substance use problems: Correlates and service needs. Journal of Offender Rehabilitation, 49, 126–145.

51.

Sykes

(1958). The society of captives. Princeton University Press.

52.

Thornberry

T. P.

Krohn

M. D.

(2000). The self-report method of measuring delinquency and crime. In U.S. National Institute of Justice (Ed.), Measurement and analysis of crime and justice: Criminal justice series (Vol. 4, pp. 33–83). National Institute of Justice.

53.

Torrey

E. F.

Zdanowicz

M. T.

Kennard

A. D.

Lamb

H. R.

Eslinger

D. F.

Biasotti

M. I.

Fuller

D. A.

(2014). The treatment of persons with mental illness in prisons and jails: A state survey. Treatment Advocacy Center.

54.

Treatment Advocacy Center. (2016). Serious Mental Illness (SMI) prevalence in jails and prisons (Criminalization). Office of Research & Public Affairs.

55.

Tung

Hsiao

Shen

Huang

(2019). The prevalence of mental disorders in Taiwanese prisons: A nationwide population-based study. Social Psychiatry and Psychiatric Epidemiology, 54, 379–386.

56.

Van Dorn

R. A.

Desmarais

S. L.

Rade

C. B.

Burris

E. N.

Cuddeback

G. S.

Johnson

K. L.

Mueser

K. T

. (2017). Jail-to-community treatment continuum for adults with co-occurring substance use and mental disorders: Study protocol for a pilot randomized controlled trial. Trials, 18(1), Article 365.

57.

Wood

S. R.

(2018). Co-occurring serious mental illnesses and substance use disorders as predictors of assaultive infraction charges among adult male jail inmates. The Journal of Forensic Psychiatry & Psychology, 29(2), 189–210.

58.

Wright

E. M.

Salisbury

E. J.

Van Voorhis

(2007). Predicting the prison misconducts of women offenders: The importance of gender-responsive needs. Journal of Contemporary Criminal Justice, 23(4), 310–340.

59.

Yang

Yeh

Hwu

(2012). Prevalence of schizophrenia, bipolar disorder, and depressive disorders in community between Taiwan and other countries. Taiwanese Journal of Psychiatry, 26(2), 77–87.

60.

Tsay

(2013). Current status of substance abuse and HIV in Taiwan. Journal of Food and Drug Analysis, 21(4), S27–S32.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.34 MB