Childhood Trauma and Risk Factors for Institutional Violence

Michael S. Martin University of Ottawa
Kwame McKenzie
 Centre for Addiction and Mental Health, Toronto, Ontario, Canada, and University of Toronto
Gordana Eljdupovic
 Correctional Service of Canada, Ottawa, Ontario, Canada
Ian Colman University of Ottawa

Inmates who experienced childhood trauma have higher rates of institutional violence. However, the potential intermediate roles of co-occurring mental health and substance use needs and early justice involvement have not previously been considered. The current study examined the relationships between trauma, mental health, substance abuse, youth criminal charges, and institutional violence during the first 180 days of incarceration. As secondary aims, we explored whether these associations differed by sex or differed for inmates of Aboriginal ethnicity. Secondary data from prison records for all 5,154 inmates admitted to a federal prison during 2011 were collected. Path analysis was used to estimate the direct and indirect associations between trauma and institutional violence. Approximately 45% of inmates reported childhood trauma, which was associated with a higher prevalence of co-occurring mental health and substance abuse needs, and youth criminal charges. Although mental health, substance abuse, and youth criminal charges interacted with one another in predicting violence, their associations were similar for those with and without histories of trauma. A direct association between trauma and institutional incidents remained (Relative Risk [RR] ? 1.38, 95% CI [1.07, 1.78]) after accounting for indirect associations through these co-occurring risk factors. There was insufficient evidence to suggest that these associations differed between men and women or between Aboriginal and non-Aboriginal inmates. Given the high co-occurrence of multiple health and behavioral risk factors for inmates with traumatic histories, clarifying which factors are causally associated and reversible is needed to inform effective trauma informed care.

trauma, violence, prisons, mental health, substance abuse

“Inmates with childhood trauma are more likely to engage in violent behaviors while incarcerated. These associations between trauma and violence appear quite strong in unadjusted models, but are substantially attenuated when controlling for important co- occurring needs such as criminal history, substance abuse, and mental health. While there has been a growing interest in trauma-informed care in correctional settings to better respond to the needs of these inmates, the processes underlying the associations between childhood trauma and violence in prison remain speculative. This article explores the associations between trauma, mental health, substance abuse, and age of onset of criminal history to better understand how these variables are associated with one another and the risk of violence within prisons” (p. 614).

Trauma

“Inmates with trauma histories have higher rates of important risk factors for violence in prison. Lake (1995) reported that, on average, inmates with histories of physical abuse by their parent were approximately 3 to 4 years younger at the time of first arrest, had approximately four more arrests, and committed approximately 12 additional criminal acts compared with inmates who were not physically abused. Specific to violent behaviors, Dutton and Hart (1992) reported that the odds of having an adult conviction for physical or sexual violence were between 2 and 7 times greater among inmates who reported that they were the victim of physical or sexual abuse in childhood or reported that they witnessed family violence. Similarly, the odds of substance use and mental health issues have been estimated to be roughly 2 to 4 times greater among those with histories of childhood trauma. Systematic reviews have reported that criminal history and substance abuse are among the strongest predictors of institutional violence and misconduct, whereas personal distress and mental health needs have weaker associations” (p. 615).

“Although interactions between risk factors and trauma have received less attention in the correctional literature, trauma may influence adult behaviors such as violent behavior as a result of impacts of childhood trauma on developmental processes. Individuals with traumatic pasts may develop a response that is specific to a given situation, which, over time, may develop into a more permanent response set. The individual may begin to perceive and interpret new situations as threatening, which may shape interactions with others. For instance, an individual may interpret new situations as hostile and respond with violence, aggression, and/or fear. These tendencies may be magnified in a highly regimented and confined environment such as prison. Individuals who experience childhood adversity also perceive stressors to be more stressful than those who did not experience childhood adversity. Thus, stressful aspects of the prison environment may be more distressing to those who have experienced childhood adversity” (p. 615).

“Men and women differ in the nature of traumas experienced and in terms of behavioral manifestations…Similarly, in the Canadian context, in which the current research was conducted, Aboriginal people have experienced unique historical events, such as the removal from families and placement in residential schools in which many students were victims of physical and sexual abuse…In light of these differences between the sexes and ethnic groups, there have been calls for sex- and culturally responsive services and risk assessment. However, there is limited empirical evidence that the associations between various risk factors and institutional violence differ between men and women or between different cultural or ethnic groups… Many of the ‘gender specific’ risk factors that have been identified have been based on findings from samples of women only, or from studies that have compared risk factors for men and women solely on the basis of statistical significance (i.e., a factor is significant for one sex but not the other)” (p. 616).

The Current Study

“The present study aimed to estimate the association between childhood trauma and violence in prison, including indirect associations with psychological distress, substance abuse, and age at first arrest. As a secondary objective, we explored whether these associations differed for women or Aboriginal inmates” (p. 616).

This study utilized secondary data from 5,154 federal Canadian inmates, representing nearly all admissions into federal prison, during the 2011 calendar year. Numerous data sources were collected throughout the inmates’ first 180 days within the prison. The dependent variable in this study was the amount of recorded violent incidents, if any, the inmate had within their first 6 months of incarceration. The researchers excluded threats from their analyses and violent incidents were only included when the inmate was the instigator.

“This large study of a cohort of all admissions to a national prison service showed that childhood trauma is very common among inmates, and that trauma often co-occurs with other risk factors such as substance abuse, youth crime, and distress” (p. 619). The results of this study indicate that inmates with a history of trauma were more likely to exhibit multiple risk factors than those without a reported history of trauma. Inmates with a history of trauma were less likely to have no risk factors and they were less likely to report solely distress, substance abuse, or youth criminal charges. “After accounting for distress, substance abuse, and youth criminal charges, individuals with trauma had a small but significantly increased risk of violence” (p. 618).

The researchers tested for “interactions between substance abuse, distress, and youth criminal charges, given that interactions among these factors were most strongly supported by past literature, and that the distribution of these factors appeared to differ based on inmate’s trauma histories… There was a significant improvement in model fit when interactions between distress, substance abuse, and youth criminal history were included compared with the model with only main effects of these variables, indicating that there were significant interactions among at least some of these factors” (p. 618). Additionally, when each interaction term was analyzed independently, they all led to a statistically significant improvement in model fit.

Additionally, they “considered possible interactions between trauma and the three intermediate risk factors, to test whether risk factors were more strongly associated among those with histories of trauma compared with those who did not have such histories… there was insufficient evidence to suggest that the relative risk of violence associated with the various combinations of distress, substance abuse, and youth criminal charges differed for inmates who had experienced childhood trauma from those who had not” (p. 618-619). “Thus, our findings suggest a higher prevalence of these intermediate risk factors among inmates with histories of trauma, but that their associations with institutional violence appear to be similar for all inmates” (p. 620).

Lastly, “the model fit did not statistically significantly improve with the inclusion of interactions with either sex or Aboriginal status compared with models without these interactions” (p. 619). “Although women and Aboriginal inmates generally had higher rates of the risk factors studied and higher rates of violence in prisons, the associations between these factors and institutional violence was similar across all groups” (p. 620).

“Although risk factors were equally predictive regardless of inmates’ childhood histories, the majority of inmates with a history of trauma also had co-occurring mental health, substance abuse, and/or youth criminal charges (and, conversely, the majority of those with co-occurring needs had a history of childhood trauma). This highlights the multiple social, health, and behavioral needs of inmates who are at highest risk of violence that should be considered as treatment targets. It also reinforces the need for interventions for inmates with trauma histories to be integrated (i.e., target multiple needs simultaneously rather than sequentially by requiring inmates to address one need before progressing to the next treatment target) in order to be effective” (p. 619).

The findings in this study shine light on “the need to consider the joint associations between many variables rather than considering them in isolation. In particular, in light of the ongoing debate about the extent to which mental health needs increase the risk of violence or criminal behavior, it is noteworthy that distress alone was unrelated to institutional violence. However, distress interacted with other factors, including synergistic interaction between distress and youth criminal charges, and between distress and substance abuse” (p. 619).

“Differences in how variables interact in the community versus in prison may highlight the importance of social- environmental predictors of violence by persons with mental health needs in particular. These differences also reinforce the importance of considering the dynamic nature of risk factors and considering the individual’s current situation (or risk state) compared with their historical risk factors” (p. 620).

“Our findings highlight the challenges of addressing the multiple needs of inmates. They highlight that the majority of inmates with co-occurring risk factors that attract considerable attention of correctional institutions, such as mental health, substance abuse, and youth criminal charges, also have histories of childhood trauma. Clarifying developmental pathways to criminal behavior following traumatic experiences is needed to identify modifiable risk factors to be targeted through programming and to disentangle causal relationships from associations. Evaluating interventions that seek to address these risk factors is a necessary next step to determine the extent to which risk of violence in prison by inmates with histories of childhood trauma can be mitigated. As childhood trauma was associated with increased risk of institutional violence among all inmates, regardless of sex or ethnicity, all inmates who experienced childhood trauma may benefit from such research and practice efforts to develop and implement trauma informed care” (p. 621).

“Given that there was no evidence of differences in the fit of the final model by either sex or ethnicity, there may be benefits to targeting the co-occurring trauma, mental health, and substance abuse needs of all inmates. In this regard, services in corrections that are responsive to the unique needs of women and cultural/racial minorities may be primarily a question of quantity of services available (to account for differences in prevalence) as opposed to a difference in the types of services. Further research to replicate these findings may be warranted, given that the small number of women who are incarcerated limits power to detect differences in patterns of associations between the sexes” (p. 620).

“Our findings may inform further work to test mediation models more fully, or to test interventions that may be appropriate for inmates with multiple co-occurring treatment needs and risk factors for violent behavior. They may also suggest areas for further research among community samples to examine potential risk and protective factors that distinguish those who experience childhood trauma who initiate criminal behavior from those who do not. Research such as this may be of great value to identify preventative measures to reduce the incidence of criminal behavior among people with histories of trauma” (p. 620).

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