New Zealand prisoners’ prior exposure to trauma
Research Adviser, Department of Corrections
Marianne Bevan is a Research Adviser in the Department of Corrections Research and Analysis team. She started at Corrections in May 2014 and has completed a range of projects related to the offending, treatment and management of female offenders. Prior to working at Corrections, she conducted research and implemented projects on gender and security sector reform in Timor-Leste,Togo, Ghana and Liberia.
This study quantifies prisoners’ lifetime exposure to potentially traumatising events using data from the recent study on the comorbidity of mental health and substance abuse disorders (Indig, Gear and Wilhelm, 2016). The study identified high rates of lifetime exposure to potentially traumatising events – for example, over half (57 percent) of prisoners have experienced sexual and/or family violence and the rate was higher for women at 75 percent compared to 56 percent for men. This could have implications for the management and treatment of people in prison, including the need for trauma-informed practice and for further work exploring the relationship between family violence victimisation and perpetration.
The recently completed survey on the comorbidity of substance use disorders and mental health disorders among New Zealand prisoners showed that 52 percent of female prisoners, and 22 percent of male prisoners have a lifetime diagnosis of Post-Traumatic Stress Disorder (PTSD) (Indig, Gear and Wilhelm, 2016).
This suggests high rates of trauma within the prison population. Trauma occurs when an event, or series of events, is experienced by the individual as emotionally harmful or threatening. Trauma can be caused by a range of events, although there is debate about what types of events this can encompass (May and Wisco, 2016). However, traumatic events are generally defined as events experienced or witnessed which cause “actual or threatened death or serious injury, or a threat to the physical integrity of self or others” (May and Wisco, 2016, pp 233). This commonly includes physical or sexual abuse, the sudden death of a family member, or exposure to a natural disaster (SAMSHA, 2014).
People experience events differently; an event such asa life-threatening accident which may cause lasting adverse effects on one individual’s functioning, will not have the same effect on others.
Trauma exposure can have a range of long-term impacts including the development of PTSD, or other mental disorders such as depression and anxiety, substance abuse, and inter-personal problems (Tam and Derkzen, 2014). There has been no causal link made between trauma and criminality. However, exposure to traumatic events causes a range of other maladaptive coping strategies, which are, in turn, associated with criminality. Studies have shown that for some female offenders, the psychological and behavioural impacts of childhood victimisation contribute directly to criminogenic needs; the depression, anxiety, and substance abusing behaviour stemming from trauma can be direct contributors to offending (Salisbury and Van Voorhis, 2009, p.561). Prior exposure to traumatic events such as witnessing or experiencing family violence as a child has also been shown to play a role in family violence offending for men and women (Dutton, 2006).
There is limited international research on prisoners’ exposure to trauma events prior to their incarceration, although the research that does exist has consistently shown high rates of trauma exposure, particularly for women (Tam and Derkzen, 2014). For example, one cited study showed that 60 percent of life sentenced women had experienced sexual abuse compared to 8 percent of men in a U.S. prison sample (Leigey and Reed, 2010, cited in Tam and Derkzen, 2014). Recent qualitative studies into the female prison population in New Zealand have suggested high rates of exposure to potentially traumatising events, specifically sexual and family violence (Bevan, Lynch & Morrison, 2016).
However, as yet, there has been no attempt to quantify the extent of prior exposure to potentially traumatising events among New Zealand prisoners.
Purpose of the study
The aim of this analysis was to quantify the prevalence of prior exposure to potentially traumatising events within the New Zealand prison population. This study is based on data from the recent study on the comorbidity of mental health and substance abuse disorders which was completed in 2015. The comorbidity study used the Composite International Diagnostic Interview 3.0 (CDDI 3.0) and the Personality Diagnostic Questionnaire 4+ (PDW-4) to assess the prevalence of mental health and substance use disorders. The study used a representative sample of 1,209 randomly selected New Zealand prisoners (113 women and 1,096 men), who have recently been received into prison, across 13 prisons. The sample was predominately men (91 percent). Māori comprised 51 percent of the sample, Europeans 35 percent, Pacific Peoples 10 percent, and other/not recorded 4 percent.
The PTSD module of the CDDI 3.0 contains 28 potentially traumatic events (these are listed in Appendix 1 of this paper). Participants were asked whether they had experienced particular types of potentially traumatic events, at what age, and with what frequency. They were then assessed as to whether their experience of any of these events led to PTSD symptoms.
The data on the potentially traumatic events was used to identify offenders’ prior exposure to potentially traumatic events – such as family and sexual violence– for this study. Therefore it does not only include people who developed PTSD symptoms, but includes all incidences of these events within the whole comorbidity study sample. This study did not include the data on which events led to PTSD symptoms. As described above, whether or not an event is considered traumatic depends on the individual’s reaction to it.
Therefore while this study quantifies the extent to which participants had been exposed to potentially traumatising events, it does not identify what proportion experienced trauma symptoms as a result. This was done for two reasons. Firstly, the way the data was collected meant that it was not possible to identify which specific events led to PTSD symptoms. Secondly, this would have limited the scope of the study. As explained above, exposure to potentially traumatising events can have a range of negative outcomes beyond the development of PTSD. One of the aims of the study was to quantify overall exposure to different types of violence, particularly family and sexual violence, within the prison population, as there is not a current reliable figure of this. Only including events which led to PTSD would not have allowed the study to provide these figures. As a result, the study refers to potentially traumatising events.
The study compares rates of exposure to potentially traumatic events between men and women, and different ethnicities. Significance tests were conducted to determine whether the difference in proportions were statistically significant to the 95 percent confidence level. Statistically significant differences are identified throughout the report.
For analysis, the types of potentially traumatic events were grouped into five categories. These were:
- Violence (family violence, sexual violence, general violence)
- War/conflict/civil unrest-related events
- Exposure to serious accident and/or disasters
- Exposure to illness and death
- Event causing injury or death.
The current analysis has a particular focus on prisoners’ experiences of family violence. However, the definition of family violence used in the study was limited. The study includes three types of violence grouped as family violence: being badly beaten up by parents or the people who raised you; witnessing serious physical fights at home; or being badly beaten up by a spouse or romantic partner. There are other events such as stalking and kidnapping which, in some cases, are likely to be family violence, but because information about the perpetrator was not collected, it was not possible to identify this and so they were grouped as “general violence”. There are types of violence generally considered family violence – such as childhood neglect, or psychological, economic or emotional coercion and control exerted in a relationship – data on which was not collected in the study. This means the rates of family violence reported are likely to be an under-representation of the true exposure rates.
The data was based solely on self-reports of trauma exposure. Disclosing sexual violence is a sensitive subject, particularly for male offenders where the social stigma around reporting sexual violence victimisation is arguably higher. There were also types of sexual violence that were not captured in the study, for example, sexual contact between a minor and someone over the age of 16 which may be perceived as consensual, but which is illegal. These factors could mean reported rates were lower than actual rates of victimisation (Holmes, Offen & Walker, 1997).
Exposure to violence
Three types of violence were identified: family violence, sexual violence, and general violence. The study looked at overall incidences of violence, along with the prevalence of the different types of violence.
A high proportion of prisoners have experienced violence within their lifetime
Over three quarters of prisoners experienced some type of violence (including family violence, sexual violence or other community violence). Women experienced violence at slightly higher rates than men (81% compared to 77%), and more commonly experienced sexual and family violence, compared to men who were more likely to experience general violence (see Figure 2).
Exposure to violence
Exposure to violence by gender (Statistically significant differences between men and women are in blue and italicised.)
There was not a lot of variability in the rates of violence victimisation by ethnicity, although Māori did experience the highest rates of violence and this difference between Māori and non-Māori was statistically significant (see Figure 3).
Exposure to violence by ethnicity (Statistically significant differences between Māori and non-Māori are in blue and italicised.)
Prisoners were exposed to a high concentration of family and sexual violence
Over half of prisoners have experienced sexual and/ or family violence (57 percent). This rate was higher for women than for men, with 75 percent of women experiencing sexual or family violence compared to 56 percent of men and this difference was statistically significant. Women had also been exposed to a higher concentration of these violence types (see Figure 4); the majority (73 percent) of men had only been exposed to one type of violence (including family violence as a child, Intimate Partner Violence (IPV), and sexual violence), whereas a higher number of women had been exposed to at least two forms of family and/or sexual violence.
In terms of ethnicity, Māori offenders were most likely to experience sexual and/or family violence (63 percent), followed by European offenders (55 percent), Pacific Peoples offenders (48 percent), and lastly unknown/other offenders (31 percent). The difference in rates of sexual and/or family violence between Māori and non-Māori was statistically significant.
Concentration of violence by gender (family violence as a child, IPV, sexual violence)
Statistically significant differences between men and women are in blue and italicised.
The prevalence of family violence was high
Fifty-three percent of prisoners had experienced family violence in their lifetime. Rates of family violence were highest amongst Māori at 60 percent. Just under half of Pacific Peoples and European prisoners had experienced family violence. The difference in rates of family violence exposure between Māori and non-Māori were statistically significant.
The exposure rate was higher for women at 68 percent compared to 52 percent for men. Men had marginally higher rates of family violence exposure as a child* than women (48 percent compared to 44 percent), although the rates of IPV in adulthood were noticeably higher for women at 61 percent compared to 10 percent. The 2014 New Zealand Crime and Safety Survey (NZCASS)** reported that lifetime experience of IPV was 26.1 percent for women, and 13.8 percent for men. This means rates of IPV were much higher for female prisoners than in the general population, whereas the reverse was true for men.***
Rates of family violence victimisation
Rates of family violence victimisation by gender (Statistically significant differences between men and women are in blue and italicised.)
Over half of women who experienced IPV experienced it for a sustained period of time. This was the case for 15 percent of men. Women were also more likely to experience IPV from a younger age than men. This means women reported much greater exposure to IPV than men.
Nearly one-fifth of prisoners had experienced sexual assault (including rape). Over half of women (53 percent) had been sexually assaulted. The NZCASS lifetime experience of sexual violence for women was 23.8 percent which means the sexual violence prevalence rates for female prisoners are over double those for women in the community. Forty percent of women in the prison sample had been raped, and this was nearly four times the rate reported in the NZCASS4 where it’s 11.4 percent. While the rates were much lower for men, with 15 percent sexually assaulted, and 9 percent raped, such results are not insignificant and are much higher than rates within the general population – NZCASS showed men’s lifetime experience of sexual violence to be 5.6 percent. Differences in exposure to sexual assault between men and women were statistically significant.
Rates of sexual assault
Rates of sexual assault by gender
Of those who reported being sexually assaulted, over half (57 percent) experienced their first sexual assault when under the age of ten, and a further 34 percent under the age of twenty. Over a quarter (28 percent) of prisoners who had been sexually assaulted experienced it as part of a sustained period of abuse.
Other violence included being stalked, kidnapped, beaten up (excluding family violence), mugged or threatened with a weapon. Over half of prisoners had experienced this type of violence (61 percent), most commonly being beaten up. Men experienced this form of violence at higher rates than women (62 percent compared to 53 percent). There were variations in the types of general violence women experienced compared to men. For example, men were mugged or beaten up at higher rates than women, whereas women were stalked or kidnapped at higher rates. Rates of general violence were similar by ethnicity, although marginally higher for European offenders at 65 percent compared to 62 percent for Māori offenders and 47 percent for Pacific Peoples.
General violence by gender (Statistically significant differences between women and men are in blue and italicised.)
Exposure to other potentially traumatic events
Prisoners had also experienced high rates of other potentially traumatising events including exposure to serious accidents and/or disasters; exposure to serious illness and death; exposure to war, conflict and civil unrest; and causing serious injury and/or death of others. Over three quarters of prisoners had been exposed to serious illness and death including having a life-threatening illness or having someone close to them die unexpectedly, for example through murder or suicide. Over half had been exposed to serious accidents and/or disasters such as natural or manmade disasters, or life-threatening accidents. Men were more likely to have experienced all of these types of events than women, as evidenced in the graph below. For example 55 percent of men had experienced a serious accident or disaster compared to 40 percent of women.
Exposure to other potentially traumatising incidents by gender (Statistically significant differences between women and men are in blue and italicised.)
Rates of victimisation for family violence and sexual offenders
Over half (58 percent) of prisoners in the study had a past conviction for a family violence offence. The proportion was higher amongst male prisoners than female prisoners, at 60 percent (n=622) and 36 percent (n=41) respectively. It was relatively common for those with family violence convictions to have also experienced family violence victimisation, and was the case for 58 percent of prisoners in the sample. This is similar to international findings where the rates were much higher for women: 80 percent (n=33) with a family violence conviction had family violence victimisation, compared to 56 percent of men. A small proportion of offenders were serving sentences for sexual offences (6 percent, n=73). Of those, just under one-fifth had previously experienced sexual victimisation, which is the same rate as within the wider sample. This shows a greater congruence between victimisation and perpetration for family violence than for sexual violence.
Proportion of family violence offenders with family violence victimisation
Conclusion and implications
Prisoners have a high lifetime exposure to potentially traumatising events
This analysis shows high rates of lifetime exposure to potentially traumatising events within the New Zealand prison population. Three quarters of prisoners had been exposed to serious violence in their lifetime and over half had experienced family and/or sexual violence.
The rates of family and sexual violence were, in most incidences, higher than those experienced within the general population.
Prisoners’ experience of violence is gendered
The overall rates at which prisoners had experienced violence are similar for men and women. However, the types of violence men and women most commonly experienced differed; women were much more likely to experience “private” violence like sexual violence, interpersonal violence, stalking and kidnapping, whereas men were more likely to experience more “public” violent crime.
Māori experienced marginally higher rates of violence
There was not a lot of variation in the rates of exposure to violence between different ethnic groups. However, Māori experienced higher rates of violence overall, and experienced higher rates of family and/or sexual violence.
Female prisoners have experienced high rates of victimisation
It is commonly assumed that rates of victimisation within the female prison population are high, and we now have more definitive evidence of this. Three- quarters of the female sample had experienced family and/or sexual violence – nearly half had been raped and nearly two-thirds had experienced Intimate Partner Violence. For many of these women the abuse started when they were young, and was part of a sustained period of violence. There is work underway to implement trauma-informed practice within the women’s prisons (see McGlue, 2016) and this research emphasises the importance of this work. There is emerging international evidence that experiences of victimisation have a noticeable impact on women’s pathways into crime. It was not within the remit of this study to explore how victimisation may or may not contribute to offending. However, the high rates of victimisation suggest further research is needed on how past victimisation is addressed within women’s rehabilitation pathways.
The rates of victimisation within the male prison population are cause for attention
Over half of male offenders had experienced sexual and/or family violence; nearly half of the male prison population experienced violence as a child, and while the rates of sexual abuse were much lower than for women, they were still higher than the general population. Results show men may also need adequate access to victims’ services. Further work is also needed exploring the long-term impact family violence victimisation may have had on offending patterns; for example alcohol and drug use, and family violence perpetration. With regards to the latter, more work is needed exploring the relationship between family violence victimisation and perpetration.
*The question within the comorbidity study about childhood experience of family violence did not specify what age counted as a “child”.
**The NZCASS is a nationwide, face-to-face survey of New Zealand residents aged 15 years and over. A total of 6,943 adults was interviewed between February and June 2014 about crime that happened in 2013, and in their lifetime.
***IPV was defined as assault, threat (of force), damage (to property), threat (to damage property), which is a more expansive definition than that used in this study.
Appendix 1: Summary of rates of exposure to potentially traumatising events by prisoners over their lifetime, by gender
* indicates that differences in proportions between men and women are statistically significant to 95% confidence levels
Type of Violence
Family violence as a child
Intimate partner violence*
Any family violence*
Any sexual violence (including rape)*
Any family and/or sexual violence*
Mugged or threatened with a weapon
Beaten up (not including family violence)*
Any general violence
Any violence (family, sexual, general)
Other potentially traumatising incidents
Serious accident and/or disasters*
Exposure to serious illness or death
Cause injury and/or death of others*
Appendix 2: Summary of rates of exposure to potentially traumatising events by prisoners over their lifetime, by ethnicity
* indicates that differences in proportions between Māori and non-Māori are statistically significant to 95% confidence levels
Type of Violence
Pacific Peoples %
Family violence as a child*
Intimate partner violence
Any family violence*
Any sexual violence (including rape)
Any family and/or sexual violence*
Mugged or threatened with a weapon
Beaten up (not including family violence)
Any general violence
Any violence (family, sexual, general)*
Other potentially traumatising incidents
Serious accident and/or disasters
Exposure to serious illness or death
Cause injury and/or death of others
Bevan, M., Lynch, E., and Morrison, B. (2016) Towards an Understanding of Female Family Violence Perpetrators: A Study of Women in Prison, Practice: The New Zealand Corrections Journal, 4(2), December 2016, 29-34
Dutton, D. G. (2006) Rethinking Domestic Violence. Vancouver: UBC Press.
Holmes, G. R., Offen, L., & Walker, G. (1997). See no evil, hear no evil, speak no evil: Why do relatively few male victims of childhood sexual abuse receive help for abuse-related issues in adulthood? Clinical Psychology Review, 17, 69-88
Indig, D., Gear, C., and Wilhelm, K. (2016) Comorbid substance use disorders and mental health disorders among New Zealand prisoners, New Zealand Department of Corrections
McGlue, H. (2016) Trauma Hiding in Plain View: The Case for Trauma-Informed Practice in Women’s Prisons, Practice: The New Zealand Corrections Journal, 4(2), December 2016, 22015
May, C.L., and Wisco, B.E. (2016) Defining trauma: How level of exposure and proximity affect risk for post traumatic stress disorder, Psychol Trauma, 8(2), 233-240
Ministry of Justice (2014) New Zealand Crime and Safety Survey, New Zealand Ministry of Justice
Salisbury, E. and Van Voorhis, P. (2009) Gendered Pathways: A Quantitative Investigation of Women Probationers’ Paths to Incarceration, Criminal Justice and Behaviour, 36(6), 451-566
Substance Abuse and Mental Health Services Administration (SAMSHA) (US), (2014) Trauma Informed Care in Behavioural Health Services, Treatment Improvement Protocol Series no 57, Rockville (MD), accessed at www. ncbi.nlm.nih.gov/books/NBK207191
Tam, K. and Derkzen, D. (2014) Exposure to Trauma among Women Offenders: A Review of the Literature, Correctional Service of Canada