Recidivism Rate Among the Juvenile Population with Mental Health Issues

Introduction

Mental health issues are very common among children and youths in the juvenile justice system. Although it is difficult to effectively rehabilitate these groups of persons, it is very important to develop effective ways of rehabilitating and treating them (Hovey et al., 2017). Multiple programs have, in the past, been used to rehabilitate juveniles with psychiatric illnesses. Nonetheless, the size of the juvenile population with mental illness continues to take an upward trend. As of 2013, offenders with mental health issues made up about 50% of the U.S prison population, with a significant part of this population requiring mental health services (Russell, 2017). Roughly half to three-quarter of the 2 million youths in the juvenile justice system fulfill the criteria for mental health disorder (Russell, 2017). Between 40 to 80 percent of the juvenile population having a minimum of one diagnosable mental health disorder.

Incarceration of juveniles without taking them through proper mental health support increases their likelihood of being repeat offenders upon the transition into the community. In other words, the lack of mental health support for juveniles who are diagnosed with behavioral and emotional disorders that need psychiatric services results in recidivism (Russell, 2017). To effectively prevent recidivism, there is a need to set up provisions to address the primary risk factors that are prevalent among juvenile delinquents suffering from mental health problems. To propose effective prevention programs, it is imperative to assess the risk factors associated with recidivism. According to “The Juvenile Justice and Delinquency Prevention Act of 1974,” the primary objective of juvenile justice was to prevent the transition of juvenile offenders to the adult justice system (Russell, 2017). While juvenile facilities have psychiatrists who conduct medical assessments, most of these facilities do not offer in-depth psychiatric stabilization to help address youths’ needs in a therapeutic environment. The study, through the analysis of the recidivism rate among juveniles with mental illness, can help outlines the relevant factors and how to manipulate these factors to reduce the rate of recidivism.

Study Purpose/Research Objectives

The objective of this quantitative research is to explore the aspects that intensify the risk of juvenile recidivism among children and youth with mental issues. Through the analysis of archive data obtained from the Harris County Juvenile Probation Department (HCJPD), various risk factors are assessed to investigate the correlation between mental health disorder and recidivism (Russell, 2017). The primary objective of the research is to explore the link between juveniles and their associated characteristics, such as age, gender, ethnicity, mental health treatment records, type of crime committed, and their duration of attending the psychiatric unit, to facilitate the description of the state of recidivism among the juvenile population with mental health issues.

Research Questions and Hypotheses

Research Question 1

Is there a connection between the type of mental health issue and reentry into juvenile detention facilities within one year of attending psychiatric sessions?

The primary hypotheses associated with the above research question include:

Null hypothesis, H1o : There is no correlation between the kind of mental illness and reentry into the juvenile justice system for at least a misdemeanor within the first year of leaving a psychiatric center (Russell, 2017). 

Alternative hypothesis, H1a : There is a correlation between the type of mental illness and recidivism due to at least a misdemeanor within the first year of leaving a psychiatric center (Russell, 2017).

Research Question 2

Can the duration of attending psychiatric centers and mental health diagnoses be used to accurately foreshadow the probability of reentry into the juvenile justice system for at least a misdemeanor within one year of leaving a psychiatric center, beyond the effect of statistically significant covariates?

The primary hypotheses associated with the above research question include:

Null hypothesis, H2o : The duration of attending psychiatric centers and mental health diagnoses cannot be employed accurately to foreshadow the probability of reentry into juvenile correctional facilities for at least a misdemeanor beyond the effect of statistically significant covariates (Russell, 2017).

Alternative hypothesis, H2a : The duration of attending psychiatric centers and mental health diagnoses can be used accurately to predict the probability of reentry into juvenile detention facilities for at least a misdemeanor beyond the influence of statistically significant covariates (Russell, 2017).

Methodology

Harris County International Review Board archived data was used for this study. The archived data employed in the study were specifically for youths who attended the Harris County Psychiatric Center (HCPC) (Russell, 2017). To acquire the rate of recidivism, the archived data from HCPC, the archived data were fed into the database used by the “Harris County Juvenile Probation.” From 2007-2015, the data about subjects who had attended HCPC were acquired and fed into the “Harris County Juvenile Probation” (Russell, 2017). That way, it was possible to group the data into two, namely participants who reentered the juvenile detention facility within the time frame and participants who were not arrested. The data in the HCPC database linked to participants who were detained were applied to collect the selected risk aspects, particularly age, gender, ethnicity, mental health diagnosis, duration of attending the psychiatric hospital, and criminal offense.

Variables

The archived data provided vital information about various variables, which include the duration of attending the mental health facility, age, gender, ethnicity, criminal offense, and mental health diagnosis. The variables were grouped into independent variables (IVs), covariates, and dependent variables, as shown below:

IVs . There are two IVs in this study, namely, mental health issue or diagnosis and stay duration at the psychiatric facility. While diagnosis is given during admission and discharge, only the diagnosis at discharge was used in the study. The duration of attendance at the psychiatric center is measured by the number of days the subject takes attending the psychiatric center.

Covariate . Covariates for this research are age, ethnicity, gender, and criminal offense. Age is measured and described as the age of the subject upon release from the psychiatric center. Gender and ethnicity are defined as the participant’s sex and race, respectively. The criminal offenses are grouped into either a misdemeanor or a higher offense.

Dependent Variable . Recidivism was the sole dependent variable for this research. Recidivism, in the context of the study, is defined as any participant who successfully reentered the juvenile detention facility as a result of a misdemeanor or a higher offense within the first year of discharge from the psychiatric center.

Statistical Test Used to Test Hypotheses

The “interrupted time series” (ITS) analysis is a vital research design used in quasi studies. It analyses the longitudinal effects of interventions through regression modeling. ITS plays a significant role in the analysis of observational data that lack randomization or a case-control design (Kontopantelis et al., 2015). Aside from fully utilizing the longitudinal nature of data, the selected method also considers the pre-intervention trends. In the analysis of the second hypothesis of the study, regression analysis can be employed to evaluate the association between the reentry and covariates of interests, including age, ethnicity, gender, and criminal offense.

For the hypothesis associated with the research question 1, one chi-square analysis will be conducted to investigate the relationship between recidivism within the first year and mental health diagnosis. The mental health diagnosis variable represents the independent variable for the first research question. Therefore, to test the first null hypothesis, a chi-square analysis was completed for the IV of mental health diagnosis (Russell, 2017). A binary logistic regression model was applied to analyze the second research question and its associated hypotheses. The binary logistic regression model is based on the logit combination of predictor variable values. The analysis relied on the X 2 coefficient instead of the F coefficient that is associated with linear regression (Russell, 2017). In sum, the testing of the hypotheses relied on regression-based ITS to quantify the impact of mental health disorder and duration of attending the psychiatric center on reentry into the juvenile detention facility.

Analysis/Results

Table 1

Demographic Information

Demographic	Population Size, n	%
Sex Female Male	375 627	37.4 62.6
Race Black Latino White Other	455 280 260 7	45.4 27.6 25.9 0.7
Mental health Diagnosis Behavioral Disorder Mood Disorder Substance abuse disorder Thought Disorder Other	265 614 17 32 10	26.4 62.3 1.7 3.2 1.0
Recidivated within one year No Yes	857 145	85.5 14.5
Recidivism offense (n=145) Misdemeanor Felony	99 46	9.9 4.6

Table 2

Demographic Data for the Sample’s Continuous Descriptors (Russell, 2017)

Demographic	Min.	Max.	M	SD
Age at release (Years)	11	17	15.14	1.23
Time in psychiatric facility (Months)	2	182	39.78	23.97
Days from release to recidivism (n=145)	5	365	145.58	94.76

The primary result of the research includes recidivism within one year and the likely covariates, including age, sex, race, and criminal offenses. To investigate the aforementioned variables, various analyses were conducted. For the continuous variables, a t-test was conducted, with chi-square analysis being conducted on a categorical variable. The outcome of the t-test for age was statistically significant, t (1000) = 6.85, p < .001, showing that it is not only explainable by chance, and thus included in the analysis of recidivism (Russell, 2017). The chi-square analyses of sex to be employed as a covariate predicting recidivism with one year was also statistically significant X 2 (1) = 437, p= .037, suggesting that sex ought to be used as a covariate. A chi-square analysis on race showed that it was statistically significant with X 2 (2) = 6.31, p = .043.

For the first research question, a chi-square analysis outcome was statistically nonsignificant with X 2 (3) = 3.59, p = .309. This means that recidivism within one year is not related to mental health diagnosis during release. Therefore, the null hypothesis of the study cannot be rejected. As such, mental health diagnosis cannot be used in the analysis of the second research question.

Table 3

Chi-Square Analysis of Mental Health Diagnosis and Recidivism (Russell, 2017)

Recidivism within one year	Behavior Disorder	Mood Disorder	Other	Thought disorder
No	219 [228]	536 [528.2]	24 [23.2]	28 [27.5]
Yes	46 [37]	78 [85.8]	3 [3.8]	4 [4.5]

For the second research question, the analysis depends on previous analyses. The outcome of the binary logistic regression was statistically significant, X 2 (4) = 54.19, p < .001, indicating that a logit combination of age, sex, and race can be used to predict the probability of reentry into the juvenile detention facility. 85.4% of the participant in the sample were accurately predicted using the logistic regression model. The examination shows an inverse relationship between recidivism and age and a positive relationship between recidivism and sex. Black participants are also more likely to recidivate compared to other races.

Table 4

Classification Table for Logistic Regression Predicting Recidivism (Russell, 2017)

Observed		Predicted Recidivated within one year		% Correct
		No	Yes
Recidivated within one year	No Yes	850 144	1 0	99.9 0.0
Overall Percentage				85.4

Table 5

Finding for Each Predictor for Logistic Regression Predicting Recidivism

Variables	B	S.E.	Wald	d f	p	O.R.	C.I for O.R.
							Lower	Upper
Age	-0.466	0.072	42.26	1	.001	0.627	0.545	0.722
Sex	0.494	0.202	6.005	1	.014	1.639	1.104	2.434
Race (black)	0.607	0.243	6.22	1	.013	1.835	1.139	2.956
Race (Latino)	0.265	0.274	0.934	1	.334	1.303	0.762	2.229

Implications

The data acquired from the study can be used to track juveniles under system supervision, particularly behavioral health improvement, and subsequently, the reduction of the rate of recidivism among the juvenile populations. Through the identification of the aspects that contributes to the increase in the risk of recidivism among youth delinquents who are diagnosed with mental illness, the study can help in the development of preventive measures that will facilitate the intervention of risk factors identified to be statistically significant in the analysis (Palmer, 2019). The adoption of such intervention with the risk factors in mind can help reduce or prevent juvenile delinquency. The outcome of the study can be used as a reference to guide in the adoption of programs meant to prevent criminal behaviors among youths, such as multi-systematic therapy and functional family therapy. Although the aforementioned programs are meant to prevent criminal behavior and recidivism, they have failed in the past. Rather, clinicians should focus on ways to improve the mental health of the juvenile population along with the improvement of their personal trajectory, which helps in the reduction of delinquency (Russell, 2017). Furthermore, the outcome of the study can help guide the aftercare process upon the release from juvenile detention facilities. Follow-up activities after release increase the chances of successful juvenile stabilization.

Recommendation

The primary recommendations from the study include (Walsh et al., 2014):

Measuring recidivism among the juvenile population with the objective of identifying the multiple ways in which such a group may have subsequent contact with the juvenile system.

Analysis of recidivism data to explore the risk factors, along with the primary characteristics and variables.

The development and maintenance of infrastructure to facilitate the collection, analysis, and report of recidivism data.

The application of recidivism data to guide the policies, practices, and resource allocation in the juvenile system.

Limitations

The analysis of the study is based on a single juvenile detention facility and a single psychiatric center. Rather than treating a mental disorder, the facilities are only involved in the stabilization of youth. The services offered by the psychiatric center are only temporary, with no assurance that there will be a continuation of outpatient visits. Furthermore, the study does not include a violation of probation and technical violations in its variables. Also, there are instances where external factors may contribute to criminal behavior among juveniles. Failure to include external factors, such as childhood upbringing, socioeconomic status, and family support, among other factors, may affect the credibility of the study outcome.

Conclusion

Based on the outcome of the study, the primary risk factors that affect juvenile recidivism include age, gender, and race. Mental health disorders and the duration of attending psychiatric facilities were also variables in the study. Although the aforementioned variable does not have a significant effect on the continued criminal behavior among the juvenile population who have undergone psychiatric stabilization, it is vital to expose this population to ongoing mental health services. Interventions developed to address criminal justice, and the mental health needs of juvenile offenders can help reduce criminal recidivism. The outcome of the study can help in the development as well as in the implementation of programs designed to address the factors that lead to juvenile recidivism and thus minimize the rate of reentry into the juvenile justice systems. 

References

Hovey, K. A., Zolkoski, S. M., & Bullock, L. M. (2017). Mental Health and the Juvenile Justice System: Issues Related to Treatment and Rehabilitation.  World Journal of Education ,  7 (3), 1-13.

Kontopantelis, E., Doran, T., Springate, D. A., Buchan, I., & Reeves, D. (2015). Regression based quasi-experimental approach when randomisation is not an option: interrupted time series analysis.  bmj ,  350 , h2750.

Palmer, R. P. (2019).  A Cognitive-behavioral Intervention and Juvenile Recidivism: An Administrative Data Analysis  (Doctoral dissertation, Ashford University).

Russell, K. C. (2017). Recidivism Rates Among Juveniles With Mental Illness.

Walsh, N., Weber, J., John D. and Catherine T. MacArthur Foundation, & United States of America. (2014). Measuring and using juvenile recidivism data to inform policy, practice, and resource allocation.