Prevalence and Incidence of Nonsuicidal Self-Injury Among Federally Sentenced Women in Canada

Abstract

Nonsuicidal self-injury (NSSI) is a complex behavior that is not uncommon in the general population, yet little is known about the prevalence of this behavior among incarcerated women. Two studies were conducted to determine the prevalence and incidence of NSSI in federally sentenced Canadian women. In Study 1, a mixed-methods design that included a qualitative interview and a written questionnaire with a sample of 150 incarcerated women was used. In Study 2, archival data were analyzed for a random sample of 400 incarcerated women. Results indicated lifetime prevalence rates of NSSI ranging from 24% to 38%. Incidence of self-injury in a federal institution over a 1-year period was found to be 3.6 per 27.4 person-years (i.e., number of years incarcerated). Both studies indicated that for the majority of women in both samples, NSSI was first initiated in the community, prior to incarceration in a federal correctional institution.

Keywords

nonsuicidal self-injury women offenders mental health prison

Although women only comprise 16% and 18% of the incarceration population in Canada and the United States, respectively, increases in the female incarceration rate have far outstripped the corresponding male rate during the past two to three decades in both countries (Guerino, Harrison, & Sabol, 2011; Harrison & Beck, 2005; Kong & AuCoin, 2008; Sinclair & Boe, 2002). While Canada continues to experience such gender disparities, for the first time since 1972 the overall U.S. incarceration rate declined for both genders, dropping 0.6% for females and 0.3% for males between 2009 and 2010 (Guerino et al., 2011; Public Safety Canada, 2011).

Not surprisingly, there has also been an unprecedented interest in female-centered research and correctional practice during the past two decades (for reviews, see Blanchette & Brown, 2006; Hubbard & Matthews, 2008; Van Voorhis, 2012). During this time, a myriad of empirical research and theoretical debates germane to girls, women, and corrections have emerged. One issue in particular that continues to garner attention is the extent to which nonsuicidal self-injury (NSSI) among women offenders affects reintegration efforts as well as prison safety for staff and inmates alike.

NSSI may be defined as deliberate bodily harm or disfigurement without suicidal intent and for purposes not socially sanctioned (Klonsky & Muehlenkamp, 2007). It may include behaviors such as cutting, ligature use, burning, head banging, hitting, swallowing sharp or indigestible objects, and inserting and removing objects from the body. Many other terms have been used to describe the behavior such as self-mutilation, deliberate self-harm, parasuicide, and self-injurious behavior (SIB). NSSI is used here for enhanced clarity and to emphasize the lack of suicidal intent in the behavior. In practice, however, discerning suicidal intent can be challenging. Therefore, in situations where the lack of suicidal intent is unclear or unknown, the term SIB is used. Conversely, a clear absence of suicidal intent warrants the use of the term NSSI. Given this usage, all NSSI would fall under the umbrella term of SIB. NSSI poses a serious threat to the safety and well-being of staff and offenders within correctional settings. Consequently, research efforts that seek to better understand the nature, prevalence, and incidence of this behavior within women offender prison settings is warranted.

Mental health symptoms and diagnoses are not uncommon in individuals who self-injure, but individuals who self-injure are a heterogeneous group that exhibit an array of mental health issues (Klonsky, Oltmanns, & Turkheimer, 2003; Nock, Joiner, Gordon, Lloyd-Richardson, & Prinstein, 2006). Much of the research examining what factors are associated with NSSI has been correlational in nature. For example, NSSI has been found to correlate with borderline personality disorder as well as substance abuse disorders in various populations ranging from military samples to federally sentenced women in Canada (Andover, Pepper, Ryabchenko, Orrico, & Gibb, 2005; Borrill, Snow, Medlicott, & Paton, 2003; Klonsky et al., 2003; Langbehn & Pfohl, 1993; Wichmann, Serin, & Abarcen, 2002; Young, Justice, & Erdberg, 2006). Additionally, an association between depression and NSSI has been found in community and forensic populations (Andover et al., 2005; Darche, 1990; Klonsky et al., 2003; Roe-Sepowitz, 2007; Ross & Heath, 2002), including among federally sentenced women in Canada (Power, 2011). Anxiety has also been found to be higher in individuals who engage in NSSI compared to those who do not (Andover et al., 2005; Haines, Williams, Brain, & Wilson, 1995; Klonsky et al., 2003; Roe-Sepowitz, 2007; S. Ross & Heath, 2002; Simeon et al., 1992; Stanley, Gameroff, Michalsen, & Mann, 2001). Moreover, NSSI correlates with specific anxiety disorders. While there is some preliminary evidence that suggests symptoms of obsessive compulsive disorder co-occur with NSSI (Davis & Karvinen, 2002; Paul, Schroeter, Dahme, & Nutzinger, 2002; Yaryura-Tobias, Neziroglu, & Kaplan, 1995), there is stronger evidence for an association between NSSI and posttraumatic stress disorder (Albach & Everaerd, 1992; Kisiel & Lyons, 2001; Prinstein et al., 2008; Salina, Lesondak, Razzano, & Weilbaecher, 2007; Weaver, Cahrd, Mechanic, & Etzel, 2004; Weierich & Nock, 2008; Zlotnick, Mattia, & Zimmerman, 1999).

Research has also found support for a relationship between NSSI and impulsivity, aggression, and/or anger. Simeon et al. (1992) found that individuals with an axis II diagnosis who engaged in NSSI had significantly higher ratings on mood and trait variables such as anger, hostility, impulsivity, anger, and aggression and were more antisocial than a group of matched controls who did not self-injure. Additionally, male offenders who are highly impulsive reportedly engage in more NSSI than those who are less impulsive (Carli et al., 2010).

Males who self-injure in both psychiatric hospitals and correctional facilities have been found to engage in more frequent verbal and physical aggression compared to other patients who do not self-injure (Chowanec, Josephson, Coleman, & Davis, 1991; Hillbrand, Krystal, Sharpe, & Foster, 1994; Matsumoto et al., 2005). It is hypothesized that poor nonverbal problem-solving skills within these populations may contribute to aggressive acting out in difficult situations (Chowenac at al., 1991). Although women are substantially less violent than men (Blanchette & Brown, 2006), studies have found a relationship between internalizing and externalizing violence among women offender samples. Importantly, three individual studies have found that a history of SIB and/or attempted suicide is predictive of general and/or violent recidivism among women offender samples (Blanchette & Motiuk; 1995; Bonta, Pang, & Wallace-Capretta, 1995; Rettinger, 1998).

In addition to examining the correlates of SIB and NSSI, a large portion of theorizing and research in this field has been devoted to understanding the motivations behind these behaviors. The motivations-focused literature base has resulted in a variety of theoretical models. Unfortunately, many of these models are speculative and have little empirical support (Nock & Cha, 2009). Klonsky (2007) completed a systematic review of empirical research addressing the motivations for engaging in NSSI and found the strongest support for the affect regulation model (i.e., the use of NSSI to alleviate negative emotion; e.g., Brown, Comtois, & Linehan, 2002; Nixon, Cloutier, & Aggarwal, 2002; Rodham, Hawton, & Evans, 2004). Modest support was found for several other models, including the anti-dissociation model (i.e., to reduce feelings of depersonalization), interpersonal-influence (i.e., to seek help or other external rewards), anti-suicide (i.e., to avoid, or substitute for, a suicidal impulse), sensation-seeking (i.e., excitement), and interpersonal boundaries functions (i.e., to assert personal autonomy; Klonsky, 2007). After affect regulation, however, the strongest support was found for the self-punishment model, which purports that NSSI is used as a method of expressing anger toward the self or as a method of derogating the self. The proportion of study participants who have been found to endorse the self-punishment model varies widely from approximately 10% to 85% (Briere & Gil, 1998; Herpertz, 1995; Nock & Prinstein, 2004; Rodham et al., 2004). Interestingly, research suggests that the use of NSSI for self-punishment purposes may be more common in females than in males (Claes, Vandereycken, & Vertommen, 2007; Rodham et al., 2004).

The importance of external influences, and particularly participation in NSSI in response to others’ engagement in such behavior, is especially relevant for incarcerated populations, as there is concern that an individual’s risk of engaging in NSSI is elevated by being incarcerated with others who self-injure. Incidents of NSSI that occur after witnessing others engage in such behavior is referred to as the contagion effect (Walsh & Rosen, 1985). Reports of self-mutilation contagion have been occurring for nearly a century, with Holdin-Davis’s (1914) description of an epidemic of trichotillomania in an orphanage likely being the first report. Since then, a number of other accounts of similar outbreaks have been published (Cookson, 1977; Matthews, 1968; Menninger, 1938; Offer & Barglow, 1960; Rosen & Walsh, 1989; Walsh & Rosen, 1985). Additionally, laboratory research suggests that individuals may imitate self-aggressive behavior modeled by another person (Berman & Walley, 2003), lending further support to the existence of this effect.

Ross and McKay (1979) were among the first to describe NSSI as an epidemic within correctional facilities based on their examination of a “training school for girls” in Canada. Years later, Heney (1996) found that federally sentenced women and staff at the Prison for Women in Kingston, Ontario, reported outbreaks of self-injury, although the explanation for these outbreaks was not in line with the contagion effect. Many of the offenders and staff believed that “tension” or situational factors were largely to blame for outbreaks of NSSI. Thus, in most cases the women were not copying the behavior of others, but women were experiencing the same stressors simultaneously, and thus their NSSI incidents coincided with this trigger. While these outbreaks were described qualitatively, they were not confirmed quantitatively. Some debate remains regarding the existence of the contagion effect as rigorous empirical evidence is limited and has not definitively demonstrated the existence of this phenomenon. Studies have found that the majority of participants (73%-91%) report that they simply thought of the idea to self-injure themselves (i.e., they did not get the idea from another person, media, or literature; Favazza & Conterio, 1989; Nixon et al., 2002; Nixon, Cloutier, & Jansson, 2008). However, in a study of 64 adolescents with a history of NSSI, Deliberto and Nock (2008) found that 38% reported getting the idea to self-injure from peers (the most common response provided). It is possible that individuals are influenced by the behavior of others even though they do not acknowledge the influence or are unaware the influence has occurred.

While it is apparent that incarcerated populations have an elevated risk for engaging in NSSI compared to the general population, it is unclear whether incarceration promotes NSSI or whether incarcerated individuals are simply more likely than nonincarcerated individuals to have a history of NSSI prior to entering correctional facilities. Very few studies have investigated when NSSI was initiated, nor have they included large samples of incarcerated women. Snow (1997) found that of the 11 women interviewed in custody in England, 10 first self-injured prior to their incarceration. Additionally, using file information, Jones (1986) found that the SIB group was significantly more likely to have had wrist and arm scars upon entry to the correctional institution, which may indicate a history of NSSI. It is, of course, impossible to randomly assign individuals to an incarcerated setting and thus causal statements cannot be made in this type of research. However, the limited research conducted to date suggests that incarcerated women may be at increased risk for NSSI before ever entering a correctional facility.

One of the last major areas of investigation in the field of SIB and NSSI research to receive focused attention involves establishing reliable prevalence and incidence rates. However, this is a tremendously complex task. One significant issue that has plagued the field is the inconsistencies in the definition of self-injury, making the comparison of numbers from different studies challenging. For example, if one study includes drug overdoses as SIB and another excludes this behavior, the prevalence rates will artificially vary based on the inclusion criteria, rather than true prevalence in the samples. Similarly, in community populations, incidents are rarely recorded because the vast majority do not require medical treatment and individuals are generally ashamed and secretive about their behavior, making even self-report measures likely to produce an underestimation of behavior. Given these challenges, research on NSSI has generated many disparate prevalence rates. However, best estimates in the general population suggest that approximately 4% of adults have engaged in NSSI at some point in their lives (Briere & Gil, 1998; Klonsky et al., 2003).

Not all types of NSSI cause the same physical damage or have the same potential lethality. Currently, there is no well-defined method of categorizing NSSI based on its seriousness. A given type of NSSI could cause very different levels of damage in different incidents (e.g., cutting could involve a superficial scratch or the cutting of a major vein). Lohner and Konrad (2006) suggest that the seriousness of SIB should be assessed using two dimensions: (a) medical seriousness (i.e., the severity of the injuries and potential lethality) and (b) motivational seriousness (e.g., how strong the intent to die was at the time of the act). A wide variety of behaviors, with a range of potential seriousness, falls within the definition of NSSI used here, and most individuals who self-injure use more than one method of NSSI (Favazza & Conterio, 1989; Gratz, 2001; Herpertz, 1995; Whitlock, Eckenrode, & Silverman, 2006). The vast majority of studies report that cutting is the most common type of NSSI (e.g., Briere & Gil, 1998; De Leo & Heller, 2004; Favazza & Conterio, 1989; Heney, 1990; Howard League, 1999; Langbehn & Pfohl, 1993; Maden, Chamberlain, & Gunn, 2000; Nixon et al., 2002; Rodham et al., 2004), although seriousness of the cutting is generally not considered.

NSSI within correctional facilities is particularly challenging and results in substantial monetary and human costs (DeHart, Smith, & Kaminski, 2009). When considering medical costs (e.g., first aid, antibiotics, surgery), decontamination, staff time for paperwork, transportation, hospital supervision, and institutional equipment for offenders at risk for self-injuring, the financial cost can be staggering. There are also considerable human costs through the physical harm (and possible unintentional death) of offenders who engage in these behaviors, the negative impact on other offenders who witness the behaviors or who are kept in unnecessarily restrictive environments while staff respond to an incident of NSSI, and the potential for burnout and vicarious trauma in staff who respond to these offenders. Thus, additional resources and support are likely needed for offenders and staff who are dealing with NSSI.

Calculating prevalence and incidence rates within correctional facilities provides further challenges. While NSSI within correctional facilities is more likely to be recorded than those in the community, institutions lack accurate and consistent methods of recording NSSI. In addition to the shame individuals often experience regarding their behavior, there is an added fear of punishment for the behavior that may make offenders even more likely to hide it. Acts of low lethality may be less likely to be documented by correctional staff and may go undetected altogether. Alternatively, there is a unique potential for external rewards within a correctional facility for engaging in NSSI that may lead to reporting of the behavior. Whatever the motivation for the behavior, most rates for correctional facilities can be assumed to be underestimates.

Once the number of NSSI incidents or number of offenders (prevalence rate) who engaged in NSSI has been obtained, there are other inconsistencies to consider in regards to the reporting of these numbers. Prevalence rates may be calculated based on the average number of occupied beds, the number of admissions to the facilities, the average daily population, the average length of stay, or the number of person-years of the population (O’Toole, 1997). The use of varying methods, of course, results in varying rates that appear to provide very different pictures of the same situation. There are certain advantages and disadvantages for using each of these methods, and as such, there is no definitive consensus on the best way in which to report these rates. Authors use various methods, making the comparison of rates between studies even more tenuous.

Despite these extensive caveats, there are some prevalence rates of NSSI within correctional populations available; incidence rates, however, have not been a focus in previous research. Prevalence rates for SIB during incarceration in male offenders range from less than 2% to 18% (Appelbaum, Savageau, Trestman, Metzner, & Baillargeon, 2011; Carli et al., 2010; Fotiadou, Livaditis, Manou, Kaniotou, & Xenitidis, 2006; Maden, Swinton, & Gunn, 1994; Smith & Kaminski, 2011; Young et al., 2006). Lifetime prevalence rates for male offenders range from 15% to 35% (Fotiadou et al., 2006; Maden et al., 1994, 2000; Sakelliadis, Papadodima, Sergentanis, Giotakos, & Spiliopoulou, 2009).

Establishing prevalence rates for incarcerated women is even more difficult than for incarcerated men because there are usually very small samples of women included in the studies (Howard League, 1999; Shea & Shea, 1991). Research is very limited, but one study in the United Kingdom did find that 23% of women in custody for at least 2 years reported that they engaged in NSSI at least once during their sentence (Howard League, 1999; Office for National Statistics, 1997). Maden et al. (1994) found a much lower rate of 5% for self-injury while incarcerated and 32% with a lifetime history of this behavior. Rates may be even higher in offenders with serious psychological disorders. One study found that 53% of offenders with mental disorders had engaged in SIB, although the sample size was small (N = 34, n = 8 women).

A lack of studies that include comparable samples of men and women makes meaningful gender differences a tenuous prospect. Research on gender differences among nonincarcerated samples is more readily available, with some research finding higher rates of SIB among females (Hawton, Fagg, Simkin, Bale, & Bond, 2000; Whitlock et al., 2006). Males, however, may engage in NSSI more than previously thought, as several studies in the general population have failed to find a gender difference in prevalence rates in nonincarcerated samples (e.g., Briere & Gil, 1998; Cooper et al., 2006; Horrocks, Price, House, & Owens, 2003; Klonsky et al., 2003).

Studies on incarcerated populations generally examine only one gender at a time for two reasons: (a) Men and women are housed in separate facilities in most countries and (b) the number of men incarcerated far exceeds the number of women incarcerated. As such, when studies do include both genders, the sample is usually comprised of mostly men. For example, Jones (1986) conducted a matched control study of 67 offenders who engaged in SIB while incarcerated in an American prison with 68 matched comparison offenders, with only 4 women in each group. Smith and Kaminski (2010) recently examined incident reports from South Carolina correctional institutions and found slightly higher prevalence rates of SIB in men than women (.83% vs. .71%). Although the sample size was extremely large (n = 1,560 women, n = 21,393 men), the number of men included still far exceeded the number of women.

The Present Studies

In sum, NSSI is a serious health and safety concern within women offender prisons. The behavior exacts substantial costs—fiscal, emotional, physical—to staff and offenders alike. Prevalence and incident rates for NSSI for women offenders in Canada are not currently available. Similarly, although there is anecdotal support for the hypothesis that the prison environment itself is related to NSSI among women offenders, there is a paucity of corresponding empirical evidence. Consequently, the primary aim of the research is twofold: (a) to determine the prevalence and incidence of NSSI among federally sentenced women in Canada and (b) to examine whether or not there is support for the hypothesis that simply being in prison increases the likelihood of NSSI. This hypothesis will be examined by comparing the prevalence and incidence of NSSI prior to and during incarceration. This article utilizes a field study (face-to-face interviews and self-report questionnaires with a sample of incarcerated women offenders) as well as an archival study (an examination of electronic official files) to address the research objectives.

Study 1: Prevalence Field Study

Method

Design and participants

The field study used a mixed-methods design in which participants were asked to take part in a semi-structured interview and a questionnaire aimed at assessing personal history of NSSI. Participants were women in custody at various federal correctional institutions across Canada. In Canada, sentences of 2 years or more are administered by the Correctional Service of Canada (CSC), Canada’s federal correctional organization. One hundred and fifty incarcerated women aged 19 to 65 years (M = 35.7, SD = 10.7) were recruited from all federal women’s correctional institutions across Canada. More than half of the sample was Caucasian (54.0%), 37.3% were Aboriginal, and 8.7% were of other ethnicities. The women’s major admitting offenses for the sentence being served were homicide or manslaughter (26.0%); robbery (16.7%); drug offenses (15.3%); assault (11.3%); break and enter or theft (8.7%); fraud, forgery, or impersonation (7.3%); and other offenses (14.7%). In this sample, 36.0% were serving sentences of less than 3 years, 36.7% were serving sentences of 3 to 6 years, 12.0% were serving sentences of more than 6 years, and 15.3% were serving life sentences.

All offenders who were residing in the institutions during the study period were eligible to participate. Women were given the option of participating in English or French. There were some women, however, who the staff determined should not be interviewed for their well-being. These women were in segregation or considered to have current mental health issues that precluded their safe participation. The number of women who were ineligible for this reason is unknown. It is important to note that some women who were in segregation did participate in the study. At the time of the recruitment, there were on average 478 women residing in these institutions (CSC, 2010). One participant only completed a small proportion of the study and was consequently dropped. The total sample, therefore, contained 150 women, which represents approximately 31.4% of the population at the time of data collection. Compared to the entire population of women offenders, the sample appears to have more Aboriginal women; more women with a conviction of homicide, manslaughter, or attempted murder; and more women with longer sentences. Overall, however, the sample is comparable to that of the in-custody women offender population during the year of the study.

Materials

Semi-structured interview

The interview protocol for this study was designed specifically for this study.¹ The protocol assessed the history of NSSI in depth, including when NSSI was initiated, frequency of NSSI, types of injury, precipitating events, differences (if applicable) between NSSI before and after incarceration, and other factors thought to be associated with the behavior. As part of the interview, women were asked if they had ever injured themselves without trying to kill themselves. The purpose of this question was to distinguish between suicide attempts and NSSI. Follow-up questions probed participants to describe specific NSSI events and the circumstances in which they took place. In the majority of cases, interviews were audio recorded and transcribed. The duration of the semi-structured interview designed for this study varied considerably. For those who did not have a history of SIB, the semi-structured interview was often less than 5 minutes. For some women who had a history of SIB and/or other abuse and mental health issues, the interview lasted over an hour. The average semi-structured interview lasted about 15 minutes. All interviews were conducted by female researchers. The interviews were analyzed by the first and third author. In a random sample of 10% of the interviews (n = 15), there was perfect agreement between raters on whether the individual had engaged in NSSI.

Questionnaire

The Offender Self-Injurious Behavior Inventory (OSIBI) is a paper-and-pencil measure and was developed for this study to assess history of NSSI. The OSIBI was developed to gather information on SIB in offenders before and after incarceration. The questionnaire asks “Have you ever injured yourself without trying to kill yourself?” with an option to check yes or no. Additionally, the questionnaire includes questions such as “What kinds of injury do you do? (check all that apply),” with a list of behaviors that were each checked as now, before you entered the institution, and/or never. The following questionnaire items were correlated with the corresponding information from the interview (with Spearman rho correlations) to assess the concurrent validity: history of NSSI (.87), has used cutting as NSSI (.83), has used burning as NSSI (.62), has used hair pulling as NSSI (.57), and has used head banging as NSSI (.65). The OSIBI was also translated into French. The Spearman’s rho correlations for the French version are: history of suicide attempts (.68), history of NSSI (.79), has used cutting as NSSI (.77), and has used burning as NSSI (.58).²

Procedure

Ethical approval for this research was obtained from the Carleton University Ethics Committee for Psychological Research and the CSC Research Branch. Prior to commencement of interviewing, participants received a verbal description of the study, a consent form, and an opportunity to have questions answered about participation. No women declined to participate after reviewing the informed consent, although two women declined to be audio recorded. Detailed notes were taken in these two cases. No compensation or incentives were provided in return for participation.

All participants took part in the study individually in private rooms within the institution. Women completed the semi-structured interview first, followed by the OSIBI. Upon request, questions were read aloud by the interviewer. No time limits were imposed on participants for either the interview or the questionnaire. After completion, women were given a verbal and written debriefing.

Results

Each participant’s history of NSSI and suicide attempts was determined by combining the interview and questionnaire data. In 93.3% of cases (n = 140), both types of data were available and in agreement. Three cases did not have semi-structured interview data, and therefore their group membership was determined based solely on their questionnaire data. Two cases did not have the relevant questionnaire data, and therefore interview data were used. In five cases, there was a discrepancy between the questionnaire data and interview data. In these cases, the interview data were taken to be the most accurate response (i.e., group membership was determined based on semi-structured interview data). The interview data were believed to be more reliable due to the depth of the interviews. That is, in the interviews, the incidents and motivations (including presence or absence of suicidal intent) for self-injury were explored in detail and a more complete picture of the behavior was obtained. The interview asked participants to focus on particular incidents of self-injury that had been self-identified (e.g., first incident in the community, subsequent incident, first incident while incarcerated) and asked detailed questions about the incidents, such as “Why did you do it?”, “Did something happen to trigger the event?”, “How did you feel immediately before you did it?”, and “How did you feel immediately after you did it?” Prompts were used as needed to obtain a detailed understanding of the behavior.

Fifty-seven participants reported at least one incident of NSSI ever, which corresponds to a lifetime prevalence rate of 38%. Similarly, 61% of the participants reported engaging in at least one suicide attempt or an NSSI incident during their lifetime. Results are presented in Table 1. Cutting was by far the most common type of NSSI engaged in (77.2%), followed by head banging (19.3%) and burning and hitting inanimate objects (12.3% each; see Table 2).

Table 1:

Women Offender Field Results: Prevalence of Nonsuicidal Self-Injury and/or Suicide Attempts

	N = 150 % (n)
At least one suicide attempt or nonsuicidal self-injury incident ever	60.7 (91)
At least one incident of nonsuicidal self-injury ever	38.0 (57)
At least one suicide attempt ever	49.3 (74)
Suicide attempts only (no nonsuicidal self-injury)	37.4 (34)
Nonsuicidal self-injury only (no suicide attempts)	18.7 (17)
Both suicide attempt and nonsuicidal self-injury	44.0 (40)

Note. Prevalence rates based on self-report information taken from face-to-face interviews or self-report questionnaires; see text for further details.

Table 2:

Women Offender Field Results: Type of Nonsuicidal Self-Injury

Type of Nonsuicidal Self-Injury	N = 57 % (n)
Cutting or scratching	77.2 (44)
Head banging	19.3 (11)
Burning	12.3 (7)
Hitting inanimate objects	12.3 (7)
Hair pulling	8.8 (5)
Hitting or slapping	7.0 (4)
Ligature	5.7 (3)
Inserting objects	3.5 (2)
Piercing	3.5 (2)
Pinching	3.5 (2)
Tattooing	1.8 (1)
Run car over foot	1.8 (1)

Note. Interview data were only available for 56 participants. The Offender Self-Injurious Behavior Inventory used to determine prevalence for one participant.

The majority of women (85.5%) self-injured for the first time while in the community. Twenty-three women (41.1%) had engaged in NSSI at some point while in a federal correctional institution.³ Thirty-four women (59.6%) had only engaged in NSSI prior to being admitted to a federal correctional institution. Similarly, the majority of women who had attempted suicide did so for the first time prior to being admitted to such an institution (see Table 3).

Table 3:

Women Offender Field Results: Location of Nonsuicidal Self-Injury Incidents

	N = 57 % (n/N)
Location of first nonsuicidal self-injury
Community (not in any institution)^a	85.5 (47/55)
In a non-CSC correctional institution^a	5.6 (3/55)
CSC institution^a	7.3 (4/55)
Psychiatric non-CSC institution^a	1.8 (1/55)
Nonsuicidal self-injury prior to CSC	93.0 (53/57)
Nonsuicidal self-injury while in a CSC institution^b	41.1 (23/56)
Nonsuicidal self-injury both before CSC and while in a CSC institution^b	33.9 (19/56)

Note. CSC = Correctional Service of Canada. ^an = 2 missing. ^bn = 1 missing.

In sum, while it is difficult to compare these results to the limited women’s prevalence rates available, the 38% of women with a lifetime prevalence rate is similar to the 32% lifetime prevalence found by Maden et al. (1994). Given that 86% of women who had a history of NSSI first self-injured in the community, the being in prison increases the likelihood of NSSI hypothesis was not supported.

Study 2: Archival Prevalence And Incidence Study

Method

Design and participants

Four hundred federally sentenced women aged 19 to 69 years (M = 36.3, SD = 10.0) who were incarcerated between April 1, 2008, and March 31, 2009, were randomly selected for inclusion. The sample was also stratified by region to ensure that women from each of Canada’s five regions—Pacific, Prairie, Ontario, Quebec, and Atlantic—were proportionately represented. There were a total of 901 women who were incarcerated in a Canadian federal institution at some point during this time period. The sample used in this study therefore represents about 44% of the population.

Forty-one of the participants in the archival study also participated in the field study. This overlap is unsurprising given the relatively small number of women offenders incarcerated in the federal correctional system in Canada. Given the different method used for participant selection and data collection, overlap in the samples was anticipated.

More than half of the sample was Caucasian (52.0%), 35.8% were Aboriginal, and 12.3% were of other ethnicities. The women’s major admitting offenses for the sentence being served were drug offenses (27.0%); homicide, manslaughter, and attempted murder (19.8%); robbery (16.0%); assault (12.3%); break and enter or theft (6.8%); fraud, forgery, or impersonation (5.3%); and other offenses (6.0%). In this sample, 49.5% were serving sentences of less than 3 years, 34.3% were serving sentences of 3 to 6 years, 7.0% were serving sentences of more than 6 years, and 9.3% were serving life sentences. The study sample did not differ from the population on the variables examined.

Materials

Archival data were retrieved from an automated database used by CSC to manage information on federal offenders. Data contained in this system includes historical information, demographic information, incident reports, institutional program participation, institutional employment records, and psychological reports.

Data were collected according to the coding manual developed for this study. Variables for the coding manual were selected based on previous research and the hypotheses of the study. The coding manual contained the following sections: demographic information, criminal history, mental health and support variables, and suicide attempts and self-injurious behavior.

Procedure

Data were coded by four research assistants who had undergraduate or master’s degrees in psychology or criminology and were trained by the principal author on the use of the coding manual. A random sample of 10% of the files was selected and coded to test for interrater reliability. Given that all variables were categorical in nature, kappa was used to assess interrater agreement. For 85.2% of the variables, a kappa coefficient could not be calculated. In these cases, there was perfect agreement between the coders. For the remaining variables, kappa coefficients ranged from 0.79 to 1.00.

Results

Because reports of NSSI or suicide attempts in archival data can often be ambiguous with regards to intent, the authors devised coding guidelines. Any description of SIB occurring prior to incarceration in a Canadian federal correctional institution was taken to be accurate, as this is typically reported by the offender herself and not subject to staff interpretation. Such SIB incidents were further classified as either NSSI or suicide attempts based on the woman’s description of intent as indicated in the automated file information. Conversely, correctional staff members report on incidents of SIB occurring within the institution. These descriptions were often unclear in terms of whether or not the recording of events was based on reports by the woman herself or by the staff member interpreting the event. As well, details regarding suicidal intent were often absent. Consequently, it was determined that accepting the file description of intent could lead to inaccurate classification. Therefore, any recorded incident of NSSI or suicide attempt that occurred in the institution during the 1-year timeframe was coded only as SIB. No further classification into NSSI or suicide attempt was conducted.

Prevalence Results

Following the aforementioned procedures, there was a lifetime prevalence of 23.8% for SIB. Seventy-six of these 95 women (80.0%) had evidence in their files that they had engaged in NSSI prior to admission to a federal correctional institution. Results can be found in Table 4. Of the 54.6% of women who had an incident of SIB on file, cutting was the most common type of SIB recorded. The second most common type of SIB was ligature at 17.5%, followed by head banging at 9.2% (see Table 5).

Table 4:

Women Offender Archival Results: Prevalence of Self-Injurious Behavior and Suicide Attempts Prior to and/or After Admittance to a CSC Institution

	N = 400 % (n)
At least one suicide attempt, nonsuicidal self-injury incident, or self-injurious behavior incident	46.3 (185)
At least one incident of nonsuicidal self-injury prior to CSC or self-injurious behavior in CSC	23.8 (95)
At least one suicide attempt prior to CSC	40.5 (163)
Suicide attempts only (no nonsuicidal self-injury prior to CSC/self-injurious behavior in CSC)	22.5 (90)
Nonsuicidal self-injury prior to and self-injurious behavior in CSC	6.0 (24)
Suicide attempt and nonsuicidal self-injury incident prior to CSC and/or self-injurious behavior in CSC	17.8 (71)
Nonsuicidal self-injury prior to CSC	19.0 (76)
Self-injurious behavior while in a CSC institution	10.8 (43)
Nonsuicidal self-injury prior to CSC and self-injurious behavior while in CSC	6.0 (24)
Self-injurious behavior only after being admitted to a CSC	4.8 (19)
Suicide attempt prior to CSC and self-injurious behavior while in CSC	8.3 (33)

Note. CSC = Correctional Service of Canada.

Table 5:

Women Offender Archival Results: Types of Self-Injurious Behavior Participants Engaged in While in the Custody of the Correctional Service of Canada

Type of Self-Injurious Behavior Reported	Ever in Correctional Service of Canada N = 227 % (n)
Cutting or scratching	54.6 (125)
Ligature	17.5 (39)
Head banging	9.2 (21)
Biting	1.3 (3)
Inserting objects	0.9 (2)
Swallowing objects	1.3 (3)
Burning	2.6 (6)
Hitting	1.3 (3)
Poison	0.9 (2)
Pinching	0.4 (1)
Stabbing	0.4 (1)
Hair pulling	0.4 (1)
Not specified	9.2 (21)

Note. Swallowing objects refers to swallowing items other than food or drugs (e.g., glass, pins).

Incidence Results

Incidence rates of SIB while in federal custody were also calculated for this sample. The total number of self-injury incidents that occurred while in a federal correctional institution between April 1, 2008, and March 31, 2009, was recorded. Fifteen of the 400 women (3.8%) engaged in at least 1 SIB incident during the year of study. In total, these 15 women accounted for 29 incidents of SIB. Eight women each engaged in 1 incident, 4 women engaged in 2 incidents, 1 woman engaged in 3 incidents, and 2 women engaged in 5 incidents each.

Since not all of the women in the sample were incarcerated for the full year, the incidence rate was calculated based on the amount of time the women were incarcerated, a concept also known as person-years or person-days. A person-year, in this case, is measured by one person incarcerated for one year. The women in the sample spent a total of 81,324 days in a federal correctional institution between April 1, 2008, and March 31, 2009. The number of days per person ranged from 2 to 365, with an average of 203.31 days (SD = 123.32). There were .00036 incidents per person-day or 3.6 incidents per 27.4 person-years. Interpreted another way, if these 400 women were followed for 25 days, there would be 3.6 incidents of SIB. This rate is weighted in the sense that offenders who were incarcerated for longer periods of time contribute more heavily to the rate, thus effectively accounting for time at risk.

Discussion

The aim of the current research was twofold: first, to determine the prevalence and incidence of NSSI in Canadian federally sentenced women and second, to test the being in prison increases the likelihood of NSSI hypothesis. These objectives were met utilizing two different methodological approaches. In Study 1, prevalence rates and the being in prison increases the likelihood of NSSI hypothesis were examined using a mixed-method field approach where 150 incarcerated women were interviewed face to face and completed a self-report questionnaire about the extent and nature of their SIB. In Study 2, prevalence and incidence rates were determined using an archival methodology that entailed extracting data from 400 randomly selected electronic files maintained by the Correctional Service of Canada.

In sum, the lifetime prevalence rate of NSSI was 38% in the field study. In contrast, the lifetime prevalence rate of SIB for women in the archival study was 24%. While it is difficult to precisely account for the difference in prevalence rates found between studies, the slightly lower prevalence rate found in the archival study is likely due to underreporting of incidents in file data for incidents that occurred while incarcerated and especially for incidents that occurred prior to incarceration. Further replication of both types of studies is needed.

The prevalence rates found in the studies presented here are considerably higher than the best estimates for the general population, which are about 4% (Briere & Gil, 1998; Klonsky et al., 2003). However, taken together, the rates found here are comparable to the lifetime SIB prevalence rate of 32% found in a representative sample of federally sentenced women offenders in England and Wales that used a combination of interview and archival data (Maden et al., 1994). These rates are also similar to the lifetime SIB prevalence rates of incarcerated men, at 15% to 35% (Fotiadou et al., 2006; Maden et al., 1994, 2000; Sakelliadis et al., 2009).

The current study is one of the first to investigate NSSI in women offenders prior to and after incarceration using a precise definition of NSSI. The results do not support the being in prison increases the likelihood of NSSI hypothesis. Specifically, 80% (archival study estimate) to 93% (field study estimate) of the women first engaged in NSSI prior to being admitted to a federal institution. Moreover, approximately two-thirds of these women with a history of self-injury did not self-injure after being admitted. These results are consistent with past research from the United States and England that has attempted to assess the role that the correctional environment plays in the initiation and maintenance of SIB. For example, Snow (1997) found that 10 out of 11 women interviewed in custody in England had first self-injured prior to incarceration. Similarly, Jones (1986) reported that based on file information, 32 out of 67 offenders who engaged in SIB while incarcerated had wrist and arm scars upon entry to the correctional institution. However, only 4 of the offenders in the sample were women.

While some authors have proposed that the correctional environment causes women to engage in NSSI (e.g., Kilty, 2011), the finding that most women first engage in NSSI in the community suggests that the correctional environment does not play a direct, causal role in the onset and maintenance of NSSI. Thus, the correctional environment does not appear to be causally related to NSSI. Within the literature on motivations for engaging in NSSI, the affect regulation model (i.e., individuals engage in NSSI in order to regulate negative emotions, resulting in a period of relief following the NSSI) has the strongest empirical support (for a systematic review, see Klonsky, 2007). Even if women offenders experience negative emotions while incarcerated and use NSSI to deal with these emotions, NSSI was most likely learned as a coping strategy prior to entering an institution. However, our research does not address whether the contagion effect plays a role in initiating and/or maintaining NSSI that occurs in correctional institutions. More specifically, this study could not address whether or not women with a history of NSSI prior to incarceration are more likely to reinitiate NSSI upon seeing peers engage in the behavior while in the correctional institution.

There are many ways in which incidence rates can be calculated in correctional facilities, including using the average number of occupied beds, the number of admissions to the facilities, the average daily population, the average length of stay, or the number of days incarcerated (i.e., person-days) to calculate prevalence rates, all of which have advantages and disadvantages (Gallagher & Dobrin, 2007; O’Toole, 1997). There are, however, considerable advantages to accounting for time at risk when calculating incidence, as it is the most standardized way to report incident numbers, particularly within correctional populations. One way of incorporating time at risk is through the use of person-years, which avoids these problems. The incidence rate found in the archival study corresponds to 3.6 incidents per 27.4 person-years incarcerated. The current research indicates that self-injury while incarcerated is not all that common given the number of individuals incarcerated, the high turnover of women in federal institutions due to short sentences, and the high proportion of women who have mental health issues. Comparable incidents rates could not be found and therefore it is impossible to determine how these findings related to those of other jurisdictions. That is not to say, however, that NSSI is not a serious concern within correctional facilities. Any amount of NSSI has human and financial costs. Every incident has the potential to cause serious damage to the woman engaging in the behavior or even result in her death. Additionally, witnessing and responding to NSSI can be challenging, and even traumatizing, for staff, and therefore every incident has the potential for secondary ramifications for individuals other than the woman engaging in NSSI. Additionally, the potential effect of just one woman who engages in repeated, serious self-injury could cause the numbers to change considerably.

Limitations

There are limitations to the methodologies used for the studies presented here. In the field study, the participants were not randomly selected and therefore there may have been response bias in that women with a history of self-injury may have been more or less likely to participate. It is impossible to determine if the women who agreed to participate were in some way different from those who chose not to participate. While the recruitment posters and messages conveyed that all women (i.e., regardless of NSSI history) were eligible to participate, ethically it was necessary to emphasize the fact that NSSI would be discussed during the interviews. Some women likely interpreted this information to mean that only those who had self-injured were eligible to participate, thus biasing the sample with more women who had a history of NSSI than those who did not have this history. Some women who had serious mental health issues and/or were currently housed in segregation cells were deemed to be unsafe to participate in the field study by institutional staff. Therefore, the sample in the field study cannot be considered to be representative of all federal women offenders in Canada. Additionally, due to the sensitive nature of the topic, there may have been socially desirable responding that resulted in an underreporting of history of self-injury, and the validity of the prevalence of NSSI as determined through self-report has not yet been established.

In the archival study, the true incidence and prevalence of SIB is probably underestimated as not all information on SIB is recorded. It is likely that some SIB incidents that occurred in correctional institutions were not recorded, and it is even more likely that history of SIB was underreported, as this information is most often obtained from self-report at the time of admission to the institution. Women who are flagged as having a history of NSSI likely receive extra attention from staff members, resulting in more detailed files on the psychological history. Women who can remain under the radar, however, likely receive less attention and are less likely to have detailed psychological files (the quantity of psychological files varied considerably). Therefore, there could have been a bias in terms of more information being available for those who were considered to be more problematic.

Implications And Conclusions

Despite these limitations, this research represents the most in-depth study of NSSI among Canadian federally sentenced women to date and can be considered the most accurate estimate of prevalence and incidence in this population so far determined in the literature. It can be concluded from these studies that federally sentenced woman are at increased risk for NSSI compared to the general population, with results indicating a lifetime prevalence rate of SIB ranging from 24% to 38%. Incidence of self-injury in a federal intuition over a 1-year period was found to be 3.6 per 27.4 person-years. Both studies indicated that for the majority of women in both samples, NSSI was first initiated in the community, prior to incarceration in a federal correctional institution. For treatment purposes, it is important to consider that most women are not initiating NSSI in response to their incarceration; rather, their NSSI is most likely a continuation of a pattern of behavior initiated prior to entering the correctional facility. Future research should attempt to replicate our main findings with other populations of women offenders, preferably with a longitudinal research design that could assess self-injury prior to and after being admitted to a correctional facility.

Footnotes

This research was supported by the Research Branch at the Correctional Service of Canada. The views and opinions expressed in this article are those of the authors and do not necessarily reflect the policies and perspectives of the Correctional Service of Canada. The research presented here represents a portion of the primary author’s dissertation completed at Carleton University.

Notes

Jenelle Power, PhD, is a research manager at the Correctional Service of Canada and an adjunct professor with the Department of Psychology at Carleton University, Ottawa, Canada.

Shelley L. Brown, PhD, is an associate professor within the Department of Psychology, Carleton University, Ottawa, Canada.

Amelia M. Usher, MEd, is a doctoral student at Ryerson University in Toronto, Canada. She previously worked as a research officer at the Research Branch at the Correctional Service of Canada.

References

Albach

Everaerd

(1992). Posttraumatic stress symptoms in victims of childhood incest. Psychotherapy and Psychosomatics, 57, 149-151.

Andover

M. S.

Pepper

C. M.

Ryabchenko

K. A.

Orrico

E. G.

Gibb

B. E.

(2005). Self-mutilation and symptoms of depression, anxiety, and borderline personality disorder. Suicide and Life-Threatening Behavior, 35, 581-591.

Appelbaum

K. L.

Savageau

J. A.

Trestman

R. L.

Metzner

J. L.

Baillargeon

(2011). A national survey of self-injurious behaviour in American prisons. Psychiatric Services, 62, 285-290.

Berman

M. E.

Walley

J. C.

(2003). Imitation of self-aggressive behavior: An experimental test of the contagion hypothesis. Journal of Applied Social Psychology, 33, 1036-1057.

Blanchette

K. D.

Brown

S. L.

(2006). The assessment and treatment of women offenders: An integrative perspective. Chichester, UK: John Wiley and Sons.

Blanchette

K. D.

Motiuk

L. L.

(1995, June). Female offender risk assessment: The case management strategies approach. Poster session presented at the Annual Convention of the Canadian Psychological Association, Charlottetown, Prince Edward Island.

Bonta

Pang

Wallace-Capretta

(1995). Predictors of recidivism among incarcerated female offenders. The Prison Journal, 75, 277-294.

Borrill

Snow

Medlicott

R. T.

Paton

(2003). Learning from “near misses”: Interviews with women who survived an incident of severe self-harm in prison. The Howard Journal, 44, 57-69.

Briere

Gil

(1998). Self-mutilation in clinical and general population sample: Prevalence, correlates, and functions. American Journal of Orthopsychiatry, 68, 609-620.

10.

Brown

M. Z.

Comtois

K. A.

Linehan

M. M.

(2002). Reasons for suicide attempts and nonsuicidal self-injury in women with borderline personality disorder. Journal of Abnormal Psychology, 111, 198-202.

11.

Carli

Jovanovic

Podlesek

Roy

Rihmer

Maggi

Sarchiapone

(2010). The role of impulsivity in self-mutilators, suicide ideators and suicide attempters—A study of 1265 male incarcerated individuals. Journal of Affective Disorders, 123, 116-122.

12.

Chowanec

G. D.

Josephson

A. M.

Coleman

Davis

(1991). Self-harming behavior in incarcerated male delinquent adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 30, 202-207.

13.

Claes

Vandereycken

Vertommen

(2007). Self-injury in female versus male psychiatric patients: A comparison of characteristics, psychopathology and aggression regulation. Personality and Individual Differences, 42, 611-621.

14.

Cookson

H. M.

(1977). A survey of self-injury in a closed prison for women. British Journal of Criminology, 17, 332-347.

15.

Cooper

Husain

Webb

Waheed

Kapur

Guthrie

Appleby

(2006). Self-harm in the UK: Differences between South Asians and Whites in rates, characteristics, provision of service and repetition. Social Psychiatry and Psychiatric Epidemiology, 41, 782-788.

16.

Correctional Service of Canada. (2010). Unpublished raw data. Retrieved September 1, 2010 from Correctional Service of Canada Corporate Reporting System.

17.

Darche

M. A.

(1990). Psychological factors differentiating self-mutilating and non-self-mutilating adolescent inpatient females. The Psychiatric Hospital, 21, 31-35.

18.

Davis

Karvinen

(2002). Personality characteristics and intention to self-harm: A study of eating disordered patients. Eating Disorders, 10, 245-255.

19.

DeHart

D. D.

Smith

H. P.

Kaminski

R. J.

(2009). Institutional responses to self-injurious behavior among inmates. Journal of Correctional Health Care, 15, 129-141.

20.

De Leo

Heller

T. S.

(2004). Who are the kids who self-harm? An Australian self-report school survey. Medical Journal of Australia, 181, 140-144.

21.

Deliberto

Nock

M. K.

(2008). An exploratory study of correlates, onset, and offset of non-suicidal self-injury. Archives of Suicide Research, 12, 219-231.

22.

Favazza

A. R.

Conterio

(1989). Female habitual self-mutilators. Acta Psychiatrica Scandinavica, 79, 283-289.

23.

Fotiadou

Livaditis

Manou

Kaniotou

Xenitidis

(2006). Prevalence of mental disorders and deliberate self-harm in Greek male prisoners. International Journal of Law and Psychiatry, 29, 68-73.

24.

Gallagher

C. A.

Dobrin

(2007). Risk of suicide in juvenile justice facilities: The problem of rate in high-turnover populations. Criminal Justice and Behavior, 34, 1363-1376.

25.

Gratz

K. L.

(2001). Measurement of deliberate self-harm: Preliminary data on the Deliberate Self-Harm Inventory. Journal of Psychopathology and Behavioral Assessment, 23, 253-263.

26.

Guerino

Harrison

P. M.

Sabol

W. J.

(2011). Prisoners in 2010. U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics. Retrieved from http://www.bjs.gov/content/pub/pdf/p10.pdf

27.

Haines

Williams

C. L.

Brain

K. L.

Wilson

G. V.

(1995). The psychophysiology of self-mutilation. Journal of Abnormal Psychology, 104, 471-489.

28.

Harrison

P. M.

Beck

A. J.

(2005). Prison and jail inmates at midyear 2004 (NCJ 208801). Retrieved from http://bjs.ojp.usdoj.gov/content/pub/pdf/pjim04.pdf

29.

Hawton

Fagg

Simkin

Bale

Bond

(2000). Deliberate self-harm in adolescents in Oxford, 1985-1995. Journal of Adolescence, 23, 47-55.

30.

Heney

(1990). Report on self-injurious behaviour in the Kingston Prison for Women. Ottawa, Canada: Correctional Service of Canada.

31.

Heney

(1996). Dying on the inside: Suicide and suicidal feelings among federally incarcerated women. Ottawa, Canada: Carleton University.

32.

Herpertz

(1995). Self-injurious behaviour: Psychopathological and nosological characteristics in subtypes of self-injurers. Acta Psychiatrica Scandinavica, 91, 57-68.

33.

Hillbrand

Krystal

J. H.

Sharpe

J. S.

Foster

H. G.

(1994). Clinical predictors of self-mutilation in hospitalized forensic patients. Journal of Nervous and Mental Disease, 182, 10-13.

34.

Holdin-Davis

(1914). An epidemic of hair-pulling in an orphanage. British Journal of Dermatology, 26, 207-210.

35.

Horrocks

Price

House

Owens

(2003). Self-injury attendances in the accident and emergency department. British Journal of Psychiatry, 183, 34-39.

36.

Howard League for Penal Reform. (1999). Scratching the surface: The hidden problem of self-harm in prisons. London: Author.

37.

Hubbard

Matthews

(2008). Reconciling the differences between “gender-responsive” and the “What Works” literatures to improve services for girls. Crime and Delinquency, 54, 225-258.

38.

Jones

(1986). Self-mutilation in prison: A comparison of mutilators and nonmutilators. Criminal Justice and Behavior, 13, 286-296.

39.

Kilty

J. M.

(2011). On self-injurious behaviour in prison. Journal of Prisoners on Prisons, 20(1), 102-107.

40.

Kisiel

C. L.

Lyons

J. S.

(2001). Dissociation as a mediator of psychopathology among sexually abused children and adolescents. American Journal of Psychiatry, 158, 1034-1039.

41.

Klonsky

E. D.

(2007). The functions of deliberate self-injury: A review of the evidence. Clinical Psychology Review, 27, 226-239.

42.

Klonsky

E. D.

Muehlenkamp

J. J.

(2007). Self-injury: A research review for the practitioner. Journal of Clinical Psychology: In Session, 63, 1045-1056.

43.

Klonsky

E. D.

Oltmanns

T. F.

Turkheimer

(2003). Deliberate self-harm in a nonclinical population: Prevalence and psychological correlates. American Journal of Psychiatry, 160, 1501-1508.

44.

Kong

AuCoin

(2008). Female offenders in Canada, Juristat (Catalogue no. 85-002-XIE). Ottawa, Canada: Canadian Centre for Justice Statistics, Statistics Canada.

45.

Langbehn

D. R.

Pfohl

(1993). Clinical correlates of self-mutilation among psychiatric patients. Annals of Clinical Psychiatry, 5, 45-51.

46.

Lohner

Konrad

(2006). Deliberate self-harm and suicide attempt in custody: Distinguishing features in male inmates’ self-injurious behaviour. International Journal of Law and Psychiatry, 29, 370-385.

47.

Maden

Chamberlain

Gunn

(2000). Deliberate self-harm in sentenced male prisoners in England and Wales: Some ethnic factors. Criminal Behaviour and Mental Health, 10, 199-204.

48.

Maden

Swinton

Gunn

(1994). A criminological and psychiatric survey of women serving a prison sentence. British Journal of Criminology, 34, 172-191.

49.

Matsumoto

Yamaguchi

Asami

Okada

Yoshikawa

Hirayasu

(2005). Characteristics of self-cutters among male inmates: Association with bulimia and dissociation. Psychiatry and Clinical Neurosciences, 59, 319-326.

50.

Matthews

P. C.

(1968). Epidemic self-injury in an adolescent unit. International Journal of Psychiatry, 14, 125-133.

51.

Menninger

K. A.

(1938). Man against himself. Oxford, UK: Harcourt, Brace.

52.

Nixon

M. K.

Cloutier

P. F.

Aggarwal

(2002). Affect regulation and addictive aspects of repetitive self-injury in hospitalized adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 41, 1333-1341.

53.

Nixon

M. K.

Cloutier

P. F.

Jansson

S. M.

(2008). Nonsuicidal self-harm in youth: A population-based survey. Canadian Medical Association Journal, 178, 306-312.

54.

Nock

M. K.

Cha

C. B.

(2009). Psychological models of nonsuicidal self-injury. In Nock

M. K.

(Ed.), Understanding nonsuicidal self-injury (pp. 65-77). Washington, DC: American Psychological Association.

55.

Nock

M. K.

Joiner

T. E.

Gordon

K. H.

Lloyd-Richardson

Prinstein

M. J.

(2006). Non-suicidal self-injury among adolescents: Diagnostic correlates and relation to suicide attempts. Psychiatry Research, 144, 65-72.

56.

Nock

M. K.

Prinstein

M. J.

(2004). A functional approach to the assessment of self-mutilative behavior. Journal of Consulting and Clinical Psychology, 72, 885-890.

57.

Offer

Barglow

(1960). Adolescent and young adult self-mutilation incidents in a general psychiatric behavior. Archives of General Psychiatry, 3, 194-204.

58.

Office for National Statistics. (1997). Mortality statistics: Injury and poisoning 1995. London: The Stationery Office.

59.

O’Toole

(1997). Jails and prisoners: The numbers say they are more different than generally assumed. American Jails, 2(2) 27-31.

60.

Paul

Schroeter

Dahme

Nutzinger

D. O.

(2002). Self-injurious behavior in women with eating disorders. American Journal of Psychiatry, 159, 408-411.

61.

Power

(2011). Non-suicidal self-injury in federally sentenced women: Prevalence, nature, motivations, and pathways (Doctoral dissertation). Retrieved from ProQuest Dissertations & Theses A&I. (NR81588)

62.

Prinstein

M. J.

Nock

M. K.

Simon

Aikins

J. W.

Cheah

C. S. L.

Spirito

(2008). Longitudinal trajectories and predictors of adolescent suicidal ideation and attempts following inpatient hospitalization. Journal of Consulting and Clinical Psychology, 76, 92-103.

63.

Public Safety Canada. (2011). Corrections and conditional release statistical overview 2011: Annual report 2011. Retrieved from http://www.publicsafety.gc.ca/res/cor/rep/_fl/2011-ccrso-eng.pdf

64.

Rettinger

J. L.

(1998). A recidivism follow-up study investigating risk and need within a sample of provincially sentenced women (Unpublished doctoral dissertation). Carleton University, Ottawa, Canada.

65.

Rodham

Hawton

Evans

(2004). Reasons for deliberate self-harm: Comparison of self-poisoners and self-cutters in a community sample of adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 43, 80-87.

66.

Roe-Sepowitz

(2007). Characteristics and predictors of self-mutilation: A study of incarcerated women. Criminal Behaviour and Mental Health, 17, 312-321.

67.

Rosen

P. M.

Walsh

B. W.

(1989). Patterns of contagion in self-mutilation epidemics. American Journal of Psychiatry, 146, 656-658.

68.

Ross

R. R.

McKay

H. B.

(1979). Self-mutilation. Lexington, MA: Lexington Books.

69.

Ross

Heath

(2002). A study of the frequency of self-mutilation in a community sample of adolescents. Journal of Youth and Adolescence, 31, 67-77.

70.

Sakelliadis

E. I.

Papadodima

S. A.

Sergentanis

T. N.

Giotakos

Spiliopoulou

C. A.

(2009). Self-injurious behavior among Greek male prisoners: Prevalence and risk factors. European Psychiatry, 25, 151-158.

71.

Salina

D. D.

Lesondak

L. M.

Razzano

L. A.

Weilbaecher

(2007). Co-occurring mental disorders among incarcerated women: Preliminary findings from an integrated health treatment study. Mental Health Issues in the Criminal Justice System, 45, 207-225.

72.

Shea

S. J.

Shea

M. C.

(1991). Self-mutilatory behavior in a correctional setting. Corrective and Social Psychiatry and Journal of Behavior Technology Method and Therapy, 37, 64-67.

73.

Simeon

Stanley

Frances

Mann

J. J.

Winchel

Stanley

(1992). Self-mutilation in personality disorders: Psychological and biological correlates. American Journal of Psychiatry, 149, 221-226.

74.

Sinclair

R. L.

Boe

(2002). Canadian federal women offender profiles: Trends from 1981 to 2002 (Revised) (Research Report R-131). Ottawa, Canada: Research Branch, Correctional Service of Canada.

75.

Smith

H. P.

Kaminski

R. J.

(2010). Inmate self-injurious behaviors: Distinguishing characteristics within a retrospective study. Criminal Justice and Behavior, 37(1), 81-96.

76.

Smith

H. P.

Kaminski

R. J.

(2011). Self-injurious behaviors in state prisons: Findings from a national survey. Criminal Justice and Behavior, 38, 26-41.

77.

Snow

(1997). A pilot study of self-injury amongst women prisoners. Issues in Criminological & Legal Psychology, 28, 50-59.

78.

Stanley

Gameroff

M. J.

Michalsen

Mann

J. J.

(2001). Are suicide attempters who self-mutilate a unique population? American Journal of Psychiatry, 158, 427-432.

79.

Van Voorhis

(2012). On behalf of women offenders: Women’s place in the science of evidence-based practice. Criminology and Public Policy, 11, 111-145.

80.

Walsh

B. W.

Rosen

P. M.

(1985). Self-mutilation and contagion: An empirical test. American Journal of Psychiatry, 142, 119-120.

81.

Weaver

T. L.

Cahrd

K. M.

Mechanic

M. B.

Etzel

J. C.

(2004). Self-injurious behavior, PTSD arousal, and general health complaints within a treatment-seeking sample of sexually abused women. Journal of Interpersonal Violence, 19, 558-575.

82.

Weierich

M. R.

Nock

M. K.

(2008). Posttraumatic stress symptoms mediate the relation between childhood sexual abuse and nonsuicidal self-injury. Journal of Consulting and Clinical Psychology, 76, 39-44.

83.

Whitlock

Eckenrode

Silverman

(2006). Self-injurious behaviours in a college population. Pediatrics, 117, 1939-1948.

84.

Wichmann

Serin

Abracen

(2002). Women offenders who engage in self-harm: A comparative investigation Ottawa, Canada: Correctional Service of Canada.

85.

Yaryura-Tobias

J. A.

Neziroglu

F. A.

Kaplan

(1995). Self-mutilation, anorexia, and dysmennorrhea in obsessive compulsive disorder. International Journal of Eating Disorders, 17, 33-38.

86.

Young

M. H.

Justice

J. V.

Erdberg

(2006). Risk of harm: Inmates who harm themselves while in prison psychiatric treatment. Journal of Forensic Science, 51, 156-162.

87.

Zlotnick

Mattia

J. I.

Zimmerman

(1999). Clinical correlates of self-mutilation in a sample of general psychiatric patients. Journal of Nervous & Mental Disease, 187, 296-301.