Abstract
Recently, the Interpersonal Theory of Suicide has offered some understanding as to why inmates are at a disproportionately high risk to die by suicide. The present study was designed to investigate how one aspect of the theory, the acquired capability for suicide, may differ between prison inmates with violent offenses versus those without. Data from 343 male prisoners within the Mississippi Department of Corrections were used to examine potential differences between these two groups on scales of the Acquired Capability for Suicide Scale. Contrary to the hypothesis that violent offenders would evidence higher levels of acquired capability, only the Spectator Enjoyment of Violence subscale was significantly different between the two groups, with nonviolent offenders producing higher scores. Potential explanations for these findings and implications for correctional mental health care providers are discussed.
Suicide is a chronic problem in U.S. prisons, constituting the second leading cause of death among prison inmates from 2001 to 2013 (Noonan, Rohloff, & Ginder, 2015). Prison inmates may be at a greater risk for suicide compared with people who are not incarcerated due to environmental factors specific to incarceration. This may include social isolation due to single cell placement, separation from social support systems in the community (i.e., family and friends), family conflict that arises from the impacts of one’s incarceration, and an experience of hopelessness about one’s future when facing a long sentence and knowing that on release, one will still have a criminal record (Smith et al., 2013).
Although risk factors for suicide have been well established, both in general and among specific populations such as inmates, the causal mechanisms of suicide are still uncertain. Although a number of theories have been offered to explain suicide, the Interpersonal Theory of Suicide (ITS) is a relatively recent and promising theory that attempts to explain two main components underlying suicide: why people form a desire to die by suicide, and what enables them to engage in a serious suicide attempt (Joiner, 2005). The ITS states that people develop suicidal ideation when they believe that they are a burden on other people (perceived burdensomeness) and feel disconnected from those around them (thwarted belongingness) over time (Van Orden et al., 2010). An individual, however, becomes particularly at-risk for suicide when both of these components are present, and the person has a diminished fear of death and self-harm behavior (acquired capability).
Given the unique circumstances that many inmates experience, the ITS appears to be particularly relevant to this population (Mandracchia & Smith, 2015). Prison inmates may be at increased risk for experiencing thwarted belongingness simply by the fact that they are being separated from their support system in the community (e.g., family, friends, coworkers, neighbors). Given the finding that prison inmates who die by suicide often had been residing in a single cell at the time of their death (Patterson & Hughes, 2008), it may be that thwarted belongingness is exacerbated if inmates are additionally separated from their support systems that they have established “on the inside.” Inmates may also be at increased risk for experiencing perceived burdensomeness given that they may no longer be able to provide financial, emotional, or other support to their families (e.g., caring for the household and/or children), and that their families and friends may feel pressure to send money to the correctional institution for the inmate’s use (e.g., for commissary/canteen purchasing) or feel compelled to travel a significant distance to visit an inmate on a regular basis (Mandracchia & Smith, 2015).
The ITS suggests that when a person experiences high levels of both perceived burdensomeness and thwarted belongingness, they will experience the most severe form of suicidal desire (Van Orden et al., 2010). Given the circumstances that incarcerated offenders are likely to experience, then, it is unsurprising that they are more likely to experience a desire to die by suicide than among people who are not incarcerated (i.e., 33% of incarcerated individuals [Mandracchia & Smith, 2015], compared with 2% to 10% of the U.S. general population [Nock et al., 2008]). As explained by the ITS, though, only a portion of those who experience a desire to die by suicide actually go on to make a serious suicide attempt, and even fewer individuals will die by suicide (Joiner, 2005). This discrepancy between the prevalence of a desire to die by suicide and the incidence of a serious suicide attempt is suggested to exist due to people’s instinctive drive toward self-preservation (Van Orden et al., 2010). Humans have acquired a fear system through evolution to identify threats to survival (Öhman & Mineka, 2001), which includes painful and life-threatening events, even when such events are self-inflicted. The few who are able to overcome this instinct acquire the capability to do so through repeated exposure to painful and/or provocative events (i.e., experiences that threaten one’s life or safety; Joiner, 2005).
The acquired capability for suicide, then, is composed of both increased physical pain tolerance and a reduced fear of death (Van Orden et al., 2010). This capability is developed when the fear-causing stimuli continues to be present; the person’s experience of fear decreases and is replaced by experience of relief (opponent process). In other words, with repeated exposure to painful and provocative events, the fear response becomes extinct (i.e., habituation; Van Orden et al., 2010). Previous researchers have found numerous factors that appear to promote the development of acquired capability, including having committed acts of interpersonal violence (Wolford-Clevenger et al., 2015), previous suicide attempts, childhood maltreatment, combat exposure, and even frequent exposure to suicide behaviors engaged in by other people (Van Orden et al., 2010).
Compared with the general population, prison inmates seem to be exposed to various risk factors for developing acquired capability that are specific to incarceration (Smith, Wolford-Clevenger, Mandracchia, & Jahn, 2013). However, certain inmates may be at greater risk for suicide than others. One such factor that may be particularly relevant to offenders is experiences of violence. Experiencing violence in one’s environment during childhood and adolescence has been shown to be related to increased suicide risk as an adult (Moberg et al., 2014). Also, individuals with a history of both self- and/or other-directed violence have been shown to be at a higher risk for suicide (Van Dulmen et al., 2013). Pretrial offenders who were perpetrators of interpersonal violence reported being more likely to experience suicidal ideation than those who had not engaged in violent behavior (You, Swogger, Cerulli, & Conner, 2011).
Regarding risk of suicide specific to prisoners, a case-controlled study conducted in Austria evaluated differences between inmates who had died by suicide versus “matched controls” (i.e., inmates matched by institution, custodial status, time of admission, age, gender, and nationality who did not die by suicide), and found that having been serving a sentence for a “highly violent index offense” was a postdictive factor for death by suicide (Fruehwald, Matschnig, Koenig, Bauer, & Frottier, 2004). Given that having a violent index offense appears to be associated with suicide risk, the World Health Organization (WHO) identified this as one of many risk factors that correctional employees (i.e., mental health care providers as well as administrators and correctional officers) should be aware of in considering an inmate’s overall level of risk for engaging in suicidal behavior (WHO, 2007). However, simply being aware of known risk factors does not promote an understanding of why certain risk factors are associated with death by suicide, nor does it offer ideas as to why having a violent index offense may be contributing to the causal mechanisms underlying serious suicide attempts.
Despite the link between violent behavior and suicide risk (in general, as well as specifically among incarcerated offenders), and the link between perpetration of violence and the acquired capability for suicide, to our knowledge, no studies have been conducted that investigate the specific relation between having been convicted of a violent offense and the development of acquired capability among prison inmates. Smith et al. (2016) found that prison inmates who endorsed having had experiences that were labeled as “aggression” endorsed higher levels of overall acquired capability, general fearlessness and perceived pain tolerance, and enjoyment of viewing violence committed by others. These findings suggest that having committed a violent criminal offense, specifically, may promote the development of acquired capability for suicide; however, the aggressive life experiences that were included by Smith et al. were not offense-specific; they included acts of aggression committed against animals as well as acts of aggression that did not constitute a criminal offense. As such, no study to date that we are aware of has specifically addressed the possibility that an inmate’s offense type (i.e., a violent offense) may be an indicator that he or she has an increased level of acquired capability.
Researchers have studied the influence of person-to-person violence (e.g., combat exposure) on acquired capability in military personnel (Bryan & Cukrowicz, 2011). Similar to prison inmates, military personnel have been found to be at a higher risk for suicide than the general population, and like inmates, they experience separation from loved ones (i.e., which may promote thwarted belongingness) and may view their circumstances (i.e., deployment, frequent relocation) as causing problems or disadvantages for their families and friends (i.e., which may promote perceived burdensomeness). Interestingly, even though acquired capability among military personnel increases with exposure to combat, acquired capability among military personnel is higher than the general population even prior to deployment (i.e., before exposure to combat; Bryan, Morrow, Anestis, & Joiner, 2009), which suggests that even the process of undergoing military training may be sufficient to increase acquired capability (Bryan, Cukrowicz, West, & Morrow, 2010).
Existing research has demonstrated increased levels of acquired capability and suicidal ideation in prison inmates (Smith et al., 2016; Smith et al., 2013). As stated above, though, no studies that we are aware of have examined the relation between violent crime perpetration and acquired capability for suicide among incarcerated prison inmates. Therefore, the aim of the present study was to explore the potential relationship between having been convicted of a violent crime and the acquired capability for suicide in incarcerated males. We hypothesized that inmates who reported having been incarcerated for a violent offense would endorse higher levels of acquired capability than inmates with exclusively nonviolent offenses.
Method
Participants
In total, 399 adult male prisoners from two Mississippi Department of Corrections (MDOC) facilities participated in the present study. Most (i.e., nearly 95%) participants were incarcerated in a large state facility (i.e., capacity of more than 3,200 inmates), with the remaining participants being housed in a smaller regional facility in which both prison and jail inmates were housed (with only those classified as prisoners being asked to participate in the present study). Fifty-six cases were excluded due to improper provision of responses (i.e., not providing a response about the offense[s] for which they were incarcerated or not providing a sufficient number of responses on the measure of acquired capability for suicide). Of the 343 participants whose data were included in the analyses, their age ranged from 19 to 63 years (M = 34.46 years, SD = 10.39). The majority of the participants identified themselves as Black/African American (56.3%; n = 193), with the remaining participants identifying as either White/Caucasian (37.6%; n = 129), or Other (5.3%; n = 18), with 0.9% (n = 3) who provided no response regarding their race/ethnicity.
Most participants (66.8%; n = 229) reported having been convicted of only one crime for which they were currently incarcerated (i.e., index offense), while some reported two index offenses (28.6%; n = 98), three index offenses (n = 12; 3.5%), four index offenses (n = 2; 0.6%), or five index offenses (n = 2; 0.6%). The participants’ sentence length ranged from 4 months to 60 years (M = 9.28 years, SD = 9.58 years), with a mean of about 4 years of their sentence already served (M = 4.01 years, SD = 5.34 years).
Measures
Demographic Questionnaire
A demographic questionnaire was created for use in this study to obtain basic demographic and incarceration-related information about the sample.
Acquired Capability for Suicide Scale (ACSS)
The ACSS is a self-report questionnaire designed to determine one’s fearlessness of death and their perceived pain tolerance (Van Orden, Witte, Gordon, Bender, & Joiner, 2008). The response format of the ACSS consists of a 5-point Likert-type scale in which participants indicate to what degree each statement describes them (0 = not at all like me to 4 = very much like me). A factor analysis of the 20-item ACSS was conducted by Smith et al. (2013) using the present sample of incarcerated offenders. This produced a three-factor model of acquired capability, which included General Fearlessness and Perceived Pain Tolerance (four items), Fearlessness of Death (five items), and Spectator Enjoyment of Violence (three items). For these three factor scores and the corresponding 12-item total score, higher values represent higher levels of acquired capability. Recently, however, additional, independent modifications of the original ACSS have resulted in a 7-item ACSS Fearlessness About Death (FAD) scale (Ribeiro et al., 2014), which appears to currently be the most widely supported version of this measure. As such, we also calculated the 7-item ACSS-FAD score to include in our analyses along with scores from the three-factor and total scales identified by Smith et al. (2013).
Procedure
The present study utilized existing data that were previously approved by the Institutional Review Boards of the MDOC and the university at which the primary researcher was affiliated at the time data collection was conducted. Written informed consent was obtained from all participants prior to participation in the study. Incarcerated inmates with basic English literacy who agreed to participate in the study completed the questionnaire in a group setting and in accordance with MDOC policies and procedures to ensure the safe and efficient functioning of the institutions from which participants were solicited.
Results
Prior to the analyses, participants’ reported index offenses (i.e., the “grouping” variable) were individually categorized as either violent or nonviolent. This dichotomous categorization was conducted by having two undergraduate research assistants independently classify each offense reported by each offender as either violent or nonviolent. Interrater reliability was good (i.e., more than 95% agreement in classifying offenses). All categorized responses were reviewed by the principal researcher, and all inconsistencies in categorization were resolved via discussion with the two research assistants who had categorized the responses, a third research assistant, and the principal researcher.
Of the 343 participants whose data were included in the analysis, 114 reported at least one violent index offense (33.2%), and 229 (66.8%) reported no violent index offenses. The mean scores of the five scales of the ACSS described above were compared between these two groups (i.e., those with and those without a violent index offense) using a multivariate analysis of variance (MANOVA). These five ACSS scales included Ribeiro et al.’s (2014) FAD scale as well as Smith et al.’s (2013) Fearlessness and Pain Tolerance subscale, Fearlessness of Death subscale, Spectator Enjoyment of Violence subscale, and 12-item total scale. Of the five ACSS scale scores compared between the two groups, there was a statistically significant difference only for the Spectator Enjoyment of Violence Subscale (F = 4.346, p = .038). Specifically, the inmates with only nonviolent index offenses had significantly higher scores on Spectator Enjoyment of Violence compared with inmates with one or more violent index offense. Means and standard deviations of these ACSS scales by offender group (i.e., violent and nonviolent) as well as more complete data from the MANOVA are provided in Table 1.
Means, Standard Deviations, and Tests of Between-Group Differences for the Acquired Capability for Suicide Scale Scores.
Note. GF and PPT = general fearlessness and perceived pain tolerance; FD = fearlessness of death; SEV = spectator enjoyment of violence; FAD = fearlessness about death.
Statistically significant between-group mean score difference, p < .05.
Discussion
The present study is the first that we are aware of, to date, to explore the relationship between having committed a violent crime and the acquired capability for suicide in prison inmates. Our hypothesis was not supported by the results: inmates who reported having been incarcerated for at least one violent offense did not endorse higher levels of acquired capability. For most of the subscales of the ACSS, there was no significant difference between the two groups. However, the results indicated a significant difference in Spectator Enjoyment of Violence. Specifically, and contrary to our hypothesis, participants who were convicted of only nonviolent crimes scored higher on the Spectator Enjoyment of Violence subscale than those with a violent crime.
Regarding this finding that nonviolent offenders endorsed higher scores for Spectator Enjoyment of Violence, it may be that observing violence in media may be perceived or interpreted differently between these two groups. Those who have not committed violent offenses may associate viewing violence in television shows, movies, and sports (the latter of which being the focus of the items from this subscale of the ACSS) with excitement and enjoyment as a form of entertainment. Those who have committed a violent offense may, instead, find viewing violent behavior intended for entertainment purposes as more aversive. Consider, for instance, that one of the criteria for posttraumatic stress disorder includes “persistent avoidance of the stimuli associated with the traumatic event(s)” (American Psychiatric Association, 2013, p. 217). Similarly, it may be that those offenders who have engaged in violent behavior toward others had been affected by that experience such that they maintain a desire to avoid viewing any violent behavior, even in the context of “entertainment.” Those offenders who have not engaged in violent behavior, on the other hand, may not share this same desire to avoid viewing violence, such that they may consider the violence present in television shows, movies, and sports to contribute to their entertainment value.
Overall, however, the present results suggest that being a perpetrator of violence does not build acquired capability for suicide. The finding that four of the five ACSS scales were not significantly different between those with and those without a violent index offense could be because individuals who experience violence as the perpetrator (i.e., not the victim) might not experience this violence as an event that promotes a lessened fear of death or a perceived tolerance to pain. This may be because committing an act of violence toward another person may be “provocative” but at the same time may not cause pain to or threaten the safety of the perpetrator. For instance, if an offender were to have harmed someone, that offender may have experienced intense emotions as a result of engaging in that violent behavior, but the act of harming someone, by itself, may not increase the offender’s perceived likelihood of being harmed. More generally, it is possible that hurting/killing someone else does not increase the perpetrator’s perceived pain tolerance or decrease their fear of their own death; if there is no direct threat to the perpetrator, the process of habituation and extinction required to develop the acquired capability for suicide does not take place.
Even though the findings of the present study are inconsistent with our hypotheses as derived from the ITS, these findings may add to a more nuanced understanding of how acquired capability is formed (or not formed) among specific populations and with exposure to specific experiences or by engaging is specific forms of behavior. It may simply be that being a perpetrator of a violent crime, specifically, may not contribute to a lowered fear of death and/or increased perception of pain tolerance for the reasons considered above. It stands to reason that a decreased fear of one’s own death and a heightened perception of pain tolerance would be achieved through experiencing serious threats to one’s own life or one’s own experience of pain. This may be especially true for prison inmates, who have been shown to have lower levels of empathy (van Langen, Wissink, van Vugt, Van der Stouwe, & Stams, 2014). Having a decreased ability to understand or relate to another’s experience of having their life threatened or feeling pain may diminish an offender’s building of acquired capability for suicide while the offender is committing a violent act against another person. In a sense, then, while offenders’ lower levels of empathy may put them at an increased risk for violent behavior against others, it may at the same time provide some amount of protection against engaging in suicidal behavior by not enabling their violent actions toward others to contribute to their own acquired capability for suicide.
Our results do not promote any recommendations for correctional mental health care practitioners regarding inmates’ risk of suicidal behavior in relation to the inmates having a violent index offense. That is, knowing if an inmate is serving time for a violent offense does not appear to be an important component in determining suicide risk, based on the present findings. However, it is important to note that our findings only relate to one aspect of the ITS: the acquired capability to engage behavior that constitutes a serious suicide attempt. Our findings do not address any possible differences in a desire to die by suicide among inmates with a violent index offense versus those without a violence index offense. It may, or may not, be that having engaged in violent behavior against others contributes to an inmate’s experience of perceived burdensomeness and/or thwarted belongingness. It would be erroneous, therefore, for correctional mental health care providers to assume on the basis of the present study that inmates who have a violent index offense are “not more suicidal” than inmates with only nonviolent offenses.
The findings of the present study should be considered in light of a few limitations. The ACSS is a self-administered survey, and only responses from inmates who were motivated to complete the measures appropriately were included in the analysis. This may have resulted in a self-selection bias that undermines a true cross-sectional sample of the general population of inmates. Perhaps a more important limitation of this study is the certainty of the nature of the crimes for which the participating inmates were incarcerated. Previous research has shown that offenders’ self-report of their index offenses is generally accurate (Kroner, Mills, & Morgan, 2007). However, even if a participant had not been convicted of a violent offense for which they were serving time, and they reported this honestly on the demographic form, this does not rule out the possibility that they had actually engaged in violent behavior. For example, some participants may have made a plea bargain and had a violent offense reduced to a nonviolent offense. Furthermore, some participants who were serving time for a truly nonviolent offense may have previously engaged in violent behavior; we did not ask in the present study if the participant had previously served time for a violent offense or if they had engaged in violent behavior for which they were never convicted. Simply stated, having a group of inmates with no current violent index offenses does not infer that individuals in this group had never committed violent offenses against others.
There are many ways in which further research on this topic may extend upon the present findings and address the limitations of the present study. It may prove useful to include a more comprehensive assessment of inmates’ history of perpetrating violence, including prior convictions that led to sentences that have already been served as well as acts that did not result in a conviction as a violent offense. It may also be important to investigate whether there is a relation between prison inmates’ level of acquired capability for suicide and their having been the victim of violent acts, both prior to and during their present period of incarceration (i.e., in-prison violent victimization). Finally, other factors that may influence the relationship between having committed violent acts against others and the acquired capability, such as the inmates’ level of empathy (as discussed above), may point to important moderating variables that influence how acquired capability is differentially formed among incarcerated offenders.
Despite its limitations, the present study is important in that it is the first to examine the potential relationship between having committed a violent crime and the acquired capability for suicide. Only recently has the ITS garnered any attention in relation to its applicability to inmates despite this population being particularly at-risk for suicide. However, the few studies that do consider the applicability of the ITS to inmates tend to focus on the two factors of the theory that are said to create the desire to die by suicide (i.e., thwarted belongingness and acquired capability). The present study is an important step in expanding the investigation of the applicability of the third component of the theory, acquired capability, to inmates, and the first that considers specific, offense-related factors that may contribute to acquired capability. Given the challenges in conducting research with this population and the relative newness of the interpersonal theory, continued research is needed to bolster correctional mental health care providers’ understanding of how the acquired capability for suicide may form through experiences that are unique to the inmates with whom they work.
Footnotes
Authors’ Note
The research contained in this document was coordinated in part by the Mississippi Department of Corrections. The contents of this document reflect the views of the authors and do not necessarily reflect the views or policies of the Mississippi Department of Corrections.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Funding for this research was provided by Missouri Western State University’s Program of Research, Teaching, and Applied Learning (PORTAL).