Violence against women and more specifically, intimate partner violence (IPV) is a national and international public health concern that affects women of all racial, ethnic, and socioeconomic backgrounds (Abramsky et al. 2011). African American women experience IPV more than other populations and this is most true for those from a lower socioeconomic background (Rennison and Planty 2003; Yoshihama et al. 2006). When compared to Caucasian women, African American women experience IPV at higher frequencies resulting in more serious and lethal injuries and more mental health problems (Ellsberg et al. 2008; Iverson et al. 2013; Taft et al. 2009). In the African American community, IPV is linked to posttraumatic stress, depressive symptoms, and suicidal behavior (Bradley et al. 2005; Devries et al. 2011, 2013; Leiner et al. 2008; McLaughlin et al. 2012). For example, in a study that examined 451 African American female patients who were approached in an emergency department waiting room, 6 % of survivors of IPV endorsed current suicidal ideation (Houry et al. 2006), and compared to their nonabused counterparts, African American women were 2.5–2.8 times more likely to attempt suicide (Kaslow et al. 2000). Fortunately, evidence-based and culturally-informed interventions with impoverished African American women have shown promise in decreasing depressive symptoms and general distress, in addition to less severe suicidal ideation when exposed to physical/sexual and emotional/psychological IPV (Kaslow et al. 2010; Zhang et al. 2013).

Despite the negative sequalae associated with IPV, many individuals are resilient in the face of this major interpersonal stressor (Nathanson et al. 2012). This has led to efforts to determine resilience factors that help to boost psychological well-being, and one of these candidate resilience factors is spirituality (Chapman and Steger 2010; Gullatte et al. 2010). Spirituality is characterized by a sense of relatedness or connectedness to others and the development of well-being as a safeguard (Hawks et al. 1995). It is a factor of well-being similar to positive self-esteem, assertiveness, and purposefulness (Koenig 2001a). There are two components of spiritual well-being: one’s purpose in life, which pertains to the existential well-being, and one’s connection with God, which refers to religious well-being (Ellison 1993).

Existential and religious well-being can lead to improved physical and mental health and strong social support (Holt-Lunstad et al. 2011; Stevens-Watkins et al. 2014). Moreover, existential and religious well-being are key factors associated with resilience in African Americans (Ellison et al. 2007; Gillum et al. 2006; Hirsch et al. 2014; Stevens-Watkins et al. 2014). Indeed, many abused women indicate that their spirituality and trust in a higher power comforts them and is integral to their healing (Gillum et al. 2006; Potter 2007; Stennis et al. 2015; Wang et al. 2009). A recent qualitative analysis revealed that spiritual coping strategies played a critical role in female IPV survivors’ self-efficacy (Drumm et al. 2013). Spiritual well-being, namely existential and religious well-being, instills a sense of purpose, empowerment, faith, self-awareness, and connection in abused women (Hampton and Gullotta 2006; Kaslow et al. 2010; Yick 2008). Furthermore, given the significant roles that existential and religious well-being play in the lives of African American women, these culture-specific strategies enhance these women’s capacity to reduce the negative impact that IPV may have on their physical and mental health (Farley 2007; Ward et al. 2009; Watlington and Murphy 2006). Attention paid to existential and religious well-being in stressful contexts related to African Americans has consistently shown to be beneficial (Lamis et al. 2014a; b; Stevens-Watkins et al. 2014; Utsey et al. 2005).

Several studies (e.g., Edmondson et al. 2008; Kleftaras and Katsogianni 2012; Lamis et al. 2014c; Utsey et al. 2007; Zhang et al. 2013) have examined existential and religious well-being as mediators of the relation between various forms of life stress and psychological well-being. One study showed that religious well-being was a more significant mediator in the relation between predictors (demographic variables, culturally-specific coping) and psychological health, rather than existential well-being (Utsey et al. 2007). However, in general, studies support a stronger relationship between existential well-being and psychological outcomes, rather than religious well-being and psychological outcomes. This includes quality of life, psychological adjustment, hopelessness, and despair (Edmondson et al. 2008; Kleftaras and Katsogianni 2012; Lamis et al. 2014b, c). Consistent with this, existential well-being, but not religious well-being, has been found to serve as a mediator of treatment outcomes for abused, suicidal African American women (Zhang et al. 2013).

The current study is the first to determine if existential well-being and religious well-being separately mediated the relation between IPV and one critically important psychological outcome, namely suicidal ideation (i.e., repeated thoughts of self harm). On the basis of previous research, we hypothesized that: 1) IPV would be negatively associated with existential and religious well-being and positively related to suicidal ideation; 2) existential well-being and religious well-being would be negatively associated with suicidal ideation, and 3) existential well-being and religious well-being would both significantly mediate the relation between IPV and suicidal ideation, however, existential well-being would serve a more powerful mediational role.

Method

Participants

The sample consisted of 111 African American women of low socioeconomic status between the ages of 18 and 56 (M = 37.05, SD = 10.31) who presented to a large public sector hospital for medical or psychiatric reasons. Most respondents were unemployed (86.0 %) and reported less than $500 monthly household income (66.0 %). In addition, 42 % of the women did not graduate high school or complete a General Education Diploma (GED) program and 54 % classified themselves as homeless (living on the streets, staying in a shelter, or temporarily staying with family/friends). All participants had attempted suicide in the year prior to presentation and also endorsed being in an abusive interpersonal relationship, either currently or within the past year. Of the women in our sample, approximately 98 91, and 70 % reported experiencing psychological aggression, physical assault, and sexual coercion, at the time of the assessment based on their responses on the Conflict Tactics Scales-2 (CTS2). Women were excluded from the study if they were determined to have significant intellectual or cognitive impairment, as assessed by a score of less than 24 out of 30 on the Mini-Mental State Exam (MMSE) (Folstein et al. 2001); if they were determined to be functionally illiterate by scoring less than 18 on the Rapid Estimate of Adult Literacy in Medicine (REALM) (Williams et al. 1995); or if they were unable to complete the assessment battery due to active psychotic symptoms. Appropriate support was provided to participants, as requested, including reading or clarifying statements listed among the assessment battery.

Measures

Demographic Data Form

This questionnaire, developed for use in previous studies, includes questions about demographics (e.g., age, education level, employment status, monthly income) and living situation (homeless vs. not homeless). All of these demographic factors have been found to be associated with suicidal ideation and attempts in other research (Eynan et al. 2002; Kaslow et al. 2005; Nock et al. 2008) and thus were included as covariates in all analyses.

Conflict Tactics Scales-2 (CTS-2)

IPV was assessed via the CTS-2 (Straus et al. 1996), which consists of 78 self-report items arranged in 39 item pairs, assessing positive and negative relationship behaviors that may occur in the context of relationship conflict. The CTS-2 includes five subscales designed to tap into abusive behaviors: psychological aggression, physical aggression, sexual coercion, positive conflict resolution strategies (negotiation), and outcomes associated with physical forms of abuse (injury). The paired items on the CTS-2 ask participants to report acts that they have committed towards a partner (perpetration) and acts committed by a partner towards them (victimization) during the past year. Given the high occurrence rate of involvement in abusive relationship behaviors, we used the sum of the total items on the five subscales (i.e., psychological aggression, physical aggression, sexual coercion, negotiation, and injury) that assessed IPV victimization to derive a continuous index, as recommended in the literature (Straus et al. 2003). Sample items on the CTS-2 include “my partner pushed or shoved me,” “my partner called me fat or ugly,” and “my partner used threats to make me have sex.” Each question is rated on a scale of 0 to 6 (has never happened, happened 1 time, 2 times, 3–5 times, 6–10 times, 11–20 times, more than 20 times). In this study, the raw responses were summed and averaged to reflect the frequency of IPV. The measure has good internal consistency reliability, test-retest reliability, and construct validity (Connelly et al. 2005; Straus et al. 1996; Vega and O’Leary 2007), in previous studies with African Americans (Kocot and Goodman 2003). The internal consistency reliability estimate for women’s reports of IPV victimization on the CTS-2 was 0.94.

Spiritual Well-Being Scale (SWBS)

The SWBS (Paloutzian and Ellison 1991) was used to determine the participants’ sense of spiritual well-being. It measures the perception that one’s life has meaning (existential well-being) and the affirmation of life in relationship with a god (religious well-being). This 20-item Likert scale has two subscales: existential well-being (EWB; 10 items) and religious well-being (RWB; 10 items); each item has six response options anchored by strongly disagree and strongly agree, with higher scores on each subscale indicating higher existential and religious well-being (Genia 2001; Murray et al. 2014). The EWB component assesses the existential notions of life purpose, life satisfaction, and positive or negative life experiences; whereas, the RWB subscale measures the degree to which one perceives and reports the well-being of his or her spiritual life in relation to God. A sample item on the EWB subscale is “I feel very fulfilled and satisfied with life,” and an example of an item on the RWB subscale is “I believe that God loves me and cares about me.” Previous studies have found evidence of good construct validity and internal consistency reliability with the coefficient alpha for existential well-being between 0.78 and 0.86 and for religious well-being between 0.82 and 0.94 (Bufford et al. 1991; Gow et al. 2011). In the current study, the internal consistency reliability was 0.83 and 0.87 for the EWB and RWB, respectively.

Beck Scale for Suicide Ideation (BSS)

The BSS (Beck and Steer 1991) is a 21-item self-report questionnaire measuring suicide ideation. The items provide participants three response options (e.g., “I have no wish to die,” “I have a weak wish to die,” or “I have a moderate to strong wish to die”) and are rated on a scale from zero to two, based on intensity. Scores are summed to provide a total score indicative of suicide risk (Brown 2000). The BSS demonstrates good internal consistency reliability across multiple samples (Beck and Steer 1991; Pinninti et al. 2002), including African Americans seeking care in an emergency department (α = 0.88) (Leiner et al. 2008). In addition, it evidences good convergent validity, with high correlations with the self-report and clinician rated measures of suicidal ideation (Beck et al. 1979; Healy et al. 2006; Leiner et al. 2008). In the current study, the internal consistency reliability estimate was 0.90.

Procedure

Participants were recruited and screened using two methods. First, African American women ages 18 to 64 who presented to the hospital’s medical or psychiatric emergency rooms after experiencing IPV or following a serious suicide attempt requiring medical attention and/or reporting significant suicidal intent were recruited and screened for study participation. Women were recruited and assessed for eligibility by a member of the research team (undergraduate or graduate student, predoctoral intern, postdoctoral fellow) once they were medically stable. The team member explained the study and answered any relevant questions. Second, women were also recruited by research assistants from other medical and psychiatric clinics in the hospital. Potential participants were administered questions regarding their experiences of IPV and suicidal behavior over the course of the preceding 12 months. They had to report IPV and a suicide attempt within the past year to qualify for the study. Women who did not meet study criteria were provided with information regarding various community resources and support groups as applicable.

Women deemed eligible for study participation were assessed immediately or scheduled for an assessment within a week of initial screening. Following the obtainment of informed consent, the assessments were administered verbally by a trained member of the research team and took 2 to 3 h to complete. Upon completion of the battery, which consisted of 29 measures, each participant received $20 and a roundtrip fare for use on the city transit system. If at any time during the interview a woman was identified as imminently suicidal, homicidal, severely depressed, or as having other acute psychiatric difficulties (e.g., psychotic symptoms), she was immediately referred for appropriate psychiatric intervention (e.g., evaluation, hospitalization, medication, psychotherapy).

Data Analytic Strategy

Path analysis was used to test the first two hypotheses. To test the predictive relations among study constructs, we examined these relations as paths, adjusting for sociodemographic covariates, which were modeled as exogenous predictors of the study variables. Results were obtained by fitting two separate saturated (i.e., just-identified) path analytic models with existential and religious well-being as mediating variables. Model fit indices are not presented within the current study due to the just-identified nature of the models. As recommended by MacKinnon and colleagues (MacKinnon 2008; MacKinnon et al. 2012), we examined existential well-being and religious well-being as potential mediators in separate models given the high correlation (r = 0.63) between the two variables.

The third and primary hypothesis, which focused on the mediation of the link from IPV to suicidal ideation by existential and religious well-being, also was tested using path analysis, with mediated paths and total effects being tested as the product of coefficients in separate saturated path models estimated in Mplus v.7.0 (Muthen and Muthen 1998-2012), using the software’s facility for maximum likelihood estimation in the context of missing data. The models were conventional three-variable mediation systems, as described in any standard treatment of indirect effects (MacKinnon 2008), with the addition of the suite of covariates. The null hypothesis was that the sum of the two indirect paths—from the predictor (IPV) to the mediator (existential well-being, religious well-being) and from the mediator to the outcome (suicidal ideation)—was equal to zero, indicating no indirect effect. We tested for the significance of indirect (mediated) effects using the percentile bootstrap with 3000 draws to generate empirical confidence intervals for the products of the coefficients composing the mediated paths, one of the methods recommended for specific indirect effects.

Results

Descriptive statistics and bivariate correlations among the primary study variables – IPV, existential well-being, religious well-being, and suicide ideation - are presented in Table 1.

Table 1 Correlation matrix, means, and standard deviations of study measures
Full size table

With the exception of the association between religious well-being and IPV, all of the bivariate correlations were significant in the expected direction, p < .05, and provide support at the bivariate level for the first two hypotheses (IPV would be negatively associated with existential well-being and religious well-being, and positively related to suicide ideation; existential well-being and religious well-being would be negatively associated with suicide ideation).

Hypotheses 1 and 2 concerned predictive relations among study constructs in the context of the mediational models, with the inclusion of covariates (age, education, employment, income, homelessness status) and standardized coefficients shown. As depicted in Fig. 1, in the first model and consistent with our hypotheses, the following path coefficients were significant: (1) IPV and suicidal ideation, b = 0.81, SE = 0.25, Est./SE = 3.24, p = .001; (2) IPV and existential well-being, b = −0.62, SE = 0.23, Est./SE = −2.68, p = .007; and (3) existential well-being and suicidal ideation, b = −.41, SE = 0.09, Est./SE = −4.49, p < .001.

Fig. 1
figure 1

Model with standardized regression coefficients examining existential well-being as a mediator in the relation between IPV and suicide ideation

Full size image

Similarly, as shown in Fig. 2, in the second model and as anticipated, the following path coefficients were significant: (1) IPV and suicidal ideation, b = 0.98, SE = 0.24, Est./SE = 4.11, p < .001; and (2) religious well-being and suicidal ideation, b = −.28, SE = 0.10, Est./SE = −2.79, p = .005. However, the path coefficient between IPV and religious well-being was not significant, b = −0.26, SE = 0.25, Est./SE = −1.06, p = .290.

Fig. 2
figure 2

Model with standardized regression coefficients examining religious well-being as a mediator in the relation between IPV and suicide ideation

Full size image

Mediational Analyses

In the first model examining existential well-being as a potential mediator of the relation between IPV and suicidal ideation in low income African American women, the total effect of IPV on suicide ideation was positive and significant, with a point estimate of 1.063 [95 % CI: 0.58–1.55], standardized estimate of 0.44. Consistent with the hypothesis, this effect was significantly mediated by existential well-being, ab = 0.25 [95 % CI: 0.06–0.51]. The confidence interval excluded zero, indicating a significant indirect effect of IPV on suicide ideation via existential well-being, supporting the mediation hypothesis.

In the second model examining religious well-being as a mediator of the relation between IPV and suicidal ideation, the total effect of IPV on suicide ideation was also positive and significant, with a point estimate of 1.056 [95 % CI: 0.57–1.55], standardized estimate of 0.44. However, contrary to hypothesis, this effect was not significantly mediated by religious well-being, ab = 0.08, [95 % CI: −0.11–0.25] as indicated by the confidence interval including zero, revealing no significant indirect (mediated) effect of religious well-being in the IPV-suicide ideation link.

Discussion

This is the first investigation to examine the relations among IPV, existential well-being, religious well-being, and suicidal ideation. The major finding was that existential well-being mediated the association between IPV and suicidal ideation, underscoring the powerful significance of having a sense of meaning in one’s life. When abused, African American women have a sense of purpose in their lives (i.e., existential well-being), it positively impacts their psychological well-being, as manifested by a reduction in their suicidal ideation.

Hypothesis 1 proposed that IPV would be negatively associated with existential well-being and religious well-being and positively related to suicidal ideation. This hypothesis was partially supported, as IPV was significantly associated with existential well-being and with suicidal ideation. These results are in keeping with previous findings from recent large-scale studies and reviews that have found a consistent association between IPV and suicidality in women around the world (Devries et al. 2011, 2013). That is, there is considerable evidence that IPV exposure is associated with higher levels of suicidal ideation and more suicide attempts. Moreover, such a link has been found repeatedly among low-income African American women (Kaslow et al. 2002; Leiner et al. 2008; Taft et al. 2009). However, the findings expand our knowledge base by demonstrating for the first time that there is a link between IPV and existential well-being, such that higher levels of IPV exposure were associated with lower levels of existential well-being. This finding suggests that when women are traumatized by violence in the context of an intimate relationship, they may feel hopeless and helpless (Thompson et al. 2002), and as a result have a more limited view of the value and significance of their lives. Contrary to what was hypothesized, IPV was not associated with religious well-being, suggesting that one’s belief in God may not be impacted negatively or positively by one’s exposure to IPV. This may be particularly true in a sample of low-income African American women, as a belief in God is a strong, consistent, and sacred cultural value (Banks-Wallace and Parks 2004).

Hypothesis 2, in which we predicted that existential well-being and religious well-being would be negatively associated with suicidal ideation, was fully supported, highlighting the relevance of these constructs to suicidality in the African American community. Moreover, these findings are in keeping with a burgeoning literature on the significant linkages between existential and religious well-being and suicidal behavior (Kaslow et al. 2004; Lamis et al. 2014a; Nad et al. 2008; Taliaferro et al. 2009), including decreased odds of a suicide attempt (Rasic et al. 2009), across demographic groups. However, it adds to our knowledge base by being the only empirical investigation to consider separately existential and religious well-being and their associations with suicidal behavior.

Hypothesis 3, the primary focus of the investigation, predicted that existential well-being and religious well-being would significantly mediate the IPV- suicide ideation link. While other studies have looked at overall spiritual well-being as a mediator of the IPV-suicidal behavior link (Meadows et al. 2005), this is the first time the two components of spiritual well-being were examined separately as mediators. We found that only existential well-being mediated this association. The salient role of existential, as opposed to religious well-being, vis-à-vis suicidal ideation has been found in other samples, such as college students (Taliaferro et al. 2009) and war veterans (Nad et al. 2008). The finding regarding religious well-being is inconsistent with prior literature that has shown that religious coping and well-being serves as a critical resilience factor against adverse outcomes following exposure to trauma (Bierman 2006; Ellison et al. 2008), that religious involvement plays a strong protective role against the negative impacts of IPV within the African American community (Ellison et al. 2007), and that spirituality and a view that God are a source of strength or comfort in the lives of IPV survivors (Gillum et al. 2006; Potter 2007; Stennis et al. 2015; Wang et al. 2009). Additionally, in light of the result that IPV was not associated with religious well-being in the bivariate analyses, it may not be surprising that religious well-being did not mediate the association between IPV exposure and suicidal ideation.

Limitations

This study has several limitations and associated future directions to consider. First, the ability to draw conclusions regarding the causal relationship among the variable of interest is restricted due to the cross-sectional research design. Thus, longitudinal designs are needed in the future to provide greater clarification with regard to the causal links among these variables. Second, the sample consisted entirely of low-income African American women, which limits the generalizability of these results to other populations. While the focus on one particular demographic group (e.g., gender, race/ethnicity, social class) enhances the cultural relevance of the work, research with samples that differ on these and other critical demographic factors will shed light on the cultural uniqueness versus the generalizability of the results. Third, the use of self-report measures may create social desirability bias, particularly as it relates to spiritual beliefs. Studies in the future, therefore, should incorporate data from multiple methods and multiple informants. Finally, there may be a number of additional variables that are critical in understanding these associations that were not examined. For example, social support may be a potential confound in this model. It is possible that those participants with higher levels of social support endorsed higher levels of existential and religious well-being due to more active community participation (Holt-Lunstad et al. 2011). Future examination of other models that consider multiple constructs may offer a more comprehensive and nuanced approach to interpretation.

Implications

The findings provide valuable information for assessing and treating low-income African American women who have survived IPV and attempted suicide. Such assessments should emphasize the importance of existential well-being as an effective coping strategy. In addition, for those African American women for whom spirituality is important, interventions should incorporate attention to bolstering spiritually-based approaches that are relevant for African Americans with histories of trauma and abuse (Ano and Vasconcelles 2005; Gillum et al. 2006; Koenig 2001b). Psychological interventions also may need to be done collaboratively across settings, as many IPV survivors prefer faith-based resources to more traditional mental health and shelter services (Fowler et al. 2011).

In addition, the findings of the powerful role of one particular facet of spiritual well-being, namely existential well-being, underscore the importance of conducting interventions designed to enhance abused, suicidal women’s sense of meaning in their lives. Narrative therapy is one approach that can serve as an invaluable model for assisting women exposed to IPV in this process (White 2007; White and Epston 1990). Through sharing their narratives, women’s personal experiences can be transformed in a way that enhances a sense of meaning in their lives and helps shape their identities. When hearing these women’s stories, it behooves the therapist to empower the women by highlighting their strengths, capacity for independent functioning, and emotional resilience in the face of oppression and suffering (Draucker 1998; Gullatte et al. 2010). Having a greater sense of purpose in one’s life is likely to be associated with a reduction in suicidal behavior, even in the face of adversity such as IPV (Heisel and Flett 2004). For the women in this sample, it is essential that such narrative approaches are conducted in a fashion that is attuned to the intersectionalities of gender, race, and class (Sokoloff and Dupont 2005). To this end, often times, religion/spirituality is one culturally relevant way to assist these women in constructing meaning in times of adversity (Mattis 2002). This may entail encouraging these women to engage actively in their religious and spiritual communities, which can serve as powerful support systems for at-risk African American survivors of trauma (Chatters et al. 2011; Mattis et al. 2013).