Coping Styles and Suicide in Racially and Ethnically Diverse Lesbian, Bisexual, and Queer Women

Abstract

The purpose of the current study was to examine whether coping strategies are associated with past suicidal ideation and history of suicide attempts among sexual minority women (SMW). Participants were 150 racially and ethnically diverse lesbian, bisexual, queer, or “other” nonheterosexual-identified cisgender women who were recruited as part of a national online survey on the experiences of SMW. Simultaneous multiple regressions suggested that coping styles significantly explained 20.3% of the variance in past suicidal ideation and 30.4% of the variance in lifetime history of suicide attempts. Within these regressions, self-blame coping positively predicted past suicidal ideation, and religious coping and venting coping were associated with a higher lifetime history of suicide attempts. Based on these findings, directions for future research and interventions to reduce suicide risk among SMW are discussed.

This literature review will discuss the high rates of suicidality in sexual minority populations, frame these high rates through a minority stress lens, and make an argument for looking specifically at suicidality in racially and ethnically diverse sexual minority women (SMW) and the coping styles that are associated with increased or decreased suicidality. Relative to their heterosexual counterparts, sexual minority adults (SMA) are at increased risk for a variety of mental health problems including substance use (Cochran, Ackerman, Mays, & Ross, 2004), depression, and anxiety (Gilman et al., 2001). Perhaps the most pernicious mental health issue prevalent among SMA is suicide, as SMA report higher levels of suicidal ideation (Hill & Pettit, 2012) and attempts (Gilman et al., 2001; Herrell et al., 1999) than heterosexual adults. In a recent systematic review, King and colleagues (2008) found that sexual minorities were nearly two and a half times more likely to have attempted suicide relative to heterosexuals. Similarly, Fergusson, Horwood, Ridder, and Beautrais (2005) found that over a 4-year period, 71.4% of gay men (vs. 10.9% of heterosexual men) and 30% of lesbians (vs. 9.7% of heterosexual women) endorsed suicidal ideation.

The most widely used model to help explain why sexual minorities may be at increased risk for negative mental health outcomes relative to their heterosexual counterparts is the minority stress model (Meyer, 2003) . This model posits that individuals from oppressed social groups (e.g., racial and sexual minorities) experience high levels of stress as well as negative life events as a product of their marginalized status within society. Sexual minorities are subject to high levels of stressors including sexual orientation discrimination, stigma, and pressures to conceal their sexual identity. In turn, multiple and varied forms of stressors may increase psychological distress (Kaysen et al., 2014). This notion is bolstered by several studies indicating that stigma-related stressors are related to psychological distress among sexual minorities (Hatzenbuehler, Nolen-Hoeksema, & Erickson, 2008; King et al., 2008; Meyer, 2003).

In line with Meyer’s (2003) minority stress model, SMW may be at especially high risk for negative mental health outcomes by virtue of multiple minority identities, facing both heterosexism and sexism. Not surprisingly, several studies have demonstrated relationships between experiences of heterosexism and sexism, and high levels of psychological distress among SMW (Piggot, 2004; Szymanski & Henrichs-Beck, 2014; Szymanski & Kashubeck-West, 2008; Szymanski & Owens, 2008). Thus, SMW represent an especially high-risk subpopulation of sexual minority individuals with respect to negative mental health outcomes. SMW of color may be at even greater risk for poor mental health outcomes in contending with experiences of racial and/or ethnic oppression, sexism, and heterosexism (Cole, 2009). For instance, Balsam and colleagues (2015) found that relative to other racial/ethnic groups, African American SMW reported greater exposure to life stressors, which in turn may be associated with greater psychological distress (Hatzenbuehler et al., 2008; King et al., 2008; Meyer, 2003).

Coping strategies represent an important process implicated in both risk for and protection against mental health problems generally (Aldwin & Revenson, 1987; Coyne, Aldwin, & Lazarus, 1981). Research on coping has often distinguished between two broad coping styles: emotion-focused and problem-focused coping (Lazarus & Folkman, 1984). In emotion-focused coping, an individual attempts to manage one’s affective response to perceived stress. Problem-focused coping by contrast is characterized by efforts to alter or eliminate the source of perceived stress (Lazarus & Folkman, 1984). Endler and Parker (1996) in their extended conceptualization of coping assert that in addition to problem-focused and emotion-focused coping, individuals may also engage in avoidance coping. Avoidance coping strategies are further classified as either person-oriented (e.g., social diversion) or task-oriented (e.g., distraction) (Edwards & Holden, 2003). Because coping responses play a crucial role in determining how perceived stress will impact an individual (Endler & Parker, 1990), coping strategies can also be distinguished between those that are adaptive or promote well-being (e.g., positive mental health outcomes) in the face of stressors, and those that are maladaptive or confer risk for negative mental health outcomes in response to stressors (Zeidner & Saklofske, 1996).

Significant research has examined relationships between coping strategies and mental health in adult populations (e.g., Bosmans, Hofland, De Jong, & Van Loey, 2015; Edwards & Holden, 2003; Marty, Segal, & Coolidge, 2010; McNeill & Galovski, 2015). Notably, some literature to date has examined relationships between coping strategies and suicidality (e.g., Edwards & Holden, 2003; Marty et al., 2010). Research with general adult populations indicates that coping strategies involving suppression of unwanted thoughts and emotions (Lynch, Cheavens, Morse, & Rosenthal, 2004) and avoidance coping broadly (Wong, Chang, Yu, Chan, & Chan, 1994) are associated with increased risk for suicidal ideation and suicide attempts. Research also indicates that deficits in problem-solving, a facet of coping (Pollock & Williams, 2004), and restricted coping repertoires (i.e., few coping strategies from which to draw in times of stress; Orbach, Bar-Joseph, & Dror, 1990) represent potent risk factors for suicidality.

A relatively small body of research examining relationships between coping strategies and mental health among sexual minorities has burgeoned (e.g., Kaysen et al., 2014; Szymanski, 2009; Szymanski, Dunn, & Ikizler, 2014). For instance, Lehavot (2012) found that among SMW, behavioral disengagement and self-blame coping were associated with reduced physical and mental health. Although relative to maladaptive coping strategies, relationships were less robust, the author also found that adaptive coping strategies such as active coping, planning, and positive reframing were inversely related to depression among SMW. Corroborating Lehavot’s (2012) former findings, Szymanski and colleagues (2014) found that rumination, detachment, and internalization coping strategies each uniquely predicted psychological distress in SMW. As well, research indicates that a variety of coping styles, including suppressive, reactive, detachment, and internalization mediate relations between heterosexist discrimination and both internalized heterosexism as well as psychological distress among SMW (Szymanski, Dunn, & Ikizler, 2014; Szymanski & Henrichs-Beck, 2014). The majority of the abovementioned coping strategies employed by SMW and discussed in the literature are emotion-focused in nature. This is in line with a trend in research conducted with general populations indicating that women are more apt to utilize emotion-focused coping strategies relative to their male counterparts (Eaton & Bradley, 2008).

There is also research to suggest that different types of religious or spiritual coping may differentially impact mental health outcomes among individuals with stigmatized identities, including sexual minorities. Whereas utilizing positive religious coping strategies (which involves leaning on a higher power for support, guidance, or forgiveness in an attempt to release stress or control over life circumstances) is tied to psychological well-being (Jeffries, Dodge, & Sandfort, 2008; Lehavot & Simoni, 2011), drawing on negative religious coping strategies (which involves seeing stressful life events as punishment or acts of malevolent powers) is associated with poorer mental health (Murr, 2013; Severson, Munoz-Laboy, & Kaufman, 2014).

Coping strategies may differ as a product of gender, race, and sexual minority status as well. Recent attention has been given to the role that so-called collective action might play in conferring risk for or buffering SMW against negative mental health outcomes (e.g., DeBlaere et al., 2014). Collective action entails “acting as a representative of the group” and engaging in actions that are “directed at improving the condition of the entire group” (Wright, Taylor, & Moghaddam, 1990, p. 995). DeBlaere and colleagues (2014) found that among SMW of color, perceived heterosexist experiences predicted psychological distress for those with low levels of sexual minority collective action, whereas for those with high levels of sexual minority collective action perceived heterosexism did not predict psychological distress.

Despite several recent studies that have examined associations between coping and mental health in SMW, no study to date has reported on relationships between coping strategies and suicidality (e.g., ideation and attempts) in this population. Given the research supporting links between maladaptive coping and negative mental health outcomes for SMW (Szymanski et al., 2014), taken together with high rates of suicidality among SMW (Fergusson et al., 2005), an examination of relationships between coping strategies and suicidality for this population has significant public health implications. This is particularly important in light of the notion that SMW may be at increased risk for suicidality by virtue of their multiple minority statuses (i.e., contending with sexism and heterosexism). If associations are present between coping and suicidality in SMW, coping could represent a viable target for suicide intervention and prevention efforts for this high-risk population.

As a result, the purpose of the current study was to examine whether coping strategies are associated with past suicidal ideation and history of suicide attempts among lesbian, bisexual, and queer women. Although no study to date has examined associations between coping strategies and suicidality in sexual minorities, past research indicates that among SMW, coping strategies characterized by avoidance of stressors are associated with mental health problems (Szymanski & Owens, 2008). Similarly, in general adult populations, avoidant coping strategies are associated with increased risk for suicidal ideation and suicide attempts (Wong et al., 1994). It is therefore hypothesized that in the current study, selfdistraction coping, denial coping, and behavioral disengagement coping (all characterized by avoidance of a stressor) will be associated with past suicidal ideation and history of suicide attempts in SMW.

As noted above, deficits in problem-solving, a facet of coping (Pollock & Williams, 2004), are also a potent risk factor for suicidality in general adult populations. Thus, it is hypothesized that coping strategies tapping problem-solving skills such as planning coping and active coping will be inversely related to past suicidal ideation and history of suicide attempts among SMW. Of note, although the present study’s measure of coping includes a number of additional coping styles, the abovementioned hypotheses are based on associations that have been found in the previous literature.

Method

Participants

Participants were lesbian, bisexual, queer, or “other” nonheterosexual-identified cisgender women who were recruited as part of a national survey on SMW, with a particular emphasis on recruiting women of color. Individuals were included in the study if they were at least 18 years of age, identified as a cisgender woman, and identified as lesbian, bisexual, queer, or an “other” nonheterosexual sexual orientation. To assess sexual orientation, participants were asked the question “Which sexual orientation best describes you?” and were provided with the following response options: Heterosexual AND Transgender, Intersex, or other Gender Identity; Heterosexual; Bisexual; Gay/lesbian; Queer; and Other. Participants who selected “other” were asked to further specify their sexual orientation via typed response. Data were automatically removed from the survey software (Qualtrics) by the authors throughout the data collection process if there was an indication of false responding or computer-generated, automated responses (i.e., completion time of less than 20 minutes or greater than 24 hours), or highly improbable response patterns (e.g., selection of the first response for each item on a scale). Data were also deleted if participants failed to correctly respond to at least 4 of 6 (66.6%) of the randomly inserted accuracy checks (e.g., “Please select strongly agree for this item”). This deletion process typically occurred weekly and before each week’s compensation day for participants to limit the chances that false participants would be fraudulently compensated with state funds. This process was mandated by the host university’s information security officer. This automatic deletion process was implemented given the high probability of obtaining false responses in conducting online research involving participant incentives, and the precise number of deleted responses was not recorded and is therefore unknown. These data cleaning procedures resulted in a final sample of 150 participants. Complete responses were required in the data software in order to complete the survey, and therefore no data points were missing.

Participants ranged in age from 18 to 66 (M = 31.9; Mdn = 28, SD = 11.95). With regard to sexual orientation, 38.7% identified as gay or lesbian, 32.7% as bisexual, and 28.6% as queer or other. With respect to race, 29.3% of participants were White/European American (non-Latino), 26.0% Black/African American (non-Latino), 16.7% Asian/Asian American/Pacific Islander, 11.3% Latino/Hispanic, 2.7% American-Indian/Native American,12.7% Multiracial/Multiethnic, and 1.3% Other. The majority (30%) of participants had earned a degree from a 4-year college, 30% a graduate degree, 9.3% a degree from a 2-year institution, 6% a high school diploma or general education diploma (GED), and 24.7% had attended some college. With regard to family income, the majority of participants (41.3%) reported $60,000–199,999, 29.3% reported $30,000–59,999, 14.7% reported $15,000–29,999, 11.3% reported $7,000–14,999, and 3.3% reported $200,000 and up.

Procedure

Individuals were recruited to participate in an online, confidential survey through a variety of Internet forums and groups. National (United States) and regional lesbian, gay, bisexual, transgender, and queer (LGBTQ) organizations (e.g., The Center Orlando, Lesbian and Bisexual Women of the Carolinas) and online LGBTQ social and community groups (e.g., Women of Color Baltimore Facebook group, LGBT People of Color Yahoo group) were contacted via e-mail and provided with recruitment information for the study. Before beginning the survey, full consent was obtained from all participants. After completion of the survey, participants entered an e-mail address where they wished for their $15 Amazon.com electronic gift card to be sent, as well as the unique compensation code that participants were told could only be entered once for compensation. The study was approved by the university’s institutional review board.

Measures

The Brief COPE (COPE; Carver, 1997) was used to assess participants’ coping strategies. This measure comprises 28 items tapping specific coping strategies across 14 conceptually distinct scales: Active coping, planning, positive reframing, acceptance, humor, religion, emotional support, instrumental support, selfdistraction, denial, venting, substance use, behavioral disengagement, and self-blame. Two items map onto each scale. For each item, informants indicate on a 4-point Likert scale the frequency with which they implement the coping strategy from 0 = I haven’t been doing this at all to 3 = I’ve been doing this a lot. The first eight subscales reflect adaptive coping strategies, whereas the final six subscales represent maladaptive coping. Psychometric properties for the Brief COPE, including reliability and validity have been found to be good (Carver, Scheier, & Weintraub, 1989). The Cronbach’s alpha for the maladaptive coping composite in the current sample was α = .639 and for the adaptive coping composite was α = .670. These values are comparable with those obtained in other studies utilizing samples of sexual minority women. For example, Lehavot (2012) reported α = .81 for adaptive coping and α = .74 for maladaptive coping. Similarly, in a sample of ethnically diverse gay men, David and Knight (2008) reported Cronbach’s alphas for maladaptive (referred to as “disengaged”) coping and adaptive (referred to as “active”) coping ranging from α = .74 to .87 across age and ethnic groups. The full score Cronbach’s alpha for the current sample was α = .81.

To assess suicidal ideation, the Suicidal Ideation subscale of the Suicide Behaviors Questionnaire (SBQ-14; Linehan, 1996) was administered to participants. This subscale contains five items and assesses the frequency of past and current suicidal ideation. Its scoring algorithm weights current suicidality more highly than past suicidality to more strongly tap the frequency of a respondent’s current ideation. Thus the bulk of the score’s index of suicidal ideation reflects ideation occurring sometime within the year during which participants completed the survey. A total score is calculated whereby higher scores indicate more suicidal ideation. The SBQ-14 is one of the most common measures of suicidality and has demonstrated good internal consistency (α = .90) in a primarily Caucasian sample of more than 600 adults ages 18 to 24 (O’Riley & Fiske, 2012), as well as good internal consistency in its initial validation (r = .73–.92; Addis & Linehan, 1989). To assess past suicide attempts, participants were asked the researcher-created question: “Over the course of your life, how many times have you attempted suicide?” Responses to this item were numerical in nature. The alpha for the Suicidal Ideation subscale in the current sample was α = .93.

Results

To examine associations between coping styles and suicidality, bivariate correlational analyses were conducted with each of the 14 Brief COPE scales and suicidal ideation and attempts (see Table 1). Past suicidal ideation was negatively associated with positive reframing coping and positively associated with behavioral disengagement coping and self-blame. Lifetime history of suicide attempts was positively associated with religious, venting, behavioral disengagement, and self-blame coping.

Table 1.

Overall Correlation Matrix and Means/Standard Deviations (SDs)

Variable	1	2	3	4	5	6	7	8	9	10	11	12	13	14	15	Mean	SD
1. Past suicidal ideation																8.91	15.39
2. Suicide attempts	.367**															.89	3.21
3. Active	−.080	.046														4.51	1.54
4. Planning	.026	.059	.546**													4.66	1.61
5. Positive reframing	−.173*	−.106	.359**	.380**												4.17	1.80
6. Acceptance	−.128	−.051	.156	.268**	.383**											4.25	1.58
7. Humor	.016	−.002	.057	−.119	.023	.133										2.94	2.18
8. Religious	.030	.253**	.269**	.209*	.389**	.051	−.023									2.03	2.26
9. Emotional support	−.155	−.107	.225**	.289**	.332**	.171*	.039	.310**								3.88	1.99
10. Instrumental support	−.029	.022	.286**	.222**	.400**	.093	.062	.363**	.600**							3.74	1.76
11. Self-distraction	.070	.035	.076	.173*	.037	.062	.064	.289**	.099	.141						4.03	1.65
12. Denial	−.006	−.041	.064	.065	.050	−.130	.186*	.210**	.106	.197*	.251**					.99	1.66
13. Venting	.146	.271**	.115	.114	.065	−.019	.181*	.180*	.185*	.300**	.092	.425**				2.52	1.82
14. Substance use	.153	−.019	−.048	−.058	−.138	−.097	.235**	−.048	.000	−.088	.185*	.178*	.135			1.61	2.00
15. Behavioral disengag	.168*	.219**	−.281**	−.151	−.343**	−.194*	.257**	−.057	−.104	−.037	.091	.377**	.298**	.274**		1.07	1.74
16. Self-blame	.338**	.186*	−.174*	.038	−.165*	−.202*	.184*	−.035	.024	.091	.174*	.266**	.259**	.313**	.463**	2.85	2.10

^*p < .05.

^**p < .01.

To calculate the effect size that a sample of 150 participants could uncover in a multiple regression with 14 predictors, a power analysis was run. With an α = .05 and power (1 – β) = .80, the current study’s sample size of 150 could uncover all medium- and large-sized effects, and some small-sized effects (with an f² ≥ .14). Two simultaneous multiple regression analyses were then run to examine the degree to which the 14 coping styles uniquely predicted suicidal ideation and attempts.

In the first regression, the 14 coping styles explained 20.3% of the variance in past suicidal ideation, F(14, 135) = 2.46, p = .004. Emotional support coping and self-blame coping were uniquely associated with past suicidal ideation. Though bivariately correlated with past suicidal ideation, positive reframing coping and behavioral disengagement coping did not uniquely predict past suicidal ideation in the regression model.

In the second regression, the overall model accounted for 30.4% of the variance in lifetime history of suicide attempts, F(14, 135) = 4.21, p < .001. Religious coping, emotional support coping, denial coping, and venting coping each uniquely predicted lifetime history of suicide attempts. However, behavioral disengagement coping and self-blame coping were not uniquely associated with lifetime history of suicide attempts as they had been in the correlation matrix.

The results of the multiple regression analyses for past suicidal ideation and lifetime history of suicide attempts are presented in Table 2. It is important to note that denial coping and emotional support coping were not correlated with lifetime history of suicide attempts in the correlation matrix, and emotional support coping was not correlated with suicidal ideation, but both types of coping became unique predictors in the regressions. This pattern is likely reflective of error attributable to suppressor effects and should not be interpreted as true effects. A suppressor effect occurs when a predictor variable in a regression is correlated with another predictor (as is typically expected in the context of a regression) but is not correlated or only weakly correlated with the criterion variable; “in this situation the regression coefficient … may be diminished or enhanced and even reversed in sign” (Ludlow & Klein, 2014, p. 1). In the current study, the suppressor effects were that two coping variables that were nonsignificant in the correlation matrix became significant predictors in the regressions. Ludlow and Klein (2014) point out that suppressors, unless they are due to an experimental intervention designed to produce that effect or are theoretically justified, can be “a statistical effect potentially devoid of substantive interpretation” (p. 2).

Table 2.

Standardized Regression Weights of Predictors of Past Suicidal Ideation and Lifetime Suicide Attempts

BRIEF COPE Scale	Past SI β	Lifetime SA β
Active	−.056	.079
Planning	.123	.060
Positive reframing	−.156	−.177
Acceptance	−.023	.038
Humor	−.010	−.035
Religious	.147	.384**
Emotional support	−.235*	−.215*
Instrumental support	.084	.010
Self-distraction	−.005	−.046
Denial	−.162	−.296**
Venting	.133	.288**
Substance use	.078	−.082
Behavioral disengagement	−.035	.188
Self-blame	.299**	.151

Note. SI = suicidal ideation; SA = suicide attempts.

^*p < .05.

^**p < .01.

Discussion

The purpose of the current study was to examine whether coping strategies are associated with past suicidal ideation and history of suicide attempts among SMW, with a particular emphasis on recruiting women of color. Simultaneous multiple regressions suggested that coping styles significantly explained 20.3% of the variance in past suicidal ideation and 30.4% of the variance in lifetime history of suicide attempts. Within these regressions, self-blame coping positively predicted past suicidal ideation, and religious and venting coping were associated with a higher lifetime history of suicide attempts.

The first regression suggested that SMW’s coping styles robustly predicted a history of suicidal ideation. Although no study to our knowledge has examined relationships between coping styles and suicidality among SMW, previous research has found that coping styles among sexual minority populations are strongly tied to both positive and negative mental health outcomes (Kaysen et al., 2014; Szymanski, Dunn, & Ikizler, 2014; Szymanski & Henrichs-Beck, 2014; Lehavot, 2012; Szymanski, 2009; Szymanski & Owens, 2008). The present study therefore extends past research in the area of coping and mental health in SMW by examining coping strategies in relation to suicidal ideation and suicide attempts.

The only unique predictor within the first regression was self-blame coping, which was positively associated with suicidal ideation. This general effect has been found in research by Lehavot (2012) among SMW, where self-blame coping was associated with reduced physical and mental health, and has also been detected in the research of Kaysen and colleagues (2014), which found that a composite of maladaptive coping techniques (comprising self-blame but not looking at it separately) was a greater contributor to negative mental health outcomes (depression and anxiety) than adaptive coping strategies. Again, the present study extends these findings by identifying unique relationships between self-blame coping and suicidal ideation in SMW, perhaps illuminating a more specific intervention target.

The second regression found that SMW’s coping styles also robustly predicted a history of suicide attempts. In the current study, religious coping was positively associated with lifetime suicide attempt history. This finding is of significant interest—in the general population, religious coping styles typically act as protective factors against suicidal behaviors and lifetime risk (Colucci & Martin, 2008; Gearing & Lizardi, 2009). Though the construct of religious coping as operationalized in the Brief COPE measure has not been previously examined in SMW, research has been conducted on the general construct of religiosity in both samples of SMW and the general sexual minority population. Research on SMW has found that religiosity does not have the protective effects against suicide risk that are seen in heterosexual samples (Mathy & Shillace, 2004), and that higher religiosity is associated with greater internalized homophobia (Clayman, 2005). Recent research by Kralovec and colleagues (2014) has further examined this relationship in a study of SMA and found that religious affiliation was associated with greater internalized homophobia but (unlike the current study) fewer suicide attempts, which implies religious affiliation can act as both a risk and protective factor. Additionally, Gattis, Woodford, and Han (2014) have found that among SMW, religious affiliation can act as either a risk or protective factor against depressive symptoms depending on whether or not the denomination endorses same-sex marriage. Research on religiosity in the general LGBT population has similarly found that conservative religious beliefs (e.g., rejection of same-sex marriage) have been associated with higher shame, guilt, and internalized homophobia (Clayman, 2005; Sherry, Adelman, Whilde, & Quick, 2010; Walker & Longmire-Avital, 2013), homonegativity (Buchanan et al., 2001), and identity conflict (Page, Lindahl, & Malik, 2013). Conversely, Lease, Horne, and Noffsinger-Frazier (2005) have found that among SMA, religious experiences that affirm sexual identity are associated with more positive mental health outcomes, and Smith and Home (2007) found that LGBT individuals affiliated with Earth-spirited faiths (e.g., Paganism, Wicca) do not experience faith conflict. Taken into account with the findings from the present study, it can be inferred that the benefits or harm of religious coping may be influenced by the religious traditions to which SMW ascribe.

Venting coping was also a unique positive predictor of lifetime history of suicide attempts. In their theoretical conceptualization of coping, Carver and colleagues (1989) highlight that coping strategies oriented toward engaging with or acting on a particular stressor are considered adaptive (e.g., may be associated with mental health benefits). Venting coping, which does not involve acting on one’s environment as much as other coping strategies, may reflect a more passive role in navigating trying circumstances. From this perspective, it is not surprising that venting coping may not have played a protective role against suicide attempts for SMW because it may have taken the place of more active strategies.

Implications

These findings have multiple implications for intervention research on suicide among SMW. Given the coping strategies that were associated with increased suicide risk in the current study, research should examine whether helping SMW to reduce self-blame, venting, and religious coping styles in specific contexts is effective. Although other coping styles were both positively and negatively associated with suicide risk in the correlation matrix, when examined in the multiple regression, the effects of these three coping styles eclipsed the effects of the other previously significant coping styles on suicide risk. This suggests that these may be three primary styles on which to focus in future research on suicide prevention in SMW. A promising way of doing this involves using cognitive–behavioral therapy (CBT), which has been adapted for use with sexual minority populations (Ross et al., 2008; Safren, Hollander, Hart, & Heimberg, 2001), and future adaptations may benefit, if shown to be effective in intervention research, to address positive and negative coping styles specific to the SMW community.

In particular, helping SMW manage the cognitive dissonance attributable to religious identity conflict and minority identification or attraction could be an important aspect of suicide prevention, as has been advocated for in the transgender community (Levy & Lo, 2013). Further, identifying religious affiliation and an individual’s attitudes toward spiritual- and faith-based experiences and messages can offer a more comprehensive understanding of religious conflicts they may face, which in turn could assist in modifying maladaptive religious coping techniques. Given that conservative religious beliefs have been identified as deterrents to successful counseling and mental health services for self-harming behavior in sexual minority youth (Tang, 2014), identifying and working with SMW’s religious beliefs may assist in therapy addressing suicide.

Limitations and Future Directions

Despite the current study’s implications for coping intervention and suicide research in SMW, there are several limitations and as a result, directions for future research. First, the data are cross-sectional, so causation cannot be inferred. Although coping styles can influence suicidality, suicidality can influence the type of coping strategies SMW draw on, and potential third variables could account for this relationship. Future research should use longitudinal cross-lagged panel designs to tease apart causality in the relationships identified in this study. Second, suicide attempts for participants in the present study may have occurred as recently as a few days before data collection or years prior. In light of seminal research suggesting that coping strategies employed by an individual change in response to specific life stressors (Folkman, Lazarus, Dunkel-Schetter, DeLongis, & Gruen, 1986), predicting lifetime suicide attempts from current coping strategies may or may not have accurately reflected the relationship between these two variables. To address this issue, future research should inquire about recent suicide attempts only to capture the relationship between current coping strategies and recent suicide attempts, although of course measures of recent attempts would be extremely skewed toward zero; longitudinal research looking at the power of coping styles to predict suicide attempts over time may be the only other more valid approach. Similarly, religious coping was positively associated in the current study with lifetime suicide attempts but not suicidal ideation. Because of the suicidal ideation computation algorithm, current and recent ideation was much more highly weighted in this variable than more remote ideation. Perhaps participants in this study currently engage in more adaptive religious coping strategies than they had in the past as they are better able to integrate over time certain religious beliefs that could conflict with diverse sexual orientations. As a result, future research should examine positive and negative religious coping styles, as opposed to religious coping more generally.

Another limitation to the present study is generalizability of results. Although the sample is nationally representative and diverse in regards to sexual orientation and race/ethnicity, the majority of participants identified as highly educated and reported a relatively high family income ($60,000–$199,999). For this reason, results from the present study are limited in their generalizability to SMW reporting lower educational attainment and/or lower family income.

As well, the present study did not include religious affiliation demographic questions, so the influence of specific religions that are supportive or condemning of same-sex attraction cannot be parsed out. Inclusion of measures of religious affiliation, religious messages received, and religious identity conflict in future research could further elucidate the differing pathways between religious coping styles and suicide risk.

Further, measures of internalized heterosexism were not included in this study, so potential mediating relationships between religious coping styles and psychological distress or suicide risk cannot be ascertained by the present study’s findings. Further research into religious coping styles should also include nonreligious spiritual coping processes, as spirituality has been identified as a protective factor among LGBT populations (Kidd & Witten, 2008; Smith & Home, 2007), an element of religious coping that was not investigated in the present study. Future research on coping styles in SMW should additionally investigate potential correlates of different coping styles (e.g., interpersonal skills, social environment, experiences with stigma) to examine how adaptive and maladaptive coping styles develop among SMW. Finally, given the prevalence of women of color in the study’s sample, future research with larger sample sizes is warranted that examines how race and sexuality may either be differentially associated with coping styles or may produce interaction effects on suicidality. Unfortunately, the present sample did not have sufficient participants in each racial/ethnic group to conduct analyses by racial/ethnic group.

Conclusion

The current study contributes to the growing body of research on adaptive and maladaptive coping styles and suicide risk among SMW. Given the increased rates of suicidal ideation and lifetime attempts by SMW, research on coping styles that may ameliorate or increase suicide risk is greatly needed and has not been examined in prior research. This study highlights how self-blame, religious, and venting coping are associated with increased suicide risk. These findings provide strong directions for future research on interventions to reduce suicide risk among SMW.