crime and mental illness essay

An official website of the United States government

The .gov means it’s official. Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

The site is secure. The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

• Publications
• Account settings

Preview improvements coming to the PMC website in October 2024. Learn More or Try it out now .

• Advanced Search
• Journal List
• Psychiatr Psychol Law
• v.24(1); 2017

Does Mental Illness Impact the Incidence of Crime and Victimisation among Young People?

Emma l cashman.

a School of Psychology, University of Wollongong, Northfields Avenue, Wollongong, NSW, Australia

Stuart DM Thomas

b Legal Intersections Research Centre, University of Wollongong, Northfields Avenue, Wollongong, NSW, Australia

c Justice and Legal Studies, School of Global, Urban and Social Studies, RMIT University, Melbourne, VIC, Australia

d Department of Psychiatry, Southern Clinical School, Monash University, Melbourne, VIC, Australia

While the high prevalence of offending and victimisation among young people is well established, no study to date has compared official crime records of young people with mental illness with those without mental illness. This case linkage study sought to determine whether young people with a formal history of mental illness were more likely to have official histories of offending and victimisation than young people who had no recorded histories of mental illness. Results suggested that young people with a history of mental illness are particularly vulnerable to violence. While a weak association was established between violent offending and mental illness, a stronger relationship was found between victimisation and mental illness, especially violent victimisation. Consistent with literature regarding the victim–offender overlap, a history of offending placed young people at a substantially higher risk of victimisation. Implications of these results are discussed in the context of how public mental health services could better protect these vulnerable young people.

While considerable research has examined the association between violent offending and mental illness, less attention has focused on the arguably stronger association between crime victimisation and mental illness, especially in the case of young people (Choe, Teplin, & Abram, 2008 ). The notion that an individual who commits a crime has a greater chance of being a victim of crime, and a victim of crime has a greater chance of committing a crime, is commonly referred to as the ‘victim–offender overlap’ (Cops & Pleysier, 2014 ). The victim–offender overlap is a robust criminological finding that dates back over half a century (e.g., von Hentig, 1948 ; Wolfgang, 1958 ). This overlap has been consistently found across time, subgroups and through various methodologies used to record and define offending and victimisation (Lauritsen & Laub, 2007 ). However, little research has focused on this apparent overlap: rather, the focus of research efforts to date has more frequently been isolated to offending or victimisation samples separately (e.g., Cops & Pleysier, 2014 ).

Empirical research has reported a strong reciprocal relationship between criminal victimisation and offending in young people (Chang, Chen, & Brownson, 2003 ; Smith & Ecob, 2007 ). The majority of the research in this area has focused on the shared socio-demographic characteristics of offenders and victims – i.e., the risks of being young, male, unemployed and unmarried. However, the question of whether victimisation occurs prior to offending or whether offending occurs prior to victimisation remains inconclusive. A greater understanding of the trajectories of youth violence is necessary, especially given the array of adverse short- and long-term consequences for those who are victims, perpetrators and perpetrating victims (Schlack & Petermann, 2013 ); gaining an understanding of their initial contacts with the criminal justice system would allow greater opportunity for potential intervention and preventive efforts. Knowledge of young people's typical ‘crime route’ may, additionally, help to target services where they would receive the maximum benefit. With respect to criminal offences, more recent evidence appears to support the notion that offending behaviour leads to subsequent victimisation in young people (Lopez & Emler, 2011 ; Ousey, Wilcox, & Fisher, 2010 ; Smith & Ecob, 2007 ).

Youth Crime

Young people are disproportionately represented in criminal offending statistics (Cops & Pleysier, 2014 ; Shaffer, 2003 ). In line with the ‘age-crime curve’, offending generally peaks in late adolescence (i.e., between 15 and 19 years of age) and declines dramatically in early adulthood (i.e., approximately 25 years of age) (Sweeten, Piquero, & Steinberg, 2013 ). A breakdown of these offence statistics in the Australian context, for example, indicates that approximately one in five young offenders commit theft or public order offences, one in six commit acts intended to cause injury (i.e., assault), and a smaller proportion commit drug-related offences (ABS, 2011 ). These figures are comparable to American statistics of juvenile offending, which report theft as the most common offence, followed by property crime and physical assault (e.g., Snyder & Sickmund, 2006 ).

Mental Illness among Offending Populations

The rate of mental illness among youth offenders has been found to be substantially higher than the rate among non-offending young people in the community (Fazel, Doll, & Langstrom, 2008 ). The prevalence of mental illness among young offenders differs greatly across the literature due to variations in sampling and the methods used to determine diagnosis. For instance, it has been reported to be anywhere between 51.9% and 89% of young people in secure juvenile justice facilities have at least one mental health diagnosis (Karnik et al., 2009 ; Teplin, Abram, McClelland, Dulcan, & Mericle, 2002 ; Wasserman, McReynolds, Schwalbe, Keating, & Jones, 2010 ). This compares to estimates that around 30% of young people in the community have used public mental health services (Rosenblatt, Rosenblatt, & Biggs, 2000 ), and approximately half report having had at least one psychiatric diagnosis (Copeland, Miller-Johnson, Keeler, Angold, & Costello, 2007 ).

Nature of Offending among Young People with Mental Illness

Young people with mental illness are often stereotyped and portrayed in the mass media as dangerous and a threat to society; however, there is little evidence to support this assertion (Coker, Smith, Westphal, Zonana, & McKee, 2014 ). In fact, there is limited published research on the nature of crimes committed by young people with mental illness; rather, much of the literature concerning youth has focused on elucidating the association between specific psychiatric disorders and crime. The available research suggests that externalising disorders, particularly conduct disorder and substance use disorders, are the most prevalent psychiatric disorders found among young offenders (Fazel et al., 2008 ; Schlack & Petermann, 2013 ).In research that has examined the nature of offences among population-based samples, the presence of a psychiatric diagnosis in young people is significantly associated with greater levels of all types of crimes, with violent and property crimes exhibiting the strongest associations among those with a mental illness (Coker et al., 2014 ; Elonheimo et al., 2007 ). While there is a limited literature reporting crimes committed by young people with mental health issues, there is even less on the association between crime victimisation and mental illness, which is arguably a more significant concern (Choe et al., 2008 ; Maniglio, 2009 ).

Youth Victimisation

The available literature suggests that young people are more likely to be victimised than any other age group (Cops & Pleysier, 2014 , Lauritsen, Sampson, & Laub, 1991 ; Ousey et al., 2010 ). For instance, young people are reportedly the victims of all forms of violence at much higher rates than adults, with some figures suggesting that the prevalence estimates are at least 10-fold higher (Danielsson, Blom, Nilses, Heimer, & Hogberg, 2009 ). While there is some research reporting that adults with mental illness are more commonly the victims of violent offences than are people without mental illness in the general population (Hodgins, Alderton, Cree, Aboud, & Mak, 2007 ; Maniglio, 2009 ), it remains unclear whether the risk between mental illness and victimisation applies equally or differentially across the lifespan. Of note, a recent study by Cater and colleagues (Cater, Andershed, & Andershed, 2014 ) found that mental health problems were significantly associated with a broad range of victimisation incidents among young people. Young people were more likely to be victimised during adolescence than during childhood, and more likely to report multiple incidents of victimisation. Specifically, they found that mental health and offending behaviour were overrepresented in young people who had experienced any type of victimisation.

Study Rationale and Hypotheses

Taken together, while there is a well-known overlap between offending and victimisation, little attention has been paid to this association among young people, nor to consideration of how mental illness potentially impacts on this association. Young people with a mental illness are over-represented in offending populations, with some studies alluding to the notion that young people with mental illness are more likely to commit violent crime and non-violent crime. Further, while young people are known to be more vulnerable to crime victimisation, it is unknown in the adult literature whether there is a greater risk of crime victimisation among those with mental illness. Against this background, the current study sought to use official records to investigate the prevalence and nature of offending and victimisation in a sample of young people with and without an established psychiatric history who have had police contact.

Based on previous research, it was hypothesised that: (1) young people would, on average, have contact with police as offenders prior to any contacts as crime victims; (2) young people with an offending history would be significantly more likely to have a victimisation history than those without an offending history; (3) young people with a history of mental illness would be more likely to have a violent and non-violent offence history than those without a psychiatric history; and (4) young people with a mental illness will be more likely to have a greater number of victimisation incidents and a violent victimisation history than young people without mental illness.

This design involved linking contact-level data from three electronic databases: (1) the Victorian Police Incident Fact Sheet Database (IFS database); (2) the Victorian Police database Law Enforcement Assistance Package (LEAP database); and (3) the Victoria Department of Human Services Psychiatric Case Register (VPCR database). The IFS database contains police reports of all incidents where police have had contact with the public. There are a variety of circumstances surrounding the police incidents (e.g., transport incidents, police pursuits and murder), and many reports do not result in the arrest of a suspect/offender. Each incident includes a narrative concerning the lead-up, the incident itself, incident outcomes and the name of offenders if known to police. A random sample of 20% of all IFS incidents during a 12-month period (July 2012 to June 2013) was extracted, leading to a potential sample of 3,088 IFS events, which formed the source database for the linkage process. From these incidents it was possible to identify 626 known offenders; these were then sorted by date of birth to identify persons aged 25 years or younger. The final eligible sample was comprised of 159 young people. For the purposes of this research, a ‘young person’ was defined as someone aged 25 years or less. Such a broad characterisation of young people is based on well-known definitions of ‘youth’, youth crime trajectories and the onset of most mental health problems in youth, which aligns with contemporary mental health service models (ABS, 2011 ; Sweeten et al., 2013 ).

The LEAP database records all contacts a person has had with Victorian police as witnesses, victims, suspects and offenders. It has been in its current form since October 1993. Information was extracted from the LEAP database regarding the frequency and type of offences, police contacts, warning flags, and victimisation incidents as well as the age of first contact with each instance. Offences were grouped into three broad categories based on previous research (Short, Thomas, Luebbers, Mullen, & Ogloff, 2013 ): violent, non-violent and sexual offences. Violent offences included offences where there was physical contact or harm (e.g., assault or armed robbery), including contact sexual offences (e.g., rape or indecent assault); sexual offences included all contact and non-contact sexual acts that were non-consensual or where consent is proscribed; non-violent offences included any offence that did not involve physical contact with a victim (e.g., property and theft). Field contacts are recorded by police officers when they approach someone and request their details because the circumstances appear suspicious or because they have contact with an individual in a specified incident (CIRCA, 2013 ). Field contacts can include person checks, vehicle checks and search warrants. Warning flags refer to narratives that police attach to individuals in order to identify risks and vulnerabilities, to help police better interact with such people. For instance, warning flags include use of weapons, arson or self-harm. Warning flags and field contacts were simply coded as being present or absent as a broad indication of police contact and prior knowledge of a person's police and mental health contact histories. Victimisation incidents were recorded in the same three broad categories; victim violent offence, victim non-violent offence and victim sexual offence.

Mental health diagnoses and age of first public mental health contact were extracted from the Victorian Psychiatric Case Register (VPCR) database. The VPCR database records all contact of the public with the statewide public mental health service, including psychiatric diagnoses, admissions, receipt and duration of treatment services. Diagnoses are recorded on the VPCR according to the International Classification of Diseases (WHO, 1992 ). Given that this database does not include individuals’ contact with primary care services (i.e., General Practitioner contacts) or the private mental health sector (e.g., private practitioners or private hospitals), such high-prevalence disorders should be considered a lower prevalence estimate (Kesic, Thomas, & Ogloff, 2012 ). ICD-10 category diagnoses were extracted for those who had a contact history recorded. Only those who had a diagnosis that fell within the ICD-10 category diagnoses (F0–F09, F10–F19, F20–F29, F30–F39, F40–F48, F50–F59, F60–F69, F70–F79, F80–F89, F90–F98) were included in the mental illness sample.

Data Linkage Process

Identifying information (full name, gender and date of birth) from the IFS database was linked with the statewide criminal records (LEAP) and public mental health databases (VPCR) via deterministic and probabilistic matching. Once extracted, the data were de-identified, and matches were then merged onto MS Access and SPSS Version 22 working files. The data-linkage methodology is a robust methodology that is consistent with guidelines proposed by the Australian National Health and Medical Research Council (NHMRC, 2007 ). Ethics was approved for this study by the University of Wollongong Human Research Ethics Committee (host institution), and the data linkage was approved and supported by Victoria Police Human Research Ethics Committee.

Approach to Analysis

Data were analysed using IBM SPSS Version 22. Descriptive statistics were performed to establish the prevalence of those with a recorded mental illness and compare the frequency of offences, police contact and victimisation incidents between samples of those with and without a mental health diagnosis. Independent t -tests were performed to compare the number of offences, victimisation incidents and warning codes as well as the average age of offending and victimisation between samples. Paired t  tests were conducted to compare age of first contact with police and public mental health services. Binary logistic regression was used to determine the likelihood of having a victimisation history between samples while controlling for the potential confounders of age and gender. Chi-squared tests of association were performed and odds ratios were obtained to examine categories within violent offences, non-violent offences and violent victimisation, between samples. Finally, effect sizes were calculated and presented alongside odds ratios for statistically significant group differences.

Sample Characteristics

The sample consisted of 159 young people; they were predominantly male (86%; n = 136), with an average age of 19.49 years ( SD = 3.76). Approximately one in three ( n = 57, 36%) of the sample had an officially recorded mental health diagnosis with public mental health services. The most common diagnostic groups were behavioural disorders, substance use disorders and anxiety disorders ( Table 1 ). There were no statistically significant differences across diagnostic groups for males and females with the exception of personality disorders, whereby the odds were six times higher for females (χ 2 = 13.32, p < .0001). There were no statistically significant differences according to age (19.6 years, SD = 3.88 vs. 19.4 years, SD = 3.71), t (157) = 3.33, p = 0.742, or sex (females 11.8% vs. males 19.3%), χ 2 = 1.68, p = .195, between those with and without a formal mental health history, respectively.

Prevalence of psychiatric diagnoses by gender.

Note: diagnoses are not mutually exclusive and therefore exceed the number of people with a mental health diagnosis ( N = 57).

Criminal History and Police Contacts

The majority of the sample ( n = 153, 96%) had an official criminal history. Those who had a mental health diagnosis had significantly more offences recorded ( M = 61.35, SD = 48.70 vs. M = 37.07, SD = 36.15), t = 3.58, p <.001, and more warning flags ( M = .95, SD = .67 vs M = .57, SD = .61), t = –F3.65, p <.001; they were younger at time of first offence ( M = 14.31, SD = 2.61 vs. M = 15.41, SD = 3.51), t = 2.06, p = .04, than those without a formally recorded mental health diagnosis. Young people with a mental health diagnosis were more than eight times more likely to have a violent offence history than young people without a mental health diagnosis (AOR = 8.18, 95% CI [1.04, 64.26], p = .046). However, statistically controlling for age and gender, the strength of this association weakened ( Table 2 ). Further exploration of offences that comprised the violent category indicated that threatening behaviour was the only offence significantly more likely among males with a mental health diagnosis. No statistical association was apparent between the presence of a mental health diagnosis and sexual offences or non-violent offences. Post-hoc analyses of the non-violent offences suggested several between-group differences, most commonly for theft, property damage and offences against the government. Individuals with a mental health diagnosis were 8.46 times more likely to commit theft, χ 2 = 10.72, p = .001, 5.20 times more likely to commit an offence against the government, χ 2 = 11.92, p = .001, and 3.54 times more likely to commit property damage, χ 2 = 7.45, p = .006, than those without a mental health diagnosis. Police were 9.80 times more likely to report a field contact with a person with a mental health diagnosis and 3.47 times more likely to place a warning flag on the database of someone with one. The three most common warning flags recorded against young people's criminal records were related to arson, weapons and drugs, with the former being three times more likely to be found in those with a mental disorder, χ 2 = 8.26, p = .004.

Prevalence and odds ratios for criminal offences and police contact.

a Adjusted odds ratio controlling for the effects of age and gender. b Adjusted odds ratio controlling for the effects of age. Note hyphens are provided for instances where odds ratios could not be calculated.

Victimisation History

Young people with a diagnosed mental illness were three times more likely to have an official police history of criminal victimisation than those with no such diagnosis. They also had a significantly greater number of victimisation incidents ( M = 2.75, SD = 3.46 vs. M = 1.29, SD = 1.60), t = –F3.63, p < .0001. After statistically controlling for the potentially confounding effects of age and gender, violent victimisation offences were significantly more common among those with a mental health diagnosis; no such association was found for sexual and other non-violent victimisation experiences. Once data were statistically controlled for age and gender, being the victim of assault was the only violent offence that remained significantly more likely to be found among those with a mental health diagnosis ( Table 3 ).

Prevalence and odds ratios for crime victimisation.

a Adjusted odds ratio controlling for the effects of age and gender.

p refers to the AOR significance level.

Victimisation and Offending

The sample of those who had officially recorded contacts with police as offenders was significantly younger than that of contacts as victims ( M = 14.45, SD = 3.05 vs. M = 15.54, SD = 4.36), t = 2.96, p = .004. When samples were examined separately, significant findings emerged only for young people with a history of mental illness. Specifically, young people were significantly younger when they first had contact with public mental health services ( M = 13.37, SD = 5.15 vs. M = 15.37, SD = 4.04), t = –F2.33, p = .025, and with police as offenders ( M = 13.93, SD = 2.22 vs. M = 15.37, SD = 4.04), t = 2.55, p = .014, than their age at first contact with police as victims.

In consideration of the entire sample of young people, victimisation and offending were significantly related. Those with a criminal offending history were more than 10 times more likely also to have a victimisation history, χ 2 = 6.78, p = .009. Those with a violent offending history were more than six times more likely also to have a violent victimisation record, χ 2 = 6.91, p = .009, four times more likely also to have a violent offending history, χ 2 = 6.29, p = .012, and ten times more likely also to have a non-violent offending history, χ 2 = 6.78, p = .009.

This study utilised data arising from a case-linkage methodology to compare official records of criminality and victimisation between a sample of young people with and without histories of mental illness. Young people with a history of mental illness had significantly more recorded offences and were more likely to have had field contacts (sub-criminal threshold policing contacts) and warning flags recorded on the police database. However, there were no significant differences with respect to committing violent, sexual or non-violent offences than young people without a mental illness history.

In relation to victimisation, the young people with a mental illness history were more likely to have a history of victimisation, to have experienced repeated victimisation incidents and to be victims of a violent offence. The finding that violent and non-violent perpetration was no more likely in those with a mental illness is inconsistent with previous research, which had found that the presence of any diagnosis increases the chances of minor, moderate and violent offending (Coker et al., 2014 ; Elonheimo et al., 2007 ). These results suggest that the link between mental illness and violent victimisation is greater than the association between mental illness and violent offending (Choe et al., 2008 ; Hodgins et al., 2007 ), as it is in adults with mental illness. Furthermore, consistent with extant literature, young people with an offending history were 10 times more likely to have a victimisation history as well, with the age trajectory suggesting that formal contact with police occurs first in the context of criminal offending and then of victimisation.

Limitations

The results from the present study need to be considered in light of several limitations. Study data obtained via case-linkage procedures carries a number of caveats. The data available were not collected for the purposes of research, and therefore the information was limited, particularly around characteristics of the sample and known risk factors that would have been useful covariates (e.g., socio-economic status, family structure, nationality) (Schlack & Petermann, 2013 ). The police database only contains offending and victimisation incidents officially recorded by the police, and so it is possible that a number of offending and victimisation incidents were not recorded. For example, data collected from self-report studies reveal that the majority of offences committed by young people do not lead to police involvement and arrests (Coker et al., 2014 ). Similarly, there is some –F albeit dated –F evidence that young people –F especially those with an offence history, who do not think the incident will be believed by police –F are less likely to report victimisation incidents (Carach, 1997 ). As a result, the true prevalence of offending and victimisation incidents is likely to be underestimated relative to that which would be seen in the community.

In addition to official records of crime, records of mental health diagnoses are also likely to be underestimated due to diagnoses being recorded via a public mental health database. Due to the nature of public health services provision, it is possible that some young people diagnosed with high-prevalence disorders (for example mood, anxiety or substance related disorders) were being managed in the primary care sector; if so, their contact histories and diagnoses would not be recorded on the database. Consequently, the findings presented here are likely an underestimate of the true prevalence of mental illness within the sample of young people. Therefore, any differences between groups can be interpreted with confidence, with the estimated prevalence of mental illnesses being considered a lower prevalence estimate.

Finally, the present study had a modest sample size that would have limited the statistical power of the analyses. Due to this, it was not possible to examine any increased risk of vulnerability beyond the ICD-10 (F00–FF99) diagnostic category level. Additionally, the data represent a secondary analysis of a subset of a larger dataset drawn from a linkage procedure that was initially carried out for a different research purpose, so the data available for analysis here were additionally limited. It is possible that additional, more nuanced, differences with respect to specific crimes and victimisation episodes would have become apparent with a larger sample.

The Victim–FOffender Overlap

The current study demonstrated that young people with an offending history had significantly greater odds of a victimisation history. This is consistent with the findings of Cops and Pleysier' ( 2014 ), who replicated the victim–Foffender overlap and found it to be more prominent for violent crimes. Of note, no such associations were found for those with a sexual offending or sexual victimisation history. This is possibly due to the small number of sexual crime incidents reported in the sample, which aligns with Australian statistics revealing that sexual crimes are less prevalent among young people than are non-violent (e.g., theft) and violent (e.g., assault) crimes (ABS, 2011 ).

The current study found that young people come into contact with police at a significantly younger age as offenders than as victims. This finding is consistent with empirical research that has found the offending-to-victim relationship to be more pronounced than the victim-to-offending relationship (Lopez & Emler, 2011 ; Smith & Ecob, 2007 ). Specifically, Cops and Pleysier ( 2014 ) found that the direct effect of victimisation on offending was mediated by the presence of a risky lifestyle; they propose that risky lifestyle leads to higher rates of offending, which can then lead to higher rates of victimisation. This assertion was not able to be tested here due to the nature of the data available, but remains an interesting avenue of exploration for future research.

Youth Offending

In contrast with previous research (e.g., Coker et al., 2014 ; Elonheimo et al., 2007 ), the present study found only a weak association between mental illness and perpetration of violent offences. Young people with a history of mental illness were more likely to commit violent offences; however, once age and gender had been taken into account, this increased likelihood was not statistically significant. The present sample differed from past research in that the young people were previously known to police in some capacity, and the overwhelming majority had a criminal history. One possible explanation for the finding presented here may be that mental illness does not discriminate rates of violent offending in young people who are already delinquent. In line with this notion, Colins and colleagues ( 2013 ) found that in samples of detained youth, psychiatric disorders were not predictive of violent recidivism. Taken together, these results suggest that contrary to common stigma and stereotypes, young people with a mental illness may be no more violent than their counterparts who do not have a mental illness history.

In post-hoc analyses of the offences that comprised the violent category, threatening behaviour was the only offence that was statistically significantly different between the two groups, and that for males only. Males with a history of mental illness were three times more likely to commit a threat offence than males without a diagnosis. In the context of related findings that these young people are at a substantially greater risk of victimisation, it is plausible that engaging in threatening behaviour might be a way of protecting themselves against potential violence. There remains controversy as to whether threatening behaviour is considered a violent offence; some researchers consider threatening behaviour a non-violent offence, others a violent offence, and yet others suggest that it should more practically be considered an ‘intermediate violent’ offence (Short et al., 2013 ; Warren, Mullen, Thomas, Ogloff, & Burgess, 2008 ). Specifically, it has recently been proposed that offences such as threats and harassment should be distinguished from other more severe violent offences like assault and homicide rather than being truncated together (Daff & Thomas, 2015 ). As indicated in the present study, such variations in the classification of violence could potentially lead to very different conclusions about young people's propensity towards violent crime. Thus it is important for future researchers to consider the definitions of violent crime when interpreting results and when grouping offence types.

In contrast with previous research, the present study failed to find a statistically significant difference in the prevalence of non-violent offending between young people with and without a mental illness. However given that all but six young people in the entire sample had a history of non-violent offending, a ceiling effect may have hindered the chance of finding an association. Post-hoc analysis of the offences comprising the non-violent category revealed many group differences. Specifically, young men with a history of mental illness were significantly more likely to be the perpetrators of theft, offences against the government (e.g., resisting police, breach of community service order and breach of bond) and property damage than were their male counterparts. The higher prevalence of theft and property offences is consistent with the literature examining youth mental illness (e.g., Coker et al., 2014 ); however, the link to government offences is a novel finding. Investigation into the offences within this category for those with a mental illness revealed that the most common offences were resisting police (60%), failure to answer bail (56%), and breaching a family violence order (38%). Recent changes towards more ‘community-based policing’ have meant that police are more commonly targeting antisocial behaviour in the community in order to address public fears and promote the perception of police legitimacy (Thomas, 2015 ). Therefore, it is possible that young people with a history of mental illness feel targeted and unfairly treated by police and so are more likely to be resistant and/or non-compliant, or that their behaviour is bringing them to the attention of the police. Furthermore, this may also explain the findings in the present study that these young people are more likely to have sub-criminal threshold field contacts with police and warning codes listed against their contact records, as their behaviours lead them to become readily identified and/or targeted by routine community safety initiatives.

Victimisation

Adding to the available literature, the present study demonstrates that having a history of mental illness significantly increases the risk of victimisation. The most alarming findings indicate that once the effects of age and gender are taken into account, those with a mental illness are three times more likely to have a recorded history of victimisation and close to three times more likely to have a recorded history of violent victimisation. Investigation of the offences within the violent victimisation grouping revealed that being the victim of assault was close to four times more likely among those young people with a history of mental illness compared to those without a mental illness. These findings are in line with Cater and colleagues ( 2014 ), who found that physical abuse was the most common form of victimisation in young people. What makes these findings particularly concerning is placing them in the context that young people, in general, are at a proposed tenfold greater risk of all forms of victimisation relative to adults (Danielsson et al., 2009 ). These findings suggest that young people with a mental illness, who are already more vulnerable, have an even greater chance of victimisation, thus compounding their existing disadvantage. This suggests that current mental health services are failing to protect this at-risk and vulnerable young population from victimisation.

Implications and Future Research

The findings presented here indicate that mental health services are currently failing to protect young people with mental illness from violence. There are a number of possible explanations for these significant ongoing service limitations. First, it could be the result of an inadequate assessment of victimisation in young people; for example Voisen ( 2007 ) suggests that standardised assessment instruments considering young people's exposure to violence should be built into routine clinical practice. While mental health clinicians routinely ask about past and present risk of harm, questioning related to criminal victimisation, particularly with respect to those incidents perpetrated by peers and other non-family members, may not be common practice. Recent recommendations (Parker, Hetrick, & Purcell, 2010 ) suggest asking questions about victimisation in the context the person's home environment and school life; however, how routinely health professionals choose to ask these ‘suggested’ questions and explore victimisation in other contexts remains unknown. One possible barrier to this is that some health professionals have not had training around trauma and may therefore lack the skills and confidence to ask questions about such experiences. A number of United States-based studies have found that therapists and primary care physicians are often unaware of, and infrequently document, reports of violence by youth (Borowsky & Ireland, 1999 ; Chaffee, Bridges, & Boyer, 2000 ; Sigel & Harpin, 2013 ). For example, Guterman & Cameron ( 1999 ) found that with the exception of sexual offences, therapists’ knowledge of clients’ violent victimisation incidents outside the family environment (e.g., assault by a non-family member) was significantly lower than young people's reported exposure to violence. This suggests that mental health professionals may be failing to screen adequately for young people's exposure to violence. Future research in this area is needed, as policy development within current health services may be a critical avenue within which these vulnerable young people could be helped.

Another possible explanation relates to young people's decisions about the disclosure of victimisation. It could be that young people are appropriately questioned about their experiences of victimisation but deny, or choose not to report, such incidents. For instance, a dated Australian review (Carach, 1997 ) revealed that the most common reasons young people give for not reporting incidents of assault include that the victim felt it was trivial, that the police could/would not do anything, or that it was considered a private matter. It could be that young people do not necessarily recognise assault by a friend or peer as a form of abuse/victimisation, especially if it occurs within their peer circle, and they are vulnerable and seeking relationship attention. In line with this, these previous results also showed that the decision to report or not also depended on whether the victim knew the perpetrator, in which case they were less likely to report and more likely to be frightened of reprisal/revenge. It is also possible that if these young people have been brought up witnessing domestic violence and living in violent environments, they may be less likely to recognise such incidents as abuse.

Taken together, these findings strongly imply a need for thorough youth mental health assessments focusing specifically on criminal victimisation. In light of recent attention towards other prevalent forms of victimisation among young people and the even more adverse effects of poly-victimisation (i.e., experiencing several types of victimisation), it is becoming more compelling to argue that it is now necessary for mental health policy to incorporate mandatory sensitive questions regarding young people's experience of an array of victimisation incidents into assessment measures.

Conclusions

The present study highlights the increased vulnerability to violence of young people with a mental illness. Of note, only a weak association was established between mental illness and violent offending. The principal finding was that young people with psychiatric illness are more likely to be violently victimised compared to their counterparts without a mental diagnosis. Taken together, these results suggest that the link between mental illness and victimisation is more significant and therefore potentially of greater concern than the link between mental illness and offending. These findings provide evidence that contrasts with the popular stereotype that young people with mental illness are excessively violent and suggests, rather, that they are much more vulnerable to violence being directed towards them.

These results strongly highlight the need for routine incorporation by clinicians of questions about a young person's victimisation history (especially non-familial violence) when conducting an assessment. These findings also apply to police working with young people, especially given that young people are likely to trivialise victimisation incidents. Given that youth violence and mental health are serious public health concerns, it is imperative that victimisation is detected early in order to protect the health and welfare of young people.

Acknowledgments

We would like to thank Superintendent Mick Williams and Sergeant Geoff Currie who facilitated access to the police data and have supported the continued research into the police-mental health nexus.

Competing Interests

The authors declare that they have no competing interests.

• Australian Bureau of Statistics (2011). Youth and their experience of victimisation and offending . 4524.0 - In Focus: Crime and Justice Statistics, September 2011 Canberra, Australia: Australian Bureau of Statistics. [ Google Scholar ]
• Borowsky I. W., & Ireland M. (1999). National survey of pediatricians' violence prevention counseling . Archives of Pediatrics & Adolescent Medicine , 153 , 1170–1176. [ PubMed ] [ Google Scholar ]
• Carcach C. (1997). Reporting crime to the police . Trends and Issues in Criminal Justice , 68 , 1–6. [ Google Scholar ]
• Cater Å. K., Andershed A-K., & Andershed H. (2014). Youth victimization in Sweden: Prevalence, characteristics and relation to mental health and behavioral problems in young adulthood . Child Abuse & Neglect , 38 , 1290–1302. [ PubMed ] [ Google Scholar ]
• Chaffee T. A., Bridges M., & Boyer C. B. (2000). Adolescent violence prevention practices among California pediatricians . Archives of Pediatric and Adolescent Medicine , 154 , 1034–1041. [ PubMed ] [ Google Scholar ]
• Chang J. J., Chen J. J., & Brownson R. C. (2003). The role of repeat victimization in adolescent delinquent behaviors and recidivism . Journal of Adolescent Health , 32 , 272–280. [ PubMed ] [ Google Scholar ]
• Choe J., Teplin L., & Abram K. (2008). Perpetration of violence, violent victimization, and severe mental illness: Balancing public health concerns . Psychiatric Services , 59 , 153–164. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Colins O. F., Boonmann C., Veenstra J., van Domburgh L., Buffing F., Doreleijers T. A.H., & Vermeiren R. R. J. M. (2013). Mental health problems and recidivism among detained male adolescents from various ethnic origins . European Child and Adolescent Psychiatry , 22 , 481–490. [ PubMed ] [ Google Scholar ]
• Coker K. L., Smith P. H., Westphal A., Zonana H. V., & McKee S. A. (2014). Crime and psychiatric disorders among youth in the US population: An analysis of the national comorbidity survey–adolescent supplement . Journal of the American Academy of Child and Adolescent Psychiatry , 53 , 888–898. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Copeland W. E., Miller-Johnson S., Keeler G., Angold A., & Costello E. J. (2007). Childhood psychiatric disorders and young adult crime: A prospective, population-based study . American Journal of Psychiatry , 164 , 1668–1675. [ PubMed ] [ Google Scholar ]
• Cops D., & Pleysier S. (2014). Usual suspects, ideal victims and vice versa: The relationship between youth offending and victimization and the mediating influence of risky lifestyles . European Journal of Criminology , 11 , 361–378. [ Google Scholar ]
• Cultural and Indigenous Research Centre (CIRCA). (2013). Review of field contact policies - final report . Retrieved from http://www.communitylaw.org.au/flemingtonkensington/cb_pages/files/CIRCA_Victoria-Police-Review-of-Field-Contact-Policy-and-Processes_Nov-2013.pdf [ Google Scholar ]
• Daff E., & Thomas S. D. M. (2015). Bipolar disorder and criminal offending: A data linkage study . Social Psychiatry & Psychiatric Epidemiology , 49 , 1985–1991. [ PubMed ] [ Google Scholar ]
• Danielsson I., Blom H., Nilses C., Heimer G., & Högberg U. (2009) Gendered patterns of high violence exposure among Swedish youth . Acta Obstetricia et Gynecologica Scandinavica , 88 , 528–535. [ PubMed ] [ Google Scholar ]
• Elonheimo H., Niemelä S., Parkkola K., Multimäki P., Helenius H., Nuutila A-M., & Sourander A. (2007). Police-registered offenses and psychiatric disorders among young males . Social Psychiatry & Psychiatric Epidemiology , 42 , 477–484. [ PubMed ] [ Google Scholar ]
• Fazel S., Doll H., & Langstrom N. (2008). Mental disorders among adolescents in juvenile detention and correctional facilities: A systematic review and metaregression analysis of 25 surveys . Journal of the American Academy of Child and Adolescent Psychiatry , 47 , 1010–1019. [ PubMed ] [ Google Scholar ]
• Guterman N. B., & Cameron M. (1999). Young clients' exposure to community violence: How much do their therapists know? American Journal of Orthopsychiatry , 69 , 382–391. [ PubMed ] [ Google Scholar ]
• von Hentig H. (1948). The criminal and his victim: Studies in the sociology of crime . New Haven: Yale University Press. [ Google Scholar ]
• Hodgins S., Alderton J., Cree A., Aboud A., & Mak T. (2007). Aggressive behaviour, victimisation and crime among severely mentally ill patients requiring hospitalisation . The British Journal of Psychiatry , 191 , 343–350. [ PubMed ] [ Google Scholar ]
• Karnik N., Soller M., Redlich A., Silverman M., Kraemer H., Haapanen R., & Steiner H. (2009). Prevalence of and gender differences in psychiatric disorders among juvenile delinquents incarcerated for nine months . Psychiatric Services , 60 , 838–841. [ PubMed ] [ Google Scholar ]
• Kesic D., Thomas S. D. M., & Ogloff J. R. P. (2012). Estimated rates of mental disorders in, and situational characteristics of, incidents of nonfatal use of force by police . Social Psychiatry & Psychiatric Epidemiol ogy , 48 , 225–232. [ PubMed ] [ Google Scholar ]
• Lauritsen J. L., & Laub J. H. (2007). Understanding the link between victimization and offending: New reflections on an old idea . Crime Prevention Studies , 22 , 55–79. [ Google Scholar ]
• Lauritsen J. L., Sampson R. J., & Laub J. H. (1991). The link between offending and victimization among adolescents . Criminology , 29 , 265–292. [ Google Scholar ]
• López E. E., & Emler N. P. (2011). Assessing the links among adolescent and youth offending, antisocial behaviour, victimization, drug use, and gender . International Journal of Clinical and Health Psychology , 11 , 269–289. [ Google Scholar ]
• Maniglio R. (2009). Severe mental illness and criminal victimization: A systematic review . Acta Psychiatrica Scandinavica , 119 , 180–191. [ PubMed ] [ Google Scholar ]
• National Health and Medical Research Council (2007). National statement on ethical conduct in human research . Canberra, Australia: National Health and Medical Research Council. [ Google Scholar ]
• Ousey G. C., Wilcox P., & Fisher B. S. (2010). Something old, something new: Revisiting competing hypotheses of the victimization-offending relationship among adolescents . Journal of Quantitative Criminology , 27 , 53–84. [ Google Scholar ]
• Parker A., Hetrick S., & Purcell R. (2010). Psychosocial assessment of young people: Refining and evaluating a youth friendly assessment interview . Australian Family Physician , 39 , 585–593. [ PubMed ] [ Google Scholar ]
• Rosenblatt J., Rosenblatt A., & Biggs E. (2000). Criminal behavior and emotional disorder: Comparing youth served by the mental health and juvenile justice systems . The Journal of Behavioral Health Services & Research , 27 , 227–237. [ PubMed ] [ Google Scholar ]
• Schlack R., & Petermann F. (2013). Prevalence and gender patterns of mental health problems in German youth with experience of violence: The KiGGS study . BMC Public Health , 13 , 1–1. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Shaffer, J. N. (2003). The victim-offender overlap: Specifying the role of peer groups (PhD thesis). The Pennsylvania State University, Crime, Law and Justice Program, Pennsylvania. [ Google Scholar ]
• Short T. B. R., Thomas S., Luebbers S., Mullen P., & Ogloff J. R. P. (2013). A case-linkage study of crime victimisation in schizophrenia-spectrum disorders over a period of deinstitutionalisation . BMC Psychiatry , 13 , 66. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Sigel E. J., & Harpin S. B. (2013). Primary care practitioners' detection of youth violence involvement . Clinical Pediatrics , 52 , 411–417. [ PubMed ] [ Google Scholar ]
• Smith D. J., & Ecob R. (2007). An investigation into causal links between victimization and offending in adolescents . The British Journal of Sociology , 58 , 633–659. [ PubMed ] [ Google Scholar ]
• Snyder H. N., & Sickmund M. (2006). Juvenile offenders and victims: 2006 national report . Washington: National Center for Juvenile Justice. [ Google Scholar ]
• Sweeten G., Piquero A. R., & Steinberg L. (2013). Age and the explanation of crime, revisited . Journal of Youth Adolescence , 42 , 921–938. [ PubMed ] [ Google Scholar ]
• Teplin L. A., Abram K. M., McClelland G. M., Dulcan M. K., & Mericle A. A. (2002). Psychiatric disorders in youth in juvenile detention . Archives of General Psychiatry , 59 , 1133–1143. [ PMC free article ] [ PubMed ] [ Google Scholar ]
• Thomas S. (2015). Policing young people . In Sheehan R. & Ogloff J. (Eds.), Working within the forensic paradigm: Cross-discipline approaches for policy and practice (pp. 126–136). London: Routledge. [ Google Scholar ]
• Voisin D. R. (2007). The effects of family and community violence exposure among youth: Recommendations for practice and policy . Journal of Social Work Education , 43 , 51. [ Google Scholar ]
• Warren L. J., Mullen P. E., Thomas S. D. M., Ogloff J. R. P., & Burgess P. M. (2008). Threats to kill: A follow-up study . Psychological Medicine , 38 , 599–605. [ PubMed ] [ Google Scholar ]
• Wasserman G. A., McReynolds L. S., Schwalbe C. S., Keating J. M., & Jones S. A. (2010). Psychiatric disorder, comorbidity, and suicidal behavior in juvenile justice youth . Criminal Justice & Behavior , 37 , 1361–1376. [ Google Scholar ]
• Wolfgang M. E. (1958). Patterns in criminal homicide . England: Pennsylvania Press. [ Google Scholar ]
• World Health Organization (1992). The international classification of diseases, 10th revision . Geneva, Switzerland: Author. [ Google Scholar ]