Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 14.
Published before final editing as: Psychol Serv. 2026 Jan 12:10.1037/ser0001016. doi: 10.1037/ser0001016

Greater Reactivity to Stressors as an Indicator of Service Utilization and Need in Veteran Women with Sexual Assault Histories

Allison Metts 1,2,3, Rahel Pearson 1,2, Kathryn Roe 1,2,3, Adam McGuire 1,2,4, Caron Zlotnick 5, Suzannah K Creech 1,2,6
PMCID: PMC12798693  NIHMSID: NIHMS2126739  PMID: 41525399

Abstract

Objective:

Veteran women with trauma histories are at greater risk for trauma-related psychopathology and continued stressor exposure. Given that services and resources for veterans are in high demand, it is imperative to identify factors relevant to veterans who are of high need and likely to engage with treatment. This study examined treatment utilization in veteran women who were more reactive (i.e., higher need) versus resilient (i.e., lower need) to stressors.

Method:

Veteran women (n=153) with sexual assault histories who took part in a randomized clinical trial were assessed three times over four months. Stressor reactivity (SR) was calculated by regressing PTSD symptoms onto stressor exposure. Outcomes were service utilization indices. Mixed effect models examined between-person (averages) and within-person SR (deviations from person averages) effects relating to outcomes over time.

Results:

Between-person SR positively related to treatment attendance (per self-report and chart review) and resource areas that veterans reported working on and needing to work on, with evidence of small to medium effects. There was also evidence for between-person SR effects in a subsample with high stressor exposure. There was no evidence of within-person SR effects for these outcomes. There was no evidence of SR effects relating to success at obtaining resources.

Conclusions:

Results indicate that veterans in highest need of treatment (i.e., more reactive) are also more likely to attend treatment and identify areas for potential intervention. Building on these results, identifying stressor reactive veterans by monitoring of stressors and symptoms may enhance care connection and prevent entrenched stressor-related psychopathology.

Keywords: stressor reactivity, trauma, stressors, treatment utilization, veteran women

Veteran women are disproportionately affected by stressors (i.e., occurrence of life events), including sexual trauma and interpersonal violence (IPV) (Booth et al., 2011; Dichter et al., 2011; Kimerling et al., 2010; Maguen et al., 2012; Pavao et al., 2013). Notably, stress (i.e., individual response to life event occurrence) experienced from IPV can be from in person interactions or through online spaces (termed “cyber IPV”), in the form of cyber stalking or digital harassment. Cyber IPV can be pervasive, impact one’s sense of safety and privacy, and is associated with negative mental health outcomes (Rogers et al., 2023). Further, those with sexual assault histories tend to have more extensive prior trauma, are at increased risk for future stressor exposure, and utilize services more compared to those without such histories (Bergman & Brismar, 1991; Davis et al., 2023; Zinzow et al., 2008). In addition, veteran women are four times more likely than non-veteran women to become homeless, with trauma exposure identified as a risk factor (Hamilton et al., 2011). Trauma exposure and related distress are also associated with involvement in the justice system and occupational impairment in veteran women (Rosenthal & Finlay, 2022; Schnurr & Lunney, 2011; Sienkiewicz et al., 2020) and unemployment in all veterans (Sripada et al., 2018). As such, there is a need to identify modifiable factors that can inform mental health care and resource connection in this population.

Stressor exposure is a robust predictor of mental health disorders, including a diagnosis of posttraumatic stress disorder (PTSD) (Goldzeweig et al., 2006; Haskell et al., 2011; Resnick et al., 1993; Suris & Lind, 2008; Stevens et al., 2021; Zinzow et al., 2008). Veteran women with PTSD diagnoses are also more likely to meet criteria for comorbid mental health conditions and have marked functional impairment (Runnals et al., 2014). However, stressor exposure in and of itself does not predict stressor-related psychopathology. In fact, some individuals are resilient and maintain adaptive functioning or recover following stressor exposure (Bonanno et al., 2005). Further, the level of stressor exposure or one’s ability to cope with such exposure can change over time (Kalisch et al., 2021). Distinguishing between those who are more stressor reactive versus resilient to ongoing stressors could serve as a useful indicator for service allocation and help to identify those veterans who are in high need of such services. Stressor reactivity (i.e., mental health symptom severity adjusting for stressor exposure) could also inform care such that it is modifiable and could be targeted by interventions.

Whereas attending to symptom level distress as an indicator of need is useful, this criterion may not fully capture the range of patients who could benefit from services and resources. For example, those with higher levels of symptoms, without considering levels of stressor exposure, may appear similar according to a symptom screener score. However, if stressor exposure is considered, these are arguably distinct clinical or need profiles. Namely, individuals with low stressor exposure and high symptoms are more reactive than those with high stressor exposure and high symptoms. Whereas both those with high symptoms could benefit from services, considerations related to their care and resource connection could be tailored to stressor reactivity. For example, some veterans may benefit from extending their care beyond interventions focused on disorder-specific symptoms by adding skills to help with coping with stressors and by attending to the need for psychosocial resource support in the case of high stressor reactivity.

Stressor reactivity may provide useful information regarding service utilization as it attends to symptom severity within the context of continued stressor exposure. This metric is particularly relevant for veteran women, as it has previously been found that veteran women are more reactive regarding PTSD symptoms than veteran men when considering military-related stressors (Metts et al., 2024a). However, it is unknown how stressor reactivity may relate to treatment and resource utilization in veteran women. Such knowledge could inform needs assessments within care settings (e.g., primary care, mental health clinics) or prevention efforts and enhance connection to care.

The Present Study

The present study is a secondary analysis that aims to examine associations between reactivity regarding PTSD symptoms against stressors and indicators of service utilization over time in veteran women with sexual assault histories. Of note, hazardous drinking data was collected but not analyzed in the present study due to focus on reactivity to stressors regarding PTSD symptoms. Stressors were operationalized as exposure to various traumatic events, IPV, and cyber IPV. Stressors are conceptualized to capture a broader range of events (e.g., loss of home, interpersonal conflict) that may or may not meet Criterion A trauma.

We study these associations within the context of data collected in a randomized controlled trial (RCT) that tested the impact of a brief computerized intervention (Safe and Health Experiences, SHE) that was designed to reduce health risks (probable PTSD diagnosis, IPV, hazardous drinking) and increase mental health treatment utilization (Creech et al., 2022). RCT results indicated that SHE led to increased treatment utilization compared to the control condition (Creech et al., 2022). We hypothesized that given greater levels of need likely associated with higher reactivity to stressors (i.e., lower resilience), higher average stressor reactivity across four months (i.e., between-person effect) would relate to higher service utilization. In line with this rationale, we also hypothesized that higher stressor reactivity relative to a participant’s average stressor reactivity (i.e., within-person effect) would relate to higher service utilization.

Methods

Participants

Participants were 153 veteran women seeking primary care at a VHA medical center. Of these women, 76 were assigned to the SHE intervention condition and 77 were assigned to the control condition. Veteran women were eligible to participate in this trial, further eligibility criteria included: (1) aged 18–65, (2) sexual assault history (i.e., at least one occurrence of unwanted sexual contact during lifetime), and (3) at least one current psychosocial health risk (probable PTSD diagnosis, hazardous drinking, IPV). Exclusion criteria were: (1) inability to understand study procedures in English and (2) active suicidal or homicidal crisis warranting immediate clinical attention. All study procedures were approved by the Central Texas Veterans Health Care System Institutional Review Board. Participants provided data between May 2017 and April 2019. Detailed demographics of the sample at baseline are in Table 1.

Table 1.

Sample Demographic Characteristics at Baseline (N = 153)

Count/Mean Percentage/SD
Age 43.55 10.10
Race
Caucasian/White 54 35.3
Black 73 47.7
Bi-Racial/Multi-Ethnic 10 6.5
Native American/Native Alaskan 2 1.3
Asian 2 1.3
Native Hawaiian/Other Pacific Islander 1 0.7
Other 9 5.9
Missing 2 1.3
Ethnicity (Hispanic/Latino/a/x/e) 23 15.0
Sexual Orientation
Heterosexual 132 86.3
Lesbian/Gay 10 6.5
Bisexual 7 4.6
Other 4 2.6
Relationship Status
Married 53 34.6
Separated 10 6.5
Divorced 54 35.3
Single, no relationship 20 13.1
Single, in a relationship 14 9.2
Single, same sex partner 2 1.3
Number of Deployments 1.14 1.13
Service-Connected Disability 141 92.2
Actively Seeking Service Connection 58 37.9
Health Risk
IPV 92 60.1
PTSD 43 28.1
Hazardous Drinking 18 11.8
Open in a new tab

Procedure

The present study’s data was collected within a previously conducted randomized controlled trial. Detailed study procedures can be found in the primary report (Creech et al., 2022). Veteran women were recruited via fliers, letters, and in-person at women’s primary care clinics. Those who were interested in participating provided written informed consent prior to completing self-report screening measures. As part of the parent study, veterans were screened for health risks. A positive screen for past-month PTSD was a score of ≥ 33 on the PTSD Checklist for Diagnostic and Statistical Manual of Mental Disorders – 5th Edition (Weathers et al., 2013). A positive screen for past-month hazardous drinking was operationalized as ≥ 4 drinks on a single day using the Graduated Frequency Measure (Hilton, 1989). A positive screen for past-year IPV was operationalized as a score ≥ 4 on the Woman Abuse Screening Tool (Brown et al., 2000). Eligible veterans were randomized to the SHE or control conditions following completion of baseline assessments (see Creech et al., 2022 for randomization details). Participants were compensated for their time at screening and all assessments (baseline, 2-month, 4-month) with gift cards.

Assessments

Participants in both conditions completed a battery of self-report assessments at baseline, 2-month follow-up, and 4-month follow-up. Most assessments were completed in person, with a small subsample (<5%) completing assessments by phone for logistical reasons. In the present study, assessments used in analyses were PTSD symptom severity, trauma history, IPV, cyber-IPV, treatment utilization, and efforts toward obtaining resources.

SHE and Control Conditions

Participants randomized to SHE were presented with 20-minute modules for each health risk (probable PTSD diagnosis, hazardous drinking, IPV) that was relevant to them. Modules included audiovisual presentations and psychoeducational handouts. Module content focused on psychoeducation on health risks and incorporated motivational interviewing techniques. Participants could not proceed to the next section of the intervention without completing the prior section of the intervention. Of the SHE participants who were eligible for a given intervention, 100% completed the hazardous drinking intervention, 91% completed the IPV intervention, and 88% completed the PTSD intervention. See Creech et al., (2022) for full description of SHE intervention content. Participants randomized to the control condition only completed assessments. Participants were provided mental health and IPV referrals in addition to assistance with referrals throughout the study period, regardless of condition.

Measures

Demographic Variables and Military History

Demographic data was collected at time of screening using a form to assess demographic variables of interest, including age, gender, ethnicity, race, marital status, education, and sexual orientation. Military history was collected at baseline assessment using a form that queried about a range of military-related topics, including service connection.

Stressor Exposure

Trauma History.

Trauma history was assessed using items from the Trauma History Screen (THS; Carlson et al., 2005). The THS items assessed exposure to 11 potentially traumatic events at baseline (lifetime) and at follow-up (with reference to the interim period). Participants rated each event with a “Yes” or “No” response. Items specified events that happened broadly. Sample events included “A really bad car, boat, train, or airplane accident” and “Hit or kicked enough to injure.” A sum score was computed ranging from 0 to 11 to indicate total trauma exposure at each timepoint. The THS has evidence of reliability and validity (Carlson et al., 2011). The KR-20 for our sample at baseline was .65, which is consistent with expectations for trauma history measures capturing heterogenous life events.

Interpersonal Violence.

The Composite Abuse Scale (CAS) is a 30-item measure of IPV (Hegarty et al., 2005). This measure was administered at baseline (with reference to the past year) and follow-up assessments (with reference to the interim period). Items specified events that occurred by the participant’s partner. Example items include “My partner beat me up” and “My partner hit or tried to hit me with something.” Participants rated items on a 6-point scale ranging from 0 (Never) to 5 (Daily). Higher scores indicate more abuse. The CAS has evidenced good psychometric properties (Hegarty et al., 2005). Internal consistency across all assessment points in the current sample ranged from α = .90 to .95.

Cyber-Interpersonal Violence.

A 6-item scale that assessed cyber-IPV behaviors was administered as an add-on to the CAS (Creech et al., 2022). The measure assessed such behaviors at baseline (over the past year) and at follow-up assessments (with reference to the interim period). Example items were “My partner became jealous after reading my social networking profile online” and “My partner checked up on my social networking profile (MySpace, Facebook, etc.) to see if someone is flirting with me.” Participants rated items on a 6-point scale ranging from 0 (Never) to 5 (Daily). This measure has been used in previous reports on cyber-IPV (Hailemariam et al., 2023). Internal consistency across all assessment points in the current sample ranged from α = .94 to .98.

Posttraumatic Stress Disorder Symptoms

PTSD symptom severity was assessed using the PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013). The PCL-5 contains 20 items to assess past-month symptoms. Participants rated how much each symptom bother them on a 0 (Not at all) to 4 (Extremely) scale, with higher scores reflecting more severe symptoms of PTSD. The PCL-5 has demonstrated excellent psychometric properties (Bovin et al., 2016). Internal consistency across all assessment points in this sample ranged from α = .95 to .96.

Stressor Reactivity

Stressor Reactivity (SR) was calculated using normative modeling of the sample’s PTSD symptoms relative to stressor exposure as measured by to various traumatic events, IPV, and cyber-IPV (Kalisch et al., 2021). We fit the stressor-PTSD symptom relationship over data from participants from baseline to 4-month follow-up. The stressor-PTSD symptom line was determined by a linear mixed model with random slopes and intercepts for participants. The fixed effect estimates for the slope and intercept were used as estimates of the sample average longitudinal stressor-PTSD symptom relationship, which represents the normative relationship between stressor exposure and PTSD symptoms in our sample (Kalisch et al., 2021). The regression residuals quantified SR, with a negative residual (i.e., negative SR) representing lower-than-expected PTSD symptom severity given an individual’s exposure to various traumatic events, IPV, and cyber-IPV (i.e., higher resilience) and a positive residual (i.e., positive SR) representing a higher-than-expected PTSD symptom severity given an individual’s various traumatic events, IPV, and cyber-IPV (i.e., higher reactivity). The use of SR has been supported in previous research (e.g., Bögemann et al., 2023; Cahill et al., 2022; Metts et al., 2024a; Metts et al., 2024b).

Service Utilization

Treatment Attendance.

Treatment attendance was assessed broadly via self-report using the Treatment Services Review. The Treatment Services Review is comprised of 15 items describing health care treatment use (e.g., individual and group therapy, 12-step group sessions, residential substance abuse treatment, psychological testing, inpatient psychiatric care, physical examination, drug testing). Participants rated attendance for the past two months at each assessment point. This measure is reported as the total number of health care appointments at each timepoint.

Treatment attendance specific to mental health was assessed through chart review methods. Study staff reviewed medical records for participants to tally mental health appointments for the past two months at each assessment point. This measure is reported as the total number of mental health care appointments at each timepoint.

Effectiveness in Obtaining Resources Scale (EOR).

The EOR measures each woman’s effort toward obtaining resources from 11 different community resources such as health care, housing, legal services, or social services over the past two months for each assessment (Sullivan & Bybee, 1999). Past literature has found the EOR to be successful in the evaluation of the level of efficacy of advocacy services used by women with IPV (e.g., Johnson et al., 2008; Stevens et al., 2015). Three subscales are derived from the EOR: (1) need for obtaining resources, (2) work on obtaining resources, (3) success at obtaining resources. On the EOR, participants are asked to select services that apply to each of the three questions (0 = No, 1 = Yes). Higher scores reflect more resources that veterans indicated needing, working on, or successfully obtaining at a given timepoint. The EOR has evidence of reliability (Sullivan & Bybee, 1999).

Data Analysis

RStudio (RStudio Team, 2024) was used for descriptive statistics, reliability analyses, and analyses (mixed effect model, outlier detection) used to derive SR scores. In the calculation of SR, we examined our sample for influential data points to reduce bias in the SR score. Outlier analysis was performed using Mahalanobis distance (Mahalanobis, 1936). Cases (n = 4) were excluded from analyses if chi-square values were significant at the p < .001 level.

Mixed effect models relating SR to treatment utilization outcomes were conducted using Bayesian estimation and Blimp 3 software (Keller & Enders, 2021). There were five focal models with SR relating to five different outcomes: (1) self-reported treatment attendance, (2) chart-reviewed mental health treatment attendance, (3) need for obtaining resources, (4) work on obtaining resources, and (5) success at obtaining resources. Following previous methods (Metts et al., 2024b), all models were repeated in a sample with high stressor exposure (i.e., top two tertiles of data according to average stressor exposure over the study period; n = 98) to examine whether the pattern of results was consistent with a stricter definition of resilience. Analyses were conducted to examine potential covariates to include. Race was significantly negatively correlated with success at obtaining resources (r = −.26, p = .002); as such, race was included as a covariate in models.

The percent missingness on variables of interest over the follow-up period were as follows: 19.7% for SR, 12.5% for EOR variables, 14.5% for self-reported treatment attendance, and 0.7% for race. There was no missing data for chart-reviewed mental health treatment attendance, condition, ethnicity, or marital status. Analyses examining whether demographic variables (marital status, race, ethnicity, sexual orientation, education) or condition were associated with missingness over the study period revealed no significant associations (all ps > .05). Bayesian estimation assumes a conditionally missing at random process. Blimp’s Markov chain Monte Carlo algorithm first estimates all model parameters given the filled-in data from the prior iterations and then uses updated model parameters to estimate missing values.

SR was disaggregated into within-person and between-person components. Given that means were not computed from the same amount of data points due to missing data, latent variable group means were constructed in Blimp for between-person variables (Keller & Enders, 2021). Between-person effects indicated the relationship between treatment utilization and an average level of SR across all assessments. Deviations at each time point from the latent mean represented the within-person component. Within-person effects indicated the relationship between treatment utilization and changes in the level SR compared to their average SR scores across all assessments.

Models also included effects of time (centered at study endpoint), condition (0 = control, 1 = SHE), race (0 = races other than White; 1 = White), time by condition interaction to adjust for treatment effects over time, grand mean centered latent mean SR (between-person effect), and group mean centered SR (within-person effect). We tested different iterations of models with random intercept and random slopes (time, SR) and determined the best model based on fit (Marginal Likelihood WAIC; Conditional Likelihood WAIC), with lowest Widely Applicable Information Criterion (WAIC) values indicating best fit. All models included the random intercept for participant. The random slopes of time and SR were included in EOR models.

For treatment attendance variables, the data for self-report (TSR) and chart-review variables were positively skewed with some zero inflation. Following methods used previously for these variables in this sample (Creech et al., 2022), variables at each assessment point were transformed to categorize participant treatment attendance according to four levels ranked according to frequency: 0 (No treatment attended), 1 (Attended treatment once a month on average), 2 (Attended treatment more than once a month on average but not more than weekly), 3 (Attended treatment more than weekly). These ordinal variables were used as outcomes.

In Blimp analyses, the potential scale reduction factor diagnostic (Gelman & Rubin, 1992) was used to establish the algorithm’s convergence (i.e., initial burn-in period). All analyses used 50,000 iterations following the initial burn-in period. This is an iterative process that produces a distribution of estimates for each model parameter. The center and spread of each distribution (i.e., posterior median and standard deviation) serve as point estimates and measures of uncertainty that are analogous to frequentist point estimates and standard errors. Standardized coefficients of the posterior medians (abbreviated as Mdn.) and corresponding standard deviations (abbreviated as SD) are reported.

Effect size is based on standardized regression coefficients, with 0.10–0.29 representing small effects, 0.30–0.49 representing medium effects, and 0.50 or greater representing large effects (Nieminen, 2022). Adjustment to p-values using the Benjamini-Hochberg false discovery rate procedure (Benjamini & Hochberg, 1995) produced analog q-values to correct for multiple comparison in the full sample analyses.

Results

Descriptive Statistics

Most participants identified as Black and non-Hispanic/Latino/a/x/e. Regarding sexual orientation and relationship status, heterosexual and divorced were the most endorsed responses. IPV represented the most common health risk in the sample. Most participants reported having a service-connected disability. Stressor exposure descriptives by timepoint are presented in Supplemental Table 1.

Self-Reported Treatment Attendance

Full Sample

Table 2 presents a summary of the results. Table 3 contains results for models ran in the full sample. There was a significant medium positive between-person effect of SR (Mdn. = 0.36, q < .001), indicating that higher average SR related to more frequent treatment attendance over four months per self-report. The within-person effect was non-significant (q = .517). There was also a significant small negative effect of race, indicating that White veteran women had lower treatment attendance at the study endpoint, adjusting for other covariates. There was evidence for small effect of time, indicating that treatment attendance decreased over four months, adjusting for other covariates. The effects of treatment condition and the time by condition interaction were non-significant.

Table 2.

Summary of Findings

Outcome Significant SR Effect in Full Sample (Effect Size) Significant SR Effect in Sample with High Stressor Exposure (Effect Size)
Treatment attendance (self-report) Between-person (0.36) Between-person (0.19)
Treatment attendance (chart review) Between-person (0.29) Between-person (0.20)
Need for Obtaining Resources Between-person (0.35) Between-person (0.29)
Working on Obtaining Resources Between-person (0.22) No
Success at Obtaining Resources No No
Open in a new tab

Note. Effect sizes are standardized point estimates.

Table 3.

Fixed Effect Estimates from Mixed Effect Models in Full Sample

Self-reported treatment attendance
Mdn. (SD) 95% CI p-value
SR (within) −0.05 (0.05) −0.15, 0.04 .279
SR (between) 0.36 (0.08) 0.19, 0.50 <.001
Time −0.14 (0.05) −0.23, −0.04 .005
Condition 0.12 (0.08) −0.04, 0.28 .144
Race −0.20 (0.07) −0.33, −0.06 .005
Time*Condition 0.01 (0.06) −0.11, 0.13 .849
Chart-reviewed treatment attendance
Mdn. (SD) 95% CI p-value
SR (within) 0.03 (0.05) −0.06, 0.12 .469
SR (between) 0.29 (0.08) 0.12, 0.44 <.001
Time −0.14 (0.05) −0.24, −0.05 .003
Condition 0.20 (0.08) 0.04, 0.35 .012
Race −0.20 (0.07) −0.33, −0.06 .005
Time*Condition 0.14 (0.06) 0.03, 0.26 .016
Need for Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) 0.09 (0.06) −0.03, 0.21 .130
SR (between) 0.35 (0.07) 0.20, 0.48 <.001
Time −0.06 (0.05) −0.17, 0.04 .260
Condition −0.07 (0.07) −0.21, 0.08 .380
Race −0.05 (0.06) −0.17, 0.08 .445
Time*Condition −0.10 (0.07) −0.23, 0.02 .113
Working on Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) 0.06 (0.06) −0.06, 0.19 .310
SR (between) 0.22 (0.08) −0.06, 0.19 .004
Time −0.17 (0.06) −0.28, −0.06 .002
Condition −0.13 (0.08) −0.27, 0.03 .104
Race −0.11 (0.06) −0.22, 0.02 .094
Time*Condition −0.04 (0.07) −0.17, 0.10 .596
Success at Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) −0.01 (0.07) −0.14, 0.13 .939
SR (between) 0.08 (0.07) −0.05, 0.22 .216
Time −0.09 (0.06) −0.22, 0.03 .144
Condition −0.11 (0.07) −0.25, 0.04 .149
Race −0.16 (0.06) −0.27, −0.04 .006
Time*Condition −0.04 (0.08) −0.19, 0.10 .563
Open in a new tab

Note. Mdn. = Posterior median (i.e., standardized point estimate). SD = standard deviation; CI = credible interval.

Sample with High Stressor Exposure

Results for models ran in the sample with high stressor exposure are reported in Table 4. The pattern of results was consistent in the sample with high stressor exposure, with evidence of a positive small between-person effect (Mdn. = 0.19, p = .045). The within-person effect was non-significant (p = .341). Effects of time and race were also consistent with results from the full sample in terms of significance and direction. The effect of treatment condition was significant, indicating that veteran women with higher stressor exposure in the SHE condition had higher treatment attendance at the study endpoint, adjusting for other covariates. The time by condition interaction was non-significant (p = .370), consistent with the full sample results.

Table 4.

Fixed Effect Estimates from Mixed Effect Models in Sample with High Stressor Exposure

Self-reported treatment attendance
Mdn. (SD) 95% CI p-value
SR (within) −0.06 (0.06) −0.17, 0.06 .341
SR (between) 0.19 (0.10) −0.00, 0.38 .045
Time −0.18 (0.06) −0.30, −0.05 .005
Condition 0.20 (0.10) 0.00, 0.39 .044
Race −0.23 (0.08) −0.38, −0.06 .006
Time*Condition 0.07 (0.08) −0.08, 0.22 .370
Chart-reviewed treatment attendance
Mdn. (SD) 95% CI p-value
SR (within) −0.01 (0.06) −0.13, 0.11 .849
SR (between) 0.20 (0.10) −0.00, 0.38 .046
Time −0.12 (0.06) −0.25, 0.00 .055
Condition 0.24 (0.10) 0.04, 0.42 .016
Race −0.24 (0.08) −0.39, −0.07 .004
Time*Condition 0.14 (0.08) −0.01, 0.29 .071
Need for Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) 0.08 (0.07) −0.06, 0.22 .253
SR (between) 0.29 (0.09) 0.11, 0.44 .001
Time −0.09 (0.07) −0.23, 0.05 .198
Condition −0.02 (0.10) −0.021, 0.17 .855
Race −0.08 (0.08) −0.23, 0.08 .339
Time*Condition −0.08 (0.09) −0.25, 0.09 .361
Working on Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) 0.08 (0.07) −0.07, 0.22 .299
SR (between) 0.09 (0.09) −0.09, 0.26 .308
Time −0.18 (0.07) −0.32, −0.03 .014
Condition −0.11 (0.10) −0.31, 0.09 .280
Race −0.15 (0.08) −0.30, 0.00 .051
Time*Condition −0.05 (0.09) −0.23, 0.12 .562
Success at Obtaining Resources
Mdn. (SD) 95% CI p-value
SR (within) −0.05 (0.08) −0.21, 0.11 .515
SR (between) 0.00 (0.08) −0.15, 0.16 .994
Time −0.11 (0.08) −0.27, 0.05 .177
Condition −0.12 (0.09) −0.29, 0.06 .179
Race −0.18 (0.07) −0.31, −0.04 .010
Time*Condition −0.02 (0.10) −0.21, 0.17 .857
Open in a new tab

Note. Mdn. = Posterior median (i.e., standardized point estimate). SD = standard deviation; CI = credible interval.

Chart-Reviewed Mental Health Treatment Attendance

Full Sample

There was a significant medium positive between-person effect of SR (Mdn. = 0.29, q < .001), indicating that higher average SR related to more frequent mental health treatment attendance over four months per chart review. The within-person effect was non-significant (q = .586). There was also a significant small negative effect of race, indicating that White veteran women had lower mental health treatment attendance at the study endpoint, adjusting for other covariates. The time by condition interaction (Mdn. = 0.14, p = .016) was significant, indicating that veteran women in SHE had higher mental health treatment attendance at the end of the follow-up period and a more positive change in this outcome over time (Table 3).

Sample with High Stressor Exposure

The pattern of results for effects of SR were consistent with evidence for a significant medium positive between-person effect (Mdn. = 0.20, p = .046) but no evidence for a within-person effect of SR (p = .849). The effect of race was also consistent. The effects of treatment condition and time, however, were inconsistent. In this subsample, there was only evidence for an effect of treatment condition (Mdn. = 0.24, p = .016), indicating that veteran women with higher stressor exposure in the SHE condition had higher treatment attendance at the study endpoint, adjusting for other covariates (Table 4).

Need for Obtaining Resources

Full Sample

There was a significant medium between-person effect of SR (Mdn. = 0.35, q < .001), indicating that higher average SR over four months related to higher need for resources. The within-person effect was non-significant (q = .517). The effects of time and race were non-significant. The effect of treatment condition and the time by condition interaction were also non-significant (Table 3).

Sample with High Stressor Exposure

The pattern of results was consistent in the sample with high stressor exposure, with evidence of a positive small between-person effect of SR (Mdn. = 0.29, p = .001). The within-person effect of SR was non-significant (p = .253). Effects of time, race, condition, and time by condition interaction were also non-significant (Table 4).

Work on Obtaining Resources

Full Sample

There was a significant small positive between-person effect of SR (Mdn. = 0.22, q = .005), indicating that higher average SR over four months related to more work toward obtaining resources. The within-person effect was non-significant (q = .517). There was evidence for small effect of time, indicating that work toward obtaining resources decreased over four months, adjusting for other covariates. The effects of race, condition, and time by condition were non-significant (Table 3).

Sample with High Stressor Exposure

The pattern of results was consistent in the sample with high stressor exposure in terms of direction. However, the between-person effect of SR reduced to non-significance (Mdn. = 0.09, p = .308). The within-person effect of SR was non-significant (p = .253). The effects of time, race, condition, and time by condition were all non-significant (Table 4).

Success at Obtaining Resources

Full Sample

The between-person effect of SR (q = .216) and within-person effect of SR (q = .939) were non-significant. There was a significant effect of race, indicating that White veteran women had lower success at obtaining resources at the study endpoint, adjusting for other covariates. The effects of time, condition, and time by condition were all non-significant (Table 3).

Sample with High Stressor Exposure

Consistent with the full sample results, there was no evidence of between-person or within-person effects of SR or significant effects of time, condition, and time by condition. The effect of race was consistent with the full sample results (Table 4).

Discussion

The present study examined associations between reactivity to stressors regarding PTSD symptoms and service utilization over four months in veteran women with histories of sexual assault, adjusting for treatment condition effects on service utilization. Notably, the effect of treatment condition and the interaction of treatment condition by time were significant when examining chart-reviewed mental health treatment attendance outcomes in the full sample, suggesting that veteran women in SHE had a more favorable trajectory of mental health treatment attendance across the 4-month study period compared to controls. There was additional evidence for an effect of treatment condition in the same direction when examining self-reported and chart-reviewed treatment attendance as an outcome in the sample with high stressor exposure.

Regarding our hypotheses, we found evidence for higher reactivity to stressors relating to higher treatment attendance at the between-person level, indicating that veteran women who displayed higher PTSD symptoms on average, adjusting for stressor exposure, were more likely to engage with treatment (broadly and mental health specifically) over four months. We also found that higher reactivity to stressors at the between-person level was associated with more work toward obtaining resources and higher need for obtaining resources, indicating that veteran women who displayed higher PTSD symptoms on average, adjusting for stressor exposure, reported more efforts toward and greater need for resources over the study period. The observed between-person associations between stressor reactivity and treatment utilization reflects stable individual differences in stressor reactivity that predicted service use, above and beyond the effects of treatment condition, time, and their interaction. In other words, even after accounting for intervention-related changes, individuals who were generally more reactive to stressors across the study period were more likely to utilize services than those who were less reactive. Result patterns were consistent overall in terms of direction with some evidence of attenuated effect sizes in the sample with high stressor exposure. This consistency suggests that treatment attendance and need for resources were robust associations that may be of particular importance to veteran women with traumatic distress in the context of elevated stressor exposure.

Higher average trauma-related distress, adjusting for stressor exposure, related to higher treatment attendance over four months. Effects were medium in strength and consistent when considering self-report and chart-reviewed outcomes. These findings indicate that veteran women who displayed a higher severity of PTSD symptoms than expected given their levels of stressor exposure were attending treatment at a higher frequency than those with lower symptoms than expected given their level of stressor exposure. Further, in the sample with high stressor exposure, the pattern of results was consistent, with evidence of significant positive associations between stressor reactivity and treatment attendance in the sample with high stressor exposure. This sample with high stressor exposure result supports the notion of stressor reactivity serving as an indicator of short-term service engagement in more vulnerable veterans. These results are in line with previous research indicated that higher symptom distress is associated with higher service utilization (Farmer et al., 2020; Ouimette et al., 2003; Ryan et al., 2015). We add to previous work by adjusting for stressor exposure and examining reactivity to stressors, as opposed to solely symptom level associations. Given that higher stressor reactivity can be associated with impairment, including within veteran interpersonal relationships (Metts et al., 2024a), stressor reactivity may be a useful tool to inform engagement with and need for treatment beyond symptom level information.

Higher average trauma-related distress, adjusting for stressor exposure, also related to more work toward obtaining resources and higher need for obtaining resources over four months. The effects for work toward obtaining resources and need for obtaining resources were small and medium in size, respectively. In the sample with high stressor exposure, the pattern of results was consistent, with further evidence of a significant positive association between stressor reactivity and need for obtaining resources in the sample with high stressor exposure. This result in the sample with high stressor exposure supports the notion of stressor reactivity serving as an indicator of need in the short-term in more vulnerable veterans. This pattern of results is consistent with evidence suggesting that traumatic distress is related to resource use (e.g., Rosenthal & Finlay, 2022). Interestingly, stressor reactivity did not relate to success at obtaining resources. Similar patterns of results are found in research on veterans needing employment services that found lower engagement with needed services in veterans who screened positive for a PTSD diagnosis (Sripada et al., 2018). Difficulties with success of engaging with resources may be attributable to barriers commonly cited for service engagement, including limited availability or difficulty navigating available resources (Cheney et al., 2018; Kotzias et al., 2019; Marshall et al., 2021). As such, continued efforts to increase resource engagement, especially in veterans who are in high need, are crucial.

In both the full sample and sample with high stressor exposure, effect sizes for average stressor reactivity on service utilization were small to medium. These effect size magnitudes indicate that higher stressor reactivity is reliably but modestly associated with greater service use. In practical terms, a one standard deviation increase in reactivity corresponds to a small but meaningful uptick in the likelihood or volume of service utilization, consistent with stressor reactivity functioning as an indicator of risk rather than a determinant of service utilization. Accordingly, targeting individuals with elevated stressor reactivity may yield incremental gains in engagement when combined with barrier-reduction or outreach strategies.

We did not find evidence that relative changes in a person’s stressor reactivity meaningfully predicted treatment utilization over four months. This pattern suggests that average levels of stressor reactivity for veteran women may be more informative than the relative change in reactivity to stressors. This finding suggests relative stability of the trauma-related distress and stressor exposure over four months. Indeed, stability of PTSD symptoms has been supported in veteran samples by previous research (e.g., von Stockert et al., 2018). It is possible that more within-person variability in stressor reactivity could be captured if symptom and stressor information were to be collected over a longer follow-up period or within the context of a study in which repeated interventions targeted symptoms or stressors over time.

The significant effects of race in models with self-reported treatment attendance, chart-reviewed mental health treatment attendance, and success at obtaining resources suggested that White veteran women had lower service utilization at the study endpoint. Differences in persistence with seeking care across groups may reflect variation in illness burden, as prior work has documented higher rates of mental health disorders in certain veteran populations (Merians et al., 2023). However, due to the uneven distribution of race in our sample, statistical power for race effects was reduced. As such, we err on the side of not overinterpreting evidence of these effects. However, future research should investigate further to better understand the potential differences in service engagement for racially diverse veteran women.

Interventions could identify stressor-reactive veteran women by routinely screening and monitoring both general stressors and those related to intimate relationships, thereby informing assessment of stressor reactivity. Veterans demonstrating high stressor reactivity could be flagged in the medical record or other tracking systems to support proactive outreach. For engagement, ongoing monitoring may enable timely contact, in-the-moment interventions, and sustained connection between visits, as well as facilitate warm handoffs to specialty care. Research on service needs for veteran women supports the need for a monitoring paradigm within the context of stressors (Bean-Mayberry et al., 2011). The VA is well-positioned to do such monitoring with technologies in place (e.g., Behavioral Health Laboratory Touch, a software VA providers use to send assessments to veterans to complete on personal devices). Within treatment, emphasizing skills for emotion regulation and problem-solving to enhance control and safety in the face of stressors could supplement evidence-based interventions for PTSD (e.g., prolonged exposure, cognitive processing therapy) and prevent entrenched stressor-related psychopathology. For retention, frequent monitoring of stressors and symptoms could help tailor interventions over time and ensure veterans’ evolving needs are continuously addressed.

Despite our study’s strengths, including consideration of symptoms in the context of continued stressors exposure, multi-method measurement of service utilization, and the longitudinal study period, our study is not without limitations. First, although our study’s data included multiple assessments over four months, the follow-up window is relatively short in duration due to constraints on the study’s budget and other goals (e.g., feedback on intervention, intervention development) that were to be accomplished during the funding period. As such, we cannot infer longer term impacts of stressor reactivity regarding PTSD symptoms on service utilization. Second, the time scales on our measures of interest varied, spanning from assessing PTSD symptoms over one month to service utilization over one year. Though we collapse across data in analyses, this mismatch may lead to difficulty concluding exact temporal relations among variables. Third, our analyses examined covariance of variables over four months, which prevents conclusions regarding stressor reactivity at a previous time point relating to future service utilization. We also examine stressor reactivity regarding PTSD symptoms only due to the nature of available measures. Future work on service utilization may consider stressor reactivity regarding other stressor-related psychopathology, including depression and anxiety. Given that nearly all participants had a service-connected disability, the findings may not fully generalize to women veterans who are not service-connected. Separately, due to the nature of our data, the stressor composite measure was based on occurrence as opposed to severity of stressors. This approach does not account for the fact that one traumatic event could be more stressful than a series of smaller stressful events. Lastly, our analyses do not distinguish VA versus non-VA services or resources pursued by veterans. Future research should consider such a distinction in analyses to more properly inform how stressor reactivity may relate to engagement and need for services within the VA healthcare system specifically.

Conclusion

Our study’s results indicating that veteran women who are more reactive to stressors regarding PTSD symptoms are more engaged in mental health treatment and indicate increased effort towards and need for resources over four months are promising. Assessing veterans for ongoing stressor context may be a useful practice to identify veterans of need and who are likely to engage with limited services and resources. Such practices can enhance care connection and allow for personalization of services regarding stressor reactivity in veterans who are highly susceptible to the harm of continued stressors.

Supplementary Material

Supplemental

Public Significance Statement.

Mental health services and resources are in high demand. It is essential to identify who is likely to engage with—and in most need of—services. This is particularly crucial for trauma-exposed veteran women, as they are at increased risk for posttraumatic stress disorder and continued stressor exposure. This study found that veteran women who were more reactive to stressors attended more treatment and reported higher need for obtaining resources over time.

Acknowledgements:

The authors wish to acknowledge Emma Harris for her feedback on an earlier version of this project.

Funding:

This work was supported by a grant from the Department of Defense, W81XWH-14-1-0368. This research was supported by a grant from the US Department of Veterans Affairs (RR&D 1IK2RX004565; PI: Pearson). ClinicalTrials.gov identifier: NCT02957747. AMe, KR, RP, AMc, and SC are also supported by the Department of Veterans Affairs VISN 17 Center of Excellence for Research on Returning War Veterans, and the Central Texas Veterans Affairs Healthcare System.

Footnotes

Disclosures: No authors have any conflicts of interest to report.

References

  1. Bean-Mayberry B, Yano EM, Washington DL, Goldzweig C, Batuman F, Huang C, Miake-Lye I, Shekelle PG. Systematic review of women veterans’ health: update on successes and gaps. Womens Health Issues. 2011. Jul-Aug;21(4 Suppl):S84–97. doi: 10.1016/j.whi.2011.04.022. [DOI] [PubMed] [Google Scholar]
  2. Benjamini Y, & Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal statistical society: series B (Methodological), 57(1), 289–300. 10.1111/j.2517-6161.1995.tb02031.x [DOI] [Google Scholar]
  3. Bergman B, & Brismar B (1991). A 5-year follow-up study of 117 battered women. American Journal of Public Health, 81(11), 1486–1489. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Bögemann SA, Puhlmann LM, Wackerhagen C, Zerban M, Riepenhausen A, Köber G, … & Kalisch R (2023). Psychological resilience factors and their association with weekly stressor reactivity during the COVID-19 outbreak in Europe: prospective longitudinal study. JMIR Mental Health, 10(1), e46518. 10.2196/46518 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bonanno GA (2005). Resilience in the face of potential trauma. Current directions in psychological science, 14(3), 135–138. [Google Scholar]
  6. Booth BM, Mengeling M, Torner J, & Sadler AG (2011). Rape, sex partnership, and substance use consequences in women veterans. Journal of Traumatic Stress, 24(3), 287–294. 10.1002/jts.20643 [DOI] [PubMed] [Google Scholar]
  7. Bovin MJ, Marx BP, Weathers FW, Gallagher MW, Rodriguez P, Schnurr PP, & Keane TM (2016). Psychometric properties of the PTSD checklist for diagnostic and statistical manual of mental disorders–fifth edition (PCL-5) in veterans. Psychological assessment, 28(11), 1379. [DOI] [PubMed] [Google Scholar]
  8. Brown JB, Lent B, Schmidt G, & Sas G (2000). Application of the Woman Abuse Screening Tool (WAST) and WAST-short in the family practice setting. Journal of family practice, 49(10), 896–903. [PubMed] [Google Scholar]
  9. Cahill S, Chandola T, & Hager R (2022). Genetic variants associated with resilience in human and animal studies. Frontiers in Psychiatry, 13, 840120. 10.3389/fpsyt.2022.840120 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Carlson E, Palmieri P, Smith S, Kimerling R, Ruzek J, & Burling T (2005). The Trauma History Screen (THS). [Measurement instrument]. Available from https://www.ptsd.va.gov [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Carlson EB, Smith SR, Palmieri PA, Dalenberg C, Ruzek JI, Kimerling R, … & Spain DA (2011). Development and validation of a brief self-report measure of trauma exposure: the Trauma History Screen. Psychological Assessment, 23(2), 463–477. 10.1037/a0022294 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cheney AM, Koenig CJ, Miller CJ, Zamora K, Wright P, Stanley R, … & Pyne JM (2018). Veteran-centered barriers to VA mental healthcare services use. BMC Health Services Research, 18, 1–14. 10.1186/s12913-018-3346-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Creech SK, Pulverman CS, Kahler CW, Orchowski LM, Shea MT, Wernette GT, & Zlotnick C (2022). Computerized intervention in primary care for women veterans with sexual assault histories and psychosocial health risks: a randomized clinical trial. Journal of General Internal Medicine, 37(5), 1097–1107. 10.1007/s11606-021-06851-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Davis JP, Prindle J, Saba S, et al. (2023). Childhood aversity, combat experiences, and military sexual trauma: a test and extension of the stress sensitization hypothesis. Psychological Medicine, 53(9), 4055–4063. 10.1017/S00332911722000733 [DOI] [PubMed] [Google Scholar]
  15. Dichter ME, Cerulli C, & Bossarte RM (2011). Intimate partner violence victimization among women veterans and associated heart health risks. Women’s Health Issues, 21(4), S190–S194. [DOI] [PubMed] [Google Scholar]
  16. Farmer CC, Rossi FS, Michael EM, & Kimerling R (2020). Psychotherapy utilization, preferences, and retention among women veterans with post-traumatic stress disorder. Women’s Health Issues, 30(5), 366–373. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Gelman A, & Rubin DB (1992). Inference from iterative simulation using multiple sequences. Statistical Science, 7(4), 457–472. 10.1214/ss/1177011136 [DOI] [Google Scholar]
  18. Goldzweig C, Balekian T, Rolón C, Yano E, & Shekelle P (2006). The state of women veterans’ health research: Results of a systematic literature review. Journal of General Internal Medicine, 21, S82. 10.1111/j.1525-1497.2006.00380.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hailemariam M, Johnson JE, Johnson DM, Sikorskii A, & Zlotnick C (2023). Computer-based intervention for residents of domestic violence shelters with substance use: A randomized pilot study. PLoS one, 18(5), e0285560. 10.1371/journal.pone.0285560 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Hamilton AB, Poza I, & Washington DL (2011). “Homelessness and trauma go hand-in-hand”: Pathways to homelessness among women veterans. Women’s Health Issues, 21(4), S203–S209. 10.1016/j.whi.2011.04.005 [DOI] [PubMed] [Google Scholar]
  21. Haskell SG, Gordon KS, Mattocks K, Duggal M, Erdos J, Justice A, & Brandt CA (2010). Gender differences in rates of depression, PTSD, pain, obesity, and military sexual trauma among Connecticut war veterans of Iraq and Afghanistan. Journal of Women’s Health, 19(2), 267–271. 10.1089/jwh.2008.1262 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Haskell SG, Mattocks K, Goulet JL, Krebs EE, Skanderson M, Leslie D,… Brandt C (2011). The burden of illness in the first year home: Do male and female VA users differ in health conditions and healthcare utilization. Women’s Health Issues, 21(1), 92–97. 10.1016/j.whi.2010.08.001 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Hegarty K, Bush R, & Sheehan M (2005). The Composite Abuse Scale: Further development and assessment of reliability and validity of a multidimensional partner abuse measure in clinical settings. Violence Vict, 20(5), 529–547. 10.1891/0886-6708.2005.20.5.529 [DOI] [PubMed] [Google Scholar]
  24. Hilton ME (1989). A comparison of a prospective diary and two summary recall techniques for recording alcohol consumption. British Journal of Addiction, 84(9), 1085–1092. 10.1111/j.1360-0443.1989.tb00792.x [DOI] [PubMed] [Google Scholar]
  25. Johnson DM, Zlotnick C, & Perez S (2008). The relative contribution of abuse severity and PTSD severity on the psychiatric and social morbidity of battered women in shelters. Behavior Therapy, 39(3), 232–241. 10.1016/j.beth.2007.08.003 [DOI] [PubMed] [Google Scholar]
  26. Kalisch R, Köber G, Binder H, Ahrens KF, Basten U, Chmitorz A, … & Engen H (2021). The frequent stressor and mental health monitoring-paradigm: a proposal for the operationalization and measurement of resilience and the identification of resilience processes in longitudinal observational studies. Frontiers in Psychology, 12, 710493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Keller BT & Enders CK (2021). Blimp user’s guide (Version 3). Retrieved from www.appliedmissingdata.com/multilevel-imputation.html
  28. Kimerling R, Street AE, Pavao J, Smith MW, Cronkite RC, Holmes TH, & Frayne SM (2010). Military-related sexual trauma among Veterans Health Administration patients returning from Afghanistan and Iraq. American Journal of Public Health, 100(8), 1409–1412. 10.2105/ajph.2009.171793 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Kotzias V, Engel CC, Ramchand R, Ayer L, Predmore Z, Ebener P, … & Karras E (2019). Mental health service preferences and utilization among women veterans in crisis: perspectives of veterans crisis line responders. The Journal of Behavioral Health Services & Research, 46, 29–42. 10.1007/s11414-018-9635-6 [DOI] [PubMed] [Google Scholar]
  30. Mahalanobis PC (1936). On the generalized distance in statistics. Proceedings of 695 the National Institute of Sciences, 2, 49–55. [Google Scholar]
  31. Maguen S, Cohen B, Ren L, Bosch J, Kimerling R, & Seal K (2012). Gender differences in military sexual trauma and mental health diagnoses among Iraq and Afghanistan veterans with posttraumatic stress disorder. Women’s Health Issues, 22(1), e61–66. 10.1016/j.whi.2011.07.010 [DOI] [PubMed] [Google Scholar]
  32. Marshall V, Stryczek KC, Haverhals L, Young J, Au DH, Ho PM, … & Sayre G (2021). The focus they deserve: improving women veterans’ health care access. Women’s health issues, 31(4), 399–407. [DOI] [PubMed] [Google Scholar]
  33. Merians AN, Gross G, Spoont MR, Bellamy CD, Harpaz-Rotem I, & Pietrzak RH (2023). Racial and ethnic mental health disparities in US military veterans: results from the National Health and Resilience in Veterans Study. Journal of Psychiatric Research, 161, 71–76. 10.1016/j.jpsychires.2023.03.005 [DOI] [PubMed] [Google Scholar]
  34. Metts A, Mendoza C, Pearson R, Creech SK. (2024a). Longitudinal associations among resilience, social isolation, and gender in United States Post-9/11 veterans. Journal of Traumatic Stress. 10.1002/jts.23111 [DOI] [PubMed] [Google Scholar]
  35. Metts A, Puhlmann LM, Zerban M, Kalisch R, Zinbarg RE, Mineka S, & Craske MG (2024b). Cross-Sectional, Longitudinal, and Dynamic Associations Among Big Five Personality Traits and Resilience in Primarily Female, Upper-Middle Class, Ethnically Diverse US Adolescents. Clinical Psychological Science, 21677026241281312. [Google Scholar]
  36. Nieminen P (2022). Application of standardized regression coefficient in meta-analysis. BioMedInformatics, 2(3), 434–458. 10.3390/biomedinformatics2030028 [DOI] [Google Scholar]
  37. Ouimette P, Wolfe J, Daley J, & Gima K (2003). Use of VA health care services by women veterans: findings from a national sample. Women & Health, 38(2), 77–91. [DOI] [PubMed] [Google Scholar]
  38. Pavao J, Turchik JA, Hyun JK, Karpenko J, Saweikis M, McCutcheon S,… Kimerling R (2013). Military sexual trauma among homeless veterans. Journal of General Internal Medicine, Suppl 2, 536–541. 10.1007/s11606-013-2341-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Resnick H, Kilpatrick D, Dansky B, Saunders B, & Best C (1993). Prevalence of civilian traumas and posttraumatic stress disorder in a representative national sample of women. Journal of Consulting and Clinical Psychology., 61, 984–991. 10.1037//0022-006x.61.6.984 [DOI] [PubMed] [Google Scholar]
  40. Rogers MM, Fisher C, Ali P, Allmark P, & Fontes L (2023). Technology-Facilitated Abuse in Intimate Relationships: A Scoping Review. Trauma Violence Abuse. 24(4):2210–2226. 10.1177/15248380221090218 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Rosenthal J, & Finlay AK (2022). Expanding the scope of forensic and other services for justice-involved veterans. The Journal of the American Academy of Psychiatry and the Law, 50(1), 106–116. 10.29158/JAAPL.210047-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. RStudio Team. (2024). Rstudio: Integrated development environment for R (Version 2024.04.2+764) [Computer software]. Posit Software, PBC. https://posit.co [Google Scholar]
  43. Runnals JJ, Garovoy N, McCutcheon SJ, Robbins AT, Mann-Wrobel MC, Elliott A, … & Strauss JL (2014). Systematic review of women veterans’ mental health. Women’s Health Issues, 24(5), 485–502. 10.1016/j.whi.2014.06.012 [DOI] [PubMed] [Google Scholar]
  44. Ryan ET, McGrath AC, Creech SK, & Borsari B (2015). Predicting utilization of healthcare services in the veterans health administration by returning women veterans: The role of trauma exposure and symptoms of posttraumatic stress. Psychological Services, 12(4), 412. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Schnurr PP, & Lunney CA (2011). Work-related quality of life and posttraumatic stress disorder symptoms among female veterans. Women’s Health Issues, 21(4), S169–S175. [DOI] [PubMed] [Google Scholar]
  46. Sienkiewicz ME, Amalathas A, Iverson KM, Smith BN, & Mitchell KS (2020). Examining the association between trauma exposure and work-related outcomes in women veterans. International journal of environmental research and public health, 17(12), 4585. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sripada RK, Henry J, Yosef M, Levine DS, Bohnert KM, Miller E, & Zivin K (2018). Occupational functioning and employment services use among VA primary care patients with posttraumatic stress disorder. Psychological Trauma: Theory, Research, Practice, and Policy, 10(2), 140–143. 10.1037/tra0000241 [DOI] [PubMed] [Google Scholar]
  48. Stevens F, Nurse JR, & Arief B (2021). Cyber stalking, cyber harassment, and adult mental health: A systematic review. Cyberpsychology, Behavior, and Social Networking, 24(6), 367–376. 10.1089/cyber.2020.0253 [DOI] [PubMed] [Google Scholar]
  49. Stevens J, Scribano PV, Marshall J, Nadkarni R, Hayes J, & Kelleher KJ (2015). A trial of telephone support services to prevent further intimate partner violence. Violence against women, 21(12), 1528–1547. 10.1177/1077801215596849 [DOI] [PubMed] [Google Scholar]
  50. Sullivan CM & Bybee DI (1999). Reducing violence using community-based advocacy for women with abusive partners. Journal of Consulting and Clinical Psychology, 67(1), 43–53. 10.1037/0022-006X.67.1.43 [DOI] [PubMed] [Google Scholar]
  51. Sullivan CM, Tan C, Basta J, Rumptz M, & Davidson WS (1992). An advocacy intervention program for women with abusive partners: Initial evaluation. American Journal of Community Psychology, 20(3), 309–332. 10.1007/BF00937912 [DOI] [PubMed] [Google Scholar]
  52. Suris A, & Lind L (2008). Military sexual trauma: A review of prevalence and associated health consequences in veterans. Trauma, Violence and Abuse, 9, 250–269. 10.1177/1524838008324419 [DOI] [PubMed] [Google Scholar]
  53. Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, & Schnurr PP (2013). The PTSD Checklist for DSM-5 (PCL-5). Scale available from the National Center for PTSD at www.ptsd.va.gov. [Google Scholar]
  54. von Stockert SH, Fried EI, Armour C, & Pietrzak RH (2018). Evaluating the stability of DSM-5 PTSD symptom network structure in a national sample of US military veterans. Journal of Affective Disorders, 229, 63–68. 10.1016/j.jad.2017.12.043 [DOI] [PubMed] [Google Scholar]
  55. Zinzow H, Gruabaugh A, Frueh B, & Magruder K (2008). Sexual assault, mental health, and service use among male and female veterans seen in Veterans Affairs primary care clinics: A multi-site study. Psychiatry Research, 159(1–2), 226–236. 10.1016/j.psychres.2007.04.008 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental
NIHMS2126739-supplement-Supplemental.docx (15.4KB, docx)

RESOURCES