Abstract
Generalized anxiety disorder (GAD) has received minimal empirical attention in the context of intimate partner violence (IPV). Furthermore, factors related to lower levels of GAD symptoms in this population have received limited focus. This study evaluated the protective role of four forms of support, spiritual, family, friend, and community, in predicting levels of generalized anxiety among women who have experienced recent IPV. Participants included 116 women who were recruited from local agencies serving IPV-exposed individuals. Participants completed measures of IPV, GAD, stressful life events, spiritual support, social support, and community support during a 1-hr interview. Findings from a hierarchical multiple regression analysis indicated that after accounting for age, income, mental health service utilization, stressful life events, and severity of IPV, lower GAD symptoms were only associated with higher spiritual support (β = −0.20, p = .02) and community support (β = −0.25, p = .01), not friend or family support, F(9, 114) = 5.10; p < .001; R2 = .30. These findings indicate that alternative sources of support (i.e., spiritual and community) may be more accessible for IPV-exposed women, contributing to their association with reduced GAD symptomatology. The current study highlights the potential for spiritual and community support to serve a beneficial role above and beyond standard social support proffered by friends and family on symptoms of GAD. Results reinforce the examination of a broad range of multiple supports among women experiencing IPV. This thorough examination of different support systems may provide further insight into novel resources that can be strengthened among IPV-exposed populations.
Among all psychiatric disorders, anxiety disorders are the most common in the United States, resulting in an economic burden exceeding 42 billion dollars per year (Kessler et al., 2005; Norrholm & Ressler, 2009). One of the most common anxiety disorders is generalized anxiety disorder (GAD), which is characterized by excessive and uncontrollable worry and anxiety that leads to other difficulties including sleep disturbance, restlessness, fatigue, irritability, muscle tension, and difficulty concentrating (American Psychiatric Association, 2013). GAD has a lifetime morbidity risk of 9% among U.S. adults (Kessler, Petukhova, Sampson, Zaslavsky, & Wittchen, 2012). Research examining gender differences in anxiety indicates that women are twice as likely as men to experience an anxiety disorder over the course of their lifetime (McLean, Asnaani, Litz, & Hofmann, 2011). In addition, the prevalence of GAD is highest among women over the age of 25 (Kessler et al., 2005).
In addition to gender differences in the prevalence of GAD, there are also gender differences in the rate of violence exposure within romantic relationships, with women being more likely than men to experience intimate partner violence (IPV; Coker, Davis, et al., 2002). IPV is characterized by psychological, physical, or sexual abuse committed by a romantic partner (Coker, Davis, et al., 2002). IPV is highly prevalent, with research suggesting that between 24% and 59% of women will experience some form of IPV during their lifetime (Black et al., 2011; Coker, Davis, et al., 2002; Rivara et al., 2007). Spiritual abuse can also occur in the context of IPV, which is characterized by the use of spirituality or religion to subordinate a partner or prevent them from leaving a potentially abusive relationship (Bent-Goodley & Fowler, 2006; Flinck, Paavilainen, & Åstedt-Kurki, 2005). Spiritual abuse can also include preventing women from attending a church service or reading spiritual literature (Flinck et al., 2005). Furthermore, women are more likely than men to experience detrimental outcomes from IPV such as poor physical health, substance use, psychopathology, and injury (Coker, Weston, Creson, Justice, & Blakeney, 2005). These problematic outcomes contribute to high health care utilization and costs, with an estimated increased economic burden of 19% for total health care costs among women exposed to IPV (Rivara et al., 2007). The economic burden of IPV also includes crisis intervention services, housing services, victim’s advocate services, legal services, and approximately 8 million lost days of paid work annually (Black et al., 2011). Despite the increased prevalence of both IPV and GAD among women, as well as the substantial physical, psychological, and financial burden associated with these experiences, GAD has received minimal empirical attention in studies of IPV-exposed women. Thus, the present study aims to assess GAD among women experiencing recent IPV by exploring potential protective factors associated with lower GAD symptomatology among these women.
The Relation Between GAD and IPV
To date, only a few studies have examined GAD in the context of IPV, each of which found a positive association between more severe IPV and greater GAD symptoms. J. G. Beck and colleagues (2014) noted the lack of empirical attention that GAD has received among IPV-exposed samples and sought to further examine this construct through two research studies. To expand upon past work, additional factors that may be associated with GAD were examined, including perceived social support (J. G. Beck et al., 2014). Among their sample, IPV severity was positively associated with GAD symptoms, with 34% of participants endorsing GAD symptomology. However, social support was not significantly associated with GAD. In a similar study, J. G. Beck, Jones, Reich, Woodward, and Cody (2015) further examined GAD among IPV-exposed women. Within this sample, 56.1% of participants met diagnostic criteria for GAD and results indicated that IPV exposure was associated with more GAD symptoms. Stuart and colleagues (2006) examined mood and anxiety disorders among women who were arrested for violence within a romantic relationship. Because female perpetrators of IPV are often also victims of IPV (Swan & Snow, 2006), these results can inform estimates of GAD in the context of IPV. Results from Stuart et al. (2006) indicated that IPV was positively associated with GAD in women, with 34% of the sample meeting diagnostic criteria for GAD. While existing research has enhanced the literature by establishing a direct relationship between IPV and GAD symptomatology, it has failed to examine factors that may mitigate psychopathology.
Protective Factors Associated With Lower Levels of GAD
There is a dearth of research on protective factors associated with lower GAD symptomatology among IPV-exposed women. Protective factors are variables that may be associated with lower risk of impairment in individuals exposed to adversity (Afifi & MacMillan, 2011). Due to the high prevalence of IPV and GAD among women, as well as the substantial difficulties that often accompany these experiences, it is essential to understand protective factors that may be most influential for this population.
Social Support
Social support is a commonly studied protective factor that is often broadly defined and measured. However, conceptual work examining social support has recommended assessing specific forms of support (i.e., family support, friend support, and community support) rather than a global measure of this construct (Barrera, 1986). One theoretical framework that aligns with this conceptualization of social support is ecological systems theory (Bronfenbrenner, 1977), which emphasizes the contribution of variables at different contextual levels, including the individual level, the relational level, and the social-contextual level. Within this model, individuals are embedded within a system of multiple interacting relationships that are represented by overarching categories. Bronfenbrenner’s model highlights the importance of examining these multiple levels of support, including support found within the individual (i.e., spiritual connection), their microsystem (i.e., friends and family), and their exosystem (i.e., neighbors and community). Lin, Ye, and Ensel (1999) also discussed the benefit of examining structural supports at different levels, highlighting the different types of support (e.g., instrumental vs. emotional) that exist within each level of support.
Consistent with this model, it is necessary to examine multiple forms of support that promote positive outcomes among IPV-exposed women. However, such research is absent from the literature with respect to GAD. Only two of the reviewed studies of GAD symptoms among IPV-exposed women examined social support, and these studies utilized a global measure of this construct (J. G. Beck et al., 2014; Woodward et al., 2013). Finally, among research that examines multiple types of social support that can be provided by others (i.e., emotional, practical, and critical), the effectiveness of this support based on the type of relationship with the individual providing the support is not distinguished (i.e., friends or family; Levendosky, Bogat, Theran, Trotter, Eye, & Davidson, 2014). These limitations of past studies have contributed to disparate findings in the literature regarding the relationship between social support and GAD among IPV survivors.
Spiritual Support
In alignment with Bronfenbrenner’s model, spiritual support can be seen through the lens of the individual system, as the private spiritual relationship with a higher power held by some individuals. Under Lin and colleagues (1999) model, spirituality would be conceptualized as perceived emotional support. The connections one finds through their spirituality (e.g., with God, religion, and the transcendent) have been shown to serve as a source of support in many studies of IPV-exposed women (Peterson, Johnson, & Tenzek, 2010; Watlington & Murphy, 2006). Spirituality can provide a source of inner strength, a source of comfort, and direct perceived support from the spiritual relationship (Gillum, Sullivan, & Bybee, 2006; Peterson et al., 2010; Watlington & Murphy, 2006). While spirituality is typically conceptualized as serving a protective role for survivors of IPV, there is some evidence to indicate that beliefs associated with spirituality (i.e., family and gender role expectations) may lead to greater tension and distress among survivors of IPV (Yick, 2008). However, in multiple studies of IPV-exposed women, spiritual support was associated with reduced psychological distress, reduced posttraumatic stress disorder (PTSD) and depression symptomology, and enhanced psychological well-being (Gillum et al., 2006; Peterson et al., 2010; Watlington & Murphy, 2006). Despite consistent research regarding spirituality and psychological well-being among IPV survivors, researchers have yet to examine spirituality as a source of support for IPV-exposed women experiencing GAD. However, spiritually has been effective in reducing symptoms among individuals diagnosed with GAD (Koszycki, Raab, Aldosary, & Bradwejn, 2010). Furthermore, spiritual support may be less affected by factors inherent to violent relationships, such as isolation, given that this individual source of support (i.e., connection to a higher power) can be called upon even within socially isolating environments.
Social Support From Friends and Family
In line with the microsystem of Bronfenbrenner’s model, we examined friend and family support at this level of the ecological system. Friend and family support represents an individual’s broad social network and intimate relationships, which typically proffer both instrumental and emotional forms of support (Lin et al., 1999). Findings regarding the protective role of friend and family support on GAD symptoms among IPV-exposed women are equivocal. In studies measuring friend and family support jointly among IPV survivors, findings indicated that social support was not associated with fewer GAD symptoms (J. G. Beck et al., 2014; J. G. Beck et al., 2015; Stuart et al., 2006; Woodward et al., 2013). However, among studies that examined friend and family support as separate constructs, researchers found that women who identified consistent support from their friends or their family members also experienced lower levels of GAD symptoms (Coker, Smith, et al., 2002). Another study that assessed friend and family support as two distinct variables also showed that women reporting more emotional support from both sources experienced fewer GAD symptoms (Levendosky et al., 2014). Studies that utilize specific measures of social support, rather than globally measuring total social support, are in the minority within the empirical literature. This prevents researchers from being able to disentangle whether the protective role of social support is specific to friends or family without a more targeted measure to assess the contribution of each type of support. To address this gap in the literature, research examining friend and family support as distinct constructs, in addition to other potential protective factors, is warranted among IPV survivors experiencing GAD.
Community Support
Finally, within Bronfenbrenner’s ecological systems framework, an additional source of support can be experienced through one’s community and social context. Community support can provide a sense of belonging and social identity, as well as opportunities for social interaction and relationships (Lin et al., 1999). Community support has been associated with increased well-being and reduced distress (Lin et al., 1999). Community cohesion, otherwise known as “collective efficacy,” is a measure of community support that examines perceptions of trust, safety, and unity among community members (i.e., social ties, holding similar values; Fagan, Wright, & Pinchevsky, 2013; Sampson, Raudenbush, & Earls, 1997). Community support is distinct from other forms of social support in that it represents the individual’s perception of support from their neighborhood and the local community. Community cohesion also differs from social support in that community cohesion is theorized to be protective against violence in communities (Sampson et al., 1997). Community safety, trust, and the availability of social networks are factors that can influence an individual’s ability to cope effectively following trauma (Johns, Aiello, Cheng, Galea, Koenen, & Uddin, 2012). In contrast, unsafe and untrustworthy communities with less “collective efficacy” confer reduced community support networks and fewer resources to help individuals cope with life stressors (Johns et al., 2012; Sampson et al., 1997).
Community support is a relatively understudied protective factor; however, the literature that has sought to examine this construct has indicated that lower levels of community support are associated with increased levels of anxiety and stress (Gary, Stark, & LaVeist, 2007; Johns et al., 2012; Zhang, Eamon, & Zhan, 2015). Consistent with the theory of collective efficacy and community cohesion proposed by Sampson and colleagues (1997) and Fagan and colleagues (2013), research among a sample of predominantly low-income African American men and women revealed that lower community support and higher neighborhood disorder were associated with greater anxiety symptoms, even after controlling for prior trauma exposure (Gapen et al., 2011). While this literature supports the influence that community support may have on anxiety symptoms within the context of trauma, currently there is not a complete understanding of the influence of community support on GAD symptoms among IPV-exposed women. Given that traditional social support networks, such as friend and family support, are often reduced within the context of IPV, alternative forms of support that display more stability in the lives of these women may serve a protective role for GAD symptoms (Lanier & Maume, 2009), although this has yet to be empirically examined.
The Current Study
While previous literature has established a solid foundation for our understanding of the connection between IPV and GAD, there is a lack of information regarding protective factors associated with lower GAD symptoms. Accordingly, the present study aimed to identify specific sources of support in the lives of women exposed to IPV that are related to fewer symptoms of GAD. The current study sought to address multiple facets of support rather than examining support as a unitary construct, with the goal of gaining a more precise understanding of different aspects of support. Guided by the ecological systems theory (Bronfenbrenner, 1977), we examined four forms of support, spiritual support, family support, friend support, and community support, as potential protective factors in the context of IPV and GAD. Given past research underscoring the protective role of each form of support in the development of GAD symptoms following IPV exposure, it was hypothesized that all four forms of support would be inversely related to GAD symptomatology.
The present study controlled for the influence of age, income, and mental health service utilization. Participant age was included because GAD occurs at the highest frequency in women above the age of 25 (Kessler et al., 2005). Income was included as a control variable because lower income levels are associated with higher rates of GAD (Baer, Kim, & Wilkenfeld, 2012; Fryers, Melzer, & Jenkins, 2003). Furthermore, bearing in mind that most of the sample was living below the poverty line, income was a relevant variable to include given its association with reduced social support (Weyers et al., 2008). Finally, utilization of mental health services was controlled for given that receiving care may impact the expression of GAD symptomatology (Gorman, 2001). Furthermore, the present study also examined the association between GAD symptoms and both stressful life events and severity of IPV. These two variables were added in the model due to their consistent association with GAD in the existing research literature (J. G. Beck et al., 2014; A. Beck, Emery, & Greenberg, 2005; J. G. Beck et al., 2015; Kendler, Hettema, Butera, Gardner, & Prescott, 2003; Stuart et al., 2006).
Method
Participants
Participants included 116 women between the ages of 22 and 59 (Mage = 32.33, SD = 6.53) who were exposed to IPV within the past 6 months. Participant reports of race/ethnicity varied, with 68% identifying as non-Hispanic Black, 14.8% identifying as Multiracial, 11.7% identifying as non-Hispanic White, 3.1% identifying as Latina, 1.6% identifying as “Other,” and 0.8% identifying as Asian. The majority of participants were experiencing poverty, with only 25% of women reporting an income above the Federal Poverty Line for a family of four (US$24,250). In addition, almost half of the sample indicated that they were unemployed (41.4%).
Procedure
Following institutional review board (IRB) approval, women were recruited from community organizations serving individuals experiencing IPV via flyers and direct referral from staff. All women signed informed consent forms prior to participating in the study. Participants completed an hour-long interview with trained research assistants. Instructions and self-report measure questions from validated self-report measures were read aloud to participants to enhance participant engagement and reduce potential literacy barriers. Interviewers recorded participant responses electronically. Participants received a US$20 gift card as compensation for their time and referral information for local mental health resources.
Measures
Demographics
Participants were administered a demographics questionnaire to obtain background information including age, education, annual income, employment status, relationship status, and race/ethnicity. Utilization of mental health services was assessed by asking participants, “Did you see a mental health provider (i.e., counselor, therapist, psychologist, and/or psychiatrist) for any mental health condition over the past year?”
Generalized Anxiety Disorder 7 (GAD 7)
The GAD-7 is a widely used seven-item self-report measure that assesses symptoms of generalized anxiety over the past 2 weeks (Spitzer, Kroenke, Williams, & Löwe, 2006). This measure utilizes a 4-point Likert-type scale ranging from 0 (not at all) to 3 (nearly every day) to address the core symptoms of GAD (e.g., “Feeling nervous, anxious, or on the edge”). Items are summed to create a total score that can range from 0 to 21, with higher scores indicating more symptoms of generalized anxiety. The GAD-7 has good internal consistency, with alpha coefficients ranging from .79 to .91, as well as strong criterion, construct, factorial, discriminant, and procedural validity (Spitzer et al., 2006). This measure has also shown a satisfactory balance between sensitivity (89%) and specificity (82%; Spitzer et al., 2006). In the present study, the internal consistency was α = .91.
The Revised Conflict Tactics Scale (CTS2)
Participants reported on the severity of the IPV they experienced using the CTS2 (Straus, 1979; Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The CTS2 is a 78-item self-report measure used to assess the severity of psychological, physical, and sexual violence experienced in a dating, cohabitating, or marital relationship. This study included the 39 items that refer to violence perpetrated by the victim’s partner. Items inquire about the frequency of different types of IPV in the past 6 months using a 7-point Likert-type scale that ranges from 0 (This has never happened) to 6 (Happened more than 20 times in the past 6 months). Responses are summed to create a total score in which higher scores reflect greater frequency of IPV. Psychometric evaluations of the CTS2 have indicated that it has good internal consistency, with alpha coefficients ranging from .79 to .95, as well as satisfactory construct and discriminant validity (Straus et al., 1996). In the present study, the internal consistency was α = .94.
Life Events Checklist (LEC)
The LEC is a 17-item self-report measure developed by the National Center for PTSD to screen for stressful events that have occurred during the participant’s lifetime (Gray, Litz, Hsu, & Lombardo, 2004). The LEC covers nine types of stressful life events including natural disasters, transportation accidents, other serious accidents, physical assault, assault with a weapon, sexual assault, combat, other life-threatening events, and the unexpected death of a loved one. Participants indicate yes/no to which events they have experienced and affirmative responses are summed to create a total stressful life events score. The LEC has demonstrated adequate psychometric properties as a standalone measure of traumatic exposure through high test–retest reliability, strong convergent validity, and satisfactory kappa coefficients (Gray et al., 2004). Given the nature of this questionnaire, a reliability coefficient is not necessary because individuals can experience one type of trauma without necessarily experiencing others; accordingly, LEC items need not be related (Gray et al., 2004).
Daily Spiritual Experience Scale (DSES)
The DSES is a 15-item self-report measure designed to assess participant perceptions of their spiritual quality of life (Underwood, 2011; Underwood & Teresi, 2002). The DSES captures the main dimensions comprising spirituality, including one’s personal intimacy with a higher power, strength and comfort, perceived divine love, inspiration or discernment, transcendence, and internal integration (e.g., “I ask for God’s help in the midst of daily activities,” “I find comfort in my religion or spirituality”; Underwood, 2011; Underwood & Teresi, 2002). The DSES primarily examines one’s individual experience with spirituality or the divine. Items are rated using a 6-point Likert-type scale ranging from 1 (many times a day) to 6 (never or almost never) based on measure developer guidelines (Underwood, 2011; Underwood & Teresi, 2002). The DSES items are reverse-scored and summed, with possible scores ranging from 15 to 90 and higher scores reflecting greater spirituality. The DSES has displayed high internal consistency with alpha coefficients ranging from .94 to .95, as well as satisfactory construct and discriminant validity (Underwood & Teresi, 2002). In the present study, the internal consistency was α = .94.
The Community Cohesion Scale (CCS)
The CCS is a six-item measure that evaluates the perceived connectedness and social ties within the participant’s communities. Participants were instructed to think specifically about the community or neighborhood in which they live while completing the CCS. The CCS was adapted by Cutrona, Russell, Hessling, Brown, and Murry (2000) from the Social Cohesion and Trust Scale developed by Sampson and colleagues (1997). This measure uses a 4-point Likert-type scale from 1 (strongly disagree) to 4 (strongly agree) to gauge participant perception of safety and unity within their neighborhood (i.e., “You live in a close-knit neighborhood”). Scores on this measure can range from 6 to 24 with higher scores indicating stronger perceptions of community support. The Social Cohesion and Trust scale, from which this measure was adapted, has strong psychometric properties indicating convergent validity with other community characteristics, such as neighborhood socioeconomic status (SES) and perceived neighborhood violence (Sampson et al., 1997). For the present study, the CCS demonstrated strong internal consistency, with an alpha coefficient of .88.
Lubben Social Network Scale–Revised (LSNS-R)
The LSNS-R is a 12-item self-report measure of perceived social support from friends and family (Lubben, 1988; Lubben, Gironda, & Lee, 2001). The LSNS-R consists of a 6-point Likert-type scale ranging from 0 (less social engagement) to 5 (more social engagement) to assess perceived social connections. The items on this scale examine perceived family support (e.g., “How many relatives do you feel at ease with, like you can talk about private or personal matters?”) and perceived friend support (e.g., “How many friends do you feel close to such that you could call on them for help?”). Scores on the LSNS-R can range from 0 to 60, with higher scores indicating greater perceived social support. The questions assessing friend and family support can be summed separately to examine friend and family support individually. The LSNS-R has demonstrated acceptable internal consistency with alpha coefficients ranging from .70 to .89, as well as convergent validity (Lubben, 1988; Lubben et al., 2006). In the present study, internal consistency was α = .84 for the Family subscale and α = .88 for the Friend subscale.
Data Analytic Plan
Hierarchical multiple regression analyses were conducted using SPSS version 25 to assess the unique contributions of spiritual support, family support, friend support, and community support on symptoms of GAD. Data were first examined for missingness. The data exhibited less than 1% of missing data (i.e., 0.09%, 11 in 127). Given the low percentage of missing data, no additional techniques were utilized to compensate for missingness, and participants with missing data were excluded from the analyses. The demographic variables of age, annual income, and mental health service utilization were entered in the first model. The second model added stressful life events and severity of IPV to assess the influence of adversity on GAD symptoms while accounting for demographics. Finally, the protective factors of spiritual support, family support, friend support, and community support were added to the final model, while accounting for demographic variables, IPV severity, and life stressors. In addition to assessing whether each successive model accounted for a greater proportion of variance in GAD symptomatology, we evaluated how each individual variable independently influenced GAD symptoms.
Results
Descriptive statistics for study variables are presented in Table 1. The majority of participants experienced severe IPV (M = 179.74, SD = 144.91), with approximately seven violent incidents occurring each week. The sample displayed significant variability in total GAD symptoms, covering the full range of possible scores on this measure (M = 11.06, SD = 6.30, range = 0-21). Intercorrelations among the study dependent variable and the continuous predictor variables ranged from r = .07 to .57 (see Table 1). GAD symptoms were positively correlated with stressful life events and severity of IPV and negatively correlated with spiritual support, family support, and community support. Analyses were then conducted to examine normality, linearity, independence of residuals, and homoscedasticity; no violations were observed. In addition, multicollinearity diagnostics were evaluated using the variance inflation factor (VIF) and all values fell within acceptable standards (VIF < 2).
Intercorrelations, Means, Standard Deviations, and Internal Consistency Among Continuous Study Variables.
Note. N = 116. GAD-7 = Generalized Anxiety Disorder–7; CTS2 = Revised Conflict Tactics Scale; LEC = Life Events Checklist; DSES = Daily Spiritual Experience Scale; CCS = Community Cohesion Scale; LSNS-R: Family = Lubben Social Network Scale–Revised Family; LSNS-R: Friends = Lubben Social Network Scale–Revised Friends.
p < .05. **p < .01.
Table 2 presents the hierarchical multiple regression model findings. Participant age, annual income, and mental health service utilization were entered in Model 1 and the model was not significant, F(3, 114) = 0.69; p = .56; R2 = .02, explaining only 2% of the variance in GAD symptoms. When total stressful life events and severity of IPV were added into Model 2, the model was significant, F(5, 114) = 4.21; p = .002; R2 = .16, and the amount of variance explained in GAD symptoms increased to 16%. In this model, IPV severity contributed significantly to the variance in GAD symptoms (β = .34; p = .001), such that more severe violence was associated with more GAD symptoms. When the protective factors of spiritual support, family support, friend support, and community support were added to Model 3, the model was also significant, F(9, 114) = 5.10; p < .001; R2 = .30, with a substantial increase in the variance explained, shifting from 16% to 30%. In this final model, income was significantly associated with GAD (β = 0.23, p = .01) as was IPV severity (β = 0.25, p = .01). As hypothesized, spiritual support (β = −0.20, p = .02) and community support (β = −0.25, p = .01) emerged as significant predictors of GAD symptomatology, in that greater community support and higher levels of spiritual support were associated with reduced GAD symptoms. However, contrary to our hypothesis, neither social support from family (β = −0.12, p = .28) nor from friends (β = −0.06, p = .57) reached significance.
Summary of Hierarchical Regression Analysis Examining Generalized Anxiety Disorder Symptoms as Measured by the GAD-7.
Note. N = 116. Mental health services is a grouping variable indicating if participants have or have not seen a mental health provider in the last year. GAD-7 = Generalized Anxiety Disorder–7; CTS2 = Revised Conflict Tactics Scale; LEC = Life Events Checklist; DSES = Daily Spiritual Experience Scale; LSNS-R: Family = Lubben Social Network Scale–Revised Family; LSNS-R: Friends = Lubben Social Network Scale–Revised Friends; CCS = Community Cohesion Scale.
p < .05. **p < .01.
Discussion
The present study concurrently examined the protective role of multiple forms of support in relation to GAD symptoms among women recently exposed to IPV. This broad examination of support systems, in line with Bronfenbrenner’s ecological systems theory, provides unique insight into the differential influence of various forms of support, which have the potential to mitigate psychopathology following IPV. This study adds to the growing body of literature suggesting that among women exposed to IPV, friend and family support are not associated with GAD symptomatology (e.g., J. G.Beck et al., 2014; Woodward et al., 2013). The current work also emphasizes the potential for alternative sources of support (i.e., community and spiritual) to be related to GAD symptomatology in the context of IPV. These findings underscore the importance of evaluating the influence of multiple forms of support, rather than relying on a broad conceptualization of support, in the lives of women exposed to IPV. Finally, the current study contributes to the growing field of diversity in research by examining the relationship between support factors and GAD in the context of IPV among a sample of women who primarily self-identified as being African American.
Based on our conceptual model and previous literature, we hypothesized that all four forms of support examined in the present study would be associated with fewer GAD symptoms. The results indicated that only two of these forms of support were in fact associated with fewer GAD symptoms—spiritual and community support. Thus, we must consider explanations as to the lack of a relationship between GAD and social support from friends or family. While social support is generally viewed as a protective factor against many forms of psychopathology, social support is often decreased among IPV-exposed women, with many women reporting increased social isolation during and after their violent relationship (Lanier & Maume, 2009). This reduced social support, particularly evident in relationships with friends and family, is the result of many factors within the violent relationship, including the abusive individual’s efforts to isolate and control their partner’s social encounters (Lanier & Maume, 2009). Survivors themselves may also disengage from their social network due to the shame and stigma often associated with IPV (Coker, Smith, et al., 2002). Furthermore, survivors may perceive a diminished availability of their social support network. This common occurrence may be explained by the deterioration model of social support. This model postulates that difficult life events (e.g., IPV), and the associated distress tied to those events (e.g., GAD), reduce both the perception and availability of social support (Wheaton, 1985). Other research with IPV-exposed women supports this model, indicating that IPV is associated with lower levels of perceived social support and increased psychological distress (Thompson et al., 2000). Furthermore, multiple studies have indicated that social support from friends and family may not serve a strong protective role against GAD symptoms among IPV-exposed women (J. G. Beck et al., 2014; J. G. Beck et al., 2015; Stuart et al., 2006; Woodward et al., 2013). In addition, research has suggested that experiencing a disadvantaged SES is associated with reduced social support and poor social networks (Weyers et al., 2008). The only exception to this positive relationship between low SES and reduced social support occurs when individuals have family members who live within close proximity to them (Weyers et al., 2008). Together these findings may explain the results seen in the present study, where friend support was not significantly associated with GAD, and family support was no longer significant after accounting for spiritual and community support. Our sample displayed low-average levels of friend (M = 13.96; possible range: 0-30) and family (M = 16.34; possible range: 0-30) support, suggesting that, while friend and family support were not at strikingly low levels among these women, their relation with GAD was not as strong once accounting for alternative forms of support.
Accordingly, alternative sources of social support may be both more accessible and more effective in protecting against increased severity of GAD symptoms following IPV. This appears to ring true in the present sample where higher levels of community and spiritual support were associated with fewer GAD symptoms. These alternative supports may be protective by promoting a sense of security, safety, and strength among IPV-exposed women. Furthermore, these alternative sources of support may exhibit greater stability in the midst of social isolation and control. This perspective is consistent with research findings among IPV survivors, of whom 97% of the sample indicated that their spirituality served as a source of strength and comfort amid their adversity (T. Gillum et al., 2006). These women described their spirituality as a fundamental component of their identity, one that withstood the effects of their violent relationship (T. Gillum et al., 2006). The findings of the present study can be further understood after taking into consideration features of the sample, including that most participants were residing in the U.S. Midsouth, as women residing in the South are more likely to utilize religious coping strategies (Chatters, Taylor, Jackson, & Lincoln, 2008). These findings support the notion that spirituality may be a more reliable protective factor for women exposed to IPV as compared to traditional sources of support from friends or family.
A second identified source of support that was associated with reduced GAD symptomatology was community support. Women in the present sample were recruited from centrally located service organizations that assist individuals experiencing violence in obtaining orders of protection and other services, such as housing and counseling. Accordingly, many participants were likely facing safety concerns and instability in their homes at the time of assessment. Thus, if their community is viewed as a stable and supportive place, then they may experience fewer symptoms of generalized anxiety. Furthermore, once IPV escalates to the level where women are seeking services, friends and family often further disengage due to fear of the violent perpetrator and concerns for their own safety (Beeble, Bybee, Sullivan, & Adams, 2009). If the wider community remains supportive when these more immediate sources of support dwindle, then women may experience less psychopathology. Thus, if women experience less engagement from their immediate supports (i.e., friends and family), but believe they can trust their neighbors and the broader community, it may enhance their own feelings of safety and lessen levels of generalized anxiety (Johns et al., 2012).
Limitations
Although the present study boasts a number of strengths, there are also limitations that should be considered when interpreting the results. The cross-sectional study design prevents the ability to make causal claims about the relationships between study variables. We can only speculate about the nature of the relationship between these forms of support and GAD symptoms among women exposed to IPV. Thus, we cannot draw predictive conclusions about the definitive role of these identified supports. Given that these measures were all gathered at one time-point, we do not know whether higher spiritual and community support precluded more severe GAD symptoms, whether the reverse is true, or whether some third variable is contributing to these associations. A second limitation comes with respect to the method of data collection being self-reported information collected in an interview format. Due to the sensitive nature of study questions, participants may not have been comfortable answering honestly, which introduces the potential for response bias. However, this method of data collection through an interview format ensures that each participant hears and understands each question, which increases the validity of the data.
Another consideration when interpreting the results is the relatively homogeneous nature of the sample with respect to race (i.e., 68% Black) and region (i.e., Midsouth), which in turn limits the generalizability of the findings. This is particularly relevant for the variable of spiritual support, given that the data were collected within what may be considered the “Bible-belt,” which may have influenced the high levels of spirituality seen in the present sample. Despite this homogeneity, the racial makeup of the current study is representative of the population of the city from which the data were collected. Finally, all of the participants in the present study were recruited through a local service organization that aids individuals experiencing violence. Accordingly, these help-seeking participants may differ from other IPV survivors not seeking formal support in that they may require support above and beyond what family and friends can offer. Furthermore, the features of this sample including that most participants were living below the poverty line may explain the high social service utilization of the current sample, which may also have influenced the current findings. Despite these limitations, the findings from the present study offer novel information about the understudied outcome of GAD in relation to multiple forms of support among an IPV-exposed sample. By being among the first studies to concurrently examine multiple forms of support in an IPV-exposed sample, we are able to gain a more detailed understanding of which factors drive GAD symptoms.
Clinical Implications
Individual responses to IPV vary widely; therefore, it is essential for health care providers to be aware of the different presentations that can result from violence exposure and the most effective ways to target accompanying symptoms. As survivors may disengage from their social network due to the shame and stigma associated with IPV, it may be particularly beneficial to connect women to other survivors of IPV who can provide support through their shared experience and understanding (Coker, Smith, et al., 2002). The present findings suggest that it may be beneficial for intervention developers to explore how they can further engage women within their communities and spiritual networks. Previous group interventions with spiritual components have shown improvements in overall mental health among women exposed to adversity (McCain et al., 2008; Taha, Zhang, Snead, Jones, Blackmom, Bryant, & Kaslow, 2015; Zhang et al., 2013). Spiritually driven group-based interventions that build upon individual’s connection within their communities may prove advantageous for IPV-exposed women experiencing symptoms of GAD. Creative ways to integrate spirituality into intervention and treatment should continue to be examined, such as having spiritual leaders take part in delivering the intervention. Of note, spirituality has disparate meanings among different individuals and cultures (Plunkett, 2014). Accordingly, interventions should reflect culturally competent care that is respectful of myriad spiritual perspectives. It may further the literature if interventions utilizing prayer and other methods to deepen spirituality, among women who identify as spiritual, are explored to assess if they are beneficial in reducing GAD symptoms (F. Gillum & Griffith, 2009). Finally, clinicians should inquire about community resources within the lives of IPV-exposed women and efforts should be explored to increase this form of support.
Future Research Directions
Given the alternative forms of support identified in this study, future research would likely benefit from examining other understudied forms of support that may serve a protective role among IPV survivors. Potential sources may include individual factors, church communities, and social ties to other IPV survivors. Furthermore, future work should assess the strength and role of these identified supports before, during, and after the violent relationship via longitudinal research. This will provide insight into the evolution of supports within the context of IPV, as well as which supports may be most beneficial to target via intervention at specific critical junctures for IPV-exposed women. In addition, given the role of spiritual support found in the present study, future work should examine how religious support (i.e., support found specifically within one’s religious community) or lack thereof and religious or spiritual abuse may impact GAD. In the present study, spiritualty was examined at the individual level. Future research may benefit from evaluating spiritual support within the relational (e.g., friends and family) and social-contextual (e.g., religious community) levels to provide a more comprehensive understanding of the role of spirituality across different contexts. Finally, future research would be advanced by including a more diverse sample, both in terms of race/ethnicity and region where the data are collected (i.e., North vs. South) to further understand the role of these demographic variables. It will also be important for future work in this area to evaluate the role of mental health care (e.g., individual and group psychotherapy) and psychotropic medication (e.g., anxiolytics) on GAD symptoms, in conjunction with these sources of support, which can be further enhanced using therapeutic strategies (e.g., social skills training, mindfulness-based techniques, and cognitive-behavioral therapy).
Conclusion
Study findings support further examination of understudied mental health difficulties, as well as factors that may serve to buffer against these difficulties, among IPV survivors. Such an avenue of research may provide untapped insight into novel protective factors that can be enhanced among individuals exposed to IPV. Gaining a more comprehensive understanding of modifiable protective factors, such as community and spiritual supports, may serve to inform treatment providers about factors to target via intervention. Given the lack of available research that has examined the role of these multiple forms of support on GAD symptoms among IPV survivors, this study advances the field’s understanding of strengths-based factors associated with improved functioning in women experiencing partner violence.
Footnotes
Acknowledgements
The authors would like to thank the mothers and community partners who made this research possible, as well as the graduate and undergraduate research assistants who helped with data collection.
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: This work was supported by institutional grants including the University of Memphis Faculty Research Grant Fund (Principal Investigator [PI]: Howell) and the University of Memphis Diversity Research Grant (PI: Thurston). This support does not necessarily imply endorsement by the University of Memphis of the study’s research conclusions. Authors’ effort on this study was also funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development R15HD089410 (PI: Howell).