Mood disorders among adolescents in conflict with the law and in custody

Abstract

The high prevalence of psychiatric symptoms among juvenile delinquents is a well-replicated international finding. This study aimed to find the prevalence of mood disorders and their relationship with serious criminal acts in a population of adolescents in conflict with the law and in custody. A total of 123 male inmates aged 14 to 17 years were interviewed and assessed. Mood disorders were diagnosed in 15% of the sample for current episode and 31% for lifetime, making them third most prevalent after dependence disorders and disruptive disorders. The psychopathological profile of the adolescents who had committed violent crimes corroborates other studies reporting a high prevalence of mood disorders in this population. Several factors have been found to influence the formation of juvenile delinquency, including absence of family structure, social inequality, lack of quality school education, alcohol and drug abuse/addiction and disruptive disorders. The present results confirm mood disorders as another such factor.

Introduction

Adolescence is one of the most important periods of human development, due to various bio-psychosocial transformations. It is a phase that can sometimes be marked by changes in behavior as the child becomes an adult, developing great conflicts and hormonal changes (Schoen-Ferreira et al., 2010). Adolescent offenders are a population associated with a high prevalence of psychiatric diagnoses (Colins et al., 2010). In most cases, these disorders are neither identified nor adequately treated; consequently, many disorders can become chronic. Epidemiological data on the adult prison population are robust (S. Fazel & Baillargeon, 2011), including in Brazil, and provide a clear profile of inmates (Andreoli et al., 2014). By contrast, information on adolescents in custody is lacking in both developed and developing countries. There are significant differences in patterns of psychiatric morbidity among adolescent offenders exposed to psychiatric services compared to adult offenders (Vreugdenhil et al., 2004). There is a growing body of scientific evidence to demonstrate the relationship between mood disorders, violent behavior and consequent conflict with the law (Teplin et al., 2002; Tijssen et al., 2010); however, although mood disorders are one of the relevant elements that predispose adults to violent behavior, research demonstrating this relationship in adolescents is scarce even in developed countries.

People with psychiatric disorders are more often victims of violence than perpetrators of violent behavior (Higgins et al., 2005). However, in some mood disorders such as the manic phase of bipolar disorder, such individuals are more prone to engaging in violent behavior, sexual promiscuity and confrontations with authorities due to the irritability, impulsivity, increased energy, euphoria and feelings of omnipotence characteristic of this phase (Fovet et al., 2015).

This study aims to find the prevalence of mood disorders and their relationship with serious criminal acts in a population of adolescents in conflict with the law and in custody in a juvenile detention center. The secondary objective is to map the profiles of other psychiatric disorders among this population. Other variables such as age, residential environment, education, school dropout status, hobbies, family history of crime and lifetime prevalence of psychiatric diagnoses will serve to map the profiles of this population.

Method

The sample comprises 123 male adolescents aged 14 to 17 years who were serving a socio-educational order imposing deprivation of liberty at the detention facilities of the Centro de Socioeducação para menores infratores (CENSE [Socio-Educational Center for Juvenile Offenders]). This facility is located in the city of Piraquara within Greater Curitiba, Paraná state, Brazil and has a current capacity for 75 adolescents. The inclusion criteria were: adolescents who were sentenced to detention in or temporarily detained at the CENSE; and both the adolescent and their legal guardian must provide written consent to take part in the study. The exclusion criteria included individuals diagnosed with intellectual disabilities and/or pervasive developmental disorders and individuals aged 18 years and over.

Instruments

The psychiatric diagnoses were evaluated by the lead researcher, who is experienced in applying the Schedule for Affective Disorders and Schizophrenia for School-Age Children – Present and Lifetime (K-SADS-PL; Kaufman and Schweder, 2004), a semi-structured psychiatric interview to screen for disorders in 6- to 18-year-olds by assessing the Axis I disorders (mental health and substance use disorders) of the DSM-IV and ICD-10 for children and adolescents (American Psychiatric Association, 2000; World Health Organization, 2004). This diagnostic tool is subdivided into three main parts. The first part of the interview entails the identification and collection of demographic data, general medical history and current medical care, development, psychiatric history, school history and social relationships. The second part screens for 82 symptoms to detect key symptoms for past and current episodes in 20 different diagnostic areas. The third part consists of complementary diagnostic scales (e.g. other affective, psychotic, anxiety, behavioral and substance abuse disorders) quantifying confirmatory diagnostic symptoms for which the screening was positive.

Statistical analysis

The study design is quantitative and cross-sectional. Statistical Package for the Social Sciences (SPSS) v23.0 was used for the construction of the statistical calculations. The level of statistical significance adopted is .05.

The dependent variables in this study are dichotomous (presence or absence of the disorder). The following predictive models of mood disorder were employed: depression, bipolar disorder type 1 and bipolar disorder type 2. The independent variables are categorical and the answers are based on the participants’ reports, including residential environment, education, school failure and current and previous offenses. Kolmogorov–Smirnov and Shapiro–Wilk tests were used to determine whether or not the continuous variables have a normal distribution. The chi-square test was applied to determine whether or not the absolute frequency observed at the intersection of some variables differs significantly from the distribution of the expected absolute frequency and to assess whether or not there is a statistically significant association between the mood disorders and serious infractions (comprising homicide, attempted homicide and armed robbery).

Results

Of the 126 adolescents interviewed, 3 were excluded, giving a final sample of 123. The sociodemographic data of the participants are described in Table 1. In relation to the educational level of the sample, most of the participants did not finish elementary school and most (70%) were not attending school regularly during their imprisonment.

Table 1.

Sociodemographic data of the participants.

Age (years)	Minimum 14	Maximum 17	Mean 16.2
Ethnicity	White (48%)	Brown (35%)	Black (15%)
Residential environment	Living with one parent (51%)	Living with both parents (21%)	Not living with either parent (14%)
Educational level*	9 years (26%)	8 years (22%)	10 years (22%)
Hobbies	Sports (38%)	Videogames (15%)	Visual/performing arts (8%)

Note: *Number of years of formal study in a school environment.

The offenses for which the current socio-educational orders with deprivation of liberty were served included drug trafficking (27%), homicide (18%), theft (16%), robbery (14%), armed robbery (9%), attempted homicide (9%), fraud (3%) and others (4%). Just over half of the participants (55%) had a family member who had prior involvement in criminal activities, and 92% of the participants were repeat offenders (i.e. they had a history of conflict with the law).

One quarter of the participants had at least one close relative diagnosed with one or more mood disorder. Only 11% of the participants did not meet the diagnostic criteria for previous mood disorders. The prevalence of previous diagnoses were drug abuse/addiction (35%), mood disorders (31%), conduct disorder (30%), alcohol abuse/addiction (22%), oppositional defiant disorder (19%) and anxiety disorders (6%; see Table 2). These manifested alone or as comorbid disorders. Only 55% of the participants previously diagnosed with a psychiatric disorder had received some form of psychotherapeutic treatment or drug therapy in the past – that is, despite presenting with psychiatric symptoms for a long time, they had never sought out mental health services to relieve these symptoms.

Table 2.

Current and lifetime prevalence of psychiatric disorders in the participants.

Psychiatric d isorder	Current (%)	95% CI	Lifetime (%)	95% CI
Disruptive d isorders	31	30.1–33.2	50	41.1–57.2
Conduct disorder	19	15.7–22.3
Oppositional defiant disorder	12	10.2–14.1
Alcohol and drug abuse/addiction	22	18.9–24.8	58	50.9–64.8
Drug abuse/addiction	14	10.7–17.3
Alcohol abuse/addiction	8	7.0–10.9
Mood d isorders	15	12.3–19.1	31	26.3–37.1
Depression	6	4.8–8.7
Bipolar disorder type I	5	4.1–7.2
Bipolar disorder type II	3	2.3–4.8
Hyperkinetic d isorders	5	3.2–8.1
Anxiety d isorders	4	2.9–6.9	6	3.9–9.7
Psychotic d isorders	1	0.5–2.1

Note: CI = confidence interval. Lifetime column grouped by syndrome.

The disorders in italic and consequently their values in bold are the Syndromes (Group of diseases). The other elements in the table are the specific diseases.

In terms of the current diagnoses, 43% of the participants had been diagnosed with a single psychiatric disorder and 30% had been diagnosed with more than one disorder (see Table 3). Conduct disorder was the most prevalent diagnosis (26%), followed by drug abuse/addiction (19%), oppositional defiant disorder (17%), alcohol abuse/addiction (12%), depression (8%), bipolar disorder type I (8%), attention deficit hyperactivity disorder (ADHD; 7%), anxiety disorders (5%), bipolar disorder type II (4%) and psychotic disorders (1%; see Table 2).

Table 3.

Prevalence of current psychiatric disorders among the participants.

Psychiatric disorders	n	%
Single	54	43
Multiple (comorbidity)	37	30
None diagnosed	32	26

Three quarters of the participants satisfied the criteria for at least one psychiatric diagnosis, but only 64% were receiving psychiatric medications at the time of the interview. A point of interest is that some of the participants who did not have a current or previous psychiatric diagnosis were also receiving medication, some of whom claimed that the medication was being administered to them as punishment for bad behavior.

Conversely, some of the participants who were clearly symptomatic were not receiving any treatment, either psychotherapeutic or medical.

A weak but statistically significant relationship was found between mood disorders and serious infractions (homicide, attempted homicide and armed robbery), which was one of the objectives of this research: χ² (2) = 4.408, p < .001.

Discussion

When examining the variables that predispose adolescents to antisocial behaviors, parental conflicts/rows and delinquency of a close family member (e.g. father or brother) are prevalent (Martins & Pillon, 2008). In the present sample, 55% of the adolescents had a close family member with prior involvement in criminal activities and almost all of them (92%) were repeat offenders, i.e. they had a history of conflict with the law. The high prevalence of repeat offenders is associated with the fact that adolescents who are deprived of their liberty have generally served a previous socio-educational order (DiLalla et al., 1988).

The most recent national data on the population serving socio-educational orders in Brazil are presented in a 2016 annual survey (Ministry of Human Rights, 2018). This survey describes a total of 26,450 patients, of whom 18,567 were inmates (70%), 2178 were detained under a semi-liberty regime (8%) and 5184 were on temporary admission (20%). The survey indicates that 47% of all infractions in Brazil are classified as theft-related, compared to 31% in the present study. The national data also show that 22% of adolescents nationwide and 16% in Paraná state were served socio-educational orders for committing infractions related to drug trafficking, compared to 27% in the present study. Homicide-related infractions recorded nationwide account for 16% of all infractions committed, compared to 22% in Paraná state and 18% in the present study (Ministry of Human Rights, 2018). One of the largest international studies, involving a sample of 3058 Swedish individuals, found a prevalence of 32% of infractions related to homicide in the population of young offenders assessed (M. Fazel et al., 2008).

The homicide-related infractions differ significantly both in the national survey between states and in the present study. These disparities might be explained by the fact that the use of lethal force and violence in Brazil varies widely across states. In the present study, the high prevalence of homicide-related offenses can be explained by the fact that this detention center – as a better equipped, more secure institution that is closer to the state capital – receives a higher proportion of serious offenders.

In the present sample, the vast majority of the adolescents (88%) met diagnostic criteria for previous mood disorders and, of these, around half of these 88% had received some form of treatment (psychotherapeutic or medical). In a sample of 1420 American children involved in a longitudinal study with a follow-up of approximately 10 years (Azlin et al., 2010), more than half of the individuals who committed crimes as adults had no history of psychiatric disorders, and many who had psychiatric disorders in adolescence had no conflict with the law in adolescence or adulthood. Taken together, these findings support the hypothesis that psychiatric disorders in adolescence are just one of many variables influencing criminality. Thus, meeting the mental health needs of children and adolescents would not prevent all young people from engaging in criminal activity (Copeland et al., 2007). However, the mental health system, the judiciary and society as a whole would benefit from better identification and treatment of adolescents with psychiatric disorders.

In a large national study of 898 Brazilian adolescents including 18-year-olds, the most prevalent psychiatric disorders in those deprived of freedom due to conflict with the law were disruptive and personality disorders (28%), psychiatric disorders resulting from alcohol and drug abuse/addiction (27%) and mood disorders (15%; see de Pinho et al., 2006). In the present sample, the same psychiatric disorders proved to be the most prevalent, with 31% diagnosed with disruptive disorders, 22% with alcohol and/or drug abuse/addiction and 21% with mood disorders. These are the most prevalent disorders found in this population by a wide range of studies conducted on all continents (de Andrade et al., 2011; Dias et al., 2014; Ministry of Human Rights, 2018). However, as observed in the Brazilian samples, the prevalence of these three disorders can exhibit significant variability (Domalanta et al., 2003). Some factors of the characteristics of the population being investigated – e.g. female samples, individuals held under semi-liberty conditions, detainees who have recently been admitted to an institution and repeat offenders, among others – are confounding. However, these biases are expected, having been observed in studies since the 1980s (Timmons-Mitchell et al., 1997).

There are significant differences in patterns of psychiatric morbidity in adolescents living in the community compared with those in custody. Among the general population, anxiety disorders are more prevalent compared to mood disorders (4% to 31% compared to 3% to 21%; see Kessler et al., 2007) for adolescents in custody the mood disorders are more prevalent (Teplin et al., 2002).

Compared with adults in custody, young people aged 15 to 21 years show higher rates of depression and lower rates of psychosis, bipolar disorders and chemical dependency (M. Fazel et al., 2008). Disruptive disorders and alcohol and drug abuse/addiction, due to their high prevalence in the population of adolescents in conflict with the law, are often regarded as diagnoses inherent to this population (Kessler et al., 2007). Thus, mood disorders are the most prevalent disorder in these individuals, justifying the relevance of this diagnosis (see Table 2).

In the present study, a statistically significant association was found between mood disorders and serious infractions. More than half of the sample diagnosed with mood disorders (55%) had committed serious infractions such as homicide. Furthermore, 36% of these adolescents had comorbid alcohol and/or drug abuse/addiction, corroborating studies demonstrating an association between mood disorders and violent crime (Teplin et al., 2002; Tijssen et al., 2010). However, in cases of comorbidity with alcohol and/or drug abuse/addiction, this association is more robust, demonstrating that addiction can act as a risk factor for committing violent crimes in adolescents with mood disorders.

In Brazil, according to the law which enacted the Sistema Nacional de Atendimento Socioeducativo (SINASE [National System of Socio-Educational Assistance]; see S. Fazel et al., 2010), article 64 states: ‘every adolescent serving a socio-educational measure presenting signs of mental disorder, mental disability, or associated disorders, should be evaluated by a multidisciplinary and multi-sectoral technical team’. The aforementioned law also states that the judge may suspend the execution of the socio-educational measure with a view to including the adolescent in a comprehensive mental healthcare program that best meets the therapeutic goals established for each specific case. This law is supposed to protect adolescents’ rights to a dignified approach to mental health, although in practice this has not been the case.

In the present sample, 64% of the adolescents were undergoing psychiatric treatment using medication, yet only 60% had a psychiatric diagnosis that was eligible for drug treatment – that is, some of the adolescents were receiving medication without a supporting psychiatric diagnosis. A hypothesis for this pattern of prescribing lies in the culture of using psychiatric medication to treat behavioral problems and conduct disorders that do not respond to drug treatment – and the participants themselves reported that psychiatric medications (particularly high-potency, first-generation antipsychotics) were often prescribed as punishment for misbehavior. The reverse of this situation was also true, since 7% of the adolescents had been diagnosed with a current psychiatric disorder that is amenable to drug treatment but were receiving no treatment – neither drugs nor psychotherapy. It should be noted that psychotherapy is not currently offered at the CENSE, and that the psychologists who work there do not implement psychotherapeutic treatments; rather, they only periodically assess the behavior of the inmates in order to find out who is able to obtain freedom. The drug treatment at the facility is also archaic and inefficient; at the time the study was conducted there was only one medical doctor on staff who had no training in psychiatry or mental health.

The limitations of this study include sample biases such as the diagnostic exclusion criteria of intellectual deficits and global developmental disorders. In addition, no female adolescents were included in the sample, precluding extrapolation of these results to female populations. Some psychiatric disorders that have a low prevalence of 1% to 4% need larger samples to yield reliable estimates; small samples such as the one used in this study may introduce selection bias. It has also been shown that adolescents – especially those with disruptive disorders – have a tendency to minimize their symptoms, predominantly when reporting mood complaints, and so underestimation of some disorders may have occurred. Psychoeducational programs could be implemented in socio-educational centers in order to help young offenders to understand that depressive symptoms are not a sign of weakness and are amenable to treatment. Being detained in custody and deprived of freedom can worsen pre-existing psychological symptoms or contribute to triggering them, so it is essential that professionals are trained on this aspect of working with incarcerated youths.

Conclusion

This study confirms the results of other research that has found a significant prevalence of mood disorders among adolescents in conflict with the law and in custody. Several factors are responsible for the formation of the individual in adolescence and consequently for whether or not they will end up in conflict with the law, and mood disorders can be regarded as one of the factors that predisposes adolescents to delinquent behavior. Mood disorders were found to be highly prevalent in the present sample, with 15.9% being diagnosed during the current episode, making them the most prevalent diagnosis after disruptive disorders and dependence disorders. The psychopathological profile observed in relation to the adolescents who committed violent crimes (homicide, attempted homicide or armed robbery) also confirms the results of other studies that have found a high prevalence of mood disorders in this cohort.

Although this study, along with others, shows an association between mood disorders, violent behaviors and consequent conflict with the law (Teplin et al., 2002; Tijssen et al., 2010), this association cannot be considered causal. Mood disorders can, however, be seen as one of several elements predisposing to offending behavior in adolescents.

On the other hand, some psychiatric symptoms may occur in reaction to the deprivation of liberty (Whitaker et al., 1990), and regardless of the etiology, adolescents in conflict with the law are at high risk for psychiatric disorders – even at a young age. Further studies, particularly longitudinal, are needed to determine why some adolescents with psychiatric disorders or more specifically mood disorders commit infractions in their youth and throughout their adult lives whereas others do not. Future work assessing whether or not adolescents with psychiatric disorders in conflict with the law are more prone to committing crimes as adults is also warranted.

Most socio-educational centers around the world where adolescents remain deprived of their liberty are ill-prepared to provide adequate mental health services for the large proportion who have psychiatric disorders (Penner et al., 2011). This shortcoming is critical and further attention should be devoted to this topic, with more research aimed at guiding mental health policies and better understanding the complex interaction between adolescence, mental health and the justice system. Failure to diagnose psychiatric disorders can deprive young offenders of appropriate treatment, which can lead to recurrences of criminal behavior. This research is intended to alert those involved in youth justice systems and society in general about the prevalence of mood disorders in delinquent adolescents and their relationship with violent crime.

Ethical standards

Declaration of conflicts of interest

Eduardo Alves Guilherme has declared no conflicts of interest.

Ricardo Alberto Moreno has declared no conflicts of interest.

Ethical approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the State Department of Justice, Labor and Human Rights of the Department of Social and Educational Services of the State of Paraná, Brazil [permit no. 14705540] and the University of Sao Paulo Research Ethics Committee [permit no. 2579568] and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

Informed consent

Informed consent was obtained from all individual participants included in the study and from their legal guardians who signed a consent form on behalf of the participants.

Funding Statement

Funding for this study was provided by the São Paulo Research Foundation (the Fundação de Amparo à Pesquisa do Estado de São Paulo [FAPESP; grant no. 2018/04981-0]).