Read time: 2 minutes, 30 seconds
Summary risk ratings from structured professional judgment (SPJ) are not immune to racial disparities. However, these disparities are not proof of bias. All risk assessment tools may reflect societal disparities more than be a source of them.
During the October 31 session of the International Forensic Psychiatry Lecture Series, Drs. Neil Hogan and Gabriela Corăbian discussed their research on racial disparities in risk assessments using SPJ. (No recording is available for this lecture.)
The researchers emphasized that for risk assessments to be valid, they must first be reliable. They recommended embracing structured approaches while emphasizing that cultural awareness can and should co-exist with risk assessment guidelines. They also recommended looking beyond the actuarial (statistical) versus SPJ debate and remembering that psychologically meaningful risk factors offer practical and ethical advantages.
Actuarial scores are produced by algorithms that rely on normative data. That makes them highly reliable but inflexible. In contrast, SPJs are memory aids for evaluators to use at their discretion while conducting violence risk assessments. While this flexibility can be beneficial, it also introduces variability.
Indigenous People in Canada
Disparities in risk assessment must be understood in the broader context of Indigenous peoples’ experiences in the Canadian criminal justice system. Public Safety Canada data shows significant overrepresentation at every stage, including rates of criminal justice involvement, prison sentences, maximum-security placements, segregated confinement, and having their parole revoked.
Even when controlling for total risk factors, a 2014 study found that Indigenous people were 30 per cent more likely to receive high-risk ratings.
Drs. Hogan and Corăbian’s Research
Drs. Hogan and Corăbian studied 190 adults referred for specialized risk or threat assessment. The study participants were 96 per cent male, with an average age of 34.6 years and an average of 21 prior convictions. While 49 per cent were white, 43 per cent were Indigenous, and eight per cent were classified as other.
Indigenous people had higher SARA-V3 risk scores, with 95 per cent receiving high-risk ratings, compared with 79 per cent of non-Indigenous individuals.
Recidivism rates were higher among Indigenous participants, with 79 per cent facing new charges, compared with 62 per cent of non-Indigenous individuals. Violent recidivism was also more common (55 per cent and 30 per cent, respectively).
Their research also found a difference in police responses. Police-reported violence resulted in no charges on at least one occasion for 67 per cent of Indigenous people and 50 per cent of non-Indigenous people.
There was little variability in the overall sample, with 86 per cent being considered high risk.
SPJ item scales are narrow. For example, “present” problems with substance use could mean intermittent alcohol use problems or daily meth use persisting for decades.
The speakers emphasized that convictions and risk factor sums do not include everything that matters for a summary risk rating, such as the severity of the offence nor the violence that is not included in the person’s criminal record.
The findings demonstrated disparity but not necessarily clear user bias. However, these findings reflect SPJ ratings in a controlled environment, with raters receiving close oversight.
Risk rating disparities do not exist in isolation. In the real world, outcomes are influenced by historical and systemic factors. The legacy of residential schools has contributed to intergenerational trauma and higher rates of substance use, mental health challenges, and suicide among Indigenous people. Communities have limited resources to support their residents.
This victimization, trauma, and discrimination have contributed to disparities in criminal justice outcomes and likely also have contributed to higher risk ratings.
Ultimately, Drs. Hogan and Corăbian emphasized that risk assessments must consider broader social and historical contexts to ensure fair and accurate evaluations.
Written by Virginia St-Denis and approved by Drs. Neil Hogan and Gabriela Corăbian.
