Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans

Nicholas J. Parr; Sarah Young; Rachel Ward; Katherine Mackey

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans

Nicholas J. Parr, PhD, MPH, Conceptualization, Methodology, Investigation, Data curation, Programming, Formal analysis, Visualization, Writing – original draft, Writing – review & editing, Supervision, Project administration, Sarah Young, MPH, Investigation, Data curation, Formal analysis, Visualization, Writing – original draft, Writing – review & editing, Rachel Ward, Investigation, Writing – review & editing, and Katherine Mackey, MD, MPP, Conceptualization, Writing – original draft, Writing – review & editing.

Author Information and Affiliations

Washington (DC): Department of Veterans Affairs (US); 2021 Dec.

The Evidence Synthesis Program (ESP) Coordinating Center is responding to a request from the Office of Mental Health and Suicide Prevention (OMHSP) for an Evidence Brief on treatment of comorbid mental health conditions, or mental health conditions comorbid with a traumatic brain injury, substance use disorder, or chronic pain among Veterans and military Service members. Findings from this Evidence Brief will be used to inform the development of a clinical provider toolkit, as required by the Commander John Scott Hannon Veterans Mental Health Care Improvement Act (S.785 2019). The toolkit is intended to enhance clinical provider training in delivery of comprehensive care for Veterans and military Service members with comorbid conditions.

PREFACE

The VA Evidence Synthesis Program (ESP) was established in 2007 to provide timely and accurate syntheses of targeted health care topics of importance to clinicians, managers, and policymakers as they work to improve the health and health care of Veterans. These reports help:

Develop clinical policies informed by evidence;
Implement effective services to improve patient outcomes and to support VA clinical practice guidelines and performance measures; and
Set the direction for future research to address gaps in clinical knowledge.

The program comprises three ESP Centers across the US and a Coordinating Center located in Portland, Oregon. Center Directors are VA clinicians and recognized leaders in the field of evidence synthesis with close ties to the AHRQ Evidence-based Practice Center Program. The Coordinating Center was created to manage program operations, ensure methodological consistency and quality of products, interface with stakeholders, and address urgent evidence needs. To ensure responsiveness to the needs of decision-makers, the program is governed by a Steering Committee composed of health system leadership and researchers. The program solicits nominations for review topics several times a year via the program website.

The present report was developed in response to a request from the Center for Women Veterans. The scope was further developed with input from Operational Partners (below) and the ESP Coordinating Center review team. Comments on this report are welcome and can be sent to Nicole Floyd, Deputy Director, ESP Coordinating Center at vog.av@dyolF.elociN.

ACKNOWLEDGMENTS

The authors are grateful to Kathryn Vela, MLIS, for literature searching, and the following individuals for their contributions to this project:

Operational Partners

Operational partners are system-level stakeholders who help ensure relevance of the review topic to the VA, contribute to the development of and approve final project scope and timeframe for completion, provide feedback on the draft report, and provide consultation on strategies for dissemination of the report to the field and relevant groups.

Elizabeth A. Estabrooks, MSW
Deputy Director
Center for Women Veterans

Peer Reviewers

The Coordinating Center sought input from external peer reviewers to review the draft report and provide feedback on the objectives, scope, methods used, perception of bias, and omitted evidence (see Appendix D in Supplemental Materials). Peer reviewers must disclose any relevant financial or non-financial conflicts of interest. Because of their unique clinical or content expertise, individuals with potential conflicts may be retained. The Coordinating Center works to balance, manage, or mitigate any potential nonfinancial conflicts of interest identified.

EXECUTIVE SUMMARY

Background

The Evidence Synthesis Program (ESP) Coordinating Center is responding to a request from the Center for Women Veterans for an Evidence Brief on the prevalence of intimate partner violence/sexual assault (IPV/SA) among Veterans and spouses/intimate partners of Veterans. Findings from this Evidence Brief will be used to respond to activities required by section 5305 of the Johnny Isakson and David P. Roe, M.D. Veterans Health Care and Benefits Improvement Act of 2020 (H.R. 7105). Activities required by the Act are intended to enhance understanding of the scope of IPV/SA among Veterans and spouses/intimate partners of Veterans.

Methods

To identify studies, we searched OVID MEDLINE®, CINAHL, Cochrane Database of Systematic Reviews, and other sources through July 2021. We used prespecified criteria for study selection, data abstraction, and rating internal validity and strength of the evidence. See the Methods section and our PROSPERO protocol for full details of our methodology.

Key Findings

Considerable variation in sampling, recruitment, and data collection methods used among available studies limits the informativeness and quality of the overall body of evidence on intimate partner violence/sexual assault (IPV/SA) among Veterans and spouses/intimate partners of Veterans.
Moderate and low strength evidence suggests that psychological/emotional IPV is the most common form of experienced and perpetrated IPV/SA among both Veteran women and men, followed by physical IPV and sexual IPV.
Most available evidence pertains to experienced IPV/SA among Veteran women and perpetrated IPV/SA among Veteran men. Experienced IPV/SA among Veteran men, IPV/SA perpetrated by Veteran women, and IPV/SA among minority Veterans and intimate partners/spouses of Veterans are understudied.
Future studies of IPV/SA prevalence among Veterans should attempt to generate prevalence estimates that are applicable to Veterans of the range of ages, sexual and gender identities, races/ethnicities, and geographic contexts present in the Veteran population. Important methods to accomplish this aim include, but are not limited to, stratified random sampling with oversampling of important subgroups, such as historically underrepresented populations.

Intimate partner violence (IPV) includes physical violence, sexual violence including sexual assault (SA), stalking, and psychological aggression by a current or former intimate partner (ie, a spouse, dating partner, or sexual partner). Individuals of all ages, gender identities, sexual orientations, educational backgrounds, and socioeconomic statuses may experience IPV/SA. Veteran women experience IPV/SA at higher rates than women in the general US population; whether Veteran men also experience IPV/SA at disproportionately higher rates than the general population has not been well studied and is unclear. The prevalence of SA among Veteran intimate partners is also not fully understood.

The present review aimed to synthesize what is known about the prevalence of experienced IPV/SA (excluding non-partner SA) among Veterans and intimate partners of Veterans by type (physical, sexual, or psychological/emotional), timing (lifetime or past-year), and sociodemographic characteristics, as well as the prevalence of past-year IPV/SA perpetration by Veterans by type and gender identity. A second aim was to describe recruitment strategies and data collection methods used in studies of IPV/SA prevalence.

Findings of this review indicate that experienced IPV/SA is prevalent among Veteran women and men (Table ES1). The strongest available evidence was for past-year experienced IPV/SA among Veteran women; limited, low strength evidence was available for Veteran men and spouses/partners of Veterans. Psychological/emotional IPV appears to be the most common form of experienced IPV/SA among Veterans, followed by physical IPV and sexual IPV. Limited and low strength evidence was available on the lifetime prevalence of experienced IPV/SA among Veteran women, while for Veteran men, no information on lifetime prevalence of specific forms of experienced IPV/SA was found. Psychological/emotional IPV was the most common form of experienced IPV/SA across the lifetime for Veteran women. Perpetrated IPV/SA may also be common, particularly among Veteran men, but available evidence is sparse and poor quality. The limited available evidence suggests that psychological/emotional IPV is also the most common form of perpetrated IPV/SA among Veteran women and men.

Very little evidence was found on the role of sociodemographic factors in IPV/SA prevalence. The small number of identified studies used random samples, but were in Veteran women only and were small to moderate in size. This evidence suggests that past-year experienced IPV/SA may decrease with age and may be more prevalent among LGB Veteran women compared with heterosexual Veteran women. A single available study found similar prevalence of past-year experienced IPV/SA among rural and urban Veteran women. Studies reporting differences in experienced IPV/SA by race/ethnicity were inconsistent in their definition of some race/ethnicity subgroups and in their reported prevalence estimates, so it is unclear whether Veterans’ race or ethnicity is associated with greater prevalence of experienced IPV/SA. Finally, no prevalence estimates were identified among gender minority (eg, transgender) Veterans.

Included studies used a variety of sampling, recruitment, and data collection methods, limiting the comparability and generalizability of available evidence. Some studies used random sampling methods to reduce biases in data collection, while others used convenience samples that are likely poorly representative of the Veteran population and could over- or under-represent the prevalence of various forms of IPV/SA among Veterans. IPV/SA prevalence was most commonly collected via surveys using validated measures of IPV/SA, but measures varied across studies and a number of studies used unvalidated ad hoc measures. Taken together, the considerable methodological variation found among included studies limits the informativeness and quality of the overall body of evidence on IPV/SA prevalence among Veterans.

Table ES1

Key Findings for Experienced IPV.

Future studies of IPV/SA prevalence among Veterans should attempt to generate prevalence estimates that are applicable to Veterans of the range of ages, sexual and gender identities, races/ethnicities, and geographic contexts present in the Veteran population. Important methods to accomplish this aim include, but are not limited to, stratified random sampling with oversampling of important subgroups, such as historically underrepresented populations. Importantly, while rigorous sampling methods are critical to the generalizability and applicability of prevalence estimates, they do not necessarily address reporting biases, such as those that may occur when IPV/SA is assessed in clinical settings. Consequently, IPV/SA prevalence estimates derived from patient-level health care data may be best interpreted in concert with evidence from well-conducted survey research. Finally, future measurement of IPV/SA in VA clinical settings could consider 1) employing brief assessment tools that minimize respondent burden (eg, the HARK questionnaire), 2) providing patients the option of answering assessments face-to-face with a trusted provider or privately using a computer, tablet, or smartphone-based assessment, and 3) ensuring that assessment tools are culturally appropriate for measuring experienced or perpetrated IPV/SA among racial/ethnic minority and LGBTQ+ Veterans.

INTRODUCTION

PURPOSE

BACKGROUND

Intimate partner violence (IPV) includes physical violence, sexual violence including sexual assault (SA), stalking, and psychological aggression by a current or former intimate partner (ie, a spouse, dating partner, or sexual partner).¹ Individuals of all ages, gender identities, sexual orientations, educational backgrounds, and socioeconomic statuses may experience IPV/SA. Despite evidence that IPV/SA is frequently underreported,²^–⁵ a 2010 national survey of US adults found that more than 1 in 3 women and more than 1 in 4 men have experienced rape, physical violence, and/or stalking by an intimate partner in their lifetime and nearly half of women and men have experienced psychological aggression.⁶ For most women and men who have experienced IPV/SA, IPV/SA first occurs under age 25.

Experiencing IPV/SA prior to, during, or after military service may lead to or worsen health problems including anxiety, depression, and substance use and result in lower quality of life.⁷^–¹⁰ The association between IPV/SA and health outcomes may vary according to the type of IPV/SA experienced, but this potential variability has not been well studied. Veteran women experience IPV/SA at higher rates than women in the general US population based on data from the Centers for Disease Control and Prevention Behavioral Risk Factor Surveillance System.⁷ Whether Veteran men experience IPV/SA at disproportionately higher rates than the general population is not fully understood. It is also unclear whether racial/ethnic minority and sexual and gender minority (LGBTQ+) Veterans experience different rates of IPV/SA than non-minority Veterans.

The Johnny Isakson and David P. Roe, M.D. Veterans Health Care and Benefits Improvement Act of 2020 (H.R. 7105) requires the VA Center for Women Veterans (CWV) to complete a baseline study of IPV/SA prevalence among Veterans and intimate partners of Veterans and recommend ways to expand services to impacted populations. The aim of the present report is to synthesize what is already known about the prevalence of experienced IPV/SA (excluding non-partner SA) among US Veterans and intimate partners of Veterans by type (physical, sexual, or psychological/emotional), timing (lifetime or past-year), and sociodemographic characteristics, as well as the prevalence of past-year IPV/SA perpetration by Veterans by type and gender identity. A second aim is to describe recruitment strategies and data collection methods used in studies of IPV/SA prevalence to inform methods used by the CWV in its baseline study.

METHODS

PROTOCOL

A preregistered protocol for this review can be found on the PROSPERO international prospective register of systematic reviews (http://www.crd.york.ac.uk/PROSPERO/; registration number CRD42021267769).

KEY QUESTIONS

The following key questions (KQs) were the focus of this review:

KQ1.: What is the prevalence of experienced IPV/SA among Veterans and spouses/ intimate partners of Veterans by type (physical, sexual, or psychological/ emotional), timing (lifetime or past-year), and sociodemographic characteristics (eg, gender identity)?
KQ2.: What is the prevalence of past-year IPV/SA perpetration by Veterans by type (physical, sexual, or psychological/emotional) and gender identity?
KQ3.: What are common recruitment strategies and data collection methods utilized in studies of IPV/SA prevalence among Veterans and spouses/intimate partners of Veterans?

ELIGIBILITY CRITERIA

The ESP included studies that met the following criteria:

Population	US Veterans and spouses/intimate partners of US Veterans
Intervention	Not applicable
Comparator	Not applicable
Outcomes	KQ1: Prevalence (proportion) of Veterans or spouses/intimate partners of Veterans who have experienced IPV/SA (excluding non-partner SA) KQ2: Prevalence (proportion) of Veterans who have perpetrated IPV/SA (excluding non-partner SA) KQ3: Recruitment strategies and data collection methods
Timing	Any
Setting	Any
Study Design	Any, but we may prioritize articles using a best-evidence approach to accommodate the Evidence Brief timeline

DATA SOURCES AND SEARCHES

We included all relevant studies from a 2013 ESP review¹¹ encompassing this topic. To identify additional articles relevant to the key questions, a research librarian searched Ovid MEDLINE and CINAHL, as well as AHRQ, Cochrane Database of Systematic Reviews, and gray literature databases from database origination through July 2021 using terms related to experienced and perpetrated IPV/SA (see Appendix A in Supplemental Materials for complete search strategies). We limited the search to published and indexed articles involving human subjects available in the English language. Study selection was based on the eligibility criteria described above. Studies in highly specialized populations, such as purposive samples entirely composed of IPV/SA victims, were excluded. Titles, abstracts, and full-text articles were reviewed by 1 investigator and checked by another. All disagreements were resolved by consensus or discussion with a third reviewer.

DATA ABSTRACTION AND ASSESSMENT

Prevalence data and sample and methodological characteristics were abstracted from all included studies. The ROBIS tool¹² was used to assess risks of bias of included systematic reviews. Primary studies were assessed using the tool developed by Hoy et al¹³ for prevalence studies, which rates studies across several characteristics including the representativeness of the sample, risks of bias common to studies of prevalence/incidence (eg, non-response bias, recall bias), directness of estimates (ie, whether prevalence was reported directly from participants or via a proxy), reliability and validity of the instrument used to assess prevalence, and apparent accuracy of reported estimates (eg, comparability of prevalence estimate denominators and study sample sizes). Because our interest was in any occurrence of IPV/SA in a given time period, rather than more granular estimates of IPV/SA frequency in that period, we did not consider recall bias a major risk and did not rate the item assessing that form of bias (Was the length of the shortest prevalence period for the parameter of interest appropriate?). All data abstraction and internal validity ratings were first completed by 1 reviewer then checked by another; disagreements were resolved by consensus or discussion with a third reviewer.

The Hoy et al tool also provides an overall rating of confidence in reported estimates, with a rating of low risk indicating that future research is very unlikely to alter confidence in the prevalence estimate, moderate risk indicating that future research is likely to alter confidence in the prevalence estimate and may change the estimate, and high risk indicating that future research is very likely to alter confidence in the prevalence estimate and will likely change the estimate.¹³ Confidence ratings from individual studies, in concert with confidence intervals from meta-analyses, were used to rate the strength of evidence contributing to overall estimates of each form of IPV/SA (eg, past-year experienced physical IPV among Veteran women) using the following general algorithm: high strength evidence consisted of multiple, large studies rated as low risk (ie, precise estimates that are very likely to be representative and minimally biased); moderate strength evidence consisted of a mix of larger and smaller studies rated as low risk or moderate risk, or a single very large and highly representative low risk study (ie, fairly precise estimates that are likely to be representative and unbiased); low strength evidence consisted of multiple smaller trials rated as moderate risk or high risk (ie, imprecise estimates that are likely unrepresentative and biased); and insufficient evidence consisted of a single trial rated as moderate risk or high risk (ie, a single imprecise and unrepresentative estimate that is very likely to be biased), or no available studies.

Strength of evidence assessment was conducted for experienced IPV/SA findings only. For perpetrated IPV/SA findings, we located a recent high-quality systematic review and meta-analysis¹⁴ that synthesized available evidence on prevalence of perpetrated IPV/SA among Veterans (see Synthesis and Literature Overview sections). We did not directly assess risks of bias of studies identified in that review, and for most forms of perpetrated IPV/SA, limited evidence was found. As a result, we did not formally rate strength of evidence for perpetrated IPV/SA findings. In lieu of strength of evidence ratings, we summarize the assessments reported in that review.

SYNTHESIS

Prevalence estimates were organized by form of IPV/SA (experience or perpetrated), type (any, physical, sexual, or psychological/emotional), timing (past-year or lifetime), and gender identity (male or female). Physical IPV was defined as any form of non-sexual physical violence (eg, hitting or restraining). Sexual IPV included any form of sexual violence, including SA and sexual coercion, among intimate partners (ie, excluding non-partner SA). Psychological/emotional IPV included any form of non-physical and non-sexual IPV, such as verbal threats, insults or degrading language, or shouting. If studies reported both past-year and more recent (eg, past 6 months) prevalence estimates, only past-year estimates were synthesized. If only past-6 months estimates were available, these estimates were synthesized with past-year estimates from other studies and sensitivity analyses excluding the shorter-term estimates were conducted. One study¹⁵ reported an estimate of physical IPV “after military service,” which was pooled with lifetime estimates. We did not limit eligibility based on type of instrument used to collect IPV/SA prevalence; however, as shown in Table 1, most studies used well-validated and widely used tools. As noted in the previous section, use of a validated instrument was also considered in strength of evidence assessments.

In general, the denominator used to calculate prevalence estimates was the total study sample size, such that estimates reflect the proportion of each type of IPV/SA in each study. Adjusted or weighted proportions were used when reported. When available, we also report prevalence estimates for the following sociodemographic subgroups: race/ethnicity, age category, rurality, and identification as heterosexual or a sexual or gender minority. For sociodemographic subgroup estimates, the total subgroup size from each study was used as the estimate denominator (ie, estimates offer within-study comparisons of relative prevalence, for example, the prevalence of past-year physical IPV among sexual or gender minority Veterans compared to heterosexual Veterans in each study).

When 2 or more estimates of experienced IPV/SA prevalence were identified, we quantitatively synthesized estimates using meta-analytic generalized random-effects logistic models. Reported estimates were transformed using the standard logit transformation for analysis, and back-transformed for interpretation and reporting. Precision of study-level and overall estimates is reported using 95% confidence intervals (CIs), and CIs were used to evaluate statistical significance of overall prevalence estimates at a significance level of .05. Heterogeneity was estimated using maximum-likelihood estimation and is presented using 95% prediction intervals (PIs). Although no syntheses included dependent estimates (ie, multiple estimates from the same study), several studies may have included overlapping participants because they derived samples from the same health system databases or registries. When it was clear that prevalence estimates reported in different studies were from an identical sample, we excluded duplicative estimates. In some cases, however, the extent of overlap among included studies was unknown, and this may have led to overestimating the precision of some overall prevalence estimates.

Whether overall prevalence estimates varied according to study sampling method (convenience or random/population) was explored using meta-analytic generalized mixed-effects logistic models (meta-regression). Studies were characterized as using a convenience sample if they used samples derived from non-randomized selection of participants from readily accessible sources (eg, recruitment from waiting rooms or among individual clinic patients). Random or population samples were derived from randomized selection of participants from a larger population, or that included all members of a population (eg, samples composed of all VHA patients at the health center, regional, or national level). Sufficient studies were available to investigate sampling method moderation for lifetime IPV/SA estimates only. Meta-analyses and moderation analyses were conducted using the metafor¹⁶ package for R (R Foundation for Statistical Computing, Vienna, Austria).

For perpetrated IPV/SA prevalence, we identified a recent high quality systematic review and meta-analysis¹⁴ that synthesized available evidence on prevalence of perpetrated IPV/SA among Veterans. To address KQ2, we summarized the relevant findings on perpetrated IPV/SA prevalence from meta-analyses and individual studies (when meta-analysis was not conducted) reported in that review. Although initial interest was in past-year perpetrated IPV/SA only, given the sparseness of evidence on perpetrated IPV/SA in general, we also included any findings on lifetime perpetrated IPV/SA reported in the review. For KQ3, we narratively synthesized information on recruitment methods and data collection methods employed in included studies.

RESULTS

LITERATURE FLOW

The literature flow diagram (Figure 1) summarizes the results of the study selection process (full list of excluded studies available in Appendix B in Supplemental Materials).

Figure 1

Literature Flowchart. Note. Count of studies excluded for ineligible outcome includes studies meeting eligibility criteria but reporting duplicative prevalence estimates. Abbreviations. CCRCT=Cochrane Central Register of Controlled Trials; CDSR=Cochrane (more...)

LITERATURE OVERVIEW

Our search identified 824 potentially relevant articles. Of these, 32 primary studies and 1 recent systematic review met eligibility criteria and reported non-duplicative prevalence estimates. Primary study characteristics are summarized in Table 1. Included primary studies generally reported prevalence of experienced IPV/SA only, and most studies provided prevalence estimates for Veteran women. Five studies⁹^,¹⁷^–²¹ reported prevalence estimates for Veteran men, and 2 studies²²^,²³ were carried out among spouses/partners of Veterans. A small number of studies reported prevalence estimates within subgroups of interest, although no studies were identified that reported prevalence of IPV/SA among gender minority Veterans (estimates were available only for Veterans identifying as lesbian, gay, or bisexual [LGB]). All primary studies used cross-sectional or cohort designs, and the median sample size was 406 participants (range: 50-8,427). Several studies used subsamples from large Veteran databases or registries, including the New England VHA Cohort (5 studies¹⁷^,²⁴^–²⁷), the GfK Knowledge Networks Panel (4 studies¹⁷^,²⁸^–³⁰), and the VHA Corporate Data Warehouse (CDW; 1 study³¹). Sampling and IPV/SA measurement methods used in included studies are discussed in detail in the Results section. Our search also identified 3 underway systematic reviews related to this topic (see Appendix E in Supplemental Materials).

Most included primary studies (k = 25) were rated as high risk, indicating they were susceptible to risks of bias and/or were unlikely to be fully representative of the Veteran population. Common sources of potential bias and/or non-representativeness were high non-response rates in survey studies, use of a limited geographical or institutional sampling frame, incomplete or partial representation of Veteran age or era of service groups, recruitment from settings in which potential participants were likely to be at greater risk of IPV/SA than the general Veteran population, use of unvalidated IPV/SA measures, and small sample size. Low risk and moderate risk studies typically had moderate to large sample sizes and high response rates, and used random sampling procedures, validated IPV/SA measures, and established case definitions of IPV/SA.

The included systematic review¹⁴ synthesized available evidence on prevalence of perpetrated IPV/SA among Veterans. Authors searched multiple databases and conducted hand searching and forward citation searches for relevant studies. Studies were eligible for inclusion in the review if they included male and/or female active duty/reserve personnel or Veterans, and used a validated self-report measure for collecting perpetrated IPV/SA prevalence (eg, Conflict Tactics Scale) or objective measure such as military records. The review included 23 studies in US Veteran samples, which were assessed for risk of bias and quality by two independent reviewers using a composite tool drawing from several preexisting assessment tools. Studies were rated as high quality if they scored at least 50% on questions related to selection bias, with only 8 of 23 studies in Veterans receiving this rating. Reviewers noted considerable variation in the types of samples employed (representative random samples and convenience samples from clinical settings) and in IPV/SA measures and assessment periods, limiting the comparability and generalizability of studies on perpetrated IPV/SA prevalence. After quality assessment, prevalence estimates were synthesized using random-effects models when 10 or more estimates were available for a given type of IPV/SA. We rated the overall risk of bias of this review as low using the ROBIS tool, corresponding to minimal concern about the review’s eligibility criteria, search and screening strategies, study appraisal and synthesis methods, and reporting of findings.

Table 1

Characteristics of Included Primary Studies.

EXPERIENCED IPV/SA

Any IPV/SA

Veteran Women

Seven studies⁷^,⁸^,¹⁰^,²⁶^,²⁹^,³⁴^,³⁵^,³⁹ reported prevalence of any lifetime experienced IPV/SA among Veteran women. Pooling estimates from a total of 2,140 respondents, the overall prevalence was 58.0% (95% CI [43.6, 71.2], 95% PI [21.5, 87.5]). Although not significantly different, overall prevalence estimates appeared to be higher among studies using convenience samples (68.0%, 95% CI [50.0, 81.9], k = 4) compared with those using random samples (43.8%, 95% CI [35.5, 52.6], k = 3). Evidence contributing to the overall estimate was rated as low strength. Available studies were small to moderate in size (N = 127-503), used a mix of convenience and random samples, and in all but 1 case were rated as high risk.

Eleven studies⁹^,¹⁷^,²⁰^,²⁵^,²⁶^,³⁰^,³¹^,³⁷^,³⁸^,⁴⁰^,⁴² provided estimates of any past-year experienced IPV/SA among Veteran women (total N = 17,328). When pooled, the overall prevalence estimate was 26.0% (95% CI [16.3, 38.8], 95% PI [4.4, 72.7]). A sensitivity analysis removing past-6 months estimates from 2 studies⁹^,³⁷ resulted in a similar but somewhat lower overall prevalence estimate of 19.8% (95% CI [13.3, 28.4], 95% PI [5.4, 51.9]). With the exception of 1 study, available studies used random sampling, and studies were rated as a mix of low, moderate, and high risk, with 2 very large studies (N = 6,046-8,427) rated as low or moderate risk. As a result, evidence contributing to the overall estimate was rated as moderate strength.

The number of studies reporting prevalence of any past-year experienced IPV/SA in sociodemographic subgroups differed for each subgroup. Five studies²⁴^,²⁵^,³¹^,³⁸^,⁴² provided prevalence estimates for non-Hispanic white Veteran women, 3 studies³¹^,³⁸^,⁴² for Black Veteran women, 1 study³⁸ for Latina Veterans, and 5 studies²⁴^,²⁵^,³¹^,³⁸^,⁴² for non-white Veteran women (reported as “non-white,” “Hispanic, multiracial, or other race,” or “other race”). Among white Veteran women, the overall prevalence of any past-year experienced IPV/SA was 16.1% (95% CI [11.3, 22.4], 95% PI [7.0, 32.9]); among Black Veteran women, 13.2% (95% CI [8.3, 20.6], 95% PI [5.2, 29.7]); among Latina Veterans, the single study reported a prevalence of 9%; and among non-white Veteran women, the overall prevalence was 19.7% (95% CI [10.3, 34.3], 95% PI [4.2, 57.7]).

One cross-sectional study⁴² using a random sample (N = 6,046) reported differences in prevalence of any past-year experienced IPV/SA among subgroups of heterosexual- and LGB-identifying Veteran women. 18.0% of heterosexual Veteran women in the study reported any past-year experienced IPV/SA, compared with 24.7% of LGB Veteran women. The same study also reported differences in prevalence by rurality, with 18.1% of urban Veteran women and 19.0% of rural Veteran women in the study reporting any past-year experienced IPV/SA. A large cohort study³¹ using a random sample (N = 8,427) reported prevalence by age subgroups. Among Veteran women in the study younger than 35, 10.4% experienced any past-year IPV/SA; among those aged 35-44, 9.2%; aged 45-54, 8.7%; aged 55-64, 6.2%; and aged 65 or older, 3.3%.

Veteran Men

Based on estimates from 2 studies¹⁸^,¹⁹ (total N = 5,025), the overall prevalence of any lifetime experienced IPV/SA was 12.6% (95% CI [8.1, 19.1], 95% PI [6.0, 24.6]). Three studies⁹^,¹⁷^,²⁰ (total N = 1,573) reported prevalence of any past-year experienced IPV/SA among Veteran men.

When pooled, the overall prevalence was 36.7% (95% CI [16.1, 63.7], 95% PI [6.0, 84.0]). The overall prevalence was somewhat lower when 1 study⁹ providing a past-6 months estimate was removed (24.4%, 95% CI [12.0, 43.1], 95% PI [6.9, 58.4]). The lower overall estimate of lifetime prevalence (compared with past-year prevalence) may be the result of methodological variation across studies, particularly the use of ad hoc IPV/SA measures in studies providing lifetime estimates. No studies reporting prevalence estimates in sociodemographic subgroups were identified. Evidence contributing to both lifetime and past-year overall estimates was rated as low strength given the small number of available studies and the rating of studies as moderate or high risk.

Spouses/Partners

No studies were identified that reported prevalence of any experienced IPV/SA among spouses/partners of Veterans.

Other Studies

One small study³² in a random sample (N = 103) did not disaggregate prevalence estimates by gender identity, and found that 32% of Veterans reported any lifetime experienced IPV/SA.

Physical IPV

Veteran Women

Seven studies⁸^,¹⁰^,¹⁵^,²⁸^,³⁶^,³⁹^,⁴¹^,⁴⁵ reported prevalence of lifetime experienced physical IPV among Veteran women (total N = 2,859). When pooled, the overall prevalence of this form of IPV/SA was 33.8% (95% CI [26.2, 42.3], 95% PI [16.0, 57.7]). Overall prevalence was somewhat higher in convenience samples (40.0%, 95% CI [26.6, 55.0], k = 3) compared with random samples (29.8%, 95% CI [23.2, 37.3], k = 4). Evidence contributing to the overall prevalence estimates was rated as low strength. Studies used a mix of convenience and random samples, ranged from small to moderate in size (N = 127-1,259), and all but 1 was rated as high risk.

Prevalence of past-year experienced physical IPV among Veteran women was reported by 8 studies⁹^,¹⁷^,²⁵^,²⁶^,²⁸^,³⁰^,³¹^,³⁷ (total N = 10,130), and when pooled was 7.6% (95% CI [4.6, 12.4], 95% PI [1.7, 28.1]). Overall prevalence was similar when 3 studies⁹^,³⁰^,³⁷ providing past-6 months estimates were removed (7.0%, 95% CI [3.2, 14.5], 95% PI [1.1, 34.3]). Studies varied from small to very large (N = 102-8,427), used random samples, and were rated low, moderate, or high risk. Consequently, evidence contributing to the overall estimate of past-year experienced physical IPV among Veteran women was considered moderate strength.

One cross-sectional study²⁸ in a random sample (N = 411) disaggregated estimates of lifetime and past-year experienced physical IPV among Veteran women by identification as heterosexual or LGB. 27.7% and 13.2% of heterosexual Veteran women in this study reported lifetime and past-year physical IPV, respectively, compared with 46.2% and 28.2% of LGB Veteran women respondents. Studies reporting prevalence of experienced physical IPV in other sociodemographic subgroups were not found.

Veteran Men

No studies were identified that reported prevalence of lifetime experienced physical IPV among Veteran men. Two studies⁹^,¹⁷ (total N = 1,049) reporting past-year estimates were located, and when pooled, the overall prevalence of past-year experienced physical IPV among Veteran men was 7.2% (95% CI [5.6, 9.1], 95% PI [5.6, 9.1]). Studies were moderate in size (N = 407-642) and used random samples, but given the small number of studies available and their rating as moderate or high risk, evidence contributing to the overall prevalence estimate was rated as low strength.

Spouses/Partners

One older study²² using a small convenience sample (N = 50) reported 81.8% of female spouses/partners of Veterans experienced past-year physical IPV. In a second study,²¹ 22.2% of male spouses/partners of Veterans experienced past-year physical IPV. A third study²³ that did not disaggregate by gender identity reported that past-year physical IPV was experienced by 21.3% of Veteran spouses/partners in the study. The latter 2 studies were small to moderate in size (N = 89-376) and used random samples. Because available evidence consisted of single studies among Veterans of different or nonspecific gender identities, evidence was insufficient to determine the strength of evidence on prevalence of experienced physical IPV among spouses/partners of Veterans.

Sexual IPV

Veteran Women

Seven studies⁸^,¹⁰^,²⁶^,²⁸^,³³^,³⁶^,³⁹^,⁴⁴ reported prevalence of lifetime experienced sexual IPV among Veteran women (total N = 2,196). When pooled, the overall prevalence was 14.2% (95% CI [7.0, 26.4], 95% PI [1.9, 58.9]). Studies used a mix of convenience and random samples, and overall prevalence was significantly higher in convenience samples (32.2%, 95% CI [27.7, 37.1]) compared with random samples (9.9%, 95% CI [4.4, 20.8]). Evidence contributing to the overall prevalence estimate was rated as low strength. Studies were small to moderate in size (N = 127-506), used a mix of random and convenience samples, and all but 1 were rated as high risk.

Prevalence of past-year experienced sexual IPV among Veteran women was reported by 8 studies⁹^,¹⁷^,²⁵^,²⁶^,²⁸^,³⁰^,³¹^,³⁷ (total N = 10,130), and when pooled across studies, the overall prevalence estimate was 8.0%, (95% CI [4.5, 13.8], 95% PI [1.5, 33.8]). When past-6 months estimates from 3 studies⁹^,³⁰^,³⁷ were removed, the overall prevalence estimate was slightly reduced (6.5%, 95% CI [2.9, 14.2], 95% PI [0.9, 35.5]). Studies providing past-year experienced sexual IPV prevalence estimates were the same as those reporting past-year experienced physical IPV prevalence, and consequently the evidence received the same rating of moderate strength.

One cross-sectional study²⁸ in a random sample (N = 411) reported differences in lifetime and past-year experienced sexual IPV prevalence among heterosexual- and LGB-identifying Veteran women. Prevalence of experienced sexual IPV among heterosexual Veteran women was 19.5% (lifetime) and 8.8% (past-year), compared with 35.9% (lifetime) and 28.2% (past-year) for LGB Veteran women enrolled in the study. No other sociodemographic subgroups were identified.

Veteran Men

No studies were found that reported lifetime experienced sexual IPV among Veteran men. Two studies⁹^,¹⁷ (total N = 1,049) reported past-year estimates, and when pooled, the overall prevalence of past-year experienced sexual IPV among Veteran men was 2.0% (95% CI [0.8, 5.0], 95% PI [0.5, 8.1]). Studies were moderate in size (N = 407-642) and used random samples, but given the small number of studies available and their rating as moderate or high risk, evidence contributing to the overall prevalence estimate was rated as low strength.

Spouses/Partners

No studies were identified that reported prevalence of experienced sexual IPV among spouses/partners of Veterans.

Psychological/Emotional IPV

Veteran Women

Four studies⁸^,¹⁰^,²⁶^,²⁸^,³⁹ (total N = 963) provided prevalence estimates for lifetime experienced psychological/emotional IPV among Veteran women. When pooled, the overall prevalence estimate was 54.1% (95% CI [34.5, 72.5], 95% PI [16.7, 87.4]). Studies used a mix of convenience and random samples, and overall prevalence of lifetime experienced psychological/emotional IPV was significantly higher in convenience samples (69.8%, 95% CI [48.1, 85.2], k = 2) compared with random samples (38.8%, 95% CI [32.8, 45.2], k = 2). Available studies were generally small (N = 127-411), used of mix of convenience and random samples, and in all but one case were rated as high risk. As a result, evidence contributing to the lifetime overall prevalence estimate was considered low strength.

Nine studies⁹^,¹⁷^,²⁵^,²⁶^,²⁸^,³⁰^,³¹^,³⁷^,⁴³ (total N = 11,371) reported past-year prevalence of experienced psychological/emotional IPV among Veteran women. When pooled, the overall prevalence estimate was 19.7% (95% CI [10.5, 33.7], 95% PI [2.4, 70.6]). Removing 3 past-6 months estimates⁹^,³⁰^,³⁷ lowered the overall prevalence somewhat (11.6%, 95% CI [7.7, 17.0], 95% PI [4.1, 28.9]). Studies providing prevalence estimates of past-year psychological/emotional IPV among Veteran women were the same as those providing estimates of past-year physical and sexual IPV prevalence (with the addition of 1 smaller moderate risk study), and consequently the evidence also received the rating of moderate strength.

One cross-sectional study²⁸ in a random sample (N = 411) reported prevalence of past-year experienced psychological/emotional IPV among Veteran women identifying as heterosexual or LGB. 40.9% and 21.7% of heterosexual Veteran women in this study reported lifetime and past-year experienced physical IPV, respectively, compared with 56.4% and 41.0% of LGB Veteran women respondents. Another cross-sectional study⁴³ using a random sample (N =1,241) reported past-year prevalence by age, although this study only enrolled participants aged 20-44. Among Veteran women in the study aged 20-29, 14.5% experienced past-year psychological/emotional IPV; among those aged 30-34, 10.4%; aged 35-39, 11.8%; and aged 40-44, 6.2%.

Veteran Men

No studies were identified that reported prevalence of lifetime experienced psychological/emotional IPV among Veteran men. Two studies⁹^,¹⁷ (total N = 1,049) reported past-year prevalence, and when pooled the overall estimate of past-year experienced psychological/emotional IPV among Veteran men was 33.2% (95% CI [7.6, 75.1], 95% PI [2.2, 91.8]). Studies were moderate in size (N = 407-642) and used random samples, but given the small number of studies available and their rating as moderate or high risk, evidence contributing to the overall prevalence estimate was rated as low strength.

Spouses/Partners

One study²³ (N = 376) employing a random sample did not disaggregate by gender identity, and reported that past-year psychological/emotional IPV was experienced by 83.8% of Veteran spouses/partners in the study. Available evidence was insufficient to determine the strength of evidence on prevalence of experienced physical IPV among spouses/partners of Veterans.

PERPETRATED IPV/SA

The included systematic review and meta-analysis¹⁴ identified limited evidence on perpetrated IPV/SA, particularly among Veteran women. One included cross-sectional study in a convenience sample reported that 33.3% of Veteran women had perpetrated physical IPV in their lifetime, while 2 cross-sectional studies using a mix of convenience and random samples reported that 18.2-22.2% of Veteran women had perpetrated physical IPV in the past year. Two additional cross-sectional studies using a mix of convenience and random samples reported prevalence of past-year perpetrated psychological/emotional IPV ranging from 62.5-76.7%.

For IPV/SA perpetrated by Veteran men, prevalence of lifetime perpetrated physical IPV was reported by 3 cross-sectional studies using convenience samples; estimates ranged from 12.0-53.5%. Eight estimates of past-year prevalence of perpetrated physical IPV among Veteran men were pooled in a meta-analytic subgroup analysis, resulting in an overall prevalence estimate of 32.0% (95% CI [24.0, 41.0]). One cross-sectional study in a convenience sample reported that 28.0% of Veteran men perpetrated sexual IPV in their lifetime, while 2 cross-sectional studies also in convenience samples reported prevalence estimates of past-year perpetrated sexual IPV among Veteran men that ranged from 27.6-40.2%. An additional cross-sectional study in a convenience sample provided an estimate of lifetime perpetrated psychological/emotional IPV, which ranged from 67.0-68.0% depending on the subtype of psychological/emotional IPV (shouted at partner, insulted/swore at partner). Six cross-sectional studies were identified that reported past-year estimates of perpetrated psychological/emotional IPV; across the included convenience samples of Veteran men, prevalence of past-year psychological/emotional IPV ranged from 66.4-100%.

Finally, 1 cross-sectional study was identified that did not disaggregate prevalence estimates by gender identity. This study, which used a convenience sample, reported that prevalence of past-year perpetrated physical IPV among Veterans was 23.1% for moderate-severity physical IPV and 9.4% for severe physical IPV.

RECRUITMENT STRATEGIES AND DATA COLLECTION METHODS

Most included studies (k = 24) used random selection or attempted to take a census of their study population. Recruitment strategies included random digit dialing, randomly selecting participants from a national registry, extracting medical records from a database, and recruiting all patients receiving care within a VA network. Seven studies used convenience sampling techniques, such as recruiting participants who had already participated in a previous survey or who presented for care at a local clinic. Paper or web-based surveys were the most common methods of data collection (k = 22), followed by phone or in-person interviews (k = 7). Only 2 studies used medical record abstraction to collect data. Medical record information was based on routine screening during in-person appointments with providers. Most studies used a validated tool to measure IPV/SA experiences (k = 23). The most common tools used were the (Revised) Conflict Tactics Scale (CTS/CTS-2, 10 studies), the Humiliation, Afraid, Rape, and Kick questionnaire (HARK, 5 studies), and the Extended-Hurt/Insult/Threaten/Scream screening tool (E-HITS, 5 studies). Six studies employed unvalidated, ad hoc questionnaires to measure IPV/SA prevalence, and 1 study did not report the data collection instrument used.

DISCUSSION

Findings of this review indicate that experienced IPV/SA is prevalent among Veteran women and men. The strongest available evidence was for past-year experienced IPV/SA among Veteran women; limited, low strength evidence was available for Veteran men and spouses/partners of Veterans. Psychological/emotional IPV appears to be the most common form of experienced IPV/SA among Veterans, followed by physical IPV and sexual IPV. Limited and low strength evidence was available on the lifetime prevalence of experienced IPV/SA among Veteran women, while for Veteran men, no information on lifetime prevalence of specific forms of experienced IPV/SA was found. Psychological/emotional IPV was the most common form of experienced IPV/SA across the lifetime for Veteran women. Perpetrated IPV/SA may also be common, particularly among Veteran men, but available evidence is sparse and poor quality. The limited available evidence suggests that psychological/emotional IPV also is the most common form of perpetrated IPV/SA among Veteran women and men, although comparisons of IPV/SA perpetration across gender identities must be interpreted with caution, as men may be more likely than women to underreport acts of physical violence against their partners.⁴⁶

LIMITATIONS

Limitations of our review methods include use of a second reviewer check during study selection, data abstraction, and quality assessment rather than dual independent review. Our search focused on databases indexing health, psychiatric, trauma, and public health literatures, and therefore may have missed research on IPV/SA published in psychological journals. However, because our interest was chiefly in epidemiological (prevalence) research – and not literature on predictors or outcomes of IPV/SA – it is likely that most relevant literature was captured by our search. Additionally, caution should be used in interpreting reported meta-analytic confidence intervals and prediction intervals, as both statistical precision and heterogeneity can be poorly estimated in small meta-analyses. Lastly, our search for recruitment and data collection methods for IPV/SA was nonsystematic because it used included studies as a convenience sample. As a result, our findings may not reflect the complete array of methods used in research on IPV/SA prevalence.

CONSIDERATIONS FOR FUTURE RESEARCH AND PRACTICE

As noted, methods for measuring IPV/SA prevalence varied considerably across included studies. Inconsistency among, and limitations of, study sampling approaches and measurement instruments, modalities, and settings may lead to inaccurate and poorly representative prevalence estimates. For example, the widely used CTS/CTS-2 instrument does not account for contextual factors such as whether violence was perpetrated in self-defense or in aggression, possibly leading to misclassification of IPV/SA.¹⁴ Moreover, terminology used in IPV/SA measures such as abuse may lead to respondents under-reporting important but more subtle forms of psychological/emotional harm such as microaggressions, which can be perpetrated within intimate partnerships.⁴⁷ Finally, occurrence of IPV/SA among racial/ethnic minority and sexual and gender minority (LGBTQ+) Veterans appears understudied, which may reflect both sampling and recruitment limitations and a lack of culturally appropriate assessment tools.

Additionally, existing reviews and studies have found conflicting evidence on the role of screening modality in IPV/SA measurement accuracy and acceptability. One available meta-analysis⁴⁸ found that in primary care settings there is no apparent difference in IPV/SA detection rates between face-to-face interviews and computer-based surveys. A second review⁴⁹ in a broader array of clinical settings found that detection rates were higher using computer-based surveys compared to paper surveys and face-to-face interviews. Qualitative evidence from a study⁵⁰ conducted in the VA found that Veteran women prefer surveys over face-to-face screening at their primary care appointments and are more likely to disclose IPV/SA experiences after building trust with their primary care provider over multiple visits. Finally, although IPV/SA recorded in clinical records may be the most readily accessible form of IPV/SA prevalence data within a large health system, IPV/SA measured in clinical settings may mispresent the true prevalence of IPV/SA in comparison with survey research (eg, because of hesitancy to disclose IPV/SA to a medical provider).⁵¹

CONCLUSIONS

Findings of this review indicate that IPV/SA is prevalent among Veteran women and men. Evidence is strongest for past-year experienced IPV/SA among Veteran women, while for Veteran men and spouses/partners of Veterans, less and lower strength evidence is available. Compared with experienced IPV/SA, evidence on perpetrated IPV/SA is more limited. Although the amount and strength of evidence varied, psychological/emotional IPV appears to be the most common form of experienced and perpetrated IPV/SA for both Veteran women and men. Future studies of IPV/SA prevalence among Veterans should attempt to generate prevalence estimates that are applicable to Veterans of the range of ages, sexual and gender identities, races/ethnicities, and geographic contexts present in the Veteran population.

REFERENCES

1.: Breiding MJ, Basile KC, Smith SG, Black MC, Mahendra R. Intimate Partner Violence Surveillance: Uniform Definitions and Recommended Data Elements, Version 2.0. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2015.
2.: Caetano R, Schafer J, Field C, Nelson SM. Agreement on Reports of Intimate Partner Violence Among White, Black, and Hispanic Couples in the United States. Journal of Interpersonal Violence. 2016;17(12):1308–1322.
3.: Chan KL. Gender differences in self-reports of intimate partner violence: A review. Aggression and Violent Behavior. 2011;16(2):167–175.
4.: Emery CR. Examining an Extension of Johnson’s Hypothesis: Is Male Perpetrated Intimate Partner Violence More Underreported than Female Violence? Journal of Family Violence. 2009;25(2):173–181.
5.: Follingstad DR, Rogers MJ. Validity Concerns in the Measurement of Women’s and Men’s Report of Intimate Partner Violence. Sex Roles. 2013;69(3–4):149–167.
6.: Black MC, Basile KC, Breiding MJ, et al. The National Intimate Partner and Sexual Violence Survey: 2010 Summary Report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention; 2011.
7.: Dichter ME, Cerulli C, Bossarte RM. Intimate partner violence victimization among women veterans and associated heart health risks. Womens Health Issues. 2011;21(4 Suppl):S190–194. [PubMed: 21724140]
8.: Dichter ME, Marcus SC, Wagner C, Bonomi AE. Associations Between Psychological, Physical, and Sexual Intimate Partner Violence and Health Outcomes Among Women Veteran VA Patients. Social Work in Mental Health. 2014;12(5/6):411–428.
9.: Iverson K, Vogt D, Maskin RM, & Smith, BN. Intimate partner violence victimization and associated implications for health and functioning among male and female post-9/11 Veterans. In. Medical Care. Vol 552017:S78–S84. [PubMed: 28806369]
10.: Iverson KM, Sayer NA, Meterko M, et al. Intimate Partner Violence Among Female OEF/OIF/OND Veterans Who Were Evaluated for Traumatic Brain Injury in the Veterans Health Administration: A Preliminary Investigation. Journal of Interpersonal Violence. 2020;35(13–14):2422–2445. [PubMed: 29294714]
11.: Gierisch JM, Shapiro A, Grant NN, King HA, McDuffie JR, Williams JW. Intimate partner violence: Prevalence among U.S. military Veterans and active duty servicemembers and a review of intervention approaches. Department of Veterans Affairs VA Evidence based Synthesis Program Reports. 2013;08:08. [PubMed: 25232637]
12.: Whiting P, Savovic J, Higgins JP, et al. ROBIS: A new tool to assess risk of bias in systematic reviews was developed. J Clin Epidemiol. 2016;69:225–234. [PMC free article: PMC4687950] [PubMed: 26092286]
13.: Hoy D, Brooks P, Woolf A, et al. Assessing risk of bias in prevalence studies: Modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012;65(9):934–939. [PubMed: 22742910]
14.: Kwan J, Sparrow K, Facer-Irwin E, Thandi G, Fear NT, MacManus D. Prevalence of intimate partner violence perpetration among military populations: A systematic review and meta-analysis. Aggression & Violent Behavior. 2020;53:101419. [PMC free article: PMC7375166] [PubMed: 32714067]
15.: Luterek J, Bittinger J, Simpson T. Posttraumatic sequelae associated with military sexual trauma in female Veterans enrolled in VA outpatient mental health clinics. J Trauma Dissociation. 2011;12(3):261–274. [PubMed: 21534095]
16.: Viechtbauer W. metafor: Meta-analysis package for R. The Comprehensive R Archive Network. https://cran.r-project.org/web/packages/metafor/index.html. Published 2020. Accessed 2020.
17.: Bartlett BA, Iverson KM, Mitchell KS. Intimate partner violence and disordered eating among male and female veterans. Psychiatry Research. 2018;260:98–104. [PubMed: 29179017]
18.: Cerulli C, Bossarte RM, Dichter ME. Exploring intimate partner violence status among male veterans and associated health outcomes. American Journal of Mens Health. 2014;8(1):66–73. [PubMed: 23832953]
19.: Cerulli C, Stephens B, Bossarte R. Examining the Intersection Between Suicidal Behaviors and Intimate Partner Violence Among a Sample of Males Receiving Services From the Veterans Health Administration. American Journal of Mens Health. 2014;8(5):440–443. [PubMed: 24574165]
20.: Combellick JL, Dziura J, Portnoy GA, Mattocks KM, Brandt CA, Haskell SG. Trauma and Sexual Risk: Do Men and Women Veterans Differ? Womens Health Issues. 2019;29 Suppl 1:S74–S82. [PubMed: 31253246]
21.: Dutra L, de Blank G, Scheiderer E, Taft C. Correlates of female Veterans’ perpetration of relationship aggression. Psychol Trauma. 2012;4(3):323–329.
22.: Gondolf EW, Foster RA. Wife assault among VA alcohol rehabilitation patients. Hospital & Community Psychiatry. 1991;42(1):74–79. [PubMed: 2026410]
23.: Savarese VW, Suvak MK, King LA, King DW. Relationships among alcohol use, hyperarousal, and marital abuse and violence in Vietnam veterans. Journal of Traumatic Stress. 2001;14(4):717–732. [PubMed: 11776419]
24.: Iverson KM, King MW, Gerber MR, et al. Accuracy of an intimate partner violence screening tool for female VHA patients: a replication and extension. Journal of Traumatic Stress. 2015;28(1):79–82. [PubMed: 25624170]
25.: Iverson KM, King MW, Resick PA, Gerber MR, Kimerling R, Vogt D. Clinical utility of an intimate partner violence screening tool for female VHA patients. Journal of General Internal Medicine. 2013;28(10):1288–1293. [PMC free article: PMC3785669] [PubMed: 23824907]
26.: Iverson KM, Pogoda TK. Traumatic brain injury among women veterans: an invisible wound of intimate partner violence. Medical Care. 2015;53(4 Suppl 1):S112–119. [PubMed: 25767964]
27.: Mahoney CT, Iverson KM. The Roles of Alcohol Use Severity and Posttraumatic Stress Disorder Symptoms as Risk Factors for Women’s Intimate Partner Violence Experiences. Journal of Women’s Health. 2020;29(6):827–836. [PMC free article: PMC7307688] [PubMed: 31905315]
28.: Dardis CM, Shipherd JC, Iverson KM. Intimate partner violence among women veterans by sexual orientation. Women & Health. 2017;57(7):775–791. [PubMed: 27322372]
29.: Huston JC, Grillo AR, Iverson KM, Mitchell KS, VA Boston Healthcare System. Associations between disordered eating and intimate partner violence mediated by depression and posttraumatic stress disorder symptoms in a female veteran sample. General Hospital Psychiatry. 2019;58:77–82. [PubMed: 30965163]
30.: Portnoy GA, Relyea MR, Street AE, Haskell SG, Iverson KM. A Longitudinal Analysis of Women Veterans’ Partner Violence Perpetration: the Roles of Interpersonal Trauma and Posttraumatic Stress Symptoms. Journal of Family Violence. 2020;35(4):361–372.
31.: Brignone E, Sorrentino AE, Roberts CB, Dichter ME. Suicidal ideation and behaviors among women veterans with recent exposure to intimate partner violence. General Hospital Psychiatry. 2018;55:60–64. [PubMed: 30412860]
32.: Bennett DC, Thomas EJ, Porter KE, Broman RB, Rauch SAM, Sexton MB. Context Matters: Posttraumatic Stress Disorder Symptom Associations With Military Sexual Trauma Event Characteristics and Dual Histories of Interpersonal Trauma. Violence & Victims. 2019;34(1):69–84. [PubMed: 30808794]
33.: Campbell R, Raja S. The sexual assault and secondary victimization of female veterans: Help-seeking experiences with military and civilian social systems. Psychology of Women Quarterly. 2005;29(1):97–106.
34.: Campbell R, Greeson MR, Bybee D, Raja S. The co-occurrence of childhood sexual abuse, adult sexual assault, intimate partner violence, and sexual harassment: a mediational model of posttraumatic stress disorder and physical health outcomes. Journal of Consulting & Clinical Psychology. 2008;76(2):194–207. [PubMed: 18377117]
35.: Caralis P, Musialowski R. Women’s experiences with domestic violence and their attitudes and expectations regarding medical care of abuse victims. South Med J. 1997;90(11):1075–1080. [PubMed: 9386045]
36.: Coyle BS, Wolan DL, Van Horn AS. The prevalence of physical and sexual abuse in women veterans seeking care at a Veterans Affairs Medical Center. Military Medicine. 1996;161(10):588–593. [PubMed: 8918119]
37.: Creech SK, Macdonald A, Taft C. Use and Experience of Recent Intimate Partner Violence Among Women Veterans Who Deployed to Iraq and Afghanistan. Partner Abuse. 2017;8(3):251–271.
38.: Creech SK, Pulverman CS, Kroll-Desrosiers A, Kinney R, Dichter ME, Mattocks K. Intimate Partner Violence Among Pregnant Veterans: Prevalence, Associated Mental Health Conditions, and Health Care Utilization. Journal of General Internal Medicine. 2021;19:19. [PMC free article: PMC8481457] [PubMed: 33464464]
39.: Dichter ME, Wagner C, True G. Timing of Intimate Partner Violence in Relationship to Military Service Among Women Veterans. Military Medicine. 2015;180(11):1124–1127. [PubMed: 26540701]
40.: Dichter ME, Wagner C, Borrero S, Broyles L, Montgomery AE. Intimate partner violence, unhealthy alcohol use, and housing instability among women veterans in the Veterans Health Administration. Psychological Services. 2017;14(2):246–249. [PubMed: 28481611]
41.: Dobie DJ, Kivlahan DR, Maynard C, Bush KR, Davis TM, Bradley KA. Posttraumatic stress disorder in female veterans: association with self-reported health problems and functional impairment. Archives of Internal Medicine. 2004;164(4):394–400. [PubMed: 14980990]
42.: Kimerling R, Iverson K, Dichter M, et al. Prevalence of Intimate Partner Violence among Women Veterans who Utilize Veterans Health Administration Primary Care. JGIM: Journal of General Internal Medicine. 2016;31(8):888–894. [PMC free article: PMC4945568] [PubMed: 27130619]
43.: Rosenfeld EA, Miller E, Zhao X, Sileanu FE, Mor MK, Borrero S. Male partner reproductive coercion among women veterans. American Journal of Obstetrics & Gynecology. 2018;218(2):239.e231–239.e238. [PMC free article: PMC5807143] [PubMed: 29056537]
44.: Sadler A, Booth B, Cook B, Doebbeling BN. Factors associated with women’s risk of rape in the military environment. Am J Ind Med. 2003;43(3):262–273. [PubMed: 12594773]
45.: Iverson KM, Vogt D, Dichter ME, et al. Intimate Partner Violence and Current Mental Health Needs Among Female Veterans. Journal of the American Board of Family Medicine: JABFM. 2015;28(6):772–776. [PubMed: 26546653]
46.: Panuzio J, O’Farrell TJ, Marshall AD, Murphy CM, Murphy M, Taft CT. Intimate partner aggression reporting concordance and correlates of agreement among men with alcohol use disorders and their female partners. Assessment. 2006;13(3):266–279. [PubMed: 16880279]
47.: Pulice-Farrow L, Brown TD, Galupo MP. Transgender microaggressions in the context of romantic relationships. Psychology of Sexual Orientation and Gender Diversity. 2017;4(3):362–373.
48.: O’Doherty L, Hegarty K, Ramsay J, Davidson LL, Feder G, Taft A. Screening women for intimate partner violence in healthcare settings. Cochrane Database Syst Rev. 2015;2015(7):Cd007007. [PMC free article: PMC6599831] [PubMed: 26200817]
49.: Hussain N, Sprague S, Madden K, Hussain FN, Pindiprolu B, Bhandari M. A comparison of the types of screening tool administration methods used for the detection of intimate partner violence: a systematic review and meta-analysis. Trauma Violence Abuse. 2015;16(1):60–69. [PubMed: 24343478]
50.: Iverson KM, Huang K, Wells SY, Wright JD, Gerber MR, Wiltsey-Stirman S. Women veterans’ preferences for intimate partner violence screening and response procedures within the Veterans Health Administration. Res Nurs Health. 2014;37(4):302–311. [PubMed: 24990824]
51.: Dichter ME, Haywood TN, Butler AE, Bellamy SL, Iverson KM. Intimate Partner Violence Screening in the Veterans Health Administration: Demographic and Military Service Characteristics. American Journal of Preventive Medicine. 2017;52(6):761–768. [PubMed: 28209282]

Supplementary Materials

APPENDIX A. SEARCH STRATEGY

SYSTEMATIC REVIEWS

1. Search for current systematic reviews Date Searched: 08-05-21
A. Bibliographic Databases:	#	Search Statement	Results
MEDLINE: Systematic Reviews Ovid MEDLINE(R) ALL 1946 to August 04, 2021	1	Spouse Abuse/ OR Domestic Violence/ OR Battered Women/ OR Violence/ OR (((spouse OR spousal OR partner OR wife OR husband OR dating OR marital OR domestic) adj1 abuse) OR intimate partner violence OR ((domestic OR dating OR partner) adj1 violence) OR assaultive behavior$1 OR ((battered OR abused) adj (woman OR women)) OR ((psychological OR emotional OR physical) adj1 abuse) OR sexual partner$1 OR boyfriend OR girlfriend OR significant other OR couple OR romantic partner OR dyad).ti,ab.	113270
	2	Veterans/ OR Military Personnel/ OR (Veteran$1 OR armed forces OR military OR army OR navy OR marines OR air force OR active duty OR navy personnel OR naval personnel OR military personnel OR army person OR reservist OR reserve force OR coast guard OR soldier$1 OR sailor$1 OR army personnel OR air force personnel OR military OR submariner$1).ti,ab.	125516
	3	Prevalence/ OR Risk Factors/ OR (prevalence OR risk OR incidence OR rate$1 OR population OR statistics OR epidemiology OR statistical data).ti,ab.	6742062
	4	(systematic review.ti. or meta-analysis.pt. or meta-analysis.ti. or systematic literature review.ti. or this systematic review.tw. or pooling project.tw. or (systematic review.ti,ab. and review.pt.) or meta synthesis.ti. or meta-analy.ti. or integrative review.tw. or integrative research review.tw. or rapid review.tw. or umbrella review.tw. or consensus development conference.pt. or practice guideline.pt. or drug class reviews.ti. or cochrane database syst rev.jn. or acp journal club.jn. or health technol assess.jn. or evid rep technol assess summ.jn. or jbi database system rev implement rep.jn. or (clinical guideline and management).tw. or ((evidence based.ti. or evidence-based medicine/ or best practice.ti. or evidence synthesis.ti,ab.) and (((review.pt. or diseases category/ or behavior.mp.) and behavior mechanisms/) or therapeutics/ or evaluation studies.pt. or validation studies.pt. or guideline.pt. or pmcbook.mp.)) or (((systematic or systematically).tw. or critical.ti,ab. or study selection.tw. or ((predetermined or inclusion) and criteri).tw. or exclusion criteri.tw. or main outcome measures.tw. or standard of care.tw. or standards of care.tw.) and ((survey or surveys).ti,ab. or overview.tw. or review.ti,ab. or reviews.ti,ab. or search.tw. or handsearch.tw. or analysis.ti. or critique.ti,ab. or appraisal.tw. or (reduction.tw. and (risk/ or risk.tw.) and (death or recurrence).mp.)) and ((literature or articles or publications or publication or bibliography or bibliographies or published).ti,ab. or pooled data.tw. or unpublished.tw. or citation.tw. or citations.tw. or database.ti,ab. or internet.ti,ab. or textbooks.ti,ab. or references.tw. or scales.tw. or papers.tw. or datasets.tw. or trials.ti,ab. or meta-analy*.tw. or (clinical and studies).ti,ab. or treatment outcome/ or treatment outcome.tw. or pmcbook.mp.))) not (letter or newspaper article).pt.	467370
	5	1 AND 2 AND 3 AND 4	37
	6	limit 5 to english language	37
CDSR: Protocols and Reviews EBM Reviews - Cochrane Database of Systematic Reviews 2005 to August 4, 2021	1	(Spouse Abuse OR Domestic Violence OR Battered Women OR Violence).kw.	23
	2	(((spouse OR spousal OR partner OR wife OR husband OR dating OR marital OR domestic) adj1 abuse) OR intimate partner violence OR ((domestic OR dating OR partner) adj1 violence) OR assaultive behavior$1 OR ((battered OR abused) adj (woman OR women)) OR ((psychological OR emotional OR physical) adj1 abuse) OR sexual partner$1 OR boyfriend OR girlfriend OR significant other OR couple OR romantic partner OR dyad).ti,ab.	47
	3	1 OR 2	61
	4	(Veterans OR Military Personnel).kw.	3
	5	(Veteran$1 OR armed forces OR military OR army OR navy OR marines OR air force OR active duty OR navy personnel OR naval personnel OR military personnel OR army person OR reservist OR reserve force OR coast guard OR soldier$1 OR sailor$1 OR army personnel OR air force personnel OR military OR submariner$1).ti,ab.	29
	6	4 OR 5	29
	7	(Prevalence OR Risk Factors).kw.	73
	8	(prevalence OR risk OR incidence OR rate$1 OR population OR statistics OR epidemiology OR statistical data).ti,ab.	7158
	9	7 OR 8	7166
	10	3 AND 6 AND 9	0

1. Search for current systematic reviews (limited to last 7 years) Date Searched: 08-05-21
B. Non-bibliographic databases	Evidence	Results
AHRQ: evidence reports, technology assessments, U.S Preventative Services Task Force Evidence Synthesis	http://www.ahrq.gov/research/findings/evidence-based-reports/search.html Search: veteran; military; intimate partner violence	0
CADTH	https://www.cadth.ca Search: veteran; military; intimate partner violence	0
ECRI Institute	https://guidelines.ecri.org/ Search: veteran; military; intimate partner violence	0
HTA: Health Technology Assessments (UP TO 2016)	http://www.ohsu.edu/xd/education/library/ See Cochrane search above	0
NHS Evidence	http://www.evidence.nhs.uk/default.aspx Search: veteran; military; intimate partner violence	91
EPPI-Centre	http://eppi.ioe.ac.uk/cms/Default.aspx?tabid=62 Use browser search function [CNTL + F] for keyword search Search: veteran; military; intimate partner violence	0
NLM	http://www.ncbi.nlm.nih.gov/books Search: veteran; military; intimate partner violence	0
VA Products - VATAP, PBM and HSR&D publications	http://www.hsrd.research.va.gov/research/default.cfm http://www.research.va.gov/research_topics/ https://va.dimensions.ai/discover/publication Search: veteran; military; intimate partner violence Intimate Partner Violence, Health, and Health Care Among Female Veterans. January 2012-December 2016. https://www.hsrd.research.va.gov/research/abstracts.cfm?Project_ID=2141701684 Intimate Partner Violence: Patient Characteristics, Service Use and Experiences. October 2015-March 2019. https://www.hsrd.research.va.gov/research/abstracts.cfm?Project_ID=2141704480 Addressing Intimate Partner Violence Among Women Veterans: Evaluating the Impact and Effectiveness of VHA’s Response. November 2019-April 2023. https://www.hsrd.research.va.gov/research/abstracts.cfm?Project_ID=2141706237 High Prevalence of Intimate Partner Violence among Women Veterans – Up to Age 55 – Using VA Primary Care. HSR&D Pub Brief. https://www.hsrd.research.va.gov/research/citations/PubBriefs/articles.cfm?RecordID=795 Overview of Intimate Partner Violence: Current State of Knowledge in Regard to Women Veterans. HSR&D Pub Brief. https://www.hsrd.research.va.gov/research/citations/PubBriefs/articles.cfm?RecordID=632	5

2. Search for systematic reviews currently under development (includes forthcoming reviews & protocols)

Date Searched: 08-05-21

D. Under development:

Evidence:

Results:

AHRQ topics in development (EPC Status Report)

Email Charli Armstrong moc.liamg@crs.gnortsmrac

Email sent on 08-05

PROSPERO (SR registry)

http://www.crd.york.ac.uk/PROSPERO/

Katherine Sparrow, Deirdre MacManus. A systematic review of the prevalence of intimate partner violence victimisation among military personnel. PROSPERO 2016 CRD42016038800 Available from: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42016038800

Sean Cowlishaw, Dzenana Kartal, Alyssa Sbisa, Isabella Freijah. Prevalence of intimate partner violence (IPV) victimization and perpetration in military and veteran populations: A systematic review of population-based studies. PROSPERO 2020 CRD42020199214 Available from: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42020199214

Gursimran Thandi, Deirdre MacManus, Nicola Fear, Simon Wessely. Risk factors associated with Intimate Partner Violence (IPV) perpetration in military populations. PROSPERO 2014 CRD42014010307 Available from: https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42014010307

PRIMARY STUDIES

5. Search for primary literature Date searched: 08-05-21
MEDLINE [Ovid MEDLINE(R) ALL 1946 to August 04, 2021]
#	Search Statement	Results
1	Spouse Abuse/ OR Domestic Violence/ OR Battered Women/ OR Violence/ OR (((spouse OR spousal OR partner OR wife OR husband OR dating OR marital OR domestic) adj1 abuse) OR intimate partner violence OR ((domestic OR dating OR partner) adj1 violence) OR assaultive behavior$1 OR ((battered OR abused) adj (woman OR women)) OR ((psychological OR emotional OR physical) adj1 abuse) OR sexual partner$1 OR boyfriend OR girlfriend OR significant other OR couple OR romantic partner OR dyad).ti,ab.	113270
2	Veterans/ OR Military Personnel/ OR (Veteran$1 OR armed forces OR military OR army OR navy OR marines OR air force OR active duty OR navy personnel OR naval personnel OR military personnel OR army person OR reservist OR reserve force OR coast guard OR soldier$1 OR sailor$1 OR army personnel OR air force personnel OR military OR submariner$1).ti,ab.	125516
3	Prevalence/ OR Risk Factors/ OR (prevalence OR risk OR incidence OR rate$1 OR population OR statistics OR epidemiology OR statistical data).ti,ab.	6742062
4	1 AND 2 AND 3	757
5	Limit 4 to English language	735
CINAHL
#	Search Statement	Results
1	TI ( (MH “Intimate Partner Violence”) OR (MH “Domestic Violence”) OR (MH “Battered Women”) OR (MH “Dating Violence”) OR (MH “Violence”) OR (((spouse OR spousal OR partner OR wife OR husband OR dating OR marital OR domestic) adj1 abuse) OR intimate partner violence OR ((domestic OR dating OR partner) adj1 violence) OR assaultive behavior$1 OR ((battered OR abused) adj (woman OR women)) OR ((psychological OR emotional OR physical) adj1 abuse) OR sexual partner$1 OR boyfriend OR girlfriend OR significant other OR couple OR romantic partner OR dyad) ) OR AB ( (MH “Intimate Partner Violence”) OR (MH “Domestic Violence”) OR (MH “Battered Women”) OR (MH “Dating Violence”) OR (MH “Violence”) OR (((spouse OR spousal OR partner OR wife OR husband OR dating OR marital OR domestic) adj1 abuse) OR intimate partner violence OR ((domestic OR dating OR partner) adj1 violence) OR assaultive behavior$1 OR ((battered OR abused) adj (woman OR women)) OR ((psychological OR emotional OR physical) adj1 abuse) OR sexual partner$1 OR boyfriend OR girlfriend OR significant other OR couple OR romantic partner OR dyad) )	65722
2	TI ( (MH “Veterans+”) OR (MH “Military Personnel+”) OR (Veteran$1 OR armed forces OR military OR army OR navy OR marines OR air force OR active duty OR navy personnel OR naval personnel OR military personnel OR army person OR reservist OR reserve force OR coast guard OR soldier$1 OR sailor$1 OR army personnel OR air force personnel OR military OR submariner$1) ) OR AB ( (MH “Veterans+”) OR (MH “Military Personnel+”) OR (Veteran$1 OR armed forces OR military OR army OR navy OR marines OR air force OR active duty OR navy personnel OR naval personnel OR military personnel OR army person OR reservist OR reserve force OR coast guard OR soldier$1 OR sailor$1 OR army personnel OR air force personnel OR military OR submariner$1) )	49890
3	TI ( (MH “Prevalence”) OR (MH “Risk Factors+”) OR (prevalence OR risk OR incidence OR rate$1 OR population OR statistics OR epidemiology OR statistical data) ) OR AB ( (MH “Prevalence”) OR (MH “Risk Factors+”) OR (prevalence OR risk OR incidence OR rate$1 OR population OR statistics OR epidemiology OR statistical data) )	1391173
4	1 AND 2 AND 3	347

APPENDIX B. EXCLUDED STUDIES

Exclude reasons: 1=Ineligible population, 2=Ineligible intervention, 3=Ineligible comparator, 4=Ineligible outcome (including duplicative prevalence estimate), 5=Ineligible timing, 6=Ineligible study design, 7=Ineligible publication type, 8=Outdated or ineligible systematic review.

Citation	Exclude Reason
Beckham JC, Feldman ME, Kirby AC, Hertzberg MA, Moore SD. Interpersonal violence and its correlates in Vietnam Veterans with chronic posttraumatic stress disorder. J Clin Psychol. 1997;53(8):859–869. doi:10.1002/(sici)1097-4679(199712)53:8<859::aid-jclp11>3.0.co;2-j [PubMed: 9403389] [CrossRef]	E4
Begić D, Jokić-Begić N. Aggressive behavior in combat veterans with post-traumatic stress disorder. Mil Med. 2001;166(8):671–676. [PubMed: 11515314]	E4
Bossarte RM. Challenges associated with the use of policy to identify and manage risk for suicide and interpersonal violence among Veterans and other Americans. Adm Policy Ment Health. 2018;45(4):692–695. doi:10.1007/s10488-018-0882-x [PMC free article: PMC6416781] [PubMed: 29789982] [CrossRef]	E7
Calhoun PS, Van Voorhees EE, Elbogen EB, et al. Nonsuicidal self-injury and interpersonal violence in U.S. Veterans seeking help for posttraumatic stress disorder. Psychiatry Res. 2017;247:250–256. doi:10.1016/j.psychres.2016.11.032 [PMC free article: PMC5191947] [PubMed: 27930966] [CrossRef]	E4
Dao J. Preventing domestic violence in families of Veterans. J Clin Psychiatry. 2013;74(10):974–980. doi:10.4088/JCP.12124co1c [PubMed: 24229747] [CrossRef]	E7
Dichter ME, Haywood TN, Butler AE, Bellamy SL, Iverson KM. Intimate partner violence screening in the Veterans Health Administration: Demographic and military service characteristics. Am J Prev Med. 2017;52(6):761–768. doi:10.1016/j.amepre.2017.01.003 [PubMed: 28209282] [CrossRef]	E4
Dichter ME, Sorrentino A, Bellamy S, Medvedeva E, Roberts CB, Iverson KM. Disproportionate mental health burden associated with past-year intimate partner violence among women receiving care in the Veterans Health Administration. J Trauma Stress. 2017;30(6):555–563. doi:10.1002/jts.22241 [PubMed: 29193289] [CrossRef]	E4
Gerlock AA, Grimesey J, Sayre G. Military-related posttraumatic stress disorder and intimate relationship behaviors: a developing dyadic relationship model. J Marital Fam Ther. 2014;40(3):344–356. doi:10.1111/jmft.12017 [PubMed: 24749950] [CrossRef]	E4
Gobin RL, Green KE, Iverson KM. Alcohol misuse among female Veterans: Exploring associations with interpersonal violence and mental health. Subst Use Misuse. 2015;50(14):1765–1777. doi:10.3109/10826084.2015.1037398 [PubMed: 26642782] [CrossRef]	E4
Iverson KM, Mercado R, Carpenter SL, Street AE. Intimate partner violence among women Veterans: previous interpersonal violence as a risk factor. J Trauma Stress. 2013;26(6):767–771. doi:10.1002/jts.21867 [PubMed: 24243652] [CrossRef]	E4
Iverson KM, Vogt D, Dichter ME, et al. Intimate partner violence and current mental health needs among female Veterans. J Am Board Fam Med. 2015;28(6):772–776. doi:10.3122/jabfm.2015.06.150154 [PubMed: 26546653] [CrossRef]	E4
Iverson KM, Stirman SW, Street AE, et al. Female Veterans’ preferences for counseling related to intimate partner violence: Informing patient-centered interventions. Gen Hosp Psychiatry. 2016;40:33–38. doi:10.1016/j.genhosppsych.2016.03.001 [PubMed: 27083252] [CrossRef]	E1
Iverson KM, Dardis CM, Pogoda TK. Traumatic brain injury and PTSD symptoms as a consequence of intimate partner violence. Compr Psychiatry. 2017;74:80–87. doi:10.1016/j.comppsych.2017.01.007 [PubMed: 28126481] [CrossRef]	E4
Mahoney CT, Iverson KM. The roles of alcohol use severity and posttraumatic stress disorder symptoms as risk factors for women’s intimate partner violence experiences. J Womens Health (Larchmt). 2020;29(6):827–836. doi:10.1089/jwh.2019.7944 [PMC free article: PMC7307688] [PubMed: 31905315] [CrossRef]	E4
Makaroun LK, Brignone E, Rosland AM, Dichter ME. Association of health conditions and health service utilization with intimate partner violence identified via routine screening among middle-aged and older women. JAMA Netw Open. 2020;3(4):e203138. Published 2020 Apr 1. doi:10.1001/jamanetworkopen.2020.3138 [PMC free article: PMC7175082] [PubMed: 32315066] [CrossRef]	E4
Maskin RM, Iverson KM, Vogt D, Smith BN. Associations between intimate partner violence victimization and employment outcomes among male and female post-9/11 veterans. Psychol Trauma. 2019;11(4):406–414. doi:10.1037/tra0000368 [PubMed: 30372099] [CrossRef]	E4
Miller TW, Veltkamp LJ. Family violence: clinical indicators among military and post-military personnel. Mil Med. 1993;158(12):766–771. [PubMed: 8108014]	E7
Montgomery AE, Sorrentino AE, Cusack MC, et al. Recent intimate partner violence and housing instability among women Veterans. Am J Prev Med. 2018;54(4):584–590. doi:10.1016/j.amepre.2018.01.020 [PubMed: 29433952] [CrossRef]	E4
Murdoch M, Polusny MA, Hodges J, O’Brien N. Prevalence of in-service and post-service sexual assault among combat and noncombat Veterans applying for Department of Veterans Affairs posttraumatic stress disorder disability benefits. Mil Med. 2004;169(5):392–395. doi:10.7205/milmed.169.5.392 [PubMed: 15186007] [CrossRef]	E4
Portnoy GA, Haskell SG, King MW, Maskin R, Gerber MR, Iverson KM. Accuracy and acceptability of a screening tool for identifying intimate partner violence perpetration among women Veterans: A pre-implementation evaluation. Womens Health Issues. 2018;28(5):439–445. doi:10.1016/j.whi.2018.04.003 [PubMed: 29885901] [CrossRef]	E4
Sayers SL, Farrow VA, Ross J, Oslin DW. Family problems among recently returned military Veterans referred for a mental health evaluation. J Clin Psychiatry. 2009;70(2):163–170. doi:10.4088/jcp.07m03863 [PubMed: 19210950] [CrossRef]	E4
Tiet QQ, Finney JW, Moos RH. Recent sexual abuse, physical abuse, and suicide attempts among male Veterans seeking psychiatric treatment. Psychiatr Serv. 2006;57(1):107–113. doi:10.1176/appi.ps.57.1.107 [PubMed: 16399970] [CrossRef]	E4

: Note. Excluded studies from the original ESP review¹ are not represented.

APPENDIX C. EVIDENCE TABLES

CHARACTERISTICS OF INCLUDED SYSTEMATIC REVIEWS

Characteristics of the Kwan et al² systematic review and meta-analysis are presented in the Literature Overview section of our Evidence Brief.

OUTCOME DATA OF INCLUDED SYSTEMATIC REVIEWS

All outcome data from Kwan et al² relevant to our current systematic review and meta-analysis are presented in the Results section of our Evidence Brief.

QUALITY ASSESSMENT OF INCLUDED SYSTEMATIC REVIEWS

Assessed SR: Kwan 2020²
Guiding Questions	Reviewer 1	Reviewer 2	Consensus	Comments
1. Eligibility Criteria
1.1 Did the review adhere to pre-defined objectives and eligibility criteria?	PY	PY	PY	PROSPERO registration number listed, but protocol not found.
1.2 Were the eligibility criteria appropriate for the review question?	Y	Y	Y
1.3 Were eligibility criteria unambiguous?	Y	Y	Y
1.4 Were all restrictions in eligibility criteria based on study characteristics appropriate (eg, date, sample size, study quality, outcomes measured)?	Y	PY	PY	Restricted to studies that used validated tools to measure IPV perpetration, which could result in over-representation of measurements with CTS, which has some issues with validity and reliability. Likely still appropriate given that these issues have been addressed/quantified for the CTS.
1.5 Were any restrictions in eligibility criteria based on sources of information appropriate (eg, publication status or format, language, availability of data)?	Y	Y	Y	Restricted to published English Language articles; likely limits generalizability to military populations in primarily English-speaking countries.
Concerns regarding specification of study eligibility criteria (LOW/HIGH/UNCLEAR)	Low	Low	Low	Rationale for concern: No major concerns.
2. Identification and Selection of Studies
2.1 Did the search include an appropriate range of databases/electronic sources for published and unpublished reports?	Y	Y	Y
2.2 Were methods additional to database searching used to identify relevant reports?	Y	Y	Y
2.3 Were the terms and structure of the search strategy likely to retrieve as many eligible studies as possible?	PY	PY	PY
2.4 Were restrictions based on date, publication format, or language appropriate?	Y	Y	Y
2.5 Were efforts made to minimize error in selection of studies?	Y	Y	Y
Concerns regarding methods used to identify/select studies (LOW/HIGH/UNCLEAR)	Low	Low	Low	Rationale for concern: No major concerns.
3. Data Collection and Study Appraisal
3.1 Were efforts made to minimize error in data collection?	NI	NI	NI
3.2 Were sufficient study characteristics available for both review authors and readers to be able to interpret the results?	Y	Y	Y
3.3 Were all relevant study results collected for use in the synthesis?	Y	Y	Y
3.4 Was risk of bias (or methodological quality) formally assessed using appropriate criteria?	Y	Y	Y
3.5 Were efforts made to minimize error in risk of bias assessment?	Y	Y	Y
Concerns regarding methods used to collect data/appraise studies (LOW/HIGH/UNCLEAR)	Low	Low	Low	Rationale for concern: Limited concerns overall; unclear whether data abstraction was checked by a second reviewer.
4. Synthesis and Findings
4.1 Did the synthesis include all studies that it should?	PY	PY	PY
4.2 Were all pre-defined analyses reported or departures explained?	PY	PY	PY
4.3 Was the synthesis appropriate given the nature and similarity in the research questions, study designs and outcomes across included studies?	Y	Y	Y
4.4 Was between-study variation (heterogeneity) minimal or addressed in the synthesis?	PY	PY	PY
4.5 Were the findings robust (eg, as demonstrated through funnel plot or sensitivity analyses)?	PY	PY	PY
4.6 Were biases in primary studies minimal or addressed in the synthesis?	PY	PY	PY	Addressed limitations of CTS.
Concerns regarding synthesis and findings (LOW/HIGH/UNCLEAR)	Low	Low	Low	Rationale for concern: Protocol does not appear to be locatable by PROSPERO ID, so cannot tell for certain whether there were departures from pre-defined analyses (a protocol that is likely this review’s but having a different ID (CRD42016038800) was located, and predefined analyses align); magnitude of heterogeneity or residual heterogeneity not reported, but relevant subgroup analyses were performed or subgroup estimates from studies were reported.
5. Risk of Bias in the Review
5.1 Did the interpretation of findings address all of the concerns identified in Domains 1 to 4?	Y	Y	Y
5.2 Was the relevance of identified studies to the review’s research question appropriately considered?	Y	Y	Y
5.3 Did the reviewers avoid emphasizing results on the basis of their statistical significance?	Y	Y	Y
Risk of bias in the review (LOW/HIGH/UNCLEAR)	Low	Low	Low	Rationale for risk of bias: Overall very few concerns. Reasonable inclusion criteria, comprehensive search, adequately described results, and appropriately contextualized findings.

: Abbreviations. NI=no information; PY=probably yes; Y=yes.

CHARACTERISTICS OF INCLUDED PRIMARY STUDIES

Primary study characteristics are presented in Table 1 of our Evidence Brief.

OUTCOME DATA OF INCLUDED PRIMARY STUDIES

A data table of prevalence estimates used in meta-analyses is available upon request by contacting vog.av@CC.PSE.

QUALITY ASSESSMENT OF INCLUDED PRIMARY STUDIES

Author Year

External validity: Target Population

1. Was the study’s target population a close representation of the national population in relation to relevant variables (eg, age, sex, occupation)?

External validity: Sampling Frame

2. Was the sampling frame a true or close representation of the target population?

External Validity: Random Selection

3. Was some form of random selection used to select the sample or was a census undertaken?

External Validity: Non-response Bias

4. Was the likelihood of non-response bias minimal (ie, ≥75% response rate or no significant demo-graphic difference between responders and non-responders)?

Internal Validity: Source of Information

5. Were data collected directly from the subjects (as opposed to a proxy)?

Internal Validity: Case Definition

6. Was an acceptable case definition used in the study?

Internal Validity: Study Instrument

7. Was the study instrument that measured the parameter of interest shown to have reliability and validity?

Internal Validity: Survey Modality

8. Was the same mode of data collection used for all subjects?

Internal Validity: Count Reports

9. Were the num. and denom. for the parameter of interest appropriate?

Overall Risk of Bias

Bartlett 2018³ (male/GFK sample)

Yes

Panel used random digit dialing and address-based sampling to generate nationally representative sample of the US pop, with sub-population of Veterans.

Current study re-sampled Veteran sub-pop, but only those who endorsed trauma exposure (2175/3157)

Yes

Random selection used.

Yes

Response rate was 76.38%

Yes

Responses completed by participants

Yes

Past-year violence or aggression by intimate partner

Yes

Used HARK instrument, which is recommended by the Institute of Medicine due to sensitivity and specificity

Yes

Participants were contacted and enrolled in various ways, but all completed the survey online (with provided internet access and hardware if needed)

Yes

Proportions are weighted, so we can’t double check calculations, but they appear consistent

Moderate

Study is likely to overestimate prevalence of IPV due to limitation of sampling to Veterans who endorsed trauma exposure

Bartlett 2018³ (female/New England VA cohort)

Target population was limited to the New England region.

Yes

Sampled randomly from a database of Veterans in the region.

Yes

Random selection used.

Only 70.73% of Veterans from an initial survey agreed to be recontacted, and only 79.8% of those who agreed to be recontacted completed the survey.

Yes

Responses completed by participants

Yes

Past-year violence or aggression by intimate partner

Yes

Used CTS-2 tool, which is widely used, validated, and aligned with CDC definition when using severe psychological/emotional aggression definitions (which the authors did)

Yes

Participants completed mail-in surveys

Numerators and denominators don’t match proportions (no mention of survey weighting)

High

Study is likely to underestimate the prevalence of IPV due to limitation of sample to New England region and high nonresponse; sample has higher proportion of White Veterans than national population

Bennett 2019⁴

Target population was limited to the Midwest and to Veterans with identified military sexual trauma

Limited to a single center in the VHA.

Yes

Census of pts treated for index military sexual trauma at the participating VHA.

Only 103/160 Veterans who completed screening had sufficient data for analysis and only 86 had IPV data. No analysis of pts with vs without missing data

Yes

Responses completed by participants

Yes

Lifetime IPV

It’s unclear which questionnaire included IPV screening and what questions the screening used.

Yes

Participants completed structured interviews in person.

Yes

Provides counts of pts who did vs did not experience lifetime IPV

High

Limitation of sample to Veterans who experienced military sexual trauma likely to have resulted in overestimate of the national prevalence of IPV.

Brignone 2018⁵

Yes

The study used data from as many VHA systems/sites as possible

It’s unclear how evenly distributed the 13 VHA sites were

Yes

Census of women treated at participating VHA sites

Study does not report the number of pts who were left out of the analysis due to missing IPV data

Yes

Responses completed by participants

Yes

Lifetime or past-year IPV

Yes

Used the E-HITS instrument, which researchers validated against the CTS-2

Yes

Participants completed in-person screening with a healthcare provider

Yes

Numerators and denominators are provided and/or can be calculated from available information

Moderate

Likely direction of potential bias is unclear. No comparisons are made between demographics of the included VHA networks and the national population of Veterans or between patients included vs excluded from the analysis

Campbell 2005⁶

Target population was predominantly racial and ethnic minority, low-income Veteran women population in the Midwest

Yes

Sampled randomly from a women’s VA clinic

Yes

Random selection used

Yes

Response rate was 88%

Yes

Responses completed by participants

May underestimate sexual IPV prevalence due to method of reporting and case definition; sexual assault was measured, which does not include all forms of sexual IPV

Yes

Used the Sexual Experiences Survey and incident report forms

Yes

Participants completed written surveys in-person at medical appointments

Proportion and denominator are reported, but not numerator

High

High proportion of low-income and racial or ethnic minorities compared to national population of Veterans likely to have resulted in overestimate of the national prevalence of IPV.

Campbell 2008⁷

Target population was predominantly racial and ethnic minority, low-income Veteran women population in the Midwest

Yes

Sampled randomly from a women’s VA clinic

Yes

Random selection used

Yes

Response rate was 88%

Yes

Responses completed by participants

Yes

Lifetime physical IPV

Yes

Study used CTS, which is validated/relia ble for physical IPV

Yes

Participants completed written surveys in-person at medical appointments

Only proportion is provided (no numerator)

High

High proportion of low-income and racial or ethnic minorities compared to national population of Veterans likely to have resulted in overestimate of the national prevalence of IPV.

Caralis 1997⁸

Target population was women treated at the VA

Sampling frame only included women attending Miami VA clinics

Study used convenience sampling (recruited pts attending clinic)

Yes

78.7% completion rate

Yes

All women were interviewed in person

Study combines physical and sexual IPV without assessing psychological IPV

Study used the Abuse Assessment Screen, which appears to be valid or abuse but unclear validity for IPV specifically

Yes

All participants were interviewed

Study reports denominators and proportions, but not denominators

High

Likely direction of potential bias is unclear given minimal information about the demographics and characteristics of the target population and sampling frame.

Cerulli 2014⁹ “Examining”

Yes

Target population was male Veterans in the US

Study only sampled from male Veteran population in upstate New York

Yes

Random selection used.

Response rate was 27%. No analysis of differences between responders and non-responders.

Yes

Responses completed by participants

Study reported “any IPV,” but questions only assessed sexual and physical IPV

Study used its own, unvalidated questions for screening

Yes

All participants interviewed by phone

Yes

Numerators and denominators are provided

High

Likely direction of potential bias is unclear. No comparison of respondents to non-respondents. No validation of study questionnaire/instrument. Use of psychological + sexual IPV may underestimate any IPV by leaving out physical IPV (though not by much, as psychological IPV generally accompanies other forms)

Cerulli 2014¹⁰ “Exploring”

Yes

Target population for the study was all men US Veterans

Only 8 states participating in the BRFSS surveyed IPV status in 2006; not representative of the Midwest or Northwest

Yes

Random selection used (random digit dialing)

Study does not report response rate or comparison between responders and non-responders

Yes

Responses answered directly by participants

Yes

Lifetime IPV

Study used a single question to assess IPV, developed by BRFSS researchers

Yes

All participants surveyed over the phone

Numerator not provided (just denominator and proportion)

High

Likely direction of potential bias is unclear. No comparisons are made between demographics of Veterans in the included states and the national population of Veterans or between respondents vs non-respondents. Study question used to assess IPV was not validated.

Combellick 2019¹¹

Target population was limited to Veterans during Operation Enduring Freedom, Operation Iraqi Freedom, and/or Operation New Dawn

Only included Veterans served at New England, Indianapolis, Los Angeles, and Durham VHAs; unclear how well these systems represent national population

Yes

Only included Took a Veterans census of served at New eligible England, women Indianapolis, Veterans and randomly sampled men

Only 1,094/9,912 Veterans responded/consented to participate and no comparison between responders and non-responders

Yes Responses answered directly by participants

Yes

IPV in the last 12 months (score of 7+ on E-HITS evaluated)

Yes

Study used E-HITS, which has been validated in Veteran populations

Yes

All participants completed written surveys

Yes

Numerators and denominators are provided and match reported proportions

High

Likely direction of potential bias is unclear. OEF/OIF/OND Veterans are likely to be younger than the national population of Veterans, and therefore more likely to report IPV than older veterans. However, no comparisons are made between demographics of Veterans in the included VHA networks and the national population of Veterans or between respondents vs non-respondents.

Coyle 1996¹²

Target population was women veterans in the Baltimore VAMC area

Yes

Surveys mailed to all women who received care in the previous 6 months

Yes

Study took a census of all pts treated in the study period

Study had a 52% response rate and respondents served longer on average than non-respondents

Yes

Surveys were completed by participants

Sexual IPV outcome was defined as rape by spouse or partner. Sexual abuse by spouse/partner was also reported, but reportedly less common despite a less stringent definition.

Study used ad-hoc questionnaire

Yes

All participants completed mail-in surveys

Numerators specific to abuse by spouse/partner were not reported, just proportions and denominators

High

Likely direction of potential bias is unclear given missing demographic info on race, ethnicity, and LGBTQ+ identity. Unclear how unvalidated survey instrument may have biased results.

Creech 2017¹³

Target population was limited to women Veterans in current intimate relationships who served during Operation Enduring Freedom, Operation Iraqi Freedom, and/or Operation New Dawn

Yes

Sampled from all women Veterans on the roster who resided within a VISN

Yes

Invited participants were randomly selected

Survey response rate was 27%

Yes

Responses were completed by participants

Yes

Past-year IPV

Yes

Used the CTS-2

Yes

Participants were first contacted by mail and provided with a link to an online survey

Yes Numerators and denominators are provided and match reported proportions

High

Likely direction of potential bias is unclear. OEF/OIF/OND Veterans are likely to be younger than the national population of Veterans, and therefore more likely to report IPV than older Veterans. However, no comparisons are made between respondents vs non-respondents.

Creech 2021¹⁴

Target population was limited to pregnant Veterans

Yes

Selected based on medical record entries indicating that Veteran was pregnant from 15 VHA hubs across the country

Yes

Study attempted to take a census of all pregnant Veterans

Only 38% of eligible patients participated in parent study (https://www.liebertpub.com/doi/10.1089/jwh.2018.7628), and only 71.3% of participants in the parent study were included in analysis, likely due to missing data from IPV screen; no comparison of responders vs non-responders provided

Yes

Responses completed by participants

Yes

Past-year IPV

Yes

Study used E-HITS, which has been validated in Veteran populations

Yes

All participants completed phone interviews

Yes

Numerators and denominators reported

High

Direction of potential sources of bias is difficult to predict without comparison between responders and non-responders.

Dardis 2017¹⁵

Yes

Panel used random digit dialing and address-based sampling to generate nationally representative sample of the US pop, with subpopulation of Veterans.

Yes

All women Veterans in the panel were invited to complete the survey

Yes

Random selection used.

Yes

Response rate was 75%

Yes

Responses completed by participants

Yes

Followed CDC recommended IPV definition (ie, physical, sexual, psychological aggression and stalking from a past or current intimate partner)

Yes

Used HARK instrument, which is recommended by the Institute of Medicine due to sensitivity and specificity

Yes

All participants completed a web-based survey

Yes

Numerators and denominators provided

Low

Dichter 2011¹⁶

Yes

Target population for the study was all women US Veterans

Only 8 states participating in the BRFSS surveyed IPV status in 2006; not representative of the Midwest or Northwest; no indication if rural and urban populations are adequately represented

Yes

Random selection used (random digit dialing)

Study does not report response rate or comparison between responders and non-responders

Yes

Responses answered directly by participants

Yes

Lifetime IPV

Study used a single question to assess IPV, developed by BRFSS researchers

Yes

All participants surveyed over the phone

Yes

Numerator and denominators are provided; proportions are adjusted for sample design and non-response, but are close to crude proportions.

High

Dichter 2014¹⁷ and 2015¹⁸

Target population was women Veterans seeking care at a single VA center

Yes

Study attempted to recruit all women visiting/seeking care at the center

Study used convenience sampling (waiting room and flyer recruitment)

Total number of women approached not reported, so nonresponse cannot be assessed

Yes

Responses completed by participants

Yes

Lifetime IPV

Yes

Study used CTS-2

Yes

All participants completed in-person interviews

Yes

Denominators and numerators not all reported in 2014 study, but they are available in 2015 and proportions align

High

Likely direction of bias is unclear. Study limited to care-seeking patients, which would typically result in overestimate of prevalence. In this case, it might have resulted in an underestimate if pts with IPV experiences avoided the triggering interview. Also unclear how demographics of sampling frame compared to national population of Veterans

Dichter 2017¹⁹ “IPV, Unhealthy alcohol use”

Target population was women Veterans treated at 2 regional VA centers

Yes

Universal IPV screening was used, and records used were consecutive

Yes

Appears to be a census (describes time period and consecutive records)

Yes

92.7% of participants had complete IPV data

Yes

Responses completed by participants

Yes

Past-year IPV

Yes

Study used E-HITS, which has been validated in Veteran populations

Yes

All participants completed paper surveys

Yes

Numerators and denominators reported; proportions match

High

Likely direction of bias is unclear given minimal information on demographics of sample.

Dobie 2004²⁰

Target population was women Veterans seen for care at the VA Puget Sound HCS

Yes

Surveys mailed to all women who received care over a 1-year period

Yes

Study took a census of all pts treated in the study period

Study had a 65% response rate and 62% completion rate; no comparisons made between respondents and non-respondents

Yes

Surveys were completed by participants

Yes

Lifetime physical IPV (study language may be misleading, as it presents results as “domestic violence by a partner” but only measured physical IPV; however, it is included in our meta-analysis as-measured, and not as-labeled).

Study used unvalidated survey question

Yes

All participants completed mail-in surveys

Yes

Numerators and denominators reported and match reported proportions

High

Unclear how well regional population of women Veterans generalizes to the national population. Unclear how survey questions may have biased results.

Dutra 2012²¹

Limited to male partners/spouses of female Vietnam Veterans

Yes

Veteran-partner dyads were sampled from the nationally representative NVVRS and NSVG studies

Sampling methods not specified (likely not random)

Response rate not reported

Data were collected from partners/spouses, but the methods are unclear. Veterans might have been present or nearby for the interviews, which could have affected the results. Double checked more detailed paper for interview methods.

Yes

Past-year physical IPV

Yes

Used the CTS, which has been validated for physical violence

Yes

All participants completed in-person interviews

Only proportions and denominators are provided

High

Likely direction of potential bias is unclear. Limitation of sample to partners of Vietnam Veterans with PTSD would like result in an overestimate of the national prevalence of IPV, but if Veterans were nearby for partner interviews, the estimated prevalence would likely be an underestimate of the true prevalence in the target population.

Gondolf 1991²²

Study limited to partners/spouses of Veterans in treatment for alcoholism

Study only included pts from 1 center and only some of their partners/spouses

Study did randomly select 50 Veteran-partner pairs, but the Veterans had to provide consent to have their partners/spouses participate. In this sense, it was not a true random sample of partners/spouses.

Yes

All partners/spouses of the Veterans who consented to their participation appear to have responded.

Data were collected from partners/spouses, but the methods are unclear. Veterans might have been present for the interviews, which could have affected the results.

Yes

Any past-year physical IPV

Yes

Study used CTS, which is validated/reliable for physical IPV

Method of data collection not described for partners/spouses

Yes

Numerators and denominators reported

High

Likely direction of potential bias is unclear. Limitation of sample to partners of Veterans in treatment for alcoholism would like result in an overestimate of the national prevalence of IPV, but Veterans had to consent to their spouse’s participation and may have been present for the interview, which likely resulted in an underestimate of the true prevalence in the target population.

Huston 2019²³

Yes

Panel used random digit dialing and address-based sampling to generate nationally representative sample of the US pop, with subpopulation of Veterans.

Yes

All women Veterans in the panel were invited to complete the survey

Yes

Random selection used.

Yes

Response rate was 34.7% but there were no differences between participants and non-participants

Yes

Responses completed by participants

Yes

Followed CDC recommended IPV definition (ie, physical, sexual, psychological aggression and stalking from a past or current intimate partner)

Yes

Used HARK instrument, which is recommended by the Institute of Medicine due to sensitivity and specificity

Yes

All participants completed a web-based survey

Yes

Numerators and denominators provided

Low

Iverson 2013²⁴ “Clinical Utility”

Target population was limited to the New England region.

Limited to Veterans who reported being in an intimate partner relationship over the past year

Yes

Random selection used.

Response rate was 63.5%, and responders were on average older than non-responders.

Yes

Responses completed by participants

Yes

Past-year violence or aggression by intimate partner

Yes

Used CTS-2 tool, which is widely used, validated, and aligned with CDC definition when using severe psychological/emotional aggression definitions (which the authors did)

Yes

Participants completed mail-in surveys

Yes

Numerators and denominators provided and reported proportions match

High

Study has unclear direction of likely bias. Limitation of sample to Veterans who were in intimate partnerships over the last year would likely result in overestimate of IPV prevalence, while limitation of sample to New England region and high nonresponse would likely result in underestimate (sample has higher proportion of older White Veterans than national population)

Iverson 2015²⁵ “Accuracy”

Target population was limited to the New England region.

Limited to Veterans who reported being in an intimate partner relationship over the past year

Yes

Random selection used.

Response rate was 50.0%, and responders were on average older than non-responders.

Yes

Responses completed by participants

Yes

Past-year violence or aggression by intimate partner

Yes

Used CTS-2 tool

Yes

Participants completed mail-in surveys

Yes

Numerators and denominators provided

High

Iverson 2015²⁶ “TBI”

Target population was limited to the New England region.

Limited to Veterans who reported being in an intimate partner relationship over the past year

Yes

Random selection used.

Response rate was 71%; study reports no difference in demographics between those who did vs did not return surveys, but there’s no comparison between those with complete vs incomplete data

Yes

Responses completed by participants

Yes

Past-year or lifetime violence or aggression by intimate partner

Yes

Used CTS-2 tool

Yes

Participants completed mail-in surveys

Not all numerators reported (just proportions and denominators)

High

Iverson 2017²⁷ “IPV”

Target population was limited to recently separated Veterans

Yes

Study sampled from a database that included all recently separated Veterans

Yes

Random selection used

No reporting of response rate at baseline; response rate at T2 was 64.2% without comparison between responders and non-responders

Yes

Responses completed by participants

Yes

6-month IPV

Yes

Study used the CTS-2

Yes

Participants completed paper/written surveys

Yes

Numerators are reported and denominators can be calculated. Proportions appear accurate, though 2 estimates may be off by <0.5%

High

Likely direction of potential bias is unclear. Recently separated Veterans are likely to be younger than the national population of Veterans, and therefore more likely to report IPV than older Veterans. However, no comparisons are made between respondents vs non-respondents.

Iverson 2020²⁸

Target population was limited to women Veterans wo experienced TBI and served during Operation Enduring Freedom, Operation Iraqi Freedom, and/or Operation New Dawn

Sampling frame only included Veterans who completed comprehensive TBI screening

Yes

All eligible pts were invited to participate (census)

Survey response rate was 16.2%

Yes

Responses were completed by participants

Yes

Lifetime IPV

Yes

Used the HARK

Participants had option to complete web or mail-in paper surveys

Numerators and denominators are provided and/or calculable, but do not always match reported proportions (seems likely that proportions are taken from pool of pts with available data for each demographic or IPV category, but missingness is not reported).

High

Likely direction of potential bias is unclear. OEF/OIF/OND Veterans are likely to be younger than the national population of Veterans, and therefore more likely to report IPV than older Veterans. However, no comparisons are made between respondents vs non-respondents or mail-in vs web-based survey respondents.

Kimerling 2016²⁹

Yes

Target population was women Veterans in the US

Yes

Study sampled from women Veterans who had at least 1 visit to the VHA in 2011

Yes

Random, representative selection used

Yes

84% participation rate

Yes

Responses completed by participants

Yes

Past-year IPV

Yes

Used the HARK

Yes

Interviews conducted over the phone

Yes

Proportions are weighted, so we can’t double check calculations, but they appear consistent and numerators and denominators are provided

Low

Luterek 2011³⁰

Target population was Veterans attending VA Puget Sound outpatient mental health clinics or specialty PTSD clinics

Yes

Sampled from database of eligible Veterans in the region

Yes

Participants were randomly selected

Yes

Response rate was 43%, but no significant demographic differences were found between responders and non-responders

Yes

Interviews were completed by participants

Case definition only refers to physical abuse and spouses (not including, non-married partner and sexual/psychological abuse)

Study used TLEQ, which is validated for trauma exposures but not specifically IPV (questions related to IPV appear minimal)

Yes

All participants completed in-person interviews

Proportions and denominators are reported, but not numerators

High

Limitation of sample to Veterans in treatment for general mental health and/or PTSD likely to have resulted in overestimate of the regional prevalence of IPV. Unclear how well regional population of women Veterans generalizes to the national population. Additionally, “spousal physical abuse” likely underestimates IPV.

Portnoy 2020³¹

Yes

Panel used random digit dialing and address-based sampling to generate nationally representative sample of the US pop, with subpopulation of Veterans.

Yes

All women Veterans in the panel were invited to complete the survey

Yes

Random selection used.

Study limited to women who responded at multiple time points (34% response rate)

Yes

Responses completed by participants

Yes

Followed CDC recommended IPV definition (ie, physical, sexual, psychological aggression and stalking from a past or current intimate partner)

Yes

Used the MST-2, MSA, and CTS-2

Yes

All participants completed a web-based survey

Yes

Numerators and denominators provided

Moderate

Study is likely to overestimate prevalence of IPV due to introduction of reporting bias in requirement that women respond at multiple time points.

Rosenfeld 2018³²

Target population was women age 18-44 who received care at the VA

Yes

Study sampled from women age 18-44 who had at least 1 visit to the VHA in a 12-month period

Yes

Random, representative selection used

Yes

Response rate was only 28%, but there were no statistical differences in demographics between participants and non-participants

Yes

Responses completed by participants

Case definition was women who experienced reproductive coercion, which is a type of IPV, but under-estimates the overall prevalence

Study did not report validation of interview questions

Yes

Computer-assisted phone interviews used for all participants

Yes

Numerators and denominators are provided and/or can be calculated from available information

High

Likely direction of potential bias is unclear. Young age range of the survey would likely result in overestimate of national prevalence of IPV, but strict limitation of case definition would likely result in underestimate.

Sadler 2003³³

Target population was women Veterans who served in the Vietnam, post-Vietnam, and the Persian Gulf eras; excluded older Veterans at time study was conducted and is outdated now that OIF/OEF/OND Veteran population has grown.

Yes

Selected from national registries of VA women from target eras

Yes

Random selection used

Only 558/2172 selected Veterans completed interviews; responders were older on average than non-responders

Yes

Responses completed by participants

Outcome definition was rape by a spouse or partner during military service

Study instrument was an ad-hoc questionnaire

Yes

All participants completed computer-assisted phone interviews

Yes

Numerators and denominators reported

High

Study likely underestimated the true prevalence of lifetime sexual IPV. Surveyed Veterans were older than the national population due to target population and demographics of responders. Additionally, case definition is more limited in scope and time period than sexual IPV.

Savarese 2001³⁴

Target population was spouses/partners of men Veterans of the Vietnam War who screened positive for PTSD

Yes

Veteranpartner dyads were sampled from the nationally representative NVVRS study

Sampling methods not specified (likely not random)

Yes

Response rate of the sub-sample was 80%

Veterans might have been nearby for the interviews, which could have affected the results if spouses who experienced IPV were did not feel safe/had concerns that their partners (Veterans) would hear their answers.

Yes

Past-year psychological and/or emotional IPV

Yes

Used the CTS, which has been validated for male-perpetrated violence

Yes

All participants completed in-person interviews

Yes

Denominators and numerators are reported and proportions align

High

: Abbreviations. BRFSS=Behavioral Risk Factor Surveillance System; CDC=Center for Disease Control and Prevention; COMFORT=Center for Maternal and Infant Outcomes and Research in Translation; CTS=Conflict Tactics Scale; denom.=denominator; DoD=US Department of Defense; E-HITS=Extended-Hurt, Insulted, Threaten, Scream; GfK=Growth from Knowledge; HARK=Humiliation, afraid, rape, kick; IPV=intimate partner violence; LGBTQ+=lesbian, gay, bisexual, transgender, queer; MSA=military sexual assault; MST=military sexual trauma; NSVG=National Survey of the Vietnam Generation; num.=numerator; NVVRS=National Vietnam Veterans Readjustment Study; OEF/OIF/OND=Operation Enduring Freedom/Operation Iraqi Freedom/Operation New Dawn; pts=patients; PTSD=posttraumatic stress disorder; subpop.=subpopulation; TBI=traumatic brain injury; TLEQ=Traumatic Life Events Questionnaire; US=United States; VA=Veterans’ Affairs; VAMC=Veterans’ Affairs medical center; VHA=Veterans’ Health Administration; WOMAN=Women’s Overall Mental Health Assessment of Needs.

STRENGTH OF EVIDENCE FOR INCLUDED STUDIES

Strength of evidence ratings are presented in the Results section of our Evidence Brief. 28

APPENDIX D. PEER REVIEW DISPOSITION

Comment #	Reviewer #	Comment	Author Response
Are the objectives, scope, and methods for this review clearly described?
1	1	Yes	None.
2	2	Yes	None.
3	3	Yes	None.
4	4	Yes	None.
5	6	Yes	None.
6	7	Yes	None.
7	8	Yes	None.
Is there any indication of bias in our synthesis of the evidence?
1	1	No	None.
2	2	No	None.
3	3	No	None.
4	4	Yes - Inherent bias in the availability of research reviewed not in the researchers presentation of the available research.	None.
5	6	No	None.
6	7	No	None.
7	8	No	None.
Are there any published or unpublished studies that we may have overlooked?
1	1	No	None.
2	2	No	None.
3	3	No	None.
4	4	No	None.
5	6	No	None.
6	7	No	None.
7	8	No	None.
Additional suggestions or comments can be provided below.
1	1	This report provides a good overview of the current state of literature regarding IPV/SA prevalence among Veterans. There are particular points that I believe warrant some additional attention/discussion/clarification, as detailed below.	None.
2	1	The estimated prevalence rate of IPV/SA among Veteran men is much higher for past-year than lifetime experience, suggesting wide variation in measurement across studies (given that lifetime experience would include past-year). The variation is noted but it might be helpful to more specifically note this disparity and address the difference in measurement more specifically that would account for this.	Thank you for this comment. We have added detail in the main text, noting the higher lifetime prevalence and that it may result from methodological variation across studies (including the use of ad hoc measures in those studies providing lifetime estimates).
3	1	Regarding the recommendation that: “Future research on the scope of IPV/SA among Veterans should collect data in safe and secure environments… Assessing IPV/SA in a safe and comfortable environment, potentially outside of participants’ homes” – more is needed on this: what is considered “safe and secure”? What are the limitations of the current studies on this? Much of the data are collected in healthcare settings that, depending on the specific context, may or may not feel safe and secure. It is not clear to the reader how an environment would be determined to be safe or secure for a respondent.	Thank you for this useful comment. We have considerably revised the future research section based on peer review feedback and clarification that the most relevant recommendations and considerations would be those that apply to prevalence assessed in clinical settings.
4	1	There is a lack of attention to the important difference between clinical screening and survey research. For example, the comment about “providing respondents the option of answering assessments face-to-face with a trusted provider or privately using a computer-, table-, or smartphone-based assessment” may be more relevant to clinical screening than research surveys (especially regarding “provider”). This distinction is also relevant to the discussion of “burdensomeness” of assessments – in a clinical setting, a shorter tool, like the HARK, may work best. In research, depending on the scope of the study, a longer tool may be beneficial and not burdensome.	See response to comment 3.
5	1	Furthermore, there are well-documented barriers to disclosure in clinical settings or reporting with documentation other than for research purposes (e.g., reporting to a military authority) and research has found much higher rates of IPV disclosure in survey research than clinical screening data. Prior literature (e.g., Dichter ME, Haywood TN, Butler AE, Bellamy SL, Iverson KM. Intimate partner violence screening in the Veterans Health Administration: demographic and military service characteristics. American journal of preventive medicine. 2017 Jun 1;52(6):761–8 [PubMed: 28209282].) has cautioned against interpreting IPV disclosure documented in clinical screening as prevalence. Given that this report uses data from studies that include both survey research and clinical data sources, this issue should be addressed in the report.	See response to comment 3. Additionally, regarding the distinction of survey vs clinically-derived estimates, in the present review we generally grouped the former into the “random/population” sample type category and the latter into the “convenience” category, then carried out sensitivity analyses to explore differences in prevalence estimates. Although these analyses were only possible for lifetime IPV estimates, and this categorization was also based on other factors than simply survey vs clinically-derived, the analyses do shed some light on the important consideration you bring up. We have also added an explicit mention of the potential misestimation of true prevalence when clinical IPV data is used, with the provided citation, to the Future Research section.
6	1	On page 10 – “objective measure such as military records” – I’m not sure that I would classify military records as “objective” (vs. a validated self-report measure). The military records are likely based on self-reporting of experience and impacted by decisions around recording / documenting these reports. I would suggest using language other than “objective” here given that there is still potential for bias or mis-/underrepresentation. The language of “objective” suggests that this data source is more factual than a “subjective” self-report measure; however, given the barriers to reporting experience to a source that would document in a military record, these records are likely to under-represent prevalence more so than the self-report measures.	Thank you for this comment. This language is used by the authors of the included review on perpetrated IPV prevalence; therefore we have maintained it.
7	1	When multiple papers reported on the same sample/dataset (duplicative estimates), how did you determine which paper to include and which to exclude? For example, Dichter Haywood et al 2017 (“Intimate partner violence screening in the Veterans Health Administration: Demographic and military service characteristics”), Dichter Sorrentino et al 2017 (“Disproportionate mental health burden associated with past-year intimate partner violence among women receiving care in the Veterans Health Administration”), and BrigNone. et al 2018 (“Suicidal ideation and behaviors among women veterans with recent exposure to intimate partner violence”) all use the same sample/dataset but only the BrigNone. article is included; the Dichter Haywood et al article, however, provides screening response data by demographic characteristics so might be most appropriate to include in this review.	As far as we are able to tell, both Brignone and Dichter Haywood break down estimates into similar or identical sociodemographic categories (in Table 1 and Table 3, respectively).
8	1	Page 21: “these metrics do not differentiate IPV/SA with a current partner vs a previous partner” – it would be helpful to clarify this statement with regards to methodological (or clinical) limitation. Given that it is common for people to experience ongoing IPV from a former partner, and that relationships may be fluid (i.e., in and out of relationship), it is not clear why differentiating the status of the relationship is particularly useful.	See response to comment 3.
9	1	Page 21: “it is clear that future research on IPV prevalence should account for risks of further IPV/SA as well as the impact disclosing and discussing IPV/SA may have on respondents” – it would be helpful to provide more discussion/clarification about this statement as well. There has been substantial research and literature on risks of revictimization and on impact of IPV/SA disclosure. These studies might not have met criteria for inclusion in this report but given their existence, the call for future research in this area may need to be reconsidered or justified.	See response to comment 3.
10	1	Page 22: “supplementing standardized assessments with a limited number of questions that more fully characterize the frequency and intensity of IPV/SA. For example, each standard item (eg, the HARK item Within the last year, have you been afraid of your partner or ex-partner?) could be accompanied by a prompt to complete a rating scale of how often in the last year this form of IPV occurred.” This recommendation may also benefit from reconsideration or further discussion. It is not clear to me that knowing the frequency of experience (especially on something like being afraid) would be particularly helpful for anything or reflect the intensity or the impact of the experience, nor that respondents would be able to recall the frequency of such experience with precision.	See response to comment 3.
11	2	Overall this review is well-written and conducted. There were a few areas that I thought could be clarified or would benefit from minor edits.	None.
12	2	1. In some IPV literature past partners who perpetrate IPV after the relationship is over are included in estimates. It would be helpful to clarify how the current review defined IPV in regards to past partners.	Thank you for the comment. We have considerably revised the Future Research section and removed mention of the past vs current partner issue based on other reviewers’ feedback. As another reviewer noted, it is not relevant to the aims of this review to distinguish past vs present partner IPV, and we made no attempt to do so.
13	2	2. It is unclear to me why the misclassification that can occur on the CTS/CTS2 would be a misclassification based on gender identity (p. 21, lines 31-33) rather than just a potential misclassification of aggression acted in defense as perpetration. People of all different gender identities could potentially act in self-defense. Please clarify or reword.	Thank you for this comment; we have clarified this sentence.
14	2	3. It appears that the recommendations made on p. 22 (lines 1-12) are primarily aimed at assessing experience of IPV. For example, do the authors have a brief measure of IPV perpetration they recommend?	Thank you for this question. We are not aware of a brief measure that covers all forms of perpetrated IPV.
15	3	The objectives, scope, and methods of the review were generally very clearly described. However, it would be helpful if the authors could clarify for what types of reasons they excluded 767 studies in the literature flowchart.	The 767 excluded studies were at the title/abstract screening stage; it is not uncommon to have many non-relevant search results and to exclude a large number of records during title/abstract screening. We conducted dual sequential title/abstract screening to ensure excluded records were appropriate for exclusion. Per the PRISMA guidelines, specific exclude reasons are provided at the full-text screening level only.
16	3	Additionally, it is not clear why the authors use the term sexual assault in addition to intimate partner violence. They do not define sexual assault, but they do define IPV, and it appears that sexual assault falls within the umbrella of IPV when between two intimate partners. When the authors stated in the background (pg. 4) that “the prevalence of SA among Veteran intimate partners and non-partners is also not fully understood,” it gave the impression that the authors were going to review sexual assault separately from IPV and were going to examine it between non-partners, but this was not the case since the authors later stated “excluding non-partner SA” on pg. 5. It would be helpful if the authors could clarify their use of SA and define it given the use of this term throughout the review.	Thank you for this comment. We have clarified the language in the Executive Summary and Background sections. The use of SA in the fashion noted is as a result of the legislative language associated with this review.
17	3	I did not find bias beyond what is already mentioned by the authors in their limitations section. Overall, I believe this is a very important review that sheds light on how much more work is needed to understand IPV in the Veteran population.	None.
18	4	page ii line 4: Change Elizabeth Estabrooks title from Acting Executive Director to Deputy Director	Corrected.
19	4	page 1 line 10: Editorial change: add “both” in front of “women and men”	Changed.
20	4	page 2 line 22: REcommended change to LGBTQ+, Here and thorughout the report. The Center for Women Veterans uses LGBTQ+, as does the VA now. This is important inclusive language that the VA adopted in the Summer, and we would like to continue that here. If this study is not inclusive of transgender Veterans, there will need to be an explanation of why please. Otherwise, let’s change this throughout to LGBTQ+ so that this Rapid Review reflects the intentional inclusivity of CWV and VA.	We did not exclude literature among gender minority Veterans but instead did not locate any prevalence estimates in this population. The studies we did find provided prevalence data among sexual minorities only (individuals self-identifying as lesbian, gay, or bisexual), and hence we felt it was more appropriate to use LGB because the available evidence provides no information on prevalence among transgender or queer/questioning Veterans (or Veterans with other sexual and/or gender minority identities).
21	4	page 3 line 49: Because transgender Veterans in particular are missing from this report, CWV would like the addition of a recommendation that identifies the need for inclusion of transgender Veterans in particular please.	Thank you for this suggestion, which has been implemented.
22	4	page 4 line 50: Change from “survey” to “study.” The language in the bill is “study.”	Corrected.
23	4	page 6 line 30: Throughout the document I changed the numbers so that all numbers under 10 are spelled out. From “1” to “one.”	ESP style conventions are to spell out numbers only when a number begins a sentence.
24	4	page 6 line 41: Change from “female” to “women.”	Corrected.
25	4	page 7 line 33, 38: Change to LGBTQ+	See response to comment 20.
26	4	page 7 line 40: “2” to “two”	See response to comment 23.
27	4	page 10 line 5: “1” to “one”	See response to comment 23.
28	4	page 10 line 11: “2” to “two”	See response to comment 23.
29	4	page 10 line 16: “1” to “one”	See response to comment 23.
30	4	page 10 line 58: “3” to “three”	See response to comment 23.
31	4	page 14 line 17: “1” to “one”	See response to comment 23.
32	4	page 14 line 24: “2” to “two”	See response to comment 23.
33	4	page 14 line 25: “1” to “one”	See response to comment 23.
34	4	page 14 line 33: “3” to “three”	See response to comment 23.
35	4	page 14 line 34: “1” to “one”; “5” to “five”	See response to comment 23.
36	4	page 14 line 45: Change to LGBTQ+-identifying	See response to comment 20.
37	4	page 14 line 46: Change to LGBTQ+	See response to comment 20.
38	4	page 14 line 58: “2” to “two”	See response to comment 23.
39	4	page 15 line 4: “1” to “one”; “6” to “six”	See response to comment 23.
40	4	page 15 line 35: “1” to “one”	See response to comment 23.
41	4	page 15 line 37: “8” to “eight”	See response to comment 23.
42	4	page 15 line 39: “3” to “three”; “6” to “six”	See response to comment 23.
43	4	page 15 line 49: Change to LGBTQ+	See response to comment 20.
44	4	page 15 line 50: Change to LGBTQ+	See response to comment 20.
45	4	page 16 line 21: “2” to “two”	See response to comment 23.
46	4	page 16 line 40: “1” to “one”	See response to comment 23.
47	4	page 16 line 43: “8” to “eight”	See response to comment 23.
48	4	page 16 line 46: “6” to “six”	See response to comment 23.
49	4	page 16 line 47: “3” to “three”	See response to comment 23.
50	4	page 16 line 54: Change to LGBTQ+	See response to comment 20.
51	4	page 16 line 56: Change to LGBTQ+	See response to comment 20.
52	4	page 17 line 42: “3” to “three”; “6” to “six”	See response to comment 23.
53	4	page 17 line 47: “1” to “one”	See response to comment 23.
54	4	page 17 line 53: Change to LGBTQ+	See response to comment 20.
55	4	page 17 line 55: Change to LGBTQ+	See response to comment 20.
56	4	page 18 line 34: “2” to “two”	See response to comment 23.
57	4	page 18 line 41: “3” to “three”	See response to comment 23.
58	4	page 18 line 46: “2” to “two”	See response to comment 23.
59	4	page 18 line 58: “1” to “one”	See response to comment 23.
60	4	page 19 line 18: “2” to “two”	See response to comment 23.
61	4	page 19 line 26: 1 to “one”	See response to comment 23.
62	4	page 20 line 28: Change to LGBTQ+	See response to comment 20.
63	4	page 21 line 36: Out of curiousity, did you examine how race and culture could impact the way a person responds to or answers the questions? Could the surveys/questions contain race/cultural bias that affect outcomes for individuals of different races and ethnicities?	This is a valuable question and research area. However, it may be best examined in primary research rather than a systematic review. We have acknowledged this possibility in the context of the next comment.
64	4	page 22 line 8: Consider a recommendation: identify assessment tools that are culturally appropriate and take into consideration race, ethnicity, gender and sexual identify (LGBTQ+ status).	Thank you for this comment; we have incorporated this suggestion.
65	6	As you will see in comments left in the report, Center for Women Veterans, and VA, uses LGBTQ+, not LGB when referencing the LGBTQ+ community, as it is a more inclusive phrase. CWV has used this for over a year, and VA has since July, 2021. We are requesting that this report align with CWV and VA language. This report, by using simply LGB limits the population, which may be perceived as excluding transgender Veterans and those who identify as Queer and beyond (identified by the + sign. In the report you have written a category of “Veterans of any Gender Identify,” but this is insufficient in identifying that researchers were inclusive of transgender Veterans. I did notice that the librarian pulled a couple of papers on transgender Veterans, and I am aware that there are likely a paucity of of data sources related to transgender Veterans and IPV/SA. However, it is important that we state the attempts and identify this as a limitation and a a need for further research in the future please.	See response to comment 20.
66	7	The authors should be commended for a rigorously conducted systematic review focused on IPV/SA prevalence in a Veteran population that is well-synthesized and presented. My specific comments/concerns are as follows:	None.
67	7	1. While the authors well-describe their search methods, and investigated a number of different platforms (e.g., MEDLINE, CINAHL, AHRQ), I wondered why they did not search PsycInfo (or other psychology or social science databases); often IPV papers are published outside of the medical literature so I worry that they may have missed some key articles.	Thank you for this important observation; it is true that our search focused on databases indexing health, psychiatric, trauma, and public health literatures, and therefore may have missed research on IPV/SA published in psychological journals. However, because our interest was chiefly in epidemiological (prevalence) research – and not literature on predictors or outcomes of IPV/SA – it is likely that most relevant literature was captured by our search. Nevertheless, we have added a note to this effect in the Limitations.
68	7	2. The Executive Summary does not well-represent their overarching findings and conclusions and should be revised (e.g., there are differences in experiences of IPV/SA between women and men, and this difference is not reflected and should be).	We respectfully disagree that the Executive Summary does not well-represent findings and conclusions. We agree that there are observed differences in prevalence by gender identity, which are noted in Table ES1, however methodological and strength of evidence variation and differences in the amount of available evidence by gender complicate this picture.
69	7	3. Though the audience for this piece may implicitly understand, the background does not clarify why prevalence of IPV/SA might be important to study in Veterans in particular. In fact, the findings (greater percentage of women versus men with histories of IPV/SA, predominance of psychological abuse, etc) largely parallel what is found over and over in civilian populations. There is no mention or reference to the huge body of literature examining IPV/SA prevalence in civilian populations and why (or whether) the authors expect rates to be different in Veterans.	Thank you for this comment. The focus of the review was to synthesize evidence on IPV prevalence among Veterans, and not to compare prevalence, predictors/explanatory factors, or outcomes of IPV to civilians. As a rapid review, we focus the background information on a general overview of the issue and the reason(s) why the review was requested.
70	7	4. In addition to the above comment, the authors only very generally touch on what is fairly well-known differences in men and women’s use of violence (e.g., men are more likely to use more severe violence and sexual violence, women often use violence in self-defense and yet the current measure do not allow for this contextual understanding). Some of this background context (which is well-described in the literature) is important background and context. Similarly, there is little justification for why they would look at sub-populations (including what is known currently in the literature).	See response to comment 69.
71	7	5. Methodologically, the biggest area of concern and confusion was the inclusion of the systematic review(s) and meta-analysis. First, the degree to which these articles are included (and why and how they are included) differs by section. On p. 10, the authors refer to 1 systematic review (and no meta-analyses) meeting inclusion criteria; in other sections, they refer to a meta-analysis and a systematic review and I believe in other sections two systematic reviews. Also, I am not used to seeing systematic reviews and meta-analyses being pulled into another systematic review in this way. Typically, I would expect that the existing reviews might be used to look for primary studies to be included in the current review. It is unclear to me the degree to which the authors just pulled in what others have published versus going back to the included studies and conducting their own analyses. This needs to be significantly clarified throughout the manuscript.	Thank you for this comment. As a rapid review, we do not duplicate recent, relevant, and high-quality systematic reviews. When the scope of a rapid review includes a research question addressed by one or more duplicative reviews, we summarize findings of the existing review(s). We have clarified in the Synthesis section that we conducted meta-analyses on experienced IPV prevalence, while the meta-analyses on perpetrated IPV were conducted by Kwan et al. and simply summarized by us.
72	7	6. On p. 14, lines 56-61 (and into the next page), I wonder if there is an error. The authors state that any lifetime IPV/SA for Veteran men was 12.6% but then report that past-year prevalence was 36.7%. These numbers do not make sense (pretty much always lifetime prevalence is higher than past year prevalence).	See response to comment 2.
73	7	7. Figures may be an efficient way to demonstrate their prevalence findings (more effective than the text alone).	Thank you for this suggestion. While we generally agree, given the many subsets of type and form of IPV, we felt a figure or set of figures presenting findings would be overly complex.
74	8	If there’s any sense of the potential overlap and/or additional value of the other systematic reviews that were found to be underway, it would be helpful to note that. Very much appreciate ESP’s work here under a rapid review mechanism.	Thank you for this comment. The information we were able to access on underway reviews is limited, therefore we cannot speak more to their content or methods. It is likely there will be some overlap, but this will largely depend on the timeline of those reviews, which is not clear from their registry information.
75	8	Page 6, Key Findings: bullet 1 (Veteran women and men, not Veterans women and men). 3rd bullet “perpetrated among Veteran men” is confusing and sounds like IPV among Veteran men vs. “by” – would recommend changing “among” to “by” for clarity.	Thank you for these suggestions, which have been implemented.
76	8	Lines 50-51, the present review was not about informing future VA research per se. Instead, it was requested in response to a Congressional mandate. While I can understand that you may not want to be that blunt here, I would focus this line on “to assess the impact of IPV/SA in the Veteran population and among Veteran partners” or some such since that was the purpose of the review going into our discussions with ESP. It may inform research as a side benefit but I think it may be problematic to make it sound like that was the primary purpose.	Thank you for this suggestion, which has been implemented.
77	8	Page 7, line 22, should probably spell out LGB on 1st use.	LGB is defined at first use (in the Synthesis section, page 7).
78	8	Page 8, bullet on line 30, am not familiar with the HARK tool and suspect other readers may not be either. Spelling out 1st use (unless it’s someone’s last name) and adding a link to a citation or resource would be helpful here. I am similarly not familiar with “planned missingness designs” so a citation or link to a resource would be helpful here as well (line 47).	Thank you for these suggestions. HARK is defined at first use in the main text, but we have included its definition in the Executive Summary as well. Based on other comments, we have removed the suggestion of planned missing designs.
79	8	Page 9, line 14, so the challenge here is that the legislation indicates that a study must be done, and VA Central Office decided against that being primary data collection. And technically, research will not be funding the work but instead operations, which means we are doing evaluation work since operations funds cannot be used for research. It may be better to say that “Findings from this Evidence Brief will be used to respond to activities required by section 5305…” Of particular importance is the need for edits to lines 49-50 – the legislation does not specify a “baseline survey” but instead a “baseline study” – this is important because VA Central Office decided to interpret that as doing a bunch of secondary analyses of existing data as being responsive. No new primary data collection is being done. I have double checked the language in section 5305, and it does indeed say “baseline study” so if you could please make that change here, that would be important and helpful.	Thank you for these suggestions, which have been implemented.
80	8	Page 14, figure 1, was the large # of excluded papers (n=767) because they were papers of IPV/SA-only studies? I re-read the section on eligibility and exclusion a couple of times, and that is all I could glean and I could imagine readers wondering the same thing. I noted that the Literature Overview came after figure 1, so I thought at first that it may contain insights into the large exclusion group, but it focuses on included articles, so nothing in that section solves the issue I raise here. Just want to make sure it is clear to readers why such a large number of studies were exited out of the review.	See response to comment 15.
81	8	Table 1 is tremendous! Results are clearly written and important contributions. Really appreciated the methodological recommendations as well.	None.

APPENDIX E. RESEARCH IN PROGRESS

Status	Study Title	Study Design	Information Resources
In progress	A systematic review of the prevalence of intimate partner violence victimisation among military personnel	Systematic Review	CRD42016038800
In progress	Prevalence of intimate partner violence (IPV) victimization and perpetration in military and veteran populations: A systematic review of population-based studies.	Systematic Review	CRD42020199214
In progress	Risk factors associated with Intimate Partner Violence (IPV) perpetration in military populations.	Systematic Review	CRD42014010307

REFERENCES

1.: Gierisch JM, Shapiro A, Grant NN, King HA, McDuffie JR, Williams JW. Intimate partner violence: Prevalence among U.S. military Veterans and active duty servicemembers and a review of intervention approaches. Department of Veterans Affairs VA Evidence based Synthesis Program Reports. 2013;08:08. [PubMed: 25232637]
2.: Kwan J, Sparrow K, Facer-Irwin E, Thandi G, Fear NT, MacManus D. Prevalence of intimate partner violence perpetration among military populations: A systematic review and meta-analysis. Aggression & Violent Behavior. 2020;53:101419. [PMC free article: PMC7375166] [PubMed: 32714067]
3.: Bartlett BA, Iverson KM, Mitchell KS. Intimate partner violence and disordered eating among male and female veterans. Psychiatry Research. 2018;260:98–104. [PubMed: 29179017]
4.: Bennett DC, Thomas EJ, Porter KE, Broman RB, Rauch SAM, Sexton MB. Context Matters: Posttraumatic Stress Disorder Symptom Associations With Military Sexual Trauma Event Characteristics and Dual Histories of Interpersonal Trauma. Violence & Victims. 2019;34(1):69–84. [PubMed: 30808794]
5.: Brignone E, Sorrentino AE, Roberts CB, Dichter ME. Suicidal ideation and behaviors among women veterans with recent exposure to intimate partner violence. General Hospital Psychiatry. 2018;55:60–64. [PubMed: 30412860]
6.: Campbell R, Raja S. The sexual assault and secondary victimization of female veterans: Help-seeking experiences with military and civilian social systems. Psychology of Women Quarterly. 2005;29(1):97–106.
7.: Campbell R, Greeson MR, Bybee D, Raja S. The co-occurrence of childhood sexual abuse, adult sexual assault, intimate partner violence, and sexual harassment: a mediational model of posttraumatic stress disorder and physical health outcomes. Journal of Consulting & Clinical Psychology. 2008;76(2):194–207. [PubMed: 18377117]
8.: Caralis P, Musialowski R. Women’s experiences with domestic violence and their attitudes and expectations regarding medical care of abuse victims. South Med J. 1997;90(11):1075–1080. [PubMed: 9386045]
9.: Cerulli C, Stephens B, Bossarte R. Examining the Intersection Between Suicidal Behaviors and Intimate Partner Violence Among a Sample of Males Receiving Services From the Veterans Health Administration. American Journal of Mens Health. 2014;8(5):440–443. [PubMed: 24574165]
10.: Cerulli C, Bossarte RM, Dichter ME. Exploring intimate partner violence status among male veterans and associated health outcomes. American Journal of Mens Health. 2014;8(1):66–73. [PubMed: 23832953]
11.: Combellick JL, Dziura J, Portnoy GA, Mattocks KM, Brandt CA, Haskell SG. Trauma and Sexual Risk: Do Men and Women Veterans Differ? Womens Health Issues. 2019;29 Suppl 1:S74–S82. [PubMed: 31253246]
12.: Coyle BS, Wolan DL, Van Horn AS. The prevalence of physical and sexual abuse in women veterans seeking care at a Veterans Affairs Medical Center. Military Medicine. 1996;161(10):588–593. [PubMed: 8918119]
13.: Creech SK, Macdonald A, Taft C. Use and Experience of Recent Intimate Partner Violence Among Women Veterans Who Deployed to Iraq and Afghanistan. Partner Abuse. 2017;8(3):251–271.
14.: Creech SK, Pulverman CS, Kroll-Desrosiers A, Kinney R, Dichter ME, Mattocks K. Intimate Partner Violence Among Pregnant Veterans: Prevalence, Associated Mental Health Conditions, and Health Care Utilization. Journal of General Internal Medicine. 2021;19:19. [PMC free article: PMC8481457] [PubMed: 33464464]
15.: Dardis CM, Shipherd JC, Iverson KM. Intimate partner violence among women veterans by sexual orientation. Women & Health. 2017;57(7):775–791. [PubMed: 27322372]
16.: Dichter ME, Cerulli C, Bossarte RM. Intimate partner violence victimization among women veterans and associated heart health risks. Womens Health Issues. 2011;21(4 Suppl):S190–194. [PubMed: 21724140]
17.: Dichter ME, Marcus SC, Wagner C, Bonomi AE. Associations Between Psychological, Physical, and Sexual Intimate Partner Violence and Health Outcomes Among Women Veteran VA Patients. Social Work in Mental Health. 2014;12(5/6):411–428.
18.: Dichter ME, Wagner C, True G. Timing of Intimate Partner Violence in Relationship to Military Service Among Women Veterans. Military Medicine. 2015;180(11):1124–1127. [PubMed: 26540701]
19.: Dichter ME, Wagner C, Borrero S, Broyles L, Montgomery AE. Intimate partner violence, unhealthy alcohol use, and housing instability among women veterans in the Veterans Health Administration. Psychological Services. 2017;14(2):246–249. [PubMed: 28481611]
20.: Dobie DJ, Kivlahan DR, Maynard C, Bush KR, Davis TM, Bradley KA. Posttraumatic stress disorder in female veterans: association with self-reported health problems and functional impairment. Archives of Internal Medicine. 2004;164(4):394–400. [PubMed: 14980990]
21.: Dutra L, de Blank G, Scheiderer E, Taft C. Correlates of female Veterans’ perpetration of relationship aggression. Psychol Trauma. 2012;4(3):323–329.
22.: Gondolf EW, Foster RA. Wife assault among VA alcohol rehabilitation patients. Hospital & Community Psychiatry. 1991;42(1):74–79. [PubMed: 2026410]
23.: Huston JC, Grillo AR, Iverson KM, Mitchell KS, System VABH. Associations between disordered eating and intimate partner violence mediated by depression and posttraumatic stress disorder symptoms in a female veteran sample. General Hospital Psychiatry. 2019;58:77–82. [PubMed: 30965163]
24.: Iverson KM, King MW, Resick PA, Gerber MR, Kimerling R, Vogt D. Clinical utility of an intimate partner violence screening tool for female VHA patients. Journal of General Internal Medicine. 2013;28(10):1288–1293. [PMC free article: PMC3785669] [PubMed: 23824907]
25.: Iverson KM, King MW, Gerber MR, et al. Accuracy of an intimate partner violence screening tool for female VHA patients: a replication and extension. Journal of Traumatic Stress. 2015;28(1):79–82. [PubMed: 25624170]
26.: Iverson KM, Pogoda TK. Traumatic brain injury among women veterans: an invisible wound of intimate partner violence. Medical Care. 2015;53(4 Suppl 1):S112–119. [PubMed: 25767964]
27.: Iverson K, Vogt D, Maskin RM, & Smith, BN. Intimate partner violence victimization and associated implications for health and functioning among male and female post-9/11 Veterans. In. Medical Care. Vol 552017:S78–S84. [PubMed: 28806369]
28.: Iverson KM, Sayer NA, Meterko M, et al. Intimate Partner Violence Among Female OEF/OIF/OND Veterans Who Were Evaluated for Traumatic Brain Injury in the Veterans Health Administration: A Preliminary Investigation. Journal of Interpersonal Violence. 2020;35(13–14):2422–2445. [PubMed: 29294714]
29.: Kimerling R, Iverson K, Dichter M, et al. Prevalence of Intimate Partner Violence among Women Veterans who Utilize Veterans Health Administration Primary Care. JGIM: Journal of General Internal Medicine. 2016;31(8):888–894. [PMC free article: PMC4945568] [PubMed: 27130619]
30.: Luterek J, Bittinger J, Simpson T. Posttraumatic sequelae associated with military sexual trauma in female Veterans enrolled in VA outpatient mental health clinics. J Trauma Dissociation. 2011;12(3):261–274. [PubMed: 21534095]
31.: Portnoy GA, Relyea MR, Street AE, Haskell SG, Iverson KM. A Longitudinal Analysis of Women Veterans’ Partner Violence Perpetration: the Roles of Interpersonal Trauma and Posttraumatic Stress Symptoms. Journal of Family Violence. 2020;35(4):361–372.
32.: Rosenfeld EA, Miller E, Zhao X, Sileanu FE, Mor MK, Borrero S. Male partner reproductive coercion among women veterans. American Journal of Obstetrics & Gynecology. 2018;218(2):239.e231–239.e238. [PMC free article: PMC5807143] [PubMed: 29056537]
33.: Sadler A, Booth B, Cook B, Doebbeling BN. Factors associated with women’s risk of rape in the military environment. Am J Ind Med. 2003;43(3):262–273. [PubMed: 12594773]
34.: Savarese VW, Suvak MK, King LA, King DW. Relationships among alcohol use, hyperarousal, and marital abuse and violence in Vietnam veterans. Journal of Traumatic Stress. 2001;14(4):717–732. [PubMed: 11776419]

Suggested citation:

Parr NJ, Young S, Ward R, Mackey K. Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans. Washington, DC: Evidence Synthesis Program, Health Services Research and Development Service, Office of Research and Development, Department of Veterans Affairs. VA ESP Project #09-199; 2021.

This report was prepared by the Evidence Synthesis Program Coordinating Center located at the VA Portland Health Care System, directed by Mark Helfand, MD, MPH, MS and funded by the Department of Veterans Affairs, Veterans Health Administration, Health Services Research and Development.

The findings and conclusions in this document are those of the author(s) who are responsible for its contents and do not necessarily represent the views of the Department of Veterans Affairs or the United States government. Therefore, no statement in this article should be construed as an official position of the Department of Veterans Affairs. No investigators have any affiliations or financial involvement (eg, employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties) that conflict with material presented in the report.

This publication is in the public domain and is therefore without copyright. All text from this work may be reprinted freely. Use of these materials should be acknowledged.

Bookshelf ID: NBK576986PMID: 35104063

PubReader
Print View
Cite this Page
Parr NJ, Young S, Ward R, et al. Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans. Washington (DC): Department of Veterans Affairs (US); 2021 Dec.
PDF version of this title (985K)

Other titles in this collection

VA Evidence-based Synthesis Program Reports

Related information

NLM Catalog
Related NLM Catalog Entries
PMC
PubMed Central citations
PubMed
Links to PubMed

Recent Activity

Clear Turn Off Turn On

Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Vet...
Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

Bookshelf

Evidence Brief: Prevalence of Intimate Partner Violence/Sexual Assault Among Veterans

Authors

Affiliations

PREFACE

ACKNOWLEDGMENTS

Operational Partners

Peer Reviewers

EXECUTIVE SUMMARY

Background

Methods

Key Findings

Table ES1

INTRODUCTION

PURPOSE

BACKGROUND

METHODS

PROTOCOL

KEY QUESTIONS

ELIGIBILITY CRITERIA

DATA SOURCES AND SEARCHES

DATA ABSTRACTION AND ASSESSMENT

SYNTHESIS

RESULTS

LITERATURE FLOW

Figure 1

LITERATURE OVERVIEW

Table 1

EXPERIENCED IPV/SA

Any IPV/SA

Veteran Women

Veteran Men

Spouses/Partners

Other Studies

Physical IPV

Veteran Women

Veteran Men

Spouses/Partners

Sexual IPV

Veteran Women

Veteran Men

Spouses/Partners

Psychological/Emotional IPV

Veteran Women

Veteran Men

Spouses/Partners

PERPETRATED IPV/SA

RECRUITMENT STRATEGIES AND DATA COLLECTION METHODS

DISCUSSION

LIMITATIONS

CONSIDERATIONS FOR FUTURE RESEARCH AND PRACTICE

CONCLUSIONS

REFERENCES

Supplementary Materials

APPENDIX A. SEARCH STRATEGY

SYSTEMATIC REVIEWS

PRIMARY STUDIES

APPENDIX B. EXCLUDED STUDIES

APPENDIX C. EVIDENCE TABLES

CHARACTERISTICS OF INCLUDED SYSTEMATIC REVIEWS

OUTCOME DATA OF INCLUDED SYSTEMATIC REVIEWS

QUALITY ASSESSMENT OF INCLUDED SYSTEMATIC REVIEWS

CHARACTERISTICS OF INCLUDED PRIMARY STUDIES

OUTCOME DATA OF INCLUDED PRIMARY STUDIES

QUALITY ASSESSMENT OF INCLUDED PRIMARY STUDIES

STRENGTH OF EVIDENCE FOR INCLUDED STUDIES

APPENDIX D. PEER REVIEW DISPOSITION

APPENDIX E. RESEARCH IN PROGRESS

REFERENCES

Suggested citation:

Views

In this Page

Other titles in this collection

Related information

Similar articles in PubMed

Recent Activity