A Rare Look Into Post-Index Offense Incidents of Aggression

Michael C. Seto, Institute of Mental Health Research, Royal Ottawa Health Care Group
Yanick Charette, Ecole de travail social et de criminology, Laval University
Tonia L. Nicholls, Department of Psychiatry, University of British Columbia
Anne G. Crocker, Departement de psychiatrie et d’addictologie, Universite de Montreal

This study identified factors that predicted which of 713 clinically documented incidents of aggression—threats to kill, assault, or sexual assault—committed by 404 forensic psychiatric patients were linked to court findings of guilt. Individuals had, on average, 1.7 aggressive incidents and were found guilty of an average of 0.3 offenses against persons during the study period. Aggressive incidents were mostly assaults, followed by uttering death threats, and sexual assaults. The victims of aggressive incidents were mainly other patients or staff, but some incidents involved family members or friends (16%) and strangers (14%). Most of the aggressive incidents (84%) did not lead to findings of guilt. Incidents of aggression linked to court findings were significantly associated with province; personality disorder; fewer prior aggressive incidents; and incidents involving strangers compared to staff or co-patients or to family or friends. These findings have implications for research in terms of understanding how criminal records
underestimate histories of aggression. These findings also point to the need for the development of more consistent policies and procedures for responding to patient aggression, including when it is necessary or productive to report to police.

Aggression; criminal justice processing; mentally disordered offending; hospital policies; NCRMD

“Much progress has been made on understanding the factors associated with aggression among forensic psychiatric patients, including the development and adoption of risk measures to assess the likelihood of future violence…These studies typically focus on the role of static and/or dynamic factors in predicting recidivism in the form of new criminal charges or convictions…or in predicting clinically documented incidents of aggression that may result in seclusion, restraint, or loss of privileges among inpatients, or rehospitalization among outpatients…Many of these incidents could meet the legal criteria for a new criminal charge, but they are not reported to police or do not result in prosecution…unless researchers have access to hospital records, these aggressive incidents are ‘invisible’ when conducting follow-up studies…” (p.365).

“…An important and essentially unexplored question among psychiatric populations is what distinguishes clinically documented incidents of aggression that lead to criminal records from those that do not. Knowing which aggressive incidents result in criminal prosecution would help link the research on clinically documented aggression and criminal recidivism, and could also inform institutional policies and practices…In the present study, we re-analyzed data from a study of individual, service, and neighborhood factors that predicted clinically documented aggressive incidents and convictions for offenses against a person in a sample of 1,491 persons found Not Criminally Responsible on account of Mental Disorder in Canada (NCRMD)…” (p.365-366).

“Here, we examine what factors predict whether a clinically documented incident of aggression results in criminal conviction, extending the work of Volavka et al. (1995) by including aggressive incidents that occurred in the community and examining dynamic factors that could vary during the follow-up period…we expected those who were subsequently convicted for aggressive incidents would be younger, male, and to have more extensive criminal histories and less extensive psychiatric histories. We also anticipated that they would have engaged in more serious aggression, such as sexual assault or assault causing injuries. We further predicted those individuals who were more difficult in terms of engaging in problematic behavior such as substance use would be more likely to be charged, because staff would be less confident, they could manage aggressive and other disruptive behavior” (p.367).

“…We identified a number of significant predictors of when clinically documented aggressive incidents involving forensic patients resulted in a guilty finding. First, consistent with our previous studies demonstrating significant provincial differences in the trajectories and outcomes of persons found NCRMD…we found that incidents were much more likely to lead to findings of guilt for persons found NCRMD in Ontario or Quebec, compared to British Columbia…Consistent with prior research, we did not find support for other variables that have been identified in prior research as associated with incidents leading to findings of guilt…We did find that individuals diagnosed with a personality disorder were more likely to have incidents resulting in findings of guilt…One possible explanation is that patients with a personality disorder are perceived as ‘belonging’ more in the criminal justice system, whereas those diagnosed with schizophrenia, but no personality disorder, are perceived as ‘belonging’ more in the mental health system…” (p.371-372).

“…These results suggest that…clinicians take into account the extent to which the aggressive patient may have acted with intent, versus acted in response to symptoms of their mental illness. Consistent with this idea, we found that individuals who had not been found guilty for prior aggressive incidents were also less likely to be convicted of a new aggressive incident than those who were found guilty for past aggression…We found that the patient’s relationship to the victim was related to the likelihood of aggressive incidents leading to findings of guilt: Incidents involving strangers were much more likely to result in findings of guilt than those involving staff, co-patients, family, or friends…” (p.372).

“We also found that aggressive incidents involving staff were significantly more likely to result in findings of guilty than when the victim was a co-patient. This is potentially concerning as it could indicate a systematic bias toward the protection of staff over patients or the availability of supports and access to guidance about how to respond following an assault…We did not find that type of aggression was associated with incidents being linked to findings of guilt. This could reflect the fact that we considered a fairly tight range of types of aggression, from threats to kill to sexual assault…” (p.372)

“There is a need for further exploration of factors that correlate with clinically documented incidents leading to prosecution; in particular, more information about the victims, the incidents, including severity, presence of witnesses or other evidence, and information from law enforcement. We would expect, for example, that more serious injury, repeated aggression committed by the same patient, and incidents with more corroborative evidence…are more likely to be prosecuted and result in findings of guilt” (p.374).

“There is clearly more work to be done regarding the entire sequence of events from incident to reporting of incident by co-patients or staff, to reporting of the incident to police to decisions to prosecute…It would be valuable to determine what variables influence the documentation of incidents. For example, gender may play a role, where both documentation of incidents and reporting to police are affected by whether the patient who was aggressive was male or female, and whether the aggression was directed at female or male targets…It would also be very interesting to see if factors that predict which aggressive incidents are reported to police differ for co-patients compared to staff, family or friends, or strangers” (p.374).

“Beyond these empirical questions, we would like to see more work on the clinical, ethical, and policy implications of decisions to prosecute clinically documented incidents of aggression. For instance, what is the impact of prosecution for aggressive incidents on the therapeutic alliance with patients and family members? When is it in the best interests of the patient, the victim, the institution, and society? On the one hand, prosecuting incidents may negatively affect the therapeutic alliance and can further ‘criminalize’ psychiatric patients, a population that is already highly stigmatized. On the other hand, it may demonstrate support and concern to patients and staff who are victimized in an inpatient setting” (p.374).

“…incidents were much more likely to lead to findings of guilty for persons found NCRMD in Ontario or Quebec, compared to British Canada. A possible explanation for this finding is that staff in British Columbia are more willing or better able to manage aggressive incidents within its integrated forensic mental health care system (one forensic hospital and six satellite clinics), compared to a network of 10 independent forensic programs in Ontario and a large decentralized network of approximately 50 hospitals in Quebec, at the time of the study. Given NCRMD legislation in Canada is national, this growing evidence of provincial differences suggests there is a need for collaboration to develop national standard and guidelines. It also suggests that additional international collaboration could be valuable for supporting best practices” (p.371-372).

As always, please join the discussion below if you have thoughts or comments to add!