Abstract
Resilience has been found to attenuate the effects of negative mental health symptomology associated with interpersonal victimization; however, existing research has largely focused on resilience traits, such as individual cognitive and environmental factors that promote resilience. In addition, empirical knowledge on the extent to which resilience mitigates suicidal symptomology associated with interpersonal violence victimization is particularly limited. This study assesses whether the relationship between interpersonal violence (i.e., IPV and nonpartner sexual violence) and mental health symptomology (i.e., depression, psychological distress, and suicidal ideation) is moderated by resilience using a general population sample of women (N = 932). A cross-sectional, observational survey was administered in four U.S. cities (Baltimore, New York City, Philadelphia, and Washington, D.C.). Bivariate results indicated that women exposed to interpersonal violence reported significantly higher rates of suicidal ideation, depression, and psychological distress compared with women without exposure to interpersonal violence. Regression models revealed significant positive associations between interpersonal violence and depression, distress, and suicidal ideation, adjusting for sociodemographics. Resilience did not significantly moderate the relationship between interpersonal violence victimization and any associated mental health outcomes. However, subgroup analyses reveal significant interaction effects between resilience and IPV within specific racial and ethnic minority subgroups, suggesting that attenuating effects of resilience on mental health symptoms (i.e., depression and psychological distress) associated with IPV likely vary across race and ethnicity. Implications for future research and clinical interventions focused on resilience among survivors of interpersonal violence are discussed.
Interpersonal violence (i.e., IPV [IPV], sexual violence) pervasively affects women in the United States. Nearly one in five (18.3%) women has been raped in her lifetime and more than a third (35.6%) of women have experienced physical violence, rape, and/or stalking by an intimate partner, with disproportionately higher rates of interpersonal violence documented among racial and ethnic minority women (Black et al., 2011). Women who have experienced IPV and sexual violence are at greater risk for a number of adverse mental health outcomes, including psychological distress, depression, and suicidal behavior (Campbell, Dworkin, & Cabral, 2009; Coker et al., 2002; Dillon, Hussain, Loxton, & Rahman, 2013; Kilpatrick, Resnick, Ruggiero, Conoscenti, & McCauley, 2007). Women exposed to IPV have 3 to 5 times greater odds of depression compared with women who have not experienced IPV (Golding, 1999). Similarly, women who have been raped have 3.5 times the odds of experiencing current major depressive disorder compared with women who have not experienced rape in general population samples (Kilpatrick, Edmunds, & Seymour, 1992). High rates of suicidal behavior have also been documented among female victims of IPV (Alhusen, Frohman, & Purcell, 2015; Devries et al., 2013; Weaver et al., 2007) and sexual violence (Kilpatrick et al., 1992; Ullman & Najdowski, 2009). In a systematic review of mental health correlates of IPV, women with exposure to IPV have 3 to 5 times the odds of having suicidal behaviors than women without exposure to IPV (Golding, 1999). Similarly, women with histories of sexual violence are over 4 times as likely to contemplate suicide than women who have not experienced sexual violence (Kilpatrick et al., 1992).
Scholars have recently called for a strengths-based approach to research on interpersonal violence and a greater focus on studying survivors’ resilience and ability to thrive despite significant adversity (Hamby, Banyard, & Grych, 2016). Existing research has suggested that women exposed to interpersonal violence typically have lower levels of resilience compared with women without interpersonal violence exposure, due to such factors as the damaging effects of increased cortisol in the brain (Anda et al., 2006) as well as the isolating impact of violence on victims’ social support networks (Coker, Watkins, Smith, & Brandt, 2003). Therefore, assessing the role of resilience in mental health symptomology among survivors of interpersonal violence is needed to better understand women’s varying responses to victimization, including the extent to which a resilient state mitigates adverse mental health outcomes.
Definitions of Resilience
Resilience in the medical and behavioral sciences has been defined in a variety of ways, including the ability to “bounce back” from or adapt to stressful events, functioning above the norm despite adversity, and resistance to illness despite encounters with stressful or adverse events (Carver, 1998; Smith et al., 2008; Tusaie & Dyer, 2004). Resilience has also been increasingly conceptualized as a modifiable state influenced by “person-environment interaction” and supportive circumstances, such as social support and family bonds (Chan, 2006; Luthar, Cicchetti, & Backer, 2000). Research on interpersonal violence has assessed resilience as a modifiable state (e.g., Jose & Novaco, 2016; Steenkamp, Dickstein, Salters-Pedneault, Hofmann, & Litz, 2012) as well as individual trait factors known to promote resilience, including social support, coping skills, perceived stress, and other cognitive and environmental factors (e.g., cognitive functioning, self-efficacy, sense of mastery, community resources; for example, Reviere et al., 2007; Rodriguez et al., 2008). Furthermore, definitions of resilience, as well as individual expressions of resilience, vary across cultural groups; that is, resilience is influenced by racial, ethnic, and cultural identity and cultural beliefs, family, and community (i.e., racial, ethnic, and/or religious communities) shape individual resilience (Tummala-Narra, 2007). Thus, the effects of resilience on negative mental health sequela associated with interpersonal violence may vary across racial and ethnic minority groups; however, this has not yet been fully examined in existing research on resilience and mental health symptomology associated with interpersonal violence. The present study conceptualizes resilience as a modifiable state, where various cognitive, environmental, and cultural factors contribute to and increase a person’s resilience and resistance to illness despite encounters with stressful or adverse events (Bonanno & Mancini, 2008; Tummala-Narra, 2007).
Effects of Resilience as a Modifiable Status on Mental Health
Among studies assessing resilience as a modifiable status, findings suggest that resilience largely mitigates the negative effects of depression, anxiety, and posttraumatic stress disorder (PTSD) symptomology, particularly among survivors of IPV (Anderson, Renner, & Danis, 2012; Ford-Gilboe et al., 2009; Humphreys, 2003; Jose & Novaco, 2016; Rodriguez et al., 2008; Schultz, Roditti, & Gillette, 2009). Although existing research, as described below, has suggested that resilience has an attenuating effect on depression, anxiety, and PTSD symptoms among women exposed to IPV and sexual violence, previous studies have relied mostly on small convenience samples and service-seeking populations of women experiencing IPV, which limits the generalizability of this knowledge to include other populations of victims.
Jose and Novaco (2016) found that resilience was inversely associated with depression and anxiety symptoms in a sample of female IPV victims seeking restraining orders. Similarly, findings from Humphreys (2003) suggest that higher resilience scores were significantly associated with lower levels of distress symptoms (i.e., depression, anxiety, repeated thoughts, impulses) in a sample of women residing in a battered women’s shelter. Few studies have compared resilience outcomes between women exposed and not exposed to interpersonal violence, which is needed to understand the nature of resilience, and its effect on mental health, among survivors of interpersonal violence compared with these effects of resilience on mental health symptomology in the general population. One study assessed resilience in a clinical sample of pregnant Latina women (Rodriguez et al., 2008) and found significant differences in resilience and mental health symptomology among women exposed to IPV compared with women not exposed to IPV. Specifically, Rodriguez et al. (2008) tested interaction effects between resilience and IPV and found that resilience significantly moderated the effects of IPV on depression among Latina women. Apart from Rodriguez et al.’s (2008) study, scant research is available on interaction effects between resilience and interpersonal violence within specific racial and ethnic groups, and thus, the extent to which resilience varies across racial and ethnic minority women exposed and not exposed to interpersonal violence is unclear.
Furthermore, compared with IPV, the role of resilience in mental health symptomology associated with sexual violence is understudied and not clearly understood. Steenkamp et al. (2012) examined latent class trajectories of PTSD in a convenience sample of sexual violence survivors and did not observe resilience trajectories in this population; however, Bonanno (2013) suggested that high levels of resilience is similarly present in populations with equivalent PTSD symptomology exposed to other forms of trauma. To our knowledge, no studies have assessed the role of resilience in attenuating depression, psychological distress, and suicidal ideation in a general population sample of women.
Effects of Social Support and Social Network on Resilience and Mental Health
Extensive research on violence has assessed individual factors that promote resilience and reduce associated depression. A particularly salient factor known to promote resilience is social support, including social networks (Bonanno & Mancini, 2008; Mancini & Bonanno, 2009). In particular, the size of a person’s social network has been linked with higher levels of resilience among individuals exposed to trauma. Specifically, larger social networks are associated with increased resilience (Mancini & Bonanno, 2009). Existing research on interpersonal violence suggests that greater levels of social support, measured as the size of a social network, the frequency of contact in a social network, and the quality of emotional support offered through social networks, are associated with decreased mental health symptomology among women exposed to IPV and sexual violence. Specifically, women exposed to IPV with higher levels of social support are less likely to experience depression, PTSD, anxiety, and suicidal ideation compared with women with lower levels of social support (Coker et al., 2003; Mburia-Mwalili, Clements-Nolle, Lee, Shadley, & Yang, 2010; Reviere et al., 2007; Rodriguez et al., 2008). Similarly, sexual violence survivors with greater amounts of social support reportedly have lower levels of PTSD and depressive symptomology (Bryant-Davis, Ullman, Tsong, & Gobin, 2011).
Although limited, some research has examined the interrelationship between social support networks, resilience, and mental health among female survivors of interpersonal violence to better understand how internal resources (i.e., resilience) interface with external resources (i.e., social support network) in mitigating mental health symptoms associated with victimization (Ford-Gilboe et al., 2009; Jose & Novaco, 2016). Findings from these studies suggest that higher levels of social support are associated with greater levels of resilience among women experiencing IPV due to feelings of connectedness in social networks (Davis, 2002; Schultz et al., 2009). In addition, Jose and Novaco (2016) found that resilience was significantly associated with decreased depression and anxiety among victims of IPV, after controlling for social support, including the size of social networks. Social support was no longer significantly predictive of mood symptoms once assessing resilience in the model (Jose & Novaco, 2016), suggesting the need for interventions focused on fostering a state of resilience among survivors beyond solely promoting social support and increasing social networks (Ford-Gilboe et al., 2009; Jose & Novaco, 2016).
Few studies have examined the role of resilience in mental health outcomes associated with sexual violence, and prior studies of both intimate partner and sexual violence have focused largely on PTSD and depression outcomes. Consequently, scant research is available on the role of resilience in suicidal behavior associated with interpersonal violence, and previous studies have focused on individual trait factors (e.g., coping, social support) in reducing suicidal symptomology (e.g., Coker, Weston, Creson, Justice, & Blakeney, 2005; Meadows, Kaslow, Thompson, & Jurkovic, 2005). To our knowledge, no studies have examined the role of resilience in mitigating the effects of suicidal ideation among women exposed to IPV and sexual violence, which is needed to inform treatment and interventions with survivors at risk for suicide.
Present Study
The purpose of this study is to assess the relationship between interpersonal violence (i.e., IPV and nonpartner sexual violence), resilience, and associated adverse mental health outcomes (i.e., depression, psychological distress, and suicidal ideation). The aims of this study are to (a) determine whether increased resilience leads to decreased levels of depression, psychological distress, and suicidal ideation among women exposed to interpersonal violence and (b) determine whether the relationship between interpersonal violence, depression, psychological distress, and suicidal ideation is moderated by resilience. Given the lack of prior research on interpersonal violence examining differences in resilience among racial and ethnic minority populations, this study also explores interaction effects between resilience and interpersonal violence exposure on mental health symptomology within racial and ethnic minority subgroups. Social network, measured as the number of individuals one can readily rely on in times of difficulty, was included in the analysis based on previous literature linking social network to both social support and resilience (Bonanno & Mancini, 2008; Mancini & Bonanno, 2009). This study builds on prior research examining the effects of resilience on depression outcomes of IPV by using a general population sample and, thus, strengthens the generalizability of this knowledge and extends these findings to include nonpartner sexual violence, in addition to IPV. In addition, to our knowledge, this is the first study to assess the effects of resilience specifically on suicidal ideation associated with IPV and nonpartner sexual violence.
Based on existing research linking adverse mental health outcomes associated with interpersonal violence, we hypothesize that women with IPV and nonpartner sexual violence exposures would demonstrate higher levels of depression, psychological distress, and suicidal ideation than women without histories of interpersonal violence. Furthermore, we hypothesize that resilience would attenuate depression, psychological distress, and suicidal ideation among women exposed to interpersonal violence based on prior studies using clinical and service-seeking samples (e.g., Humphreys, 2003; Jose & Novaco, 2016). However, we do not advance hypotheses regarding moderating effects of resilience, due to a lack of prior research testing resilience as a moderator of IPV, nonpartner sexual violence, and mental health symptomology in general population samples.
Method
Study Procedures
A cross-sectional, observational survey was administered in four U.S. cities (Baltimore, New York City, Philadelphia, and Washington, D.C.) during March and April 2016. Sampling procedures were administered by Qualtrics Panels, which provides demographically representative samples (±10% of 2010 census distributions for age, sex, and race/ethnicity in each city). Participants received a maximum compensation of US$10 by Qualtrics for completing the survey. English-speaking individuals 18 years of age or older living within the geographical boundaries of each of the four cities were eligible to participate. A total of 3,518 respondents were screened to determine eligibility and 1,153 were excluded for reasons of living outside of the geographic boundaries (n = 1,112) and being below the age of 18 (n = 41). Additional respondents were excluded for reasons of discontinuing the survey (n = 417), incorrectly responding to attention checks throughout the survey (n = 322), and responding too quickly (n = 1), resulting in a final sample size of N = 1,615 (68.6% of eligible respondents). Female participants were then selected from the sample for the purposes of the current study (N = 932). Additional information on design and sampling procedures can be found in DeVylder et al. (2017). Institutional review board (IRB) approval was obtained for this study.
Measures
Depression
The Patient Health Questionnaire (PHQ-9) was used to assess depression in the past 2 weeks. The PHQ-9 uses 4-point Likert-type response option (not at all = 1, several days = 2, more than half the days = 3, nearly every day = 4), which were summed (ranging from 9 to 36, where higher scores indicate greater levels of depressive symptomology). The PHQ-9 is widely used and demonstrates strong reliability across a variety of populations (Kroenke & Spitzer, 2002), including the current study sample (Cronbach’s alpha = .91).
Psychological distress
Psychological distress was measured using the K-6 scale, which assesses clinically significant psychological distress symptoms in the past 4 weeks (Kessler et al., 2003). The K-6 scale uses 5-point Likert-type response option (all of the time = 1, most of the time = 2, some of the time = 3, a little of the time = 4, none of the time = 5), which were reversed coded and summed (ranging from 0 to 24, where higher scores indicate greater levels of distress). The K-6 demonstrated strong reliability in the current sample (Cronbach’s alpha = .91).
Suicidal ideation
Past-year suicidal ideation was measured with the indicator “In the past 12 months, have you ever seriously thought about committing suicide?” which is consistent with measures of suicidal ideation from the National Survey on Drug Use and Health (Miller et al., 2015). Response options included yes, no, or unsure, and were recoded into a dichotomous variable (yes or no). Responses of “unsure” were recoded as “yes” to take into account under-reporting of suicidal ideation (Klonsky, May, & Saffer, 2016; Prinstein, 2008).
Sociodemographic characteristics
Age, race/ethnicity (Black or African American, Hispanic or Latino, and Other—that is, Asian American, American Indian, or Native Hawaiian), sexual orientation (heterosexual or lesbian, gay, or bisexual [LGB]), annual household income (in US$20,000 increments, up to US$100,000+), and education level (<high school, high school diploma, or General Education Development (GED), some college or technical school, college graduate, graduate or professional degree) were included as covariates in the analysis.
Lifetime interpersonal violence
Measures of IPV and nonpartner sexual violence were assessed through four dichotomous questions modified from the National Intimate Partner and Sexual Violence Survey (NISVS; National Center for Injury Prevention and Control, 2010), which are consistent with the Centers for Disease Control and Prevention (CDC) and World Health Organization (WHO) definitions of IPV and nonpartner sexual violence (Black et al., 2011; Krug, Mercy, Dahlberg, & Zwi, 2002). Intimate partner physical violence was measured with two items (“Has a romantic or sexual partner ever made threats to physically harm you?” and “Has a romantic or sexual partner ever shot at, stabbed, struck, kicked, beaten, punched, slapped, or otherwise physically harmed you?”). Intimate partner sexual violence was measured with the following question assessing forcible or coerced penetration: “Has a romantic or sexual partner ever forced or pressured you to engage in unwanted sexual activity that you did not want to do? Unwanted sexual activity includes vaginal, oral, or anal intercourse or inserting an object or fingers into your anus or vagina.” Respondents who endorsed any intimate partner physical and/or intimate partner sexual violence were coded as having experienced IPV.
Nonpartner sexual violence was included in the analyses as a separate variable using a single indicator of forcible or coerced penetration: “Has anyone else, other than a romantic or sexual partner (family member, acquaintance, or stranger) ever forced or pressured you to engage in unwanted sexual activity that you did not want to do? Unwanted sexual activity includes vaginal, oral, or anal intercourse or inserting an object or fingers into your anus or vagina.”
Resilience
The Brief Resilience Scale (BRS) consists of six items using a 5-point Likert-type scale, which were averaged to obtain a mean score (ranging from 1 = strongly disagree to 5 = strongly agree). Example items include “I tend to bounce back quickly after hard times” and “I usually come through difficult times with little trouble.” The BRS has demonstrated good internal consistency and test–retest reliability in college student and health care patient samples (Cronbach’s alpha ranging from .80 to .91; intraclass correlation coefficient [ICC] = .61 and .69 in two samples; Smith et al., 2008). The BRS demonstrated good reliability in the current study sample (Cronbach’s alpha = .85). Notably, the BRS is the only measure of resilience in its most basic meaning: to bounce back or recover from stress; whereas other measures of resilience often assess personal characteristics that promote resilience (e.g., optimism), rather than resilience itself (Smith et al., 2008).
Social network
Social network was measured with the following question: “How many people do you have near you that you can readily count on for help in times of difficulty such as to watch over children or pets, give rides to the hospital or store, or help when you are sick?” Response options provide five categories (1 = 0, 2 = 1, 3 = 2 to 5, 4 = 6 to 10, 5 = more than 10). Social network was included in the analysis based on prior research suggesting that the BRS is uniquely related to health outcomes, after controlling for individual resilience resources, including various aspects of social support (Smith et al., 2008).
Data Analysis
To address the first research aim, bivariate tests (Pearson correlations and independent-samples t tests) were conducted to examine the relationship between resilience and mental health outcomes among women exposed to interpersonal violence. To address the second research aim, a series of ordinary least squares (OLS; for depression and psychological distress outcomes) and logistic regressions (for suicidal ideation outcomes) with hierarchical variable entry were conducted to examine moderating effects of resilience for IPV and nonpartner sexual violence over five steps: (a) covariates (race—Black, race—Other, ethnicity—Latina, sexual orientation, income, and education), (b) IPV and nonpartner sexual violence, (c) social network, (d) resilience, and (e) interaction term (Resilience × IPV and Resilience × Nonpartner Sexual Violence). A growing body of literature recommends that social science research should conduct within-group analyses, rather than comparative analyses between minority and White participants, as this approach undermines the experience of minorities and assumes that experiences and outcomes among White populations are the norm (Awad, Patall, Rackley, & Reilly, 2016; Mizok & Harkins, 2012). Therefore, exploratory analyses were conducted using OLS and logistic regression testing separate models of racial and ethnic subsamples (i.e., African American and Latina women) to assess whether findings varied or generalized across racial and ethnic minority groups. All analyses were conducted in SPSS Version 22.
Missing data on all study variables were less than 5%, with the exception of resilience which had 24.9% missing (n = 232). Missing data were analyzed and demographic data (i.e., race, ethnicity, age, sexual orientation, education, and income) were compared between participants with complete (n = 700) and missing data (n = 232) on the resilience variable. Demographics between the two groups differed by race, age, and income. Compared with completers, participants with missing data were significantly more likely to be African American (33.2%, compared with 20.9%), to be older (M age = 40.60, compared with 37.96), and to have lower household incomes (M income = 3.34, compared with 3.72). Cases with missing data were handled using listwise deletion.
Results
The median age of survey participants was 35.5 (range = 18-84). Participants were predominately White non-Hispanic (55.5%), followed by Black or African American (32.6%), Hispanic or Latina (10.9%), and Other (5.6%) and were generally representative of the four cities surveyed (DeVylder et al., 2017). Approximately, 26% (n = 239) of participants reported intimate partner physical and/or sexual violence and nearly 14% (n = 127) reported nonpartner sexual violence, which is consistent with national estimates (Black et al., 2011). Sample characteristics are presented in Table 1. Bivariate results suggested statistically significant differences in depression, psychological distress, and suicidal ideation between women exposed and not exposed to IPV and nonpartner sexual violence . Specifically, mean scores on the PHQ-9 were significantly higher for IPV victims (t = –7.12, M = 7.95, SD = 6.84) and nonpartner sexual violence victims (t = –7.64, M = 9.34, SD = 7.21) compared with women without IPV (M = 4.79, SD = 5.42) and nonpartner sexual violence exposures (M = 5.03, SD = 5.55). Similarly, mean scores on the K-6 were significantly higher for IPV victims (t = –7.27, M = 7.25, SD = 5.99) and nonpartner sexual violence victims (t = –6.81, M = 8.12, SD = 6.11) compared with women without IPV (M = 4.37, SD = 5.01) and nonpartner sexual violence exposures (M = 4.66, SD = 5.18). A significantly higher proportion of interpersonal violence victims reported past-year suicidal ideation compared with women without violence exposure; that is, 23% of IPV victims and 23.6% of nonpartner sexual violence victims reported past-year suicidal ideation, compared with 8.5% of women not exposed to IPV (χ2 = 34.59, p < .001) and 10.6% of women not exposed to nonpartner sexual violence (χ2 = 17.25, p < .001). Mean scores of resilience and social network also significantly differed. Mean resilience scores were significantly lower for women exposed to IPV (t = 3.31, M = 11.15, SD = 4.44) and nonpartner sexual violence (t = 2.80, M = 10.95, SD = 4.56) compared with women without IPV (M = 12.30, SD = 3.91) and nonpartner sexual violence exposure (M = 12.17, SD = 3.98). Mean social network scores were significantly lower for women exposed to IPV (t = 2.00, M = 2.70, SD = 0.79) and nonpartner sexual violence (t = 3.13, M = 2.58, SD = 0.89) compared with women without IPV (M = 2.83, SD = 0.83) and nonpartner sexual violence exposures (M = 2.83, SD = 0.81) (see supplemental material for bivariate correlations between study variables).
Sample Sociodemographic Characteristics (N = 932).
Note. Some totals do not add to 100% due to missing data <5%.
Regression results for the final model are presented in Tables 2 (IPV) and 3 (nonpartner sexual violence) by mental health outcome. Regression results revealed significant associations between sociodemographics (i.e., age, LGB, education) and most mental health outcomes. Specifically, in both models of IPV and nonpartner sexual violence predicting mental health symptomology, LGB status was significantly associated with increased levels of depression, psychological distress, and increased odds of suicidal ideation, compared with non-LGB women. In addition, decreases in age were significantly associated with increased levels of depression and psychological distress, and increased odds of suicidal ideation in both models of IPV and nonpartner sexual violence predicting mental health symptomology. Lower levels of education were associated with increased levels of depression in both models (see Tables 2 and 3 for complete results).
Final Model for Resilience and IPV Predicting Mental Health Symptoms.
Note. CI = confidence interval; OR = odds ratio; LGB = lesbian, gay, or bisexual; IPV = IPV.
p < .05. **p < .01. ***p < .001.
Final Model for Resilience and Nonpartner Sexual Violence Predicting Mental Health Symptoms.
Note. CI = confidence interval; OR = odds ratio; LGB = lesbian, gay, or bisexual; SV = nonpartner sexual violence.
p < .05. **p < .01. ***p < .001.
Regression analyses indicated significant associations between IPV and nonpartner sexual violence and mental health symptomology. IPV exposure was significantly associated with increased levels of depression (B = 2.66, p < .001), psychological distress (B = 2.41, p < .001), and increased odds of suicidal ideation (odds ratio [OR] = 2.91, p < .001), after adjusting for sociodemographics. Similarly, nonpartner sexual violence was significantly associated with increased levels of depression (B = 3.06, p < .001) and psychological distress (B = 2.07, p < .001), after controlling for sociodemographic variables. Nonpartner sexual violence was significantly associated with increased odds of suicidal ideation in each step of the model, however, was no longer significant in the final step after adding resilience to the model (OR = 1.48, p = .16).
Resilience was added in Step 4 of each model, revealing significant associations between increased resilience and decreased depression and psychological distress and decreased odds of suicidal ideation. Interaction terms between IPV and resilience and nonpartner sexual violence and resilience were added in Step 5 of each model; however, no significant interactions were found in the model of IPV and nonpartner sexual violence, respectively, predicting depression (B = –.15, p = .18; B = .09, p = .51), psychological distress (B = .02, p = .83; B = –.01, p = .94), or suicidal ideation (B = .02, p = .73; B = .02, p = .78).
Exploratory analyses examining interaction effects across racial and ethnic minority groups suggest varying results in mental health outcomes by racial and ethnic minority subgroup. Regression results suggest a significant interaction between resilience and IPV on depression among Latina women (B = –.82, p < .05) and a significant interaction between resilience and IPV on psychological distress among African American women (B = .41, p < .05). An inverse relationship between depression and resilience was found in all Latina women and this association was stronger for women with IPV exposures than for women without IPV exposures. Similarly, a positive relationship between psychological distress and resilience was found in African American women and this association was stronger for women with IPV exposures than for women without IPV exposures. Figure 1 graphs significant interactions for (A) Latina women and (B) African American women and related mental health symptomology. To construct these figures, we manually calculated the predicted values of depression for Latina women with and without IPV histories and the predicted values of psychological distress for African American women with and without IPV histories at three values of resilience: at the mean, and one standard deviation above (high resilience) and below (low resilience) the mean. Other continuous predictors (age, education, income, and social network) were entered at their mean values, and the categorical predictor, LGB, was entered as 0 (non-LGB).

Significant interaction (A) between IPV and resilience associated with depression among Latina women and (B) between IPV and resilience associated with psychological distress among African American women.
Finally, given the lack of prior research on resilience and suicidal ideation associated with interpersonal violence and an increasing evidence linking social support to decreased suicidal behaviors among women exposed to interpersonal violence (Coker et al., 2005; Meadows et al., 2005), we tested whether social network moderated the relationship between IPV and nonpartner sexual violence and suicidal ideation; however, no significant interactions were found (B = .36, p = .23; B = –.36, p = .33, respectively).
Discussion
The purpose of this study was to investigate the role of resilience in the relationship between interpersonal violence (i.e., IPV and nonpartner sexual violence), depression, psychological distress, and suicidal ideation in a general population sample of women. Consistent with prior studies, we found disproportionately higher rates of depression, psychological distress, and suicidal ideation among women with histories of IPV (Campbell et al., 2009; Devries et al., 2013; Dillon et al., 2013) and sexual violence (Kilpatrick et al., 2007) compared with women without exposure to interpersonal violence. Previous studies have also linked higher levels of resilience to lower levels of depressive and psychological distress symptoms in samples of IPV victims (Coker et al., 2002; Jose & Novaco, 2016), suggesting that resilience attenuates depressive and distress symptomology among IPV survivors; however, existing research has not fully explored this relationship among survivors of nonpartner sexual violence, including potential attenuating effects of resilience specifically on suicidal ideation outcomes among women exposed to interpersonal violence. Overall, findings suggest an attenuating effect of resilience on associated depression, psychological distress, and suicidal ideation among women in the study’s general population sample, which is consistent with prior studies documenting significant associations between increased resilience and decreased depression and distress symptoms in clinical and service-seeking samples of interpersonal violence victims (Anderson et al., 2012; Ford-Gilboe et al., 2009; Humphreys, 2003; Jose & Novaco, 2016).
Resilience and Suicidal Ideation
Suicidal ideation is largely understudied in research on resilience and interpersonal violence. To our knowledge, this is the first study to explore moderation effects of resilience in attenuating suicidal ideation associated with IPV and nonpartner sexual violence. We did not find significant interaction effects of resilience on suicidal ideation associated with IPV and nonpartner sexual violence, including in the analyses by racial and ethnic subgroup. Although research is limited, findings may be attributed to other factors known to reduce suicidal symptomology, such as social support and feelings of connectedness (Coker et al., 2005; Meadows et al., 2005). Based on prior studies linking social support to reduced odds of suicidal ideation (Coker et al., 2005; Meadows et al., 2005), we also examined moderation effects of social network in the relationship between IPV and nonpartner sexual violence and suicidal ideation; however, no significant interaction was found. The absence of this effect may be due to the measure used for social network in our study, which consisted of a single indicator assessing the network or count of people one can rely on for help and in times of need, whereas other measures of social support consider additional factors related to perceived supports from friends, family, and community (Coker et al., 2005; Meadows et al., 2005). Future studies on suicidal ideation associated with interpersonal violence should explore individual traits (e.g., social support, feelings of connectedness) that may be associated with resilience as a modifiable status.
Moderating Effects of Resilience by Race and Ethnicity
Although results suggest an overall attenuating effect of resilience on negative mental health sequelae, findings did not suggest statistically significant interaction effects between resilience and IPV or resilience and nonpartner sexual violence and any mental health symptomology. There is a dearth of research on the effects of resilience on negative mental health sequelae using general population samples that include women both exposed and not exposed to violence, which would allow for needed comparisons in outcomes. At least one study tested moderation effects of resilience on mental health outcomes associated with interpersonal violence in a sample of women exposed and not exposed to IPV and found a significant interaction between IPV and resilience among Latina women (Rodriguez et al., 2008). Given these prior findings, we conducted exploratory analyses to test whether interaction effects varied by race and ethnicity. Consistent with Rodriguez et al. (2008), we found that resilience moderated the effects of IPV on depression for Latina women in the sample. We also conducted exploratory analyses to test interactions among African American women in the sample and found that resilience significantly moderated the effect of IPV on psychological distress for African American women. The significant interaction effects among Latina and African American women may indicate that certain contextual and cultural factors (e.g., resource access and availability, spirituality, coping strategies) influence resilience among survivors of interpersonal violence (Bradley, Schwartz, & Kaslow, 2005; Bryant-Davis et al., 2011; González-Guarda, Peragallo, Vasquez, Urrutia, & Mitrani, 2009). Future research is needed to better understand the factors that promote or hinder the development of resilience among racial and ethnic minority women to inform culturally relevant interventions with survivors.
In understanding resilience as a modifiable status shaped by person–environment interaction, scholars have recently called for research focused on resilience to address its ecological nature, that is, addressing both individual-level variables and systems-level variables that promote resilience (Shaw, McLean, Taylor, Swartout, & Querna, 2016). This approach is particularly relevant to populations who have experienced IPV and sexual violence, as many victims of interpersonal violence interact with and seek assistance from health care, criminal justice, and social service systems. Thus, the person–environment interaction that modifies and cultivates resilience likely includes these various systems-interactions, particularly for survivors of interpersonal violence. Although our study was limited to an exploration of individual-level factors, future research is needed on how systems-level variables (e.g., local systems procedures, state legislation and policy related to IPV and sexual violence), as well as survivors’ interactions with systems, can promote or inhibit the development of resilience to act as a protective measure against negative mental health sequelae.
Limitations
Several limitations in this study should be noted. Survey data are cross-sectional and causal connections between study variables cannot be established. Data were collected in four Northeastern U.S. cities, and thus, findings may not be generalizable to other cities in the United States. In addition, there were significant demographic differences between participants with missing and complete data and the largest amount of missing data were on the main study variable (i.e., resilience), which can potentially result in biased estimates. Despite these limitations, this study addresses important gaps in research on the role of resilience in attenuating negative mental health symptoms, and in particular, suicidal ideation, associated with both IPV and sexual violence. The study’s measure of social network consisted of a single item identifying the number of people one can rely on in times of need, whereas other social support measures assess additional factors such as perceived resources and perceived support (Bryant-Davis et al., 2011; Ford-Gilboe et al., 2009; Jose & Novaco, 2016). Still, our findings are consistent with Jose and Novaco (2016), who found that social support was no longer significantly associated with mood symptoms after including resilience. Finally, this study assessed intimate partner physical and sexual violence and not additional forms of IPV, such as psychological violence, which also adversely impacts mental health outcomes (e.g., Coker et al., 2002). Similarly, the study’s measure of nonpartner sexual violence included forcible or coerced penetration and did not assess other forms of unwanted sexual contact (e.g., unwanted kissing or touching), which has also been linked to poor mental health outcomes. Despite these limitations, this study is consistent with prior research linking IPV and sexual violence to negative mental health sequelae among women (e.g., Coker et al., 2002; Kilpatrick et al., 2007) and expands empirical knowledge on the role of resilience in attenuating adverse mental health outcomes, including unique effects specific to racial and ethnic minority populations.
Conclusion
Findings support the need for clinical interventions focused on fostering resilience, particularly for women with exposure to interpersonal violence to decrease risk for the development of depression, psychological distress, and suicidal ideation. Although additional research is needed, the attenuating effects of resilience on negative mental health symptomology associated with interpersonal violence vary across race and ethnicity and practitioners should be cognizant of cultural contexts that shape survivors’ resilience. Future research is needed to identify contextual and cultural factors that promote resilience among survivors of interpersonal violence, including systems-level variables (e.g., survivors’ interactions with systems, systems-level procedures, legislative policy) that influence survivors’ resilience, which is needed to inform culturally specific interventions. Findings from this study can be used to inform future research that explains individual and systems-level mechanisms influencing resilience to improve mental health and well-being outcomes among survivors of interpersonal violence.
Footnotes
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 study was provided by an intramural research grant from the University of Maryland, Baltimore (Principal Investigator: DeVylder).
Supplemental Material
Supplementary material for this article is available online.
Author Biographies
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
