Prevalence and Correlates of Suicidal Ideation among Court-Referred Male Perpetrators of Intimate Partner Violence

Abstract

Despite the documented association between intimate partner violence perpetration and suicidal ideation, few studies have examined the prevalence and correlates of suicidal ideation in men attending batterer intervention programs. This cross-sectional study examined the prevalence and correlates of suicidal ideation in 294 males court-ordered to a batterer intervention program. Twenty-two percent of the sample reported experiencing suicidal ideation within the two weeks prior to entering the batterer intervention program. Multiple linear regression indicated that depression and borderline personality disorder symptoms, but not intimate partner violence perpetration, victimization, or antisocial personality disorder symptoms, accounted for significant variance in suicidal ideation. These results suggest that symptoms of depression and borderline personality disorder observed in males attending batterer intervention programs should warrant thorough suicide risk assessment. Implications of the findings and limitations of the study are discussed.

Perpetration of intimate partner violence (IPV), defined as physical, sexual, and emotional aggression towards one’s current or former intimate partner, is associated with a number of adverse mental health symptoms (Shorey, Febres, Brasfield, & Stuart, 2012; Saltzman, Fanslow, McMahon, & Shelley, 2002). In particular, IPV perpetration has been found to be associated with suicidal ideation (Heru, Stuart, Rainey, Eyre, & Recupero, 2006; Ilgen et al., 2009; Nahapetyan, Orpinas, Song, & Holland, 2013; Rhodes et al., 2009). The development of suicidal ideation in male perpetrators is critical to understand and prevent, as it is indicative of acute risk for suicide attempt (Joiner, Walker, Rudd, & Jobes, 1999; Kessler, Berglund, Borges, Nock, & Wang, 2005), as well as risk for committing intimate partner homicide (Belfrage & Rying, 2004; Block & Christakos, 1995; Koziol-McLainet al., 2006).

Research examining the link between IPV perpetration and suicide risk has found that this relationship is especially strong in men seen in domestic violence courts (Conner, Cerulli, & Caine, 2002). Almost half of male perpetrators recruited from a domestic violence court reported threatening suicide in the past, with 70% of the threats occurring in the past six months, and 25% occurring the week prior to court (Conner et al., 2002). These data highlight the roles that interpersonal discord and related legal troubles may play in the development of suicidal ideation and related behaviors. Indeed, longitudinal data indicate that interpersonal and legal issues are strong predictors of suicide attempts above and beyond personality disorders (Yen et al., 2005). However, other studies have revealed either a negative or null relationship between IPV perpetration and suicidal ideation (Chan, Tiwari, Leung, Ho, & Cerulli, 2007; Houry et al., 2009; Peek-Asa et al., 2005). Additional research conducted on suicide risk among court-involved men is critically needed to clarify this relationship.

While the literature generally supports the relationship between IPV perpetration and suicidal ideation, little is known about the prevalence and correlates of suicidal ideation specific to men court-referred to batterer intervention programs (BIPs). A majority of BIPs utilize a blend of cognitive-behavioral and feminist-based approaches in treating partner-violent men (Stuart, Temple, & Moore, 2007). However, research demonstrating the limited efficacy of these programs (Babcock, Green, & Robie, 2004; Feder & Wilson, 2005) has led to suggestions for the use of adjunct and individualized treatments to improve BIPs (Juodis, Starzomski, Porter, & Woodworth, 2014; Stuart et al., 2007). Identifying the prevalence and correlates of suicidal ideation among men in BIPs will inform BIP service providers about the potential need for individualized assessment and adjunct treatments targeting suicide risk (Juodis et al., 2014; Stuart et al., 2007).

Important correlates of suicidal ideation among men in BIPs may include symptoms of psychopathology that are prevalent among men who are violent towards their partners. For example, male batterers differ in levels of symptoms of dysphoria, borderline personality disorder (BPD), and antisocial personality disorder (ASPD) (Hamberger, Lohr, Bonge, & Tolin, 1996; Holtzworth-Munroe, Meehan, Herron, Rehman, & Stuart, 2000; Shorey, Brasfield, Febres, & Stuart, 2011; Waltz, Babcock, Jacobson, & Gottman, 2000). Symptoms of BPD, ASPD, and depression have been identified as risk factors for suicidal ideation in the general population as well as prison populations; therefore, such symptoms may be relevant indicators of suicide risk among men mandated to BIPs (Black, Gunter, Loveless, Allen, & Sieleni, 2010; Douglas et al., 2008; James & Taylor, 2008; Kessler et al., 2005; Smith, Selwyn, Wolford-Clevenger, & Mandracchia, 2014).

Research provides indirect support for the idea that symptoms of BPD, ASPD, and depression may promote risk for suicidal ideation among male batterers. For example, male perpetrators of IPV with BPD characteristics have been found to experience higher suicidal ideation compared to perpetrators without such symptomology (Johnson et al., 2006). Similarly, in individuals receiving substance use treatment, negative affect has been found to mediate the relationship between physical violence perpetration and suicidal ideation (Ilgen et al., 2009). However, no studies have examined whether symptoms of BPD, ASPD, and depression are associated with increased suicidal ideation while controlling for the potential effects of psychological and physical IPV perpetration and victimization. Research examining the potential effects of psychiatric symptoms prevalent among male batterers will identify risk factors for suicidal ideation that are targetable for treatment in men attending BIPs.

In summary, the literature suggests that BIPs are critical settings in which male perpetrators of IPV may become at risk for suicidal ideation, and symptoms of depression, BPD, and ASPD may serve as indicators of such risk (Conner et al., 2002; Ilgen et al., 2009; Johnson et al., 2006; Yen et al., 2005). However, limitations of the research to date warrant further study. First, several of the studies investigating the link between IPV perpetration and suicidal ideation rely on dichotomous measurement of these variables and also focus primarily on physical violence perpetration, which oversimplify the conceptualization of these constructs (MacCallum, Zhang, Preacher, & Rucker, 2002). Assessing psychological IPV and using continuous measurements of IPV and suicidal ideation are critical to ensure the validity of findings, as these phenomena range in duration, frequency, and severity (Saltzman et al., 2002; Silverman, Berman, Sanddal, O’Carroll, & Joiner, 2007). Additionally, no studies have examined depressive, ASPD, and BPD symptoms as potential correlates of suicidal ideation while accounting for the potential effects of IPV perpetration and victimization. Finally, no studies, to our knowledge, have examined the prevalence of suicidal ideation in a sample of men attending BIPs.

Thus, the purpose of the current study was to extend the existing literature by examining the prevalence and correlates of suicidal ideation in male batterers attending BIPs. Given the literature demonstrating the high prevalence of risk factors for suicidal ideation, such as depression, BPD, and ASPD symptoms among male perpetrators of IPV, we expected that symptoms of depression, BPD, and ASPD would account for significant variance in suicidal ideation (Holtzworth-Munroe & Stuart, 1994; Verona, Sprague, & Javdani, 2012). We expected these correlates to emerge while controlling for the potential effects of IPV perpetration and victimization, which often co-occur and are both associated with suicidal ideation (Iverson et al., 2012; Langhinrichsen-Rohling et al., 2012; Langhinrichsen-Rohling, Misra, Selwyn, & Rohling, 2012).

Method

Participants

Participants were male perpetrators recruited from BIPs (n = 294). Nine men refused to participate in the study (3%). Participants were, on average, 33.13 years old (SD = 10.02). The racial/ethnic composition of the sample was as follows: 69% Caucasian/Non-Hispanic, 13% African American/Non-Hispanic, 9% Hispanic, 2% Asian or Pacific Islander, 2% Native American or Alaskan Native, and 5% “Other.” The average annual income of the sample was $34,054 (SD = $22,654), and the average number of years of education was 11.97 (SD = 2.25). The average number of children reported was 1.89 (SD = 1.94). A majority of the participants reported being married (27%), living with their partner (32%), or dating (19%) prior to the batterers’ intervention. The average relationship length in years was 6.31 (SD = 6.48). The average number of BIP sessions completed was 9.61 (SD = 7.12). We did not collect information about which BIP site perpetrators attended.

Measures

Demographic Questionnaire

Age, race/ethnicity, annual income, years of education, number of children (biological, adopted, and step-children), relationship status at the time of BIP entry, relationship length, and number of BIP sessions completed were collected from participants using a demographics questionnaire designed for the study. The demographics questionnaire was administered concurrent to the other measures.

Intimate Partner Violence

Psychological and physical IPV perpetration and victimization in the past year were assessed using the Conflict Tactics Scales-Revised (CTS2; Straus, Hamby, Boney-McCoy, & Sugarman, 1996; Straus, Hamby, & Warren, 2003). The CTS2 is 78-item scale that measures the frequency of sexual coercion, injury, and physical and psychological violence recorded on a 7-point Likert scale ranging from zero to more than twenty times. The CTS2 has evidence of test-retest reliability in court-mandated males (Vega & O’Leary, 2007) and construct and discriminant validity in student samples (Straus et al., 1996). The psychological and physical perpetration and victimization subscales were used for the current study and exhibited internal consistency ranging from acceptable to good (see Table 1).

Table 1.

Bivariate Correlations and Descriptive Statistics for Study Measures

	1	2	3	4	5	6	7	8
1. Physical Perpetration	.77	-	-	-	-	-	-	-
2. Psychological Perpetration	.58**	.76	-	-	-	-	-	-
3. Physical Victimization	.76**	.54**	.85	-	-	-	-	-
4. Psychological Victimization	.54**	.84**	.68**	.76	-	-	-	-
5. Depressive Symptoms	.18*	.25**	.11	.15*	.88	-	-	-
6. BPD Symptoms	.32**	37**	.25**	.25**	.47**	.70	-	-
7. ASPD Symptoms	27**	.31**	.19*	.25**	.16*	.47**	.77	-
8. Suicidal Ideation	.14*	.18*	.11*	.09	.44**	.39**	.16*	.86
Mean	8.36	29.93	15.07	33.03	3.52	2.53	2.66	.63
(SD)	(16.29)	(30.31)	(28.46)	(32.80)	(3.58)	(2.09)	(2.18)	(1.39)

Note: BPD = Borderline Personality Disorder, ASPD = Antisocial Personality Disorder, SD = Standard Deviation;

^*p < .05

^**p < .0001 (two-tailed).

Cronbach’s Alphas for each scale are presented on the diagonal.

Symptoms of Borderline and Antisocial Personality Disorders

Participants’ symptoms of BPD and ASPD occurring over the past several years were measured using the participants’ total scores on the BPD and ASPD subscales of the Personality Diagnostic Questionnaire-4, respectively (PDQ4; Hyler et al., 1988). The items assessing symptoms of BPD and ASPD loaded independently on each of the two subscales. The PDQ-4 is a screening tool designed to assess symptoms of Axis II disorders of the DSM. The PDQ-4 has demonstrated good test-retest reliability (Trull, 1993). The scale has also demonstrated high internal consistency in a patient sample (Hyler et al., 1989). The BPD and ASPD symptom subscales demonstrated acceptable internal consistency in the current study (α = .70 and .77, respectively).

Depressive Symptoms

Participants’ depressive symptoms over the two weeks prior to entering the BIP were assessed using the depression subscale of the Psychiatric Diagnostic Screening Questionnaire (PDSQ; Zimmerman & Mattia, 2001a,b). The PDSQ is an assessment used to screen various Axis I disorders from the Diagnostic and Statistical Manual-IV-TR (American Psychiatric Association, 1994). The items are on a yes-no scale with a total score ranging from 0-15. The depression subscale has demonstrated good test-retest reliability, internal consistency, and convergent and discriminant validity in psychiatric outpatients (Zimmerman & Mattia, 2001b). We only used the total score of the 15 items that did not contain suicidal content. The six suicide-related items were removed to create a suicidal ideation scale (see below). The 15 items of the depression subscale used for the current study demonstrated good internal consistency (α = .88).

Suicidal Ideation

Suicidal ideation was measured by totaling the sum of the six yes-no suicidal ideation items of the depression subscale of the PDSQ (Zimmerman & Mattia, 2001a,b). The total score ranges from 0-6. Participants were told to report on their levels of suicidal ideation for the two weeks prior to entering the BIP. To our knowledge, this set of items has not been validated as a measure of suicidal ideation. However, the subscale fits well with recent conceptualizations of suicidal ideation, with items assessing the range of severity of suicidal ideation including wishes for death, passive suicidal ideation, active suicide ideation, and serious planning and preparation for suicide (Silverman et al., 2007). The six items assessed thoughts of dying in passive ways, having wishes for death, thinking that one would be better off dead, thoughts of suicide with no intent to follow through, giving serious consideration to taking one’s life, and contemplating a suicidal plan. The six items exhibited good internal consistency in the current sample (α = .86).

Procedure

Men court-ordered to BIPs in Rhode Island were recruited for a larger study. Participants were recruited from three sites, which administered the same forty-hour curriculum in an open enrollment, group format. The number of sessions completed prior to study participation differed among the participants (M = 9.61, SD 7.12), but was not significantly correlated with IPV victimization, perpetration, symptoms of depression, BPD, ASPD, or suicidal ideation (p s > .10). Men who volunteered to participate completed a battery of measures in small groups during a batterer intervention session. No compensation for participating was provided, and information collected was completely confidential. A detailed description of the study procedures has been published elsewhere (see Stuart et al., 2008).

Results

Descriptive and Bivariate Analyses

Correlations, means, and standard deviations are displayed in Table 1. Sixty percent of the sample reported scores greater than one on both the physical perpetration and victimization subscales of the CTS2, suggesting a majority of the participants experienced bidirectional physical violence. Similarly, 90% of the sample reported scores greater than one on both the psychological perpetration and victimization subscales of the CTS2, suggesting most participants experienced bidirectional psychological aggression. Twenty-two percent of the sample reported some level of suicidal ideation within the two weeks prior to entering the BIP.

Suicidal ideation was positively correlated with physical IPV victimization, physical IPV perpetration, psychological IPV perpetration, and symptoms of BPD, ASPD, and depression. Suicidal ideation was not significantly correlated with psychological IPV victimization (see Table 1). Suicidal ideation was also not significantly correlated with any demographic variable including age, race, income, years of education, number of children, relationship status (ps > .10). Therefore, these variables were not included as covariates in the regression analysis.

Regression Analysis

Using simultaneous linear regression, we tested the hypothesis that symptoms of depression, BPD, and ASPD would account for significant variance in suicidal ideation while controlling for psychological and physical IPV perpetration and victimization. The subscales for psychological and physical IPV victimization, psychological and physical IPV perpetration, and symptoms of BPD, ASPD, and depression were entered as the predictor variables. The suicidal ideation subscale was entered as the criterion variable.

The overall model fit for the analysis was significant; R² = .307, F_{(7, 287)} = 18.20, p < .001. Standardized coefficients are reported. Our hypothesis that symptoms of BPD, depression, and ASPD would account for significant variance in suicidal ideation while controlling for IPV victimization and perpetration was generally supported. Symptoms of BPD (β = .16, t =2.63, p = .009) and depression (β = .44, t = 7.92, p < .001), but not symptoms of ASPD (β = .03, t = .537, p = .592), accounted for significant variance in suicidal ideation while controlling for IPV victimization and perpetration. Psychological IPV perpetration (β = −.06, t = −.732, p = .465), physical IPV perpetration (β = .06, t = .999, p = .319), psychological IPV victimization (β = .02, t = .286, p = .319), and physical IPV victimization (β = −.05, t = −.713, p = .477) did not significantly account for the variance in suicidal ideation.

Discussion

The current study is among the first, to our knowledge, to examine the prevalence and correlates of suicidal ideation in male batterers attending BIPs. Our results indicated that over one-fifth of the males attending BIPs experienced suicidal ideation within the two weeks prior to entering the BIP. Our results also indicated that symptoms of BPD and depression, but not symptoms of ASPD, contributed to the variance in suicidal ideation while accounting for physical and psychological IPV perpetration and victimization.

The finding that 22% of the men experienced suicidal ideation in the two weeks prior to entering the BIP corroborates past research findings that half of men involved in domestic violence court had any history of suicidal threats, with 25% of the threats occurring the week prior to court (Conner et al., 2002). Taken together, these findings indicate that men involved in domestic violence court or batterer interventions are at risk for both suicidal ideation and suicidal threats, possibly due to emotionally reactive and impulsive responses to crises (Conner et al., 2002). Considering the context of domestic violence court and batterer interventions, it is possible that the interpersonal discord and legal involvement inherent to these processes might be facilitating men’s risk for suicide as suggested by psychological autopsy and longitudinal studies (Conner, Duberstein, & Conwell, 2000; Yen et al., 2005). In addition, psychiatric symptoms that are prevalent among male perpetrators of IPV may facilitate impulsive and aggressive responses, such as suicidal ideation, to legal and interpersonal crises (Conner, Duberstein, Conwell, & Caine, 2003). Practitioners who treat male perpetrators of IPV, especially in BIP settings, should be aware of the potential and indicators for acute risk for suicide.

The current findings that depressive and BPD symptoms are indicators of suicidal ideation replicated and extended past research (Ilgen et al., 2009). This is not surprising as depressive and BPD symptoms are strong predictors of suicidal thoughts and behaviors, and have been found to be defining characteristics of a subtype of male perpetrators (Holtzworth-Munroe & Stuart, 1994; Holtzworth-Munroe et al., 2000; Johnson et al., 2006; Kessler et al., 2005; Linehan, 1993). Psychological aggression, physical violence, and depressive symptoms may constitute a construct that in itself promotes risk for suicidal ideation, such as reactive aggression or the dysphoric/borderline subtype of IPV (Conner et al., 2003; Johnson et al., 2006). However, BPD and depressive symptoms are heterogeneous, and the findings do not elucidate whether specific symptom clusters of these disorders may be driving this relationship (Conner et al., 2003). For example, men attending BIPs may exhibit the BPD cluster of inappropriate anger, and due to separation from their partner, may turn the anger onto themselves through violent, suicidal ideation. An additional explanation is that BPD and depressive symptoms may result from or be exacerbated by the legal and interpersonal difficulties associated with IPV perpetration, thus facilitating the development of suicidal ideation (Ilgen et al., 2009). Future longitudinal research is needed to examine the specific facets of BPD and depression, and the contextual factors surrounding BIP involvement that facilitate suicidal ideation in male perpetrators of IPV. Practitioners should consider depressive and borderline personality symptomology to be indicative of possible suicidal ideation.

Our hypothesis that ASPD symptoms would account for significant variance in suicidal ideation was not supported, which is inconsistent with past research (e.g., Douglas et al., 2008). Symptoms of ASPD were associated with suicidal ideation at the bivariate level but were no longer associated when accounting for IPV perpetration and victimization, and symptoms of BPD and depression. One explanation is that symptoms of BPD that have considerable overlap with ASPD symptoms (e.g., impulsivity; Paris, 1997) accounted for the variance in suicidal ideation to be explained by ASPD symptoms. Relatedly, research has found BPD to mediate the relationship between ASPD and suicidal behaviors (James & Taylor, 2008). An additional explanation is that symptoms of ASPD which appear to involve greater aggression than BPD may have a more influential role in development of risk for suicide attempts rather than ideation. Given the inconsistent findings regarding the relationship between ASPD symptoms and suicidal ideation, practitioners should not eliminate ASPD as a potential risk factor for suicide. Future work is needed to clarify the potential role of ASPD symptoms in promoting suicidal ideation among male batterers.

Limitations

The limitations of the current study provide suggestions for future research. First, participants completed the study measures in small groups during a BIP session. While facilitators did not observe the participants’ responses, the effects of facilitators’ and other group members’ presence in the room are unknown. Additionally, we did not collect data on which site participants’ data were collected from; therefore, differences among sites could not be examined. However, a longitudinal study with data collected from these BIPs demonstrated no significant differences in program outcomes that were attributable to recruitment site (Stuart et al., 2013).

Second, the cross-sectional design of this study cannot establish temporal relationships. Longitudinal data are needed to determine whether psychiatric symptoms and experiences of IPV precede, co-occur with, or follow suicidal ideation. Such data would also be helpful in understanding the trajectory of suicidal ideation over the course of court-involvement and the batterer intervention. These data would be invaluable in developing suicide risk assessment and management procedures in BIPs. Third, although this study improved upon past studies’ dichotomization of suicidal ideation, we measured suicidal ideation using items within a depression measure. Future studies should use a self-report measure or semistructured interview that was designed and validated to measure suicidal ideation.

Additionally, instead of measuring formal diagnoses of depression, BPD, and ASPD, we measured the number of symptoms endorsed for each of these disorders. Research should clarify whether discrete, diagnostic entities such as BPD or continuous constructs of emotional dysregulation, impulsivity, and negative affect common to many disorders mediate the relationship between IPV perpetration and suicidal ideation. We contend that it would be more informative to examine constructs shared by many psychiatric diagnoses that may promote suicidal ideation in male perpetrators of IPV.

Finally, we did not assess variables that directly place male perpetrators at risk for making suicide attempts. Future studies should test what correlates or aspects of IPV perpetration differentiate individuals who think about suicide from those who go on to attempt suicide. Future research that fills these gaps will advance the understanding of the relationship between IPV perpetration and risk for suicide.

Clinical Implications

Given that over one-fifth of the current sample reported experiencing suicidal ideation in the two weeks prior to entering the BIP, program service providers should routinely assess for suicidal ideation among participants. While the current findings can only generalize to BIP settings, clinicians who deliver outpatient services for couples and individuals who perpetrate IPV could also implement routine suicide risk assessment and management procedures. Thorough suicide risk assessment and management routines in BIPs and individual outpatient settings are not only critical to prevent harm to the male perpetrator but also to prevent homicide of his partner (Belfrage & Rying, 2004; Block & Christakos, 1995; Koziol-McLain et al., 2006). Targeting suicidal ideation has potential for reducing risk for homicide, as research has indicated in murder-suicides, suicidal intent occurs prior to the decision to commit murder, which possibly results from disinhibition that accompanies suicides (Baumeister, 1990; Rhine & Mayerson, 1973). Additionally, indications of suicidal ideation may necessitate a homicide risk assessment and, if necessary, the duty to warn a victim about lethality risk. Here, we first describe procedures for suicide risk assessment that may be useful to both BIP providers and outpatient clinicians. We then describe how the current findings may be applied to each of these settings specifically.

Assessing and managing suicide risk may be seen as daunting to some providers; however, helpful, detailed guides for such routines are available (e.g., Joiner et al., 1999). According to Joiner et al., (1999), providers should consider both acute and chronic risk for suicide. Individuals with high levels of acute risk will have suicidal ideation occurring at a high frequency, duration, and severity that involves thoughts about planning and preparing for one’s suicide. The current findings suggest that depressive and BPD symptoms should be considered as indicators of acute risk for suicide and should warrant further assessment. Chronic risk is indicated by a history of two or more suicide attempts (i.e., multiple attempters), as multiple attempters have been found to have a lower threshold for internal and external crises and thus are more prone to developing acute risk for suicide (Joiner & Rudd, 2000). Given that some male perpetrators may have a history of frequent suicidal threats (Campbell et al., 2007), providers should distinguish suicidal threats and suicide attempts when assessing chronic risk. Further, symptoms of BPD may also be indicative of chronic risk, as individuals with BPD often have histories of multiple suicide attempts (Linehan, 1993). Managing imminent suicide risk is beyond the scope of this article; therefore, we refer interested readers to Joiner et al., (1999) for a detailed review. However, we suggest that means-restriction is especially relevant for men in BIPs, as they may have greater access than men in the general population to weapons, medications, alcohol, and illicit substances that could be used to commit suicide (Campbell et al., 2007; Stuart et al., 2008). For a review of how to conduct means-restriction counseling, see Bryan, Stone, & Rudd (2011).

However, such extensive suicide risk assessment in BIP settings are likely difficult given these programs are limited in time, resources, and staff. Therefore, brief, cost-efficient screeners for suicide risk, such as the depression scale of the Patient Health Questionnaire (PHQ), could be used to determine whether individuals require more extensive assessment and, if necessary, referral to outpatient or emergency care (Joiner et al., 1999; Uebelacker, German, Gaudiano, & Miller, 2011). BIP providers who administer lethality measures, such as the Danger Assessment (Campbell et al., 2007), could use suicide threat or attempt history items as proxy indicators of chronic risk. Finally, to aid in the reduction of suicidal ideation, BIP providers may deliver adjunct treatments of BPD and depression symptoms, such as Dialectical Behavior Therapy (DBT; Fruzzetti & Levensky, 2000; Juodis et al., 2014; Linehan, 1993) or Interpersonal Therapy for depression (Klerman, Weissman, Rounsaville, & Chevron, 1984).

These findings may also be applicable to individual clinicians who treat couples or perpetrators of IPV in outpatient settings. For patients who present with depressive or BPD symptoms, clinicians should administer a comprehensive suicide risk assessment. Additionally, past research suggests that for individuals with comorbid depression and BPD diagnoses, the number and lethality of suicide attempts increase; therefore, clinicians should take into consideration comorbidity of these diagnoses (Soloff, Lynch, Kelly, Malone, & Mann, 2000). The literature also suggests that clinicians should consider legal and interpersonal difficulties that partner-violent men often experience, such as separation, divorce, or child custody evaluations, to be significant risk factors for suicide attempt (e.g., Yen et al., 2005). Finally, clinicians who deliver couples therapy should assess for the presence of IPV perpetration and potential co-occurring suicide risk, as over two-thirds of couples in treatment have been found to report IPV (Cascardi, Langhinrichsen, & Vivian, 1992). Such assessment is critical to protect the lives of both the perpetrators and their partners.