| Journal of Community Safety and Well-Being (2026) 11(3), 113–125. | https://doi.org/10.35502/jcswb.547 |
Sydney S. A. Rine∗, Brent R. Nixon†, Mark E. Olver∗
ABSTRACT
Given the independent nature of the provincial and territorial review boards (RBs) in Canada and observed differences in forensic populations across the country, research examining specific forensic subpopulations at the provincial level is essential. The current study is the first to examine those found not criminally responsible on account of mental disorder (NCRMD) and unfit to stand trial (UST) in Saskatchewan, Canada and focuses on sociodemographic, mental health, and criminological characteristics. Information was coded from hospital files of 138 individuals hospitalized at a forensic psychiatric hospital in Saskatchewan between November 1984 and January 2022. Although differences between those found NCRMD and UST were observed, most individuals had committed a violent offence as their most severe index offence. The predominant primary diagnosis among individuals found NCRMD was a psychotic spectrum disorder, with those found UST being diagnosed with a psychotic spectrum disorder or intellectual/cognitive disorder at equal rates. Those found NCRMD and UST were hospitalized for an average of 5.2 and 2.9 years, respectively. Overall, 18.3% of the sample recidivated, with higher rates observed among those found UST and substance use disorder diagnoses being most predictive of outcome across groups. Cross-jurisdictional comparisons of NCRMD admissions specifically demonstrated Saskatchewan to have particularly high overrepresentation of Indigenous persons, rates of certain diagnoses (e.g., psychotic spectrum, substance use) and mental health symptoms at the time of index offence (e.g., hallucinations, delusions, homicidal ideation), and patients who had a violent index offence or formal criminal history pre-verdict.
Key Words Not criminally responsible on account of mental disorder, unfit to stand trial, forensic mental health, mental health, violence.
Following the deinstitutionalization of psychiatric services which extended from the 1960s to the 1980s and changes to legislature in the 1990s regarding the management of individuals with mental disorders who come into contact with the criminal justice system, the number of individuals detained in Canadian forensic psychiatric hospitals has increased nationwide (Jansman-Hart et al., 2011; Latimer & Lawrence, 2006; Sealy & Whitehead, 2004). Detention in such hospitals can occur when an accused is found either not criminally responsible on account of mental disorder (NCRMD) or unfit to stand trial (UST; Latimer & Lawrence, 2006). Although efforts have been made to profile those detained in Canadian forensic psychiatric hospitals, population-level data are lacking, and extant research has focused disproportionately on those found NCRMD in the most populous provinces (Crocker et al., 2015d; Haag et al., 2016). The current study profiles the population of forensic psychiatric patients admitted to Saskatchewan Hospital North Battleford (SHNB) up to January 2022.
According to section 16 of the Criminal Code of Canada (1985a), a verdict of NCRMD should apply to those individuals who commit criminal acts as a result of acute symptoms of a mental disorder, and who are unable to understand the nature and quality of the act or to know that it was wrong. Despite not receiving typical criminal convictions, individuals found NCRMD are also not acquitted of the crime committed. Rather, NCRMD verdicts constitute a third option in which the accused is diverted to a provincial or territorial review board (RB) for further assessment, dispositions, and potential oversight as they may still pose a threat to the public (Latimer & Lawrence, 2006). RBs, which are specialized tribunals, function to protect the public as those under their supervision may still pose a threat despite being found not criminally responsible for their actions, while providing opportunities for the accused’s underlying mental disorder to be treated (Latimer & Lawrence, 2006). This is achieved by conducting individualized assessments of the accused and devising a disposition that weighs both the protection of the public and the treatment of the accused.
When considering what measures should be taken to ensure public safety and the well-being and treatment of those found NCRMD, three dispositions are available. The accused can receive an absolute discharge, a conditional discharge, or detention in custody in a hospital (Latimer & Lawrence, 2006). If absolute discharge is ordered, the individual is no longer involved with the system in relation to the specific offence that led to the NCRMD verdict. If the accused receives a conditional discharge, they are supervised in the community with limits placed on their freedom. In the case of detention in custody in a hospital, the hospital administrator can increase or decrease the restrictions on the accused. As such, the accused remains under the supervision of the RB until they receive an absolute discharge – which must be ordered when the individual no longer poses a significant threat to public safety. Given the indeterminate nature of the NCRMD distinction, it is of note that while the protection of the public is of great importance, to receive a conditional discharge or detention order there must be explicit evidence that the accused poses a significant risk (Latimer & Lawrence, 2006).
Like those found NCRMD, individuals deemed UST are also diverted to the RB; however, they have not yet been found guilty of an offence. Rather, the courts have determined that due to a mental disorder, the individual would be unable to defend themselves at any stage of the proceedings or to instruct counsel to do so on their behalf. Further, the individual may be unable to understand the nature of the proceedings, understand the potential consequences they face, or communicate with counsel, as outlined in section 2 of the Criminal Code of Canada (1985b). Because individuals found UST have yet to go to trial, the dispositions available to the RB are limited to conditional discharge and detention in custody; an absolute discharge cannot be granted. As such, individuals found UST are supervised by the RB until they are found fit to stand trial, the charges against them are stayed or withdrawn, or the evidence is no longer sufficient to bring them to trial and they receive an acquittal. Although the RB is unable to grant an absolute discharge, the court may order a stay of proceedings if an individual found UST is unlikely to ever become fit, does not pose a significant threat to the public, and a stay of proceedings would be in the interest of the proper administration of justice (Latimer & Lawrence, 2006).
To date, Canada’s NCRMD population has been examined in the National Trajectory Project (NTP), which investigated the trajectories of 1800 NCRMD individuals in British Columbia (BC), Ontario (ON), and Quebec (QC; Crocker et al., 2015c), and the Alberta (AB) Not Criminally Responsible (NCR) Project, which investigated 546 individuals found NCRMD in AB (Haag et al., 2016). Both studies found that most individuals who received the designation were suffering from a serious mental illness (SMI), most often a psychotic spectrum disorder, at the time of the index verdict (Crocker et al., 2015b; Haag et al., 2016). Additionally, rates of substance use disorder (SUD) diagnoses ranged from 28.7% to 39.2% and personality disorder (PD) diagnoses ranged from 9.5% (any PD) to 15.1% (antisocial personality disorder (ASPD)), depending on the province; AB had the highest rates of both substance use and PDs of the four provinces (Crocker et al., 2015b; Haag et al., 2016). Further, the NTP found that one in three individuals had been diagnosed with an SMI and a comorbid SUD or PD and that prior to the index offence, 72% had been hospitalized on at least one occasion due to their mental health (Crocker et al., 2015b).
Regarding criminological characteristics, offences ranged from those against the person (e.g., offences causing death, sexual offences, assaults) to those against property (e.g., theft), to other Criminal Code violations (e.g., prostitution, weapons offences, drug possession and/or trafficking). Although violent offences against the person accounted for the largest proportion of offences, considering the overrepresentation of sensationalized cases in the media, it is noteworthy that homicide and attempted homicide accounted for a modest proportion of violent offences in all provinces studied (Crocker et al., 2015b; Haag et al., 2016). Further, only half of individuals in the NTP sample had received a criminal conviction or been found NCRMD prior to the index offence (Crocker et al., 2015b). Although offences were most commonly committed against family members (i.e., parents being victimized most frequently, followed by romantic partners), professionals, strangers, and others known to the individual found NCRMD were also victimized (Crocker et al., 2015b).
In light of the possible NCRMD dispositions of absolute discharge, conditional discharge, and detention in hospital, both the NTP and AB NCR Project found that recommendations regarding dispositions provided in clinician reports agreed with the decision of the RB for the majority of sampled hearings (Cheng et al., 2022; Crocker et al., 2015a). The NTP found that having committed a high number of offences in the past or having a diagnosis of a psychotic spectrum disorder was found to reduce the odds of an individual receiving a conditional or absolute discharge; however, in ON, having a diagnosed mood disorder increased the odds of receiving conditional or absolute discharge (Crocker et al., 2015a). Further, if the index offence was classified as serious (i.e., offences causing death, attempt to cause death, and sexual offences), the probability of being discharged decreased compared to those who had committed other offences against a person or offences not against a person. Similarly, the odds of receiving conditional or absolute discharge were lower for those who had committed other crimes against a person when compared to those who committed other offences (Crocker et al., 2015a). Although the AB NCR Project also found the presence of a psychotic disorder to reduce the likelihood of receiving a conditional discharge, their model Cox regression model, which included variables such as criminal history, index offence severity, and diagnosis at NCR verdict, did not significantly predict conditional discharge decisions. It did, however, significantly predict absolute discharge decisions, despite no single variable significantly predicting absolute discharge on its own (Cheng et al., 2022). Further, the NTP found that NCRMD individuals were under the supervision of RBs for an average of 2.84 years before receiving an absolute discharge (Charette et al., 2015); however, those in the AB sample were supervised for twice as long, spending an average of 5.7 years under the AB RB (Haag et al., 2016).
Individuals with mental illness are often perceived by the public as dangerous and violent, a notion which is supported, and perhaps bolstered, by the representation of mental illness in the media which tends to focus disproportionately on such factors as well as criminality (Whitley & Berry, 2013). These perceptions of dangerousness likely serve to garner support for political agendas (e.g., Bill C-14) which aim to increase limits applied to those individuals with mental illness who have been found NCRMD and committed the most violent and serious crimes (i.e., high-risk accused) in the name of public safety. However, this logic only holds if high-risk accused are indeed more likely to recidivate than other NCRMD acquittees (Charette et al., 2015). Regardless of whether they were still under the purview of an RB, 16.7% committed a new offence in the 3 years following their index verdict, 20% in the 3 years following conditional discharge, and 22% in the 3 years following absolute discharge (Charette et al., 2015). Interestingly, the lowest rates of recidivism were found for individuals whose index NCRMD offence had been classified as severe – during the 3-year follow-up period only 6% had reoffended. The recidivism rate was 15.3% for those who committed a less severe index offence against a person and 21.6% for those who committed an index offence not against a person. Further, the rate of recidivism for a severe violent offence after 3 years was a mere 0.6% (Charette et al., 2015). Although individuals who had committed more criminal offences prior to their NCRMD verdict were found to be at a greater risk of recidivating, the primary psychiatric diagnosis was not related to recidivism risk; however, comorbid SUDs and PDs were both found to increase the risk of recidivism (Charette et al., 2015).
More recently, Chaimowitz et al. (2022) conducted an overview of the ON forensic mental health population, which included both those found NCRMD and UST, accounting for 91.6% and 8.4% of the sample, respectively. Similar to previous research, psychotic spectrum disorders were the most frequent diagnosis, followed by SUDs and PDs (Chaimowitz et al., 2022). Interestingly, the rates of violent index offences, including assault, sexual offences, and offences causing or attempting to cause death, were higher than those found in the NTP. Likewise, rates of deprivation of freedom offences, property offences, weapons offences, and administration of justice offences were lower when comparing to the NTP sample, which the authors suggest may be due to ON’s court diversion program which functions to divert minor offences early in the court process (Chaimowitz et al., 2022). Such findings highlight provincial differences in the characteristics, processing, and management needs of forensic psychiatric patients in Canada and underscore the importance of examining this population across all jurisdictions.
Despite decades of operation of Saskatchewan (SK)’s primary psychiatric facility and the flow of patients in and out of the provincial forensic mental health system, a formal examination of SK’s forensic psychiatric population has been notably absent from the field. The current study, therefore, profiles the population of forensic psychiatric patients admitted to SHNB up to and including January 2022 by examining sociodemographic variables, mental health symptomatology and diagnoses, and patterns of historic and post-discharge offending. This research seeks to broaden the Canadian forensic mental health literature base and is the first to examine the entire known forensic psychiatric population in SK. The following primary research questions were posed:
The current study is a retrospective archival investigation of 138 forensic psychiatric patients who had been found either NCRMD or UST in SK and were admitted to the forensic inpatient unit at SHNB between November 1984 and January 2022. The mean age of the sample at admission was 35.8 years (range = 17.0–74.9 years) with about two-thirds (64.5%, n = 89) of individuals having been found NCRMD and the remaining one-third (35.5%, n = 49) found UST.
SHNB has been the primary facility housing forensic psychiatric patients since its opening in 1913 (Smith, n.d.) and is operated by the Saskatchewan Health Authority. It is a psychiatric hospital providing inpatient mental health services, including assessment and intervention services, to voluntary, involuntary, and court-ordered admissions. Given that the facility has been in operation for over a century, it is not unexpected that program offerings have undergone significant changes since first opening while relying on the use of psychotropic medications to stabilize and treat acute symptoms of mental disorder. During the period studied, patients had consistent access to a variety of group and individual treatment services targeting criminogenic needs and which include offence-specific programming, mental health, and substance use. Patients also have access to psychoeducation modules addressing a variety of psychosocial factors, including mental illness awareness, medication management, social skills, relationships, life skills, vocational skills, recreation, life skills, court preparation, and behaviour management.
A data collection protocol was developed for the collection of key variables from file including sociodemographic information, Diagnostic and Statistical Manual (DSM) diagnoses (assigned by a staff psychiatrist or psychologist), mental status during the index offence, offence history, institutional functioning, discharge, patterns of rehospitalization, and recidivism, based on the original NTP (Crocker et al., 2015a,b,c,d). Recidivism was defined as any new criminal charge or conviction incurred post-discharge as verified through official criminal records (see the “Procedure” section for details). With the exception of one patient who reoffended violently following admission sufficient to receive outside charges (see Results), patients who had not been discharged were excluded from recidivism counts. This definition, and its breakdown into subsequent categories, is consistent with operationalizations of recidivism in correctional and forensic samples (e.g., Charette et al., 2015; Rice et al., 2013). To permit direct comparisons with the Canadian NTP, we report both overall rates of recidivism as well as 3-year rates of recidivism (i.e., per Charette et al., 2015). Violent recidivism was defined as any new offence against the person with the potential for physical or psychological harm (e.g., assault, robbery, homicide, threats), including sexual offences, while general recidivism was defined as a new charge or conviction incurred for any category of offence, violent or nonviolent.
Ethical approval to conduct the present study was provided by the Behavioural Ethics Review Board of the first and last author’s academic institution, and operational approval was provided by the Saskatchewan Health Authority. All data were coded by the first author and two research assistants (RAs) – a clinical psychology graduate student and a hospital staff therapist – from archival file information. Approximately, the first five files were coded by the first author and graduate student RA to ensure accuracy and consistency in coding. Additionally, files were co-coded periodically throughout the data collection period to help protect against rater drift. Demographic information, clinical and diagnostic information, RB hearing details, and offence information were collected from each file as information detail and quality permitted, including psychological and psychiatric assessment reports, case notes, legal and health documentation, and RB hearing documents. Recidivism data and specifics regarding offence history were coded from official criminal records, obtained from the Canadian Police Information Centre through the Royal Canadian Mounted Police. Recidivism variables were coded by the first and last authors subsequent to data collection from hospital records to avoid criterion contamination by ensuring that raters were unaware of outcomes.
Research questions were examined through analysis of descriptive information, including frequencies and means (with standard deviation (SD)) of key study variables. Formal comparisons were also conducted between individuals with NCRMD versus UST verdicts on sociodemographic, diagnostic, index mental state, offence history frequencies, and recidivism outcomes; the latter included associations between diagnosis and recidivism outcome. In addition, given that multiple trajectory projects have been conducted examining characteristics of NCRMD patients in four Canadian provincial jurisdictions –BC, ON, and QC (Crocker et al., 2015b) and AB (Haag et al., 2016) – the frequencies of these characteristics are described and the ratios of frequencies to total admissions within a catchment period are formally compared with SK across jurisdictions. All comparisons were conducted via chi-square (or t-test and analysis of variance for continuous variables), with phi (ϕ) correlations or equivalent (e.g., contingency coefficient, C) computed as a measure of the strength of association between two binary variables.
Sociodemographic characteristics for the sample as a function of verdict type are reported in Table I. The sample was comprised of 117 male (84.8%) and 21 (15.2%) female patients, with a mean age at the time of the index offence of 32.9 years (SD = 12.5; range 12.8–74.6 years). Individuals found NCRMD were slightly older (33.8 years (SD = 12.1)) than individuals found UST (31.2 years (SD = 13.3)); however, this difference was not significant, t(131) = 1.147, p = 0.232. The length of hospitalization of the sample, including those who had been discharged and those remaining in hospital, was 4.4 years. Those found NCRMD were hospitalized for longer (5.2 years (SD = 5.9)) than those found UST (2.9 years (SD = 4.0)), on average, t(122.673) = 2.885, p = 0.004. In terms of ethnic background, 44.2% of the sample was identified as White, 42.0% as Indigenous, and 7.2% as persons of colour (other than Indigenous). Those found UST were more likely to be identified as Indigenous or persons of colour than those found NCRMD, and most of the sample had been born in Canada. Overall, 38.4% of the sample had completed high school, with those found NCRMD more likely to have completed than those found UST. Prior to hospitalization, the majority of individuals were single and were receiving some form of income assistance. Roughly one-quarter of individuals were living independently prior to being admitted to hospital, about 15% were of no fixed address, and the majority were residing with a spouse, family member, or friends, or in another supervised setting (e.g., group home). Those found UST were more likely to be living in a supervised setting or with a spouse, family member, or friends, than those found NCRMD.
TABLE I Sample sociodemographic characteristics as a function of NCRMD/UST status
| Sociodemographic Characteristic | NCRMD | UST | Total | χ2 |
|---|---|---|---|---|
| % (n)/M (SD) | ||||
| Sex | 1.48 | |||
| Female | 18.0 (16) | 10.2 (5) | 15.2 (21) | |
| Male | 82.0 (73) | 89.8 (44) | 84.8 (117) | |
| Age at offence (years) | 33.8 (12.1) | 31.2 (13.3) | 32.9 (12.5) | |
| Length of hospitalization (years) | ||||
| Discharged | 5.2 (5.8) | 2.7 (3.8) | 4.3 (5.3) | |
| Not discharged | 5.3 (7.0) | 7.3 (7.2) | 5.7 (6.7) | |
| Total | 5.2 (5.9) | 2.9 (4.0) | 4.4 (5.4) | |
| Ethnicity | 20.90*** | |||
| Indigenous | 31.5 (28) | 61.2 (30) | 42.0 (58) | |
| White | 58.4 (52) | 18.4 (9) | 44.2 (61) | |
| Other POC | 4.5 (4) | 12.2 (6) | 7.2 (10) | |
| High school completed | 9.45** | |||
| Yes | 48.3 (43) | 20.4 (10) | 38.4 (53) | |
| No | 51.7 (46) | 75.5 (37) | 60.1 (83) | |
| Relationship status | 1.08 | |||
| In a relationship | 10.1 (9) | 16.3 (8) | 12.3 (17) | |
| Single | 88.8 (79) | 83.7 (41) | 87.0 (120) | |
| Country of birth | 2.17 | |||
| Canada | 91.0 (81) | 79.6 (39) | 87.0 (120) | |
| Other | 5.6 (5) | 12.2 (6) | 8.0 (11) | |
| Residential status | 25.39*** | |||
| Living alone | 38.2 (34) | 4.1 (2) | 26.1 (36) | |
| Living with spouse, family, or friends | 31.5 (28) | 44.9 (22) | 36.2 (50) | |
| Supervised setting | 10.1 (9) | 24.5 (12) | 15.2 (21) | |
| No fixed address | 11.2 (10) | 22.4 (11) | 15.2 (21) | |
| Other | 5.6 (5) | 0 (0) | 3.6 (5) | |
| Income | 3.11 | |||
| Own paid work (or partner) | 20.2 (18) | 10.2 (5) | 16.7 (23) | |
| Pension, EI, and (or) welfare | 25.8 (23) | 26.5 (13) | 26.1 (36) | |
| Disability Support Program | 18.0 (16) | 12.2 (6) | 15.9 (22) | |
| Other | 5.6 (5) | 10.2 (5) | 72 (10) | |
EI = employment insurance; NCRMD = not criminally responsible on account of mental disorder; POC = persons of colour; SD = standard deviation; UST = unfit to stand trial.
**p < 0.01, ***p < 0.001.
Table II reports sample mental health characteristics at the time of index offence and during hospital stay leading up to the first RB hearing. The most frequent diagnosis at the first RB hearing was a psychotic spectrum disorder, followed by intellectual disability or cognitive impairment; however, there were significant differences between the primary diagnoses of those found NCRMD versus UST. The majority of individuals found NCRMD were diagnosed with a psychotic spectrum disorder, with mood disorders accounting for 13.5% of the NCRMD sample, and several individuals being diagnosed with other disorders (e.g., attention deficit hyperactivity disorder, intermittent explosive disorder, autism spectrum disorder). Conversely, those found UST were equally likely to be diagnosed with a psychotic spectrum disorder or an intellectual/cognitive disorder; no individuals received a mood disorder as their primary diagnosis. On average, half of the sample had been diagnosed with a SUD, all of whom also had a concurrent diagnosis of another disorder. Moreover, there were significant differences between those found NCRMD and UST, with more than half those found NCRMD (52.8%) having received a diagnosis of a PD in contrast to a little more than a quarter (28.6%) of individuals found UST.
TABLE II Diagnostic and index offence mental state characteristics as a function of NCRMD/UST status
| Mental Health Measure | NCRMD % (n) |
UST % (n) |
Total % (n) |
χ2 |
|---|---|---|---|---|
| Clinical diagnosis | ||||
| Psychotic spectrum disorder | 83.1 (74) | 49.0 (24) | 71.0 (98) | 17.92*** |
| Mood spectrum disorder | 13.5 (12) | 0.0 (0) | 8.7 (12) | 7.24** |
| Intellectual/cognitive | 7.9 (7) | 63.3 (31) | 27.5 (38) | 48.61*** |
| Other diagnosis | 21.3 (19) | 40.8 (20) | 28.3 (39) | 5.91* |
| Substance use disorder | 55.1 (49) | 46.9 (23) | 52.2 (72) | 0.84 |
| Personality disorder | 52.8 (47) | 28.6 (14) | 44.2 (61) | 7.53** |
| Mental state at time of index offence | ||||
| Any psychotic symptom | 82.0 (73) | 30.6 (15) | 83.0 (88) | 27.73*** |
| Hallucinations – specified | 36.0 (32) | 16.3 (8) | (40) 38.5 | 2.48 |
| Delusions – specified | 70.0 (62) | 24.5 (12) | 69.8 (74) | 17.64*** |
| Suicidal ideation | 5.6 (5) | 2.0 (1) | 5.5 (6) | 0.42 |
| Suicide attempt | 0.0 (0) | 0.0 (0) | 0.0 (0) | N/A |
| Self-harm | 0.0 (0) | 2.0 (1) | 0.9 (1) | 2.46 |
| Homicidal ideation | 37.1 (33) | 12.2 (6) | 36.1 (39) | 5.29* |
| Homicide attempt | 25.8 (23) | 6.1 (3) | 24.1 (26) | 4.50* |
| Substance use and/or under influence | 15.7 (14) | 16.3 (8) | 21.4 (22) | 0.93 |
N/A = not applicable; NCRMD = not criminally responsible on account of mental disorder; UST = unfit to stand trial.
*p < 0.05, **p < 0.01, ***p < 0.001.
Among those cases mentioning the individual’s mental state at the time of the offence (see Table II), the majority of individuals were noted to have been experiencing psychotic symptoms, with hallucinations specifically mentioned in slightly over two-thirds of cases and delusions mentioned in just over one-third. Although those found NCRMD were more likely to be mentioned to have been experiencing any symptom of psychosis, they were significantly more likely to be reported to have been experiencing delusions at the time of the offence than those found UST. Suicidality and self-harm were mentioned infrequently; approximately 5% of the sample was reported to have been experiencing suicidal ideation and one individual to have engaged in self-harming behaviour, with no one in the sample being reported to have attempted suicide at the time of the offence. Homicidal ideation and attempts were mentioned in just over one-third and approximately one-quarter of cases, respectively. Those found NCRMD were significantly more likely to have been reported as experiencing homicidal ideation or having attempted to commit homicide. Approximately 20% of the sample was described as having engaged in substance use and/or being intoxicated at the time of the offence.
Table III reports the sample criminal history and index offence characteristics as a function of verdict status. Prior to being found NCRMD or UST, approximately two-thirds (68.1%) of individuals had any prior charge or conviction, while a similar proportion had a prior nonviolent offence (62.3%), approximately half (49.3%) had a prior violent offence, while the least frequent was a prior sexual offence (11/6%). Individuals found NCRMD had nonsignificantly higher rates of any prior offence or nonviolent offence than individuals found UST with small effects (ϕ = 0.14), while the two groups generally had the same rates of prior violent or sexual offences.
TABLE III Sample index offence and criminal history characteristics as a function of NCRMD/UST status
| Criminological Measure | NCRMD % (n) |
UST % (n) |
Total % (n) |
χ2 | ϕ |
|---|---|---|---|---|---|
| Index (verdict) offences | |||||
| Homicide, actual/attempted | 29.2 (26) | 6.1 (3) | 21.0 (29) | 10.15*** | 0.27 |
| Sexual offences | 5.6 (5) | 24.5 (12) | 12.3 (17) | 10.42*** | 0.28 |
| Assaults | 38.2 (34) | 55.1 (27) | 44.2 (61) | 3.66 | 0.16 |
| Threats | 16.9 (15) | 16.3 (8) | 16.7 (23) | 0.01 | 0.01 |
| Arson | 12.4 (11) | 2.0 (1) | 8.7 (12) | 4.24* | 0.18 |
| Property offences | 28.1 (25) | 46.9 (23) | 34.8 (48) | 4.95* | 0.19 |
| Weapons | 15.7 (14) | 8.2 (4) | 13.0 (18) | 1.60 | 0.11 |
| Technical violations/escape | 27.0 (24) | 51.0 (25) | 35.5 (49) | 7.99** | 0.24 |
| Other offences | 18.0 (16) | 20.4 (10) | 18.8 (26) | 0.12 | 0.03 |
| Criminal history (pre-verdict) | |||||
| Nonviolent priors | 67.4 (60) | 53.1 (26) | 62.3 (86) | 2.77 | 0.14 |
| Violent priors | 49.4 (44) | 49.0 (24) | 49.3 (68) | 0.00 | 0.00 |
| Sexual offence priors | 11.2 (10) | 12.2 (6) | 11.6 (16) | 0.03 | 0.02 |
| Any priors | 73.0 (65) | 59.2 (29) | 68.1 (94) | 2.79 | 0.14 |
Note: NCRMD n = 89, UST n = 49.
NCRMD = not criminally responsible on account of mental disorder; UST = unfit to stand trial.
*p < 0.05, **p < 0.01, ***p < 0.001.
Overall, there were 501 index offences committed by the 138 individuals comprising the current sample; Table III reports the frequency of a given offence category as a function of verdict status. Several themes were evident. First, approximately one-fifth of the sample had committed homicide or attempted homicide (21.0%), approximately one-half had an assault of some form (44.2%), and a little more than one-third a property offence (34.8%) or a technical violation (35.5%) including escapes and offences relating to the administration of justice, such as obstruction. The differences between persons found NCRMD versus UST were notable; persons found UST had significantly higher rates of several offence categories across their index verdict offences including assaults (55.1% vs. 38.2%), sexual offences (24.5% vs. 5.6%), property offences (46.9% vs. 28.1%), and technical violations (51.0% vs. 27.0%); persons found NCRMD, however, had significantly higher rates of homicide (29.2% vs. 6.1%).
Overall, 126 patients were discharged to the community and one individual, not released, reoffended violently in hospital leading to court-adjudicated criminal charges, bringing the outcome total to 127 cases. In total, 19 individuals (15%) were charged or convicted of a new violent offence and 23 individuals (18.1%) of any new offence, including violent and nonviolent. Of those found NCRMD, 10 individuals (12.7%) reoffended generally, with 7 of those individuals (8.9%) recorded as having reoffended violently. By contrast, higher rates of recidivism were observed in UST patients, with 13 (27.7%) having incurred a new charge or conviction for any new offence and 12 (25.5%) reoffending violently. In terms of recidivism outcome types, assault was most frequently observed to be the most severe re-offence, accounting for the most severe re-offences of 17 individuals (approximately 60%) who recidivated. For three individuals (approximately 10%), a sexual assault was their most serious re-offence, while a property offence was the most serious for another three patients. Two individuals (approximately 7%) had committed offences related to making threats as their most severe, and two other had committed offences against the administration of justice. Lastly, one individual’s most serious re-offence was a weapons offence and one individual had committed an offence causing, or attempting to cause, death.
Table IV reports the rates of violent and general recidivism for community discharged patients as a function of the diagnostic group. Nonsignificant inverse associations were found between psychotic spectrum disorder and violent or general recidivism, as was also the case for Cluster A diagnoses. By contrast, nonsignificantly higher rates of violent and general recidivism were found for individuals diagnosed with PDs or some variation on ASPD; the only significant effect by diagnosis was for SUDs, which had significantly higher rates of general recidivism (25.0%) relative to individuals without a substance use diagnosis (9.7%), χ2 (1, N = 126) = 5.13, p < 0.05, ϕ = 0.20.
TABLE IV Post-discharge violent and general recidivism outcomes as a function of clinical diagnosis
| Clinical Diagnosis | Recidivism Status % (n) |
χ2 | ϕ | |
|---|---|---|---|---|
| Diagnostic Group | Comparison | |||
| Violent recidivism | ||||
| Psychotic spectrum disorder | 11.5 (10/87) | 0.5 (8/39) | 1.79 | −0.12 |
| Mood spectrum disorder | 16.7 (2/12) | 14.0 (16/114) | 0.06 | 0.02 |
| Intellectual/cognitive | 22.9 (8/35) | 11.0 (10/91) | 2.91 | 0.15 |
| Other diagnoses | 18.2 (6/33) | 13.8 (13/94) | 0.36 | 0.05 |
| Substance use disorder | 18.8 (12/64) | 9.7 (6/62) | 2.11 | 0.13 |
| Personality disorder | 14.5 (8/55) | 14.1 (10/71) | 0.00 | 0.01 |
| Cluster A | 0.0 (0/11) | 15.7 (18/115) | 2.00 | −0.13 |
| ASPD or traits | 20.5 (8/39) | 11.5 (10/87) | 1.79 | 0.12 |
| General recidivism | ||||
| Psychotic spectrum disorder | 13.8 (12/87) | 5.6 (10/39) | 2.62 | −0.14 |
| Mood spectrum disorder | 33.3 (4/12) | 15.8 (18/114) | 2.32 | 0.14 |
| Intellectual/cognitive | 25.7 (9/35) | 14.3 (13/91) | 2.29 | 0.14 |
| Other diagnoses | 18.2 (6/33) | 18.1 (17/94) | 0.00 | 0.00 |
| Substance use disorder | 25.0 (16/64) | 9.7 (6/62) | 5.13* | 0.20 |
| Personality disorder | 20.0 (11/55) | 15.5 (11/71) | 0.44 | 0.06 |
| Cluster A | 0.0 (0/11) | 19.1 (22/115) | 2.55 | −0.14 |
| ASPD or traits | 23.1 (9/39) | 14.9 (13/87) | 1.24 | 0.10 |
ASPD = antisocial personality disorder.
*p < 0.05.
The final set of analyses compared sociodemographic, clinical, and criminolegal characteristics of NCRMD patients in the present SK sample to those reported in trajectory projects of these patients in four additional Canadian jurisdictions reported in Table V. Although the data capture occurred across different catchment periods (5 years for Crocker et al., 2015b; 74 years for Haag et al., 2016; 38 years for current study), the reporting of frequencies of NCRMD patients using a common unit of measurement in each study permitted comparing the ratios relative to total admissions across the five jurisdictions to examine overall trends.
TABLE V Comparative analysis of characteristics of NCRMD admissions (%) across five Canadian provinces
| Measure | SK (n = 89) | BCa (n = 222) | ABb (n = 546) | ONa (n = 484) | QCa (n =1,094) | χ2/F | C |
|---|---|---|---|---|---|---|---|
| Sex | |||||||
| Female | 18.0 (16) | 14.9 (33) | 15.9 (87) | 16.7 (81) | 15.2 (166) | 1.11 | 0.02 |
| Male | 82.0 (73) | 85.1 (189) | 84.1 (459) | 83.3 (403) | 84.8 (927) | ||
| Age at offence (M/SD, years)c | 33.8 (12.1) | 36.1 (12.5) | 35.3 (-) | 37.2 (12.0) | 36.4 (12.6) | 2.01 | - |
| Ethnicity | |||||||
| Indigenous | 31.5 (28) | 7.7 (17) | 8.6 (47) | 4.5 (22) | 1.3 (14) | 172.82*** | 0.26 |
| White | 58.4 (52) | - | 72.9 (398) | - | - | ||
| Other POC | 4.5 (4) | - | 17.3 (94) | - | - | ||
| High school completed | |||||||
| Yes | 48.3 (43) | 53.3 (112) | 30.0 (164) | 48.3 (215) | 48.4 (296) | 53.42*** | 0.17 |
| No | 51.7 (46) | 46.7 (98) | 67.0 (367) | 51.7 (230) | 51.6 (315) | ||
| Relationship status | |||||||
| In a relationship | 10.1 (9) | 16.7 (37) | 16.7 (91) | 19.9 (85) | 14.8 (149) | 8.10 | 0.06 |
| Single | 88.8 (79) | 83.3 (185) | 83.2 (454) | 80.1 (343) | 85.2 (857) | ||
| Country of birth | |||||||
| Canada | 91.0 (81) | 73.0 (157) | - | 65.6 (267) | 63.4 (322) | 35.65*** | 0.17 |
| Other | 5.6 (5) | 27.0 (58) | - | 34.4 (140) | 36.6 (186) | ||
| Residential status | |||||||
| Living alone | 38.2 (34) | 29.4 (62) | - | 20.7 (83) | 36.6 (348) | 40.82 | 0.15 |
| Living with spouse, family, or friends | 31.5 (28) | 44.0 (93) | - | 49.9 (200) | 41.3 (392) | ||
| Supervised setting | 10.1 (9) | 5.7 (12) | - | 10.2 (41) | 8.2 (78) | ||
| No fixed address | 11.2 (10) | 9.5 (20) | - | 5.2 (21) | 10.8 (103) | ||
| Other | 5.6 (5) | 11.4 (24) | - | 14.0 (56) | 3.1 (29) | ||
| Income | |||||||
| Own paid work (or partner) | 20.2 (18) | 17.1 (35) | - | 15.3 (47) | 15.7 (135) | 10.53* | 0.07 |
| Pension, EI, and (or) welfare | 25.8 (23) | 67.2 (137) | - | 67.2 (207) | 74.1 (639) | ||
| Disability Support Program | 18.0 (16) | - | - | - | - | ||
| Other | 5.6 (5) | 15.7 (32) | - | 17.5 (54) | 10.2 (88) | ||
| Clinical diagnosis | |||||||
| Psychotic spectrum disorder | 83.1 (74) | 76.5 (170) | 75.6 (412) | 79.7 (380) | 65.9 (718) | 43.26*** | 0.13 |
| Mood spectrum disorder | 13.5 (12) | 18.5 (41) | 29.1 (159) | 14.0 (67) | 28.1 (306) | 72.41*** | 0.17 |
| Intellectual/cognitive | 7.9 (7) | - | 5.1 (28) | - | - | 1.10 | 0.05 |
| Other diagnosis | 21.3 (19) | - | 26.7 (146) | - | - | 1.16 | 0.05 |
| Substance use disorder | 55.1 (49) | 39.2 (87) | 39.2 (214) | 31.7 (151) | 28.7 (312) | 43.13*** | 0.13 |
| Personality disorder | 52.8 (47) | 9.5 (21) | - | 12.2 (58) | 10.2 (111) | 139.30*** | 0.26 |
| Mental state at time of index offence | |||||||
| Any psychotic symptom | 82.0 (73) | 92.3 (205) | - | 71.9 (348) | 44.2 (483) | 253.95*** | 0.34 |
| Hallucinations – specified | 36.0 (32) | 37.4 (83) | - | 23.8 (115) | 14.5 (159) | 78.67*** | 0.20 |
| Delusions – specified | 70.0 (62) | 78.4 (174) | - | 53.1 (257) | 36.5 (399) | 174.74*** | 0.29 |
| Suicidal ideation | 5.6 (5) | 9.9 (22) | - | 5.0 (24) | 6.2 (68) | 6.46 | 0.06 |
| Suicide attempt | 0.0 (0) | 6.3 (14) | - | 2.3 (11) | 0.5 (6) | 40.71*** | 0.15 |
| Self-harm | 0.0 (0) | 2.7 (6) | - | 4.1 (20) | 0.4 (4) | 33.71*** | 0.13 |
| Homicidal ideation | 37.1 (33) | 7.2 (16) | - | 15.5 (75) | 1.6 (18) | 210.48*** | 0.32 |
| Substance use and/or under influence | 15.7 (14) | 23.9 (53) | - | 21.7 (105) | 23.6 (259) | 3.48 | 0.05 |
| Index (verdict) offences | |||||||
| Homicide, actual/attempted | 29.2 (26) | 8.1 (18) | 18.5 (101) | 11.6 (56) | 4.6 (50) | 126.95*** | 0.22 |
| Sexual offences | 5.6 (5) | 1.8 (4) | 6.2 (34) | 3.7 (18) | 1.7 (19) | 35.15*** | 0.12 |
| Assaultsd | 38.2 (34) | 32.9 (73) | 46.9 (256) | 26.2 (127) | 25.4 (278) | 150.54*** | 0.25 |
| Threats | 16.9 (15) | 23.0 (51) | - | 25.6 (124) | 29.0 (318) | 9.33* | 0.07 |
| Arson | 12.4 (11) | - | 6.6 (36) | - | - | 4.14* | 0.08 |
| Property offences | 28.1 (25) | 13.5 (30) | - | 10.7 (52) | 20.2 (222) | 30.67*** | 0.13 |
| Weapons | 15.7 (14) | 9.9 (22) | 16.6 (91) | 7.9 (38) | 4.6 (50) | 155.18*** | 0.21 |
| Criminal history (pre-verdict) | |||||||
| Nonviolent priors | 67.4 (60) | 36.9 (82) | - | 44.8 (217) | 41.5 (454) | 26.64*** | 0.12 |
| Violent priors | 49.4 (44) | 23.4 (52) | - | 37.6 (182) | 29.4 (322) | 30.31*** | 0.13 |
| Any priors | 73.0 (65) | 40.1 (89) | - | 50.2 (243) | 46.3 (506) | 30.17*** | 0.13 |
| Recidivism | |||||||
| General recidivism (3 years) | 5.8 (4) | 9.5 (21) | - | 9.3 (45) | 21.5 (229) | 51.69*** | 0.17 |
aCrocker et al. (2015b), which includes a 5-year catchment period (2000–2005) for all NCRMD admissions across British Columbia (BC), Ontario (ON), and Quebec (QC).
bHaag et al. (2016), which includes the entire NCRMD population in Alberta’s (AB) history of use of the legislation, akin to the present SK study and sample.
cCrocker et al. (2015b) reported age at NCRMD verdict, and Haag et al. (2016) reference both but notes no significant difference.
dHaag et al. (2016) reports a “total direct violence category” excluding homicide-related offences in lieu of separate assault or threats categories, but we list this as a proxy for assault as generally the highest category of offences.
Note: Percentages reported unless otherwise stated.
AB = Alberta; EI = employment insurance; NCRMD = Not Criminally Responsible on account of Mental Disorder; POC = persons of colour; SD = standard deviation; SK = Saskatchewan.
*p < 0.05, *** p < 0.001.
First, there were continuities in the frequencies of certain sociodemographic characteristics, including ratios of male to female patients, age at index offence or on admission, residential status, and relationship status; notable areas of disparities with significant differences reported included particularly low rates of high school completion for AB with Canada being the most common country of origin for SK patients, who were also most likely to have an independent source of income. Perhaps most notably, SK had the highest proportion of Indigenous NCRMD patients compared to the other four jurisdictions, ranging from 3.6 times higher than the next highest reporting province (AB) to nearly 25 times higher than the lowest reporting province (QC).
Second, clinically, SK NCRMD patients had the highest rates of psychotic spectrum disorder, SUD, and PD, while AB and QC had the highest rates of mood disorder. In terms of mental state at the time of the index offence, overwhelmingly, the patients in most jurisdictions were experiencing psychotic symptoms with the exception of QC, with SK and BC reporting the highest rates of specified hallucinations and delusions. Suicide and self-harming behaviour or ideation was very uncommon across all five jurisdictions, although slightly higher rates were observed in BC and ON. Further, SK had the highest rates of homicide ideation in its patient population, but the lowest rate of substance use or being under the influence (i.e., substance intoxication) at the time of the index offence.
Finally, in terms of criminolegal characteristics, SK had the highest base rates of most offence categories for their index offence, including actual or attempted homicide, arson, and property offences, and alongside AB, the two Prairie Provinces shared the highest rates of sexual offences, assaults, and weapon-related offences. SK patients also had the highest frequency of any formal criminal history, including for prior violent offences, nonviolent offences, and any prior offences; paradoxically, however, they had the lowest rates of 3-year post-discharge general recidivism across the reporting jurisdictions.
The present study described the NCRMD and UST populations in SK in terms of their sociodemographic, mental health, and criminological characteristics. Overall, those found NCRMD in SK shared many similarities with those from other provinces in terms of sex, age, education, and relationship status; however, more individuals were reported to be living independently in SK, and less likely to be living with a spouse, family member, or friend than observed in BC, ON, and QC (Crocker et al., 2015b). Consistent with base rates in the province, a larger proportion of the NCRMD sample was Indigenous compared to AB, BC, ON, and QC (Chaimowitz et al., 2022; Crocker et al., 2015b; Haag et al., 2016), a difference that upon formal examination was statistically significant. Despite the higher representation of Indigenous peoples among the general population in SK versus some other provinces, this still constitutes an overrepresentation of Indigenous people, although not as significant as that observed in the adult incarcerated population in SK (Clark, 2019; Statistics Canada, 2026). Given that a significantly greater proportion of the UST sample in the current study was Indigenous, it is possible that the NCRMD defence is disproportionately utilized. It is possible that the defence is simply considered less often by those representing Indigenous peoples in court, or that it is strategically avoided by legal professionals representing Indigenous clients in an attempt to avoid what is essentially an indeterminate sentence.
Those found NCRMD in SK were more likely to have been diagnosed with a psychotic spectrum disorder, SUD, PD, concurrent SUD and SMI, or concurrent PD and SMI, relative to documented NCRMD cases in BC, ON, and QC, and as verified through formal statistical comparisons (Crocker et al., 2015b). The current sample was largely similar to that described by Chaimowitz et al. (2022), with the exception of the rates of diagnosed PDs being higher in the SK sample. Notable differences were also observed between the NCRMD and UST subsamples in the current study, with rates of psychotic spectrum and mood disorders being higher among those found NCRMD, intellectual/cognitive disorders more prevalent among those found UST, and those found NCRMD more likely to be diagnosed with a PD and/or concurrent PD and SMI. Observed differences likely stem, in part, from differences in the effects of symptomatology on one’s capacity to understand right from wrong, at least in the eyes of the court. For instance, those presenting with an intellectual disability or cognitive impairment (e.g., congenital or due to brain injury) may be assessed as lacking capacity to effectively understand or participate in court proceedings while being capable of knowing that the crime committed (e.g., causing physical harm to someone) was wrong. Conversely, those found NCRMD may have had an altered perception of the morality and/or legality of their actions at the time of the offence but be capable of participating in and understanding legal proceedings once stabilized on psychotropic medications.
Rates of prior criminal justice system contact were considerably higher in the present sample than that observed in other jurisdictions (Chaimowitz et al., 2022; Crocker et al., 2015b), consistent with SK’s higher police-reported crime rate and crime severity index relative to other provinces (Statistics Canada, 2025). In addition, there were differences in the types of index offences leading to an NCRMD verdict compared to those in other provinces. Higher rates of violent offences (e.g., homicide, attempted homicide, assault) were reported among individuals found NCRMD in AB (Haag et al., 2016), whereas violent offences were more common among those found NCRMD in SK than those in BC, ON, and QC (Crocker et al., 2015b). This was the case for offences causing or attempting to cause death, sexual offences, and assault. However, offences involving threats, weapons, and dangerous driving were less frequently observed among those in SK compared to individuals found NCRMD in BC, ON, and QC (Crocker et al., 2015b). Compared to the sample described by Chaimowitz et al. (2022), the SK sample was more likely to have committed an index offence causing or attempting to cause death, a sexual offence, or a property offence, whereas the ON sample was observed to have higher proportions of assault, offences involving threats, weapon offences, and offences against the administration of justice.
Significant differences in the rates of offences causing or attempting to cause death and sexual offences among NCRMD and UST subpopulations in the current study may also be due, in part, to differences in experienced symptomatology consistent with observed mental health diagnoses. For instance, those experiencing symptoms of psychosis may be more likely to engage in severe physical violence and aggression in response to paranoia, suspiciousness, and command hallucinations, consistent with meta-analytic findings related to the relationship between violence and the positive symptoms of psychosis (Douglas et al., 2009). Conversely, those with intellectual/cognitive disorders may struggle to understand the nuances of interpersonal relationships and communication, contributing to difficulty understanding and obtaining ongoing sexual consent.
Regarding rates of recidivism, the current study’s NCRMD subsample demonstrated somewhat lower 3-year rates of general recidivism compared to those observed in BC and ON and considerably lower rates than QC, as reported in the NTP (Charette et al., 2015). Interestingly, the rate of recidivism of the UST subsample was more than double the rate among those found NCRMD in the current sample. It is possible that this could be due to differences in the length of hospitalization between subsamples (i.e., those found UST were hospitalized for a shorter duration, on average, than those found NCRMD), and/or differences in mental health symptomatology and response to psychotropic medication for purposes of stabilization. Compared to other provinces, violent offences accounted for a similar proportion of recidivistic offences among those comprising the SK sample (Charette et al., 2015). Despite public concern regarding the risk posed by those with a diagnosed mental disorder and history of offending, mental disorder diagnosis (including PDs) generally did not significantly predict either violent or general recidivism in the current sample. In fact, having been diagnosed with a psychotic spectrum disorder or Cluster A PD was associated with a reduced likelihood of reoffending, although these findings did not reach significance. However, like the findings of Charette et al. (2015), SUDs were found to be significantly associated with higher rates of general recidivism. Such findings are also consistent with the risk assessment literature which supports the prioritization of empirically supported risk/need factors, including substance use, in treatment and forensic decision making (e.g., Andrews & Bonta, 1994–2010; Bonta & Andrews, 2017–2024).
The average length of hospitalization of those found NCRMD (i.e., 5.2 years) in SK was comparable to the length of time individuals were supervised by the RB before receiving an absolute discharge in AB and over twice as long as the average length of time supervised by the RB in BC, ON, and QC (Charette et al., 2015; Haag et al., 2016). Importantly, this time period only accounts for each individual’s most recent hospitalization, not the length of RB supervision in its entirety (i.e., prior hospitalizations or time spent supervised in the community on conditional discharge prior to/following the coded period of hospitalization). As such, individuals found NCRMD in SK are likely supervised by the RB for longer periods of time than in other studied provinces. Decisions to hospitalize those found NCRMD in SK for longer may be influenced by risk level (e.g., the SK sample may be higher risk than other provinces), extraneous factors with varying relevance to risk (e.g., psychopathy, attractiveness, offence severity; Denomme et al., 2020; Wilson et al., 2015), lack of required resources and services in rural communities (e.g., approved/group homes, mental health services), or differing interpretations of what constitutes a significant threat to public safety.
This study contributes to the extant literature examining forensic mental health patients in Canada and is the first to examine the population of individuals found NCRMD and UST in SK, allowing for comparisons of the population and their oversight by the Saskatchewan RB to those of other provinces. Because the sample comprises the population of NCRMD and UST accused admitted to SHNB between November 1984 and January 2022, concerns regarding the generalizability of the data to other SHNB hospitalizations while supervised by the Saskatchewan RB are limited. The current study also utilized information coded from criminal records to examine rates of recidivism.
Several possible limitations are also present. This research was archival and retrospective in nature, meaning that data collection relied on hospital records rather than researchers’ own observations and assessments across time. Unsurprisingly, the quality of hospital records varied significantly across time, with older files having been “pruned” to only include those documents deemed essential for record keeping (e.g., psychological assessment reports). Although it was not possible to complete the entirety of the coding protocol for such cases, efforts were made to code the file as comprehensively as possible using the records available and often relying on Review Board Reasons for Decision and Disposition over Psychiatric Review Board Reports, which aided data capture. Other factors that may have impacted the quality of file information include staffing capacity for regular risk assessment completion, or perceived value of completing thorough written updates for those not observed to have made significant changes since the previous RB hearing. For these reasons, among others, it is possible that relevant information, including that related to risk (e.g., institutional functioning, mental health symptomatology) may not have been fully represented in file and subsequent data capture.
The current study also only examined those admitted to SHNB after being found NCRMD or UST and falling under the purview of the Saskatchewan RB. As such, those found NCRMD who received an absolute discharge by the courts at the time of the NCRMD verdict or those found NCRMD or UST who may have received a conditional discharge and were supervised in the community without ever being hospitalized at SHNB would not have been identifiable to the researchers of this study. Although such situations are likely to occur infrequently (Haag et al., 2016), additional data would need to be obtained directly from the Saskatchewan RB or the courts to confirm the presence of any such cases in SK’s history.
It is also noteworthy that the admission dates of those comprising the current sample span nearly 40 years, during which time a number of landmark cases and parliamentary bills (e.g., R. v. Chaulk, Bill C-30, R. v. Swain, and Winko v. British Columbia) have shaped the management of individuals with mental disorders through the criminal justice system. Most recently, Bill C-14 (passed on 11 July 2014) amended the Criminal Code of Canada to include the recognition of public safety as the paramount consideration in the RB’s decision-making process (Baillie, 2015; Dupuis, 2014). Although the current study spans changes to Part XX.1 of the Criminal Code of Canada, it was beyond its scope to evaluate the potential implications of these and other legislative changes, and warrants further research in SK and other jurisdictions.
Given differences in the length of RB supervision between provinces (Charette et al., 2015; Haag et al., 2016), future research should examine which factors (e.g., mental health, risk level, participation in programming, availability of community resources) influence RB discharge decisions. Examination of the relationships between recidivism and offence characteristics, such as the number of offences, offence type, and victims, would also be beneficial. Additionally, research collaboration with the Saskatchewan RB would provide a more fulsome illustration of the trajectories and outcomes (e.g., length of time from verdict to absolute discharge) of those found NCRMD and UST in SK.
The authors offer tremendous thanks to Saskatchewan Hospital North Battleford for their support of this research and to Dr. Anne Crocker and Dr. Malijai Caulet for sharing the National Trajectory Project (NTP) data coding protocol to facilitate data collection for several variables in the present study.
The present study was funded in part by a Social Sciences and Humanities Research Council (SSHRC) doctoral fellowship (No. 207-2019-2020-Q3-00629) awarded to the first author; the funder had no input to the design and execution of this research or the reporting of findings.
The authors have no conflicts of interest to declare.
University of Saskatchewan Behavioural Research Ethics Board (certificate No. Beh ID 2071); consent waived per TCPS-2 policy statement on secondary use of health information. Operational approval provided by the Saskatchewan Health Authority.
∗University of Saskatchewan, Saskatoon, Saskatchewan, Canada;
†Saskatchewan Hospital North Battleford, North Battleford, Saskatchewan, Canada.
Andrews, D. A., & Bonta, J. (1994–2010). The psychology of criminal conduct. LexisNexus.
Baillie, P. (2015). A valuable (and ongoing) study, the National Trajectory Project addresses many myths about the verdict of Not Criminally Responsible on Account of Mental Disorder. Canadian Journal of Psychiatry, 60(3), 93–95. https://doi.org/10.1177/070674371506000301 [CrossRef] [PubMed]
Bonta, J., & Andrews, D. A. (2017–2024). The psychology of criminal conduct. Routledge.
Chaimowitz, G., Moulden, H., Upfold, C., Mullally, K., & Mamak, M. (2022). The Ontario forensic mental health system: A population-based review. Canadian Journal of Psychiatry, 67(6), 481–489. https://doi.org/10.1177/07067437211023103 [CrossRef] [PubMed]
Charette, Y., Crocker, A. G., Seto, M. C., Salem, L., Nicholls, T. L., & Caulet, M. (2015). The national trajectory project of individuals found not criminally responsible on account of mental disorder in Canada. Part 4: Criminal recidivism. Canadian Journal of Psychiatry, 60(3), 127–134. https://doi.org/10.1177/070674371506000307 [CrossRef] [PubMed]
Cheng, J., Olver, M. E., Haag, A. M., & Wormith, J. S. (2022). Trajectories and outcomes of those not criminally responsible on account of mental disorder through a Canadian forensic system. International Journal of Forensic Mental Health, 21(4), 399–411. https://doi.org/10.1080/14999013.2021.2023708 [CrossRef]
Clark, S. (2019). Overrepresentation of Indigenous people in the Canadian criminal justice system: Causes and responses. Department of Justice Canada. https://www.justice.gc.ca/eng/rp-pr/jr/oip-cjs/index.html
Criminal Code of Canada. (1985a). Defence of mental disorder. R.S.C., 1985, c. C-46 (Section 16). https://laws-lois.justice.gc.ca/eng/acts/c-46/section-16.html
Criminal Code of Canada. (1985b). R.S.C., 1985, c. C-46 (Section 2). https://laws-lois.justice.gc.ca/eng/acts/c-46/section-2.html
Crocker, A. G., Charette, Y., Seto, M. C., Nicholls, T. L., Côté, G., & Caulet, M. (2015a). The national trajectory project of individuals found not criminally responsible on account of mental disorder in Canada. Part 3: Trajectories and outcomes through the forensic system. Canadian Journal of Psychiatry, 60(3), 117–126. https://doi.org/10.1177/070674371506000306 [CrossRef] [PubMed]
Crocker, A. G., Nicholls, T. L., Seto, M. C., Charette, Y., Côté, G., & Caulet, M. (2015b). The national trajectory project of individuals found not criminally responsible on account of mental disorder in Canada. Part 2: The people behind the label. Canadian Journal of Psychiatry, 60(3), 106–116. https://doi.org/10.1177/070674371506000305 [CrossRef] [PubMed]
Crocker, A. G., Nicholls, T. L., Seto, M. C., & Côté, G. (2015c). The national trajectory project of individuals found not criminally responsible on account of mental disorder in Canada. Canadian Journal of Psychiatry, 60(3), 96–97. https://doi.org/10.1177/070674371506000303 [CrossRef] [PubMed]
Crocker, A. G., Nicholls, T. L., Seto, M. C., Côté, G., Charette, Y., & Caulet, M. (2015d). The national trajectory project of individuals found not criminally responsible on account of mental disorder in Canada. Part 1: Context and Methods. Canadian Journal of Psychiatry, 60(3), 98–105. https://doi.org/10.1177/070674371506000304 [CrossRef] [PubMed]
Denomme, W. J., Curno, J., & Forth, A. (2020). Psychopathic traits, risk and protective factors, and attractiveness in forensic psychiatric patients: Their role in review board dispositions. Journal of Forensic Psychology Research and Practice, 20(3), 264–289. https://doi.org/10.1080/24732850.2020.1717904 [CrossRef]
Douglas, K. S., Guy, L. S., & Hart, S. D. (2009). Psychosis as a risk factor for violence to others: A meta-analysis. Psychological Bulletin, 135(5), 679–706. https://doi.org/10.1037/a0016311 [CrossRef] [PubMed]
Dupuis, T. (2014). Legislative summary of Bill C-14: An Act to amend the Criminal Code and the National Defence Act (mental disorder). Library of Parliament. https://lop.parl.ca/sites/PublicWebsite/default/en_CA/ResearchPublications/LegislativeSummaries/412C14E
Haag, A. M., Cheng, J., & Wirove, R. (2016). Describing the not criminally responsible population in Alberta’s history: Sociodemographic, mental health, and criminological profiles. Journal of Community Safety and Well-Being, 1(3), 68–74. https://doi.org/10.35502/jcswb.24 [CrossRef]
Jansman-Hart, E. M., Seto, M. C., Crocker, A. G., Nicholls, T. L., & Côté, G. (2011). International trends in demand for forensic mental health services. International Journal of Forensic Mental Health, 10(4), 326–336. https://doi.org/10.1080/14999013.2011.625591 [CrossRef]
Latimer, J., & Lawrence, A. (2006). The review board systems in Canada: An overview of results from the Mentally Disordered Accused Data Collection Study. Department of Justice Canada. https://www.justice.gc.ca/eng/rp-pr/csj-sjc/jsp-sjp/rr06_1/toc-tdm.html
Rice, M. E., Harris, G. T., & Lang, C. (2013). Validation of and revision to the VRAG and SORAG: The Violence Risk Appraisal Guide—Revised (VRAG-R). Psychological Assessment, 25(3), 951–965. https://doi.org/10.1037/a0032878 [CrossRef] [PubMed]
Sealy, P., & Whitehead, P. C. (2004). Forty years of deinstitutionalization of psychiatric services in Canada: An empirical assessment. Canadian Journal of Psychiatry, 49(4), 249–257. https://doi.org/10.1177/070674370404900405 [CrossRef] [PubMed]
Smith, C. M. (n.d.). Mental health services. In Encyclopedia of Saskatchewan. Canadian Plains Research Center, University of Regina. Retrieved from https://esask.uregina.ca/entry/mental_health_services.html
Statistics Canada. (2026). Overrepresentation of Indigenous and Black adults in provincial and federal custody. https://www150.statcan.gc.ca/n1/daily-quotidien/260114/dq260114b-eng.htm
Statistics Canada. (2025). Police-reported crime statistics in Canada, 2024. https://www150.statcan.gc.ca/n1/daily-quotidien/250722/dq250722a-eng.htm
Whitley, R., & Berry, S. (2013). Trends in newspaper coverage of mental illness in Canada: 2005-2010. Canadian Journal of Psychiatry, 58(2), 107–112. https://doi.org/10.1177/070674371305800208 [CrossRef] [PubMed]
Wilson, C. M., Crocker, A. G., Nicholls, T. L., Charette, Y., & Seto, M. C. (2015). The use of risk and need factors in forensic mental health decision-making and the role of gender and index offense severity. Behavioral Sciences and the Law, 33(1), 19–38. https://doi.org/10.1002/bsl.2162 [CrossRef] [PubMed]
*Correspondence to: Mark E. Olver, Department of Psychology and Health Studies, University of Saskatchewan, 9 Campus Drive, Arts Building Room 154, Saskatoon, Saskatchewan S7N 5A5, Canada. E-mail: mark.olver@usask.ca
This work is distributed under the Creative Commons BY-NC-ND license. To view a copy of this license, visit http://creativecommons.org/licenses/by-nc-nd/4.0/. For commercial re-use, please contact sales@sgpublishing.ca.
Journal of CSWB, VOLUME 11, NUMBER 3, September 2026