Theoretical background

In a review, Hodgins (2001; Côté & Hodgins, 2003) describes major studies which lead to an overall general conclusion that a significant link exists between major mental disorders (MMD) and criminality and/or violence. Hodgins describes four streams of investigations leading up to the authors’ strongly held conclusion. She concludes that we need to stop debating if there is an actual link between MMD and criminality and/or violence, but rather start developing prevention strategies. The four streams of investigation Hodgins describes stem from the results of numerous studies. The first type of investigation discusses results from studies of criminality among patients living in the community (i.e. Mullen, Burgess, Wallace, Palmer, & Ruschena, 2000; Volavka et al., 1997; Wessely, CasUe, Douglas, & Taylor, 1994). For the second type, the author discusses six studies of birth cohorts in which criminality of persons who develop MMD are compared to that of persons who did not develop MMD (Hodgins, 1992; Hodgins, Mednick, Brennan, Schulsinger, & Engberg, 1996; Stueve & Link, 1998; Tiihonen, Isohanni, Rasanen, Koiranen, & Moring, 1997; Wallace et al., 1998).

The author then moves on to discuss the third pathway in which studies of mental disorder among incarcerated offenders are reported. She explains that North American studies have found higher than normal preva1ence rates of MMD among both male and female offenders (Hodgins & Côté, 1995; Teplin, Abram, & McClelland, 1996). However, this phenomenon is not supported in the UK (Cooke, 1994; Gunn, Maden & Swinton, 1991 , 1992). The fourth and last stream of investigation is the study of mental disorders among samples of homicide offenders, in which Hodgins discusses research findings from five groups of researchers (Côté & Hodgins, 1992; Erb et al., 2001 ; Eronen, Tiihonen, & Hakola, 1996; Gottlieb, Gabrielsen, & Kramp, 1987; Lindqvist, 1986). Of the five studies Hodgins reVlews, the work of Côté and Hodgins (1992) is particularly interesting. They reported that the prevalence of MMD in male offenders convicted of homicide (35%) in the Province of Quebec was higher than that of those who had committed other crimes (21 %). At a closer look, there was a tendency for the homicidal group to have a higher prevalence of schizophrenia. The diagnosis of the MMD was most often present before they committed the homicide, and group differences (those convicted for homicide vs. those convicted of other crimes) were most apparent when there was a comorbid diagnosis of a substance abuse disorder or an antisocial persona lit y disorder. ln a longitudinal study conducted in Copenhagen it was reported that 20% of men found guilty of homicide were suffering from a MMD and that 41 % of those men were also substance abusers (Gottlieb et al., 1987).

In Sweden, a group of researchers found that more than half of the individuals found guilty of homicide suffered from a MMD, with 38% of them also substance abusers (Lindqvist, 1986). ln line with these findings, a more recent international systematic review (Richard-Devantoy, Olie, & Gourevitch, 2009) reporting the rates of mental disorder in people convicted of homicide found that those with MMD were at an increased risk for committing homicide (two-fold for males and six-fold for females). The authors found that a diagnosis of schizophrenia increased the risk six to 10-fold in males and eight to 10-fold in females. In another study by Swinson and Shaw (2007), it was found that 9% of those who had committed a homicide had been in contact with mental health services during the 12 months period prior to the offence and that schizophrenia was the most corn mon diagnosis (5%) among these offenders.

Regina v. Winko Much positive change was created with Bill C-30, with its c1earer and less stigmatizing vocabulary and definitions, while respecting human rights and freedoms . However, the amendment was not without flaws and was shown to still require further adjustments. Joseph Winko was arrested in 1983 for attacking pedestrians with a knife. Due to his mental illness (Auditory hallucinations-Schizophrenia) at the time, he was found NGRI for the aggravated assault and weapon-related charges. After being he Id in custody for treatment for his condition, he was then placed in a community setting with restrictions. At a Review Board hearing in 1995, Joseph Winko was rendered a conditional discharge (Sinha, 2009), despite a request for an absolute discharge. The disposition of conditional discharge was given because it was the opinion of the Review Board that Winko « ma y » possibly become a « significant » threat to the community under certain circumstances. Winko took his case to the Supreme Court of Canada dec1aring that this decision went against the Canadian Charter of Rights and Freedoms. Then, in 1999, the courts ruled that in absence of concrete evidence demonstrating that Winko could be an actual « significant » threat, Mr. Winko must be granted an absolute discharge. The ruling in the Winko case once again set a precedent and created considerable change within the system (Schneider, Glancy, Bradford, & Seibenmorgen, 2000), in that prior to Winko v. B.C (British Columbia Forensic Institute, 1999), uncertainty of threat resulted in continued follow-up within the system. Whereas after Winko v. B. C, the inability to demonstrate that the patient is an actual significant threat results in an absolute discharge (Balachandra, Swaminath, & Litman, 2004). Moreover, conditions imposed post Winko. v. B. e. were amended as weIl. First, being he Id in custody is rarely applied to a conditional discharge, and secondly, conditions are now less onerous and imposed less often than before (Latimer & Lawrence, 2006). For example, conditions such as curfew, weakly reporting and mandatory treatments were imposed on average 50% less often than prior to changes which were influenced by the Winko case (Latimer & Lawrence, 2006).

Roesch and colleagues (1997) found that in the year following the 1992 amendment, there was an increase of use and success of the NCRMD defence. Furthermore, an increase of about 20% in remands was documented. Latimer and Lawrence (2006) also reported a 102% increase from 1992 to 2004. The authors concluded that the increase was not due to total number of absolute admissions in the judicial system. In the Province of Quebec, the Review Board reported that in 2005 the number of NCRMD verdicts had more th an doubled since 1992 (N = 407 vs. N = 177, respectively), and again later in 2012 the number increased to 540 new verdicts (Carmelle Beaulieu, as mentioned in Crocker et al., 2015). Additionally, in light of the decision rendered in the Winko case, sorne have speculated that absolute discharges have been on the rise (Balachandra et al., 2004, Simpson et al., 2014). Latimer and Lawrence (2006) reported that after the Winko case, absolute discharges increased from 10% to 15%, while conditional discharges increased from 33% to 37%. Conversely, there was a decrease (56% to 47%) in detention orders imposed by the Review Boards of Canada. The study also reported fewer conditions imposed on dispositions in general. But of particular interest, they found that accused who were granted a conditional discharged were never held in detention. Published annual report figures for the T AQ shows that the amount of accused entering the Review Board system in Quebec have almost doubled since 1999 (1999, N = 1307 vs. 2017, N = 2400; Rapport d’activité 1998-2000; TAQ Rapport annuel de gestion 2016-2017). One can speculate that given the decrease in threshold for mandating continued follow-up in the system, a possible higher volume of NCRMD defences for serious crimes may have developed, as well as a higher volume of absolute discharge requests for seriously violent offences, such as homicide and attempted homicide, in hopes for a fast process and absolute discharge.

Homicide in Canada According to the Canadian Criminal Code (last amendment on July 23, 2015), a person commits homicide when « directly or indirect/y, by any means, he causes the death of a human being ». There are two types of homicide; culpable and non-culpable. Non-culpable is homicide that is not a criminal act in the sense that it occurs completely by accident, or in self-defence. Whereas, culpable homicide is murder, manslaughter or infanticide (Criminal Code, R.S.C., 1985. c.C-46). Murder is a culpable homicide that the individual (i) « means to cause the death of the victims, or (ii) means to cause the victim bodily harm that he knows is likely to cause his death, and is reckless whether death ensues or not ». A murder charge may be reduced to manslaughter if the person who committed the act did so in the « heat of passion » and was caused by « sudden provocation ». Lastly, a female may be charged with infanticide when she willfully acts or chooses not to act, causing the death ofher newly-born child. In 2014, police reported 369,359 violent Criminal Code violations in Canada. Violent crimes account for about 20% of aIl criminal offences (Miladinovic & Mulligan, 2015; CANSIM table 252-0051). Rates of homicide are thought to represent about .1 % of al! violent crime (Perreault, 2013). Also in 2014, Canadian police services reported 516 homicides, which were four more cases than in the previous year, but still 18% below average from the previous decade, which translates into a national homicide rate of about 1.45 in 100,000 population.

For the population of Quebec the homicide rate is .86 per 100,000. Statistics Canada reports that the national homicide rate is at its lowest point since 1966. Moreover, there were 617 attempted murders, which also has as weil been on a stable decline for the last five years (CANSIM table 253-0001). Almost three quarters of ail cases under mandate of the Review Board System in Canada are for violent offences (violent offences against persons, including verbal threats and harassments). Research conducted by the Department of Justice Canada compared cases involving NCRMD/UST rulings and those who have been found guilty in adult court (fiscal year of 2003/2004). They found that compared to those who were convicted in adult courts, individuals who have been found NCRMD or deemed UST are more likely to have been charged with a seriously violent o ffence , such as homicide or major assault and less likely to be charged with a non-violent offence su ch as theft (Latimer & Lawrence, 2006). While we acknowledge the low rates of homicide in Canada, we should also acknowledge that homicide is the most serious of a11 crimes, with repercussions affecting many aspects of human life (WHO, 2015), while carrying the most severe criminal sanctions.