A 3-step theory informed examination of PTSD, interpersonal challenges, hopelessness, and suicidal ideation in women veterans with military sexual trauma

ABSTRACT

U.S. military veterans are at a greater risk of suicide than non-veteran adults, and this risk is exacerbated in women veterans who have experienced military sexual trauma (MST). The Three-Step Theory (3ST) is an evidence-based model which describes the transition from suicidal ideation (SI) into action. It suggests that SI arises when physical and/or psychological pain is paired with hopelessness; SI increases when the pain and hopelessness outweigh feelings of connectedness. Informed by the 3ST, this paper examines PTSD, hopelessness, disconnection, and SI in 220 women veterans with MST who completed baseline assessments for a randomized clinical trial comparing Skills Training in Affective and Interpersonal Regulation (STAIR) to Present-Centered Therapy (PCT). In partial support of the 3ST, we found that higher PTSD severity was related to higher SI when veterans reported high hopelessness. However, we did not find that interpersonal problems exacerbated this interaction. Instead, we found that higher interpersonal problems were related to higher SI when patients reported high hopelessness. Findings highlight the importance of instilling hope, in addition to improving PTSD symptoms and interpersonal relationships, in reducing suicide risk.

Introduction

Suicide is among the most common causes of preventable mortality in the U.S., particularly among military veterans, with approximately 6,000 suicides completed each year (U.S. Department of Veteran Affairs, 2022). Several suicide theories have been developed in order to understand why people experience suicidal ideation (SI) and then take action to kill themselves. One of the most comprehensive theories, the 3-Step Theory (3ST; Klonsky & May, 2015), may be particularly applicable to women veterans who have experienced MST given the relationship challenges and psychological pain often present for them. The 3ST theory was developed in order to better understand factors that influence a person to move from SI to attempt (i.e., ideation to action). It is parsimonious with only four factors: pain, hopelessness, connectedness, and capability for suicide, but leaves room for flexible examination of those factors (Klonsky et al., 2021). For example, any form of pain (e.g., mental health, medical condition) can contribute to Step 1. In this study, PTSD symptoms are examined as the pain in Step 1. When a person experiences pain and feels hopeless (i.e., they do not see a path out of the pain) they begin to experience SI as a way out. Step 2 states that SI intensifies when pain exceeds feelings of connection. Again, connection is a broad construct that includes friends, family, job, and purpose. In this study, we examine interpersonal problems as a measure of disconnection with friends and family. In Step 3, a person progresses from ideation to attempt when SI is combined with a capability (dispositional, acquired, or practical contributor) for suicide (Klonsky et al., 2021).

Components of the 3ST theory have been examined in a range of populations including both veteran (Sokol et a., 2021) and civilian (Klonsky et al., 2018) samples and with diverse measures of pain and disconnection. Components of the 3ST have not been examined in a population of women veterans who have experienced military sexual trauma (MST), which is a particularly applicable population given their high levels of PTSD and interpersonal challenges. In this secondary analysis, we examine the first two steps of the 3ST theory with 1) PTSD symptoms combined with hopelessness and 2) the impact of interpersonal problems.

About 44.2% of female Veterans screen positive for MST (Nichter et al., 2022), and 75% report SI after experiencing MST (Monteith et al., 2019), although this is likely an underestimation. With an estimated two in five women service members experiencing MST and SI, it is critical to understand, intervene, and reduce these rates to zero. MST includes psychological or physical sexual trauma or harassment during military service (Gross et al., 2019). Women with MST represent a population that is particularly vulnerable to SI due to their higher rates of PTSD and interpersonal difficulties; two well-established risk factors for suicide (Decker et al., 2021; Klonsky et al., 2021). Further, military culture, which may discourage speaking out about sexual assault due to fear of retaliation or negative consequences for self rather than the perpetrator, may further feelings of disconnection (Stander & Thomsen, 2016) and SI. This is especially poignant in women with MST, given the established tendency of members of this population to isolate themselves from friends and family upon military discharge (Mattocks et al., 2012) Better understanding of suicide risk in this high-risk population is essential to inform prevention and treatment efforts and mitigate deaths.

Step 1 of the 3ST proposes that pain combined with hopelessness leads to SI. Hopelessness directly predicts worse PTSD symptom outcomes (Gallagher & Resick, 2012) and mediates the relation between PTSD and suicide attempts (Boffa et al., 2018). Specific experiences related to trauma (e.g., dissociation; Ozdemir et al., 2015) increase hopelessness and subsequent suicidal thoughts and behaviors (STB). Overall, research supports a strong connection between PTSD and STB if hopelessness is present.

Hopelessness also directly impacts SI independent of mental health factors (Panagioti et al., 2012), with those high in hopelessness being 11 times more likely to complete suicide than those low in hopelessness (Beck et al., 1975, 2006). Female veterans who have experienced interpersonal trauma report hopelessness and interpersonal problems, in addition to a multitude of other stress symptoms (Conard et al., 2014; Luterek et al., 2011).

According to Step 2 of the 3-Step Theory, disconnection will exacerbate SI stemming from pain and hopelessness. Maintaining social support and connection is a protective factor against developing PTSD and STB. The reverse is also true; those without social support are more likely to report worse PTSD symptoms and stronger STB. In a 2012 study on MST, veteran women reported that they chose to intentionally hide their traumatic experiences from friends and family (Mattocks et al., 2012). Military culture may inadvertently encourage moving past MST without seeking support in order to keep peace and avoid retaliation (Stander & Thomsen, 2016), with 80–90% of MSTs going unreported (Conard et al., 2014). Those with low perceived post-deployment support report more severe SI (Monteith et al., 2018). The 3ST highlights disconnection as a risk factor, along with PTSD and hopelessness, for SI.

Despite research examining pairs of these constructs (i.e., PTSD, hopelessness, interpersonal problems, and STB) the relationship between all of these constructs, in line with the 3ST Theory, in women with MST has not yet been evaluated. This project aims to fill this gap by investigating the 3ST in women with MST through a secondary analysis of baseline data from an existing RCT. In line with Step 1 of the 3ST, we hypothesize that PTSD symptoms and hopelessness will interact to predict SI, such that women with higher PTSD symptoms and higher hopelessness will have higher SI. In line with Step 2, we hypothesize that higher interpersonal problems will exacerbate the relation between PTSD and hopelessness leading to more severe SI.

Method

Participants

This manuscript analyzes data from the baseline assessments of 218 women veterans who experienced military sexual trauma and enrolled in an RCT (parent study was a trial comparing skills training in affective and interpersonal regulation (STAIR) to present-centered therapy (PCT)) (Cloitre et al., 2024). No new participants were contacted and analyses were conducted solely on previously collected baseline data. Veterans (M age = 45.4, SD = 13.1) had to be at least 18 years or older, screen positive for MST, and screen positive on a PTSD screen (i.e., PC-PTSD-5 score > 3).

Procedures

For the parent study, participants were recruited through referrals from clinical staff and flyers. Study staff reviewed their medical charts, contacted them by phone to complete a verbal informed consent and phone screen, which included providing information about the study and answering any questions. Eligible women veterans completed an informed consent, assessment battery and two clinical interviews either by phone, through video conference, or in-person at one of two west coast Veterans Affairs (VA) hospitals, then were randomized to STAIR or PCT. Participants were excluded from the baseline assessment if they did not consent to the study, had an active substance use disorder, current psychotic symptoms, unmanaged mania, cognitive impairment, or current involvement in a violent relationship defined as more than casual contact. For this secondary data analysis, no new participants were contacted and analyses were conducted solely on previously collected baseline data. The (masked for review) Institutional Review Board approved the protocol. This study was registered as a clinical trial at clinicaltrials.gov (identifier: masked for review).

Measures

Demographics Questionnaire. Included age, gender identity, sexual orientation, ethnicity, race, education status, employment status, housing status, caregiver status, and current therapy engagement.

PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2018) is a 20-item self-report measure for assessing PTSD symptoms. All items were rated on a 5-point scale indicating the severity of each symptom (0 = not at all to 4 = extremely). A score of 33 or higher for the sum of all 20 items indicates probable PTSD (Weathers et al., 2018). The full PCL-5 scale in this sample had excellent internal consistency (α = .90).

Inventory of Interpersonal Problems (IIP-32; Barkham et al., 1996) is a 32-item self-report measure that assesses interpersonal problems (e.g., I fight with other people too much.) and disconnection (e.g., It is hard for me to make friends.). Respondents rate the level of distress on a 5-point scale from 0 = not at all to 4 = extremely. Higher scores indicate more severe distress and disconnection. Internal consistencies in this sample were excellent (α = .90).

Beck Depression Inventory-II (BDI-II; Beck et al., 1996) is a 21-item self-report measure that assesses the severity of depressive symptomatology. Each response indicates the extent of a specific symptom experienced in the past 2 weeks based on a 4-point scale. Item 2 (0 = I do not feel like a failure to 3 = I feel I am a complete failure as a person) was used to assess hopelessness, and Item 9 (0 = I don’t have any thoughts of killing myself to 3 = I would kill myself if I had the chance) was used to assess SI. Internal consistencies in this sample were excellent (α = .93).

Data analytic strategy

The analyses described in this manuscript were designed for this secondary data analysis of baseline data and are distinct from the parent RCT. Data were analyzed using PROCESS (Hayes, 2013) in SPSS version 28.0. Ethnicity and race demographic variables were included in all analysis. All variables were centered within PROCESS. To test our hypotheses, and the first component of the 3-Step Theory (i.e., psychological pain coupled with hopelessness leads to higher SI), we examined a model with SI as the dependent variable and PTSD severity, hopelessness, and the interaction between the two as the independent variables. To examine the second component of the 3-Step Theory (whether lower interpersonal problems buffer this interaction), we added the measure of interpersonal problems to the model. This model included PTSD, hopelessness, interpersonal problems, all two-way interactions, and the three-way interaction between PTSD, hopelessness, and interpersonal problems. A significant two-way interaction indicates that more severe PTSD is related to higher SI at higher levels of hopelessness. A significant 3-way interaction suggests that lower interpersonal problems serve as a protective factor in the relation between PTSD, hopelessness, and SI. Alpha was set at p = .10 for interaction terms to increase sensitivity to identify differences in significant simple slopes.

Results

Participants were 23 to 77 years old and mostly female (n = 217, 98.6%) with one transgender male-to-female participant (n = 1, 0.5%). Personally identified race/ethnicity among participants was 57.7% (n = 127) White, 20.5% (n = 45) Black or African American, 15.0% (n = 33) Latinx, 7.3% (n = 16) Asian, and 4.1% (n = 9) other race. Almost half of the participants were married (n = 98, 44.3%); the remaining were single, (n = 48, 21.7%), divorced (n = 67, 30.3%) or widowed (n = 6, 2.7%). Over half identified as straight or heterosexual (n = 168, 76.4%). The majority had some college or a 2-year degree (n = 104, 47.3%), and the rest earned a 4-year college degree (n = 63, 28.6%), a postgraduate degree (n = 63, n = 28.6%), or a high school diploma (n = 15, 6.8%). Veterans reported moderate to severe PTSD symptoms (M = 52.07, SD = 15.89), severe interpersonal problems (M = 60.26, SD = 18.70), average hopelessness (M = 1.27, SD = 0.81), and low SI (M = 0.45, SD = 0.54) on average.

Full model results are displayed in Table 1. Step 1 of the 3-Step Theory was supported with the significant interaction between PTSD and hopelessness predicting SI (b = 0.005, p = .08). Simple slopes are displayed in Figure 1. At high hopelessness, there was a significant positive interaction between PTSD and SI (b = 0.01, p = .003). Therefore, individuals with more severe PTSD and higher hopelessness had higher SI. However, at low hopelessness, PTSD was not significantly related to SI (b = 0.007, p = .86). To examine Step 2 of the 3ST Theory, we tested a three-way interaction between PTSD, hopelessness, and interpersonal problems, which was non-significant (b = −0.0002, p = .31). Therefore, we did not find any indication that interpersonal problems exacerbate this relationship.

Table 1.

Results of 3-Step Theory informed examination of PTSD, interpersonal challenges, hopelessness, and suicidal ideation.

	Step 1: PTSD and hopelessness predicting SI				Step 2: PTSD, hopelessness & interpersonal problems predicting SI				Exploratory: Interpersonal problems and hopelessness predicting SI
Predictors	b	SE	95% CI		b	SE	95% CI				95% CI
			LL	UL			LL	UL	b	SE	LL	UL
Intercept	.067	.249	−.424	.559	.707	.586	−.448	1.864	.260	.267	−.267	.787
PCL	.002	.004	−.006	.010	−.002	.012	−.026	.021	.007**	.003	.002	.012
BDI_2	−.250	.179	−.603	.103	−.940	.546	−2.016	.137	−.370**	.153	−.672	−.067
IIP	–	–	–	–	−.015	.011	−.037	.007	−.008**	.003	−.015	.000
PCL*BDI_2	.005+	.003	−.001	.011	.010	.010	−.009	.030	–	–	–	–
IIP*PCL	–	–	–	–	.000	.000	−.000	.000	–	–	–	–
IIP*BDI_2	–	–	–	–	.016	.009	−.003	.034	.007***	.002	.002	.012
PCL*BDI_2*IIP	–	–	–	–	−.000	.000	−.001	.000	–	–	–	–
Hispanic	.150	.092	−.032	.332	.156	.093	−.028	.340	.150	.092	−.030	.331
White	−.045	.085	−.212	.123	−.074	.085	−.242	.094	−.072	.084	−.237	.092
Black	−.090	.103	−.293	.112	−.097	.103	−.299	.105	−.099	.101	−.299	.101

PCL = PTSD Checklist for DSM-5; BDI_2 = Beck Depression Inventory item 2 hopelessness; IIP = Inventory of Interpersonal Problems.

+ p < .10. * p < .05. ** p < .01 *** p < .001.

Figure 1.

Simple slopes of PTSD and Hopelessness predicting suicidal ideation.

In an exploratory fashion, we examined the two-way interaction between hopelessness and interpersonal problems (b = 0.007, p = .004). Simple slopes are displayed in Figure 2. At high hopelessness, there was a significant positive relation between interpersonal problems and SI (b = 0.006, p = .02) and, at low hopelessness, there was a significant negative relation (b = −0.008, p = .05).

Figure 2.

Simple slopes of interpersonal problems and hopelessness predicting suicidal ideation.

Discussion

Our findings partially supported the 3ST. In line with Step 1, PTSD was positively related to SI in women veterans reporting high hopelessness but not in those reporting low hopelessness. Step 2 was not supported. Instead, we found another 2-way interaction: a positive relation between interpersonal problems and SI when hopelessness was high and negative when hopelessness was low. In interpreting these results, it is important to consider that the findings are constrained by the quality of the measurements, as well as the cross-sectional nature of the study. Nonetheless, these results highlight the significance of hopelessness for women veterans who have experienced MST. When people believe their symptoms or relationships will improve, PTSD and interpersonal problems were either not related or inversely related to SI.

The majority of our sample reported some level of interpersonal problems, which is common for women who have experienced MST. These findings suggest that having high relationship problems and low hopelessness (perhaps believing that interpersonal problems could improve) are associated with lower SI. For these women, it was protective to acknowledge that they had interpersonal problems and recognize that their situation was not hopeless. This highlights the importance of instilling hope throughout treatment with individuals struggling with social functioning. As social connections are often time-intensive to address and improve, therapists can prioritize facilitating hopefulness and building confidence while working with patients on interpersonal skills.

Given the 2-way interaction between interpersonal problems and hopelessness, our results suggest that interpersonal problems may also be a form of pain from Step 1 of the 3ST. For example, if a woman is experiencing interpersonal problems (like a relationship conflict or separation) and feels hopeless (Step 1) but has a strong connection to other friends and her career (Step 2), then she may not experience SI. The flexibility of the 3ST may support innovative examinations of how different components of pain and connection may influence SI. However, conversely, the broad definitions of the constructs within the 3ST may also lead to less precision, creating challenges with testing the model.

Overall, these results imply that a focus on instilling hope is essential in clinical work with women veterans who endorse MST. Practitioners seeing veterans with MST should also assess for interpersonal problems and lack of hope, becoming particularly aware of the increased suicide risk when these factors are present.

Limitations

The primary limitation of this paper is the operationalization of disconnection, hopelessness, and SI. First, the IIP-2 measures interpersonal problems, which was used as a proxy for disconnection in this project. The use of more robust measures of social connectedness and loneliness may provide a better understanding of the second step of the 3ST. The parent study of this population also did not include specific measures of hopelessness and STB, so BDI-II item 2 (hopelessness) and BDI-II item 9 (SI) were used for each construct. While the BDI-II is a well-researched and validated measure, using a single item to capture complicated constructs detracts from the internal validity of the study. The BDI-II item 9 also does not capture suicidal behaviors. Although the BDI-II item 9 has been used in numerous suicide-focused studies (Beck & Steer, 1991; Pelizza et al., 2019), more thorough measures of STBs are needed. Additionally, the data used in this secondary analysis is cross-sectional; therefore, we are unable to draw conclusions about the directionality of the relationships between SI, hopelessness, disconnection, and PTSD severity. Lastly, our sample is limited to treatment-seeking women veterans who are in outpatient treatment, so results may not generalize to those outside of this population, including the broader population with STBs.

Future directions

Future intervention research may focus on incorporating strategies for instilling hope in women veterans who have experienced MST. This might look like values-based work from Acceptance and Commitment Therapy, such as envisioning an ideal future. Although this is often challenging for people who have experienced trauma, learning more about hopelessness in this population is also imperative for designing these interventions. Studies may also evaluate whether women veterans are experiencing hopelessness prior to the MST. Many women who experienced MST also experienced childhood trauma or sexual assault (Doucette et al., 2023). Identifying the severity of hopelessness may inform therapeutic strategies and session time dedicated to addressing this challenging emotion.

Additionally, incorporating interpersonal skills into PTSD treatments may also better serve the needs of women with MST. Projects may include large-scale evaluations of STAIR Narrative Therapy (Cloitre et al., 2024), which includes an additional narrative exposure component to maximize both skills-based and exposure-based benefits. Researchers may also consider evaluating the addition of a biweekly interpersonal skills group while completing trauma-focused PTSD treatments, such as Prolonged Exposure. Future research should also evaluate suicide risk in women with MST using more thorough measures of STBs and hopelessness, such as the Beck Hopelessness Scale and the Beck Scale for Suicidal Ideation or Columbia Suicide Severity Rating Scale. Examining the 3-Step Theory in women veterans longitudinally would also provide more insight on the components of each step and whether increasing connection or reducing pain supports the 3ST. Ultimately, suicide continues to be a leading cause of death among veterans; thus, continued research on suicide-related processes and related interventions in at-risk populations, including women veterans with MST, is vital.

Funding Statement

This work was supported by a Career Development Award from the Department of Veteran Affairs, Office of Research and Development, Rehabilitation Research and Development Service under Grant RX003490, and parent study was funded by the Department of Veteran Affairs, Health Services Research and Development under Grant HX002211.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data that support the findings of this study are available from the corresponding author, C.K., upon reasonable request.

Public service statement

This study suggests that higher hopelessness combined with PTSD or interpersonal problems in women veterans who have experienced Military Sexual Trauma leads to higher SI, while lower hopelessness may be protective. Instilling hopefulness while reducing PTSD and relationship distress may be particularly helpful.