Conceptual and Methodological Challenges in Examining the Relationship between Mental Illness and Violent Behaviour and Crime

There is a longstanding view within the general population and the criminal justice system that the mentally ill are more prone than the mentally healthy to violence and. This view, however, is not fully supported by empirical research, in particular due to conceptual and methodological challenges that arise when the relationship between mental illness and crime is examined. This paper reviews such challenges, reviewing areas such as the ‘criminalisation’ of the mentally ill and the ‘psychiatrisation’ of criminals, as well as the complex problem of common factors, and the mediating impact of substance abuse. Specific methodological challenges are also reviewed, including problems with conducting longitudinal and randomised research in this area, and difficulties encountered in the sampling methods used. 1 Submitted by Thomas Richardson as part fulfillment of the Psychology Undergraduate Degree Program, School of Psychology, Trinity College, Dublin, Ireland. Internet Journal of Criminology © 2009 www.internetjournalofcriminology.com 2 Throughout history and across cultures a view has persisted that the mentally ill are particularly prone to violence and crime (Monahan, 1996). The general public consistently report that the mentally ill are more dangerous (Phelan & Link, 1998), a view which is reinforced by selective media reporting (Mulvey, 1994). This relationship has been examined since the mid 19 century (Gray, 1857). However, such research has suffered from significant methodological and conceptual problems and consequentially provides results that do not entirely support this concept (Davis, 1991). This paper will outline and explain such methodological difficulties and challenges in the definition and examination of the relationship between mental illness and crime. The ‘Criminalisation’ of the Mentally Ill A significant problem in research into the relationship between crime and mentally illness is what has become known as the ‘criminalisation’ of the mentally ill. Abramson (1972) first noted that a large proportion of the mentally ill were being dealt with by the criminal justice system rather than mental health services. Since this initial argument, a large body of evidence suggests that the mentally ill are arrested, convicted and sent to prison in proportions that surpass their actual criminal behaviour (Council of State Governments, 2002). People who need mental health treatment are frequently being put into prison rather than hospital (Teplin, 1983). Changes in mental health policies may be responsible for this increasing contact between the mentally ill and justice systems (Lurigio & Fallon, 2007). The mentally ill are often ‘referred’ to the criminal justice system due to poor or inappropriate resources in the mental health sector (Borzecki & Wormith, 1985). This may be largely due to the ‘deinstitutionalisation’ seen in many countries in the last few decades (Whitmer, 1980), which shifted the emphasis of mental health care from psychiatric hospitalisation to community based settings, resulting in a dramatic reduction in the number of psychiatric hospital beds (Center for Mental Health Services, 2004). However this community mental health movement was never properly implemented (Shadish, 1989), resulting in an increased risk of violence by outpatients (Hodgins, 1994). As a result of deinstitutionalisation, the mentally ill increasingly come into contact with the police and courts, thus inflating the apparent relationship between crime and mental illness. A number of largely political factors also indirectly increase the representation of the mentally ill in the criminal justice system, with mental illness consequently complicating rather than causing their involvement (Draine, 2003). The ‘war on drugs’ is the U.S. has played a large role (Lurigio & Swartz, 2000), as it has lead to increased arrest and conviction for drug related offences (Beck, 2000), resulting in an increased proportion of the mentally ill in prisons (Swartz & Lurigio, 1999; Goldkamp & IronsGuynn, 2000). Another important factor is the role of police tactics. The recent emphasis by the western world on ‘zero-tolerance’ and ‘quality of life’ policing has led to increased arrest for minor offences such as disturbing the peace and loitering, to which the mentally ill are more prone (Fagan & Davies, 2000). Public-order policing has essentially led to patients being arrested for their symptoms (Ditton, 1999), as many of the symptoms of mental illness are behaviours considered to be antisocial or criminal (Cauffman et al., 2005). Mental illness elevates the risk of arrest as detection and subsequent calls to the police are more likely in those with such problems (Hirschfield et al., 2006), with the arrest rate being 67 times high for those Internet Journal of Criminology © 2009 www.internetjournalofcriminology.com 3 demonstrating symptoms of mental illness (Teplin, 2000). There is also a bias in convictions, as the mentally ill are more likely to be charged with misdemeanours than are the mentally healthy (Lamberti et al., 2001), and spend a longer time in jail for similar crimes (McNiel, Binder & Robinson, 2005). The mentally ill are also less likely to understand their interrogation rights, and consequently are more prone to false confessions (Redlich, 2004). Reforms in policies and laws around mental health treatment have made it increasingly difficult to section the mentally ill, which increases the likelihood that they will be arrested ‘by default’ (Davis, 1992). For example, institutions such as schools, unsure what to do, may call the police to deal with emotionally disturbed adolescents (Rice, 2003). These factors have combined so that in the U.K., 20% of mentally ill prisoners have been rejected for treatment by the National Health Service prior to conviction (Coid, 1988). As a result there is a considerable ‘flow’ of individuals between the mental health and criminal justice systems (Holley & Arboleda-Flórez, 1988). Furthermore, the criminal justice system experience may worsen mental health problems (Lurigio, Fallon & Dincin, 2000), further increasing such a flow. This becomes a problem for researchers, as it is becomes difficult to define and study the relationship, with so many indirect mechanisms by which the mentally ill are over-represented in the criminal justice system. The ‘Psychiatrisation’ of Criminals An important point is what has been referred to as the ‘psychiatrisation’ of criminals. Bad behaviour is often viewed as a symptom of psychological disorder (Flew, 1954), and for many decades it has been argued that all criminals are ‘sick’ (Silber, 1974), and that crime is a symptom of disease (Flew, 1954). As Menninger (1928) put it “The time will come, when stealing or murder will be thought of as a symptom, indicating the presence of a disease”. It has also been argued that the psychiatry is often abused in the area of crime; in the Soviet Union, and still today in China, psychiatry has been used to ‘imprison’ political dissidents (Gunn, 2006). It has been argued that the high levels of mental illness in prisons is largely due to the ‘psychiatrisation’ of criminality (Davis, 1992; Anderson, 1997), and that psychiatric concepts are commonly applied to convicted individuals due to ethical and social values rather then medical considerations (Silber, 1974). As a result, the links between the two may simply be a product of how the mentally ill and criminals are treated by such services (Wessely & Taylor, 1991). For a number mental disorders such as borderline personality disorder, sadism and intermittent explosive disorder, violent behaviour is one of the key diagnostic symptoms, and such psychiatric conceptualisations of violence as a key symptom may be increasing over time (Harry, 1985). The symptoms of specific mental illness may directly include crime or delinquency, for example in conduct disorder or oppositional defiant disorder. A particularly important diagnosis is ‘Anti Social Personality Disorder’ (ASD), as ASD is the most common diagnosis in prisoners (Henn, Herjanic & Vanderopearl, 1976). ASD has in particular being criticised, with there being controversy over whether it constitutes a mental illness per se (Bursten, 1982), and many suggesting that it is no more than a moral judgement given as a diagnostic label (Blackburn, 1998). Such research suggests that crime is increasingly being labelled as mental illness, making it a considerable challenge for researchers to distinguish between the two, and subsequently to examine their relationship. Internet Journal of Criminology © 2009 www.internetjournalofcriminology.com 4 Common risk factors There are a number of common factors shared by both criminal and psychiatric patients, which may explain correlations between mental illness and crime. Demographic variables such as age, socioeconomic status and race predict both arrest (Brownfield, Sorenson & Thompson, 2000; Beckett et al., 2005), and mental illness (Costello et al., 1998). Individuals with personality disorders share a number of common factors with criminals, usually being young men, poorly educated, unemployed, and having a deprived upbringing and disorganised home environment (Morissette, 1986). Research has consistently shown that any associations between mental disorder and criminal behaviour disappear once variables such as education and age are controlled for statistically (Abram & Teplin, 1990). Monahan & Steadman (1983) argued that relationships between the two can be accounted for almost entirely by shared demographic factors. In essence, the mentally ill are prone to the same factors which predict criminality in the mentally healthy (Wessely, 1993). Accounting for such factors is a considerable challenge in researchers trying to establish causality. There are also a number of factors which predict crime and violence in the mentally ill. Violence prior to admission to a hospital is associated with violence after discharge, as is male gender, age, increased length of stay and cognitive impairment (Shah, Fineberg & James, 1991). Violence whilst in hospital is also related to overcrowding and incompetent staff (Shah, Fineberg & James, 1991). A number of studies have found that in the mentally ill, prior criminal history predicts recidivism upon release (Teplin, Abram & McClelland, 1994), and consequently it has been argued that once criminal behaviour has begun, the most accurate predictor of future crime is criminal history (Anderson, 1997), regardless of whether an individual is mentally ill (Bonta, Law & Hanson, 1998). Studies have found that the relationship between crime and mental illness disappears entirely once previous convictions are accounted for (Abram & Teplin, 1990; Wessely & Taylor, 1991; Feder, 1991). Research suggests that the mentally ill may only be more prone to violence if they receive inadequate treatment (Hodgins, 1998), have a long-standing paranoid attitude (Arseneault et al., 2000), and are actively experiencing delusions (Taylor & Estroff, 2003). Many psychiatric medications used to treat mental illness may also increase the risk of violence (Menuck, 1983). Other risk factors for criminality and violence in the mentally ill include homelessness (McNiel, Binder & Robinson, 2005; Martell, Rosner & Harmon, 1995), family dysfunction and conflict with violent parents (Rowley, Ewing & Singer, 1987; APA, 1994), being raised by a mentally ill parent (Lewis et al., 1985) and having a history of physical abuse (Vander Stoep, Evens & Taub, 1997). Further factors that predict recidivism in psychiatric patients include a lack of responsibility, narcissistic personality, poor treatment compliance and low levels of impulse control and affect regulation (Philipse et al., 2004; Nestor, 2002). Such factors pose a significant problem in research in the area, as they must be controlled for appropriately. Furthermore, there appears to be disagreement over whether such factors are casually relevant, and therefore should not be controlled for statistically (Monahan, 1993).