Intimate partner violence (IPV), including physical, sexual, and psychological aggression, is a significant public health problem with far-reaching consequences. Experiencing IPV is associated with serious negative physical and psychological outcomes among civilians and veterans alike [1,2]. Research shows that women veterans are at increased risk of experiencing violence in relationships compared to civilian women [1]. Moreover, IPV is common, yet critically understudied, among veteran men.

Reported rates of IPV experience vary widely across studies, largely due to methodological differences [3,4]. Although US women veterans are more likely to experience lifetime IPV than men (approximately 45% vs 36%), the prevalence of past-year IPV is similar, at approximately 30% [5]. Research has found that as many as 55% of women veterans experience IPV in their lifetimes [6]. Women’s experiences of IPV are associated with adverse physical and mental health, including cardiovascular and respiratory problems, chronic pain, reproductive health challenges, posttraumatic stress, anxiety, depression, substance use, and elevated risk for suicide [2,7-10]. For veteran men, IPV experience is associated with poorer overall mental health; greater occupational impairment; and higher rates of depression, smoking, and heavy and binge drinking [11,12]. Despite high rates of IPV experience and adverse outcomes for men, little research has examined IPV screening and referral outcomes for this population. Moreover, although evidence demonstrates that transgender and nonbinary patients are at increased risk of IPV compared to cisgender patients [13,14], very little research has examined IPV screening and referral outcomes within this population [15].

The health care system plays an integral role in the detection of and response to IPV [16,17]. As such, the significant impact that IPV has on veterans across genders underscores the critical need for a comprehensive and effective health care response for patients seeking services through the Veterans Health Administration (VHA). In 2014, the VHA developed the national IPV Assistance Program to oversee and implement integrated services aimed at reducing the risk for IPV, including establishing IPV Assistance Program Coordinators at each Department of Veterans Affairs (VA) medical center across the country and providing clinical services and resources for IPV-related concerns through prevention, detection, and treatment [18]. Since the IPV Assistance Program’s inception, the implementation of IPV screening and response among women veterans has been an important priority area for the program.

The VHA policy for IPV detection and response parallels and expands on recommendations put forth by the US Preventive Services Task Force [19]. VHA policy requires annual IPV screening for women of reproductive age, as well as other patients who belong to a recognized high-risk group (eg, patients who are homeless or underhoused, those with co-occurring disorders, and patients with disability [20]) and provision of support, resources, and referrals [20]. Accordingly, enhancing the implementation of IPV screening and response among the women veteran patient population has been a major priority of VHA clinical practice and research efforts over the last decade. A robust body of literature has shown that IPV screening of women veterans during health care visits is essential as patients are unlikely to spontaneously disclose IPV but will often report their IPV experiences when asked by a clinician in a sensitive manner [21]. Evidence also demonstrates that screening increases the identification of IPV and enhances women veterans’ connections to and satisfaction with care, and women veterans report perceiving IPV screening as supportive, validating, and helpful [21-23].

However, there remain many system-level and clinician-reported barriers to screening women veterans, such as limited time and resources, discomfort in addressing IPV, lack of training, and competing priorities during health care visits [24,25]. Moreover, because the majority of IPV research and health care screening initiatives to date have targeted women, little is known about screening men for IPV, including men’s perceptions and experiences of screening, their willingness to disclose IPV during screening, and clinicians’ experiences and attitudes about screening men. These limitations highlight the need for a large-scale evaluation of IPV screening reach and effectiveness with men and patients of all gender identities to inform strategies for optimizing screening implementation across patient populations. To date, IPV screening implementation across health care systems, including the VHA, has targeted women of reproductive age and largely has occurred in primary care, obstetrics or gynecology, and urgent care settings. The recent VHA IPV screening expansion provides a unique opportunity for evaluation of the IPV screening and response protocol across VHA patient populations.

In response to growing evidence demonstrating that all VHA patient populations are at risk for experiencing IPV and associated negative health consequences (eg, men [12,26], women above reproductive age [27], and transgender and nonbinary veterans [13,14]), the VA National IPV Assistance Program has called for expanded IPV screening through the implementation of a “no wrong door” approach using the Relationship Health and Safety Clinical Reminder version 3 (RHS 3.0). This approach enables a patient-centered solution for detecting IPV such that patients are screened for IPV regardless of where they receive care within the health care system. The RHS 3.0 is a 2-part IPV screener, including a 5-item primary screen for IPV [28-30] and, if triggered, a 3-item secondary screen for risk of severe and potentially lethal IPV [31].

The RHS 3.0 was developed as a clinical reminder with a note template and approved for VHA enterprise-wide installation in August 2023. The clinical reminder prompts clinicians to screen women of reproductive age at least annually and is available and recommended for use with all patients outside of this target demographic, as well. Veterans can be screened for IPV and connected with support services wherever they present for care in the health care system, resulting in a “no wrong door” approach. The RHS 3.0 is administered using a standardized template in the electronic health record (EHR). Population health implementation support tools, such as clinical reminders and note templates, leverage standardization of the EHR to help systematize screening administration and data collection across a large patient population [32]. As the largest integrated health care system in the United States, the VHA serves over 9 million patients across 172 health care facilities [33], underscoring the importance of leveraging tools available in the EHR for reaching all patients who come into contact with the health care system.

To conduct IPV screening and complete the necessary clinical reminder steps, clinicians require training through an internet-based training module available to all VHA staff or through receiving training conducted by an IPV Assistance Program Coordinator. To support the adoption of the RHS 3.0 screening expansion, the IPV Assistance Program is combining a top-down approach, which includes disseminating education and materials throughout the VHA (eg, national live and recorded trainings, an IPV screening and response toolkit, and internet-based training modules), with facility-level implementation strategies at the discretion of local IPV Assistance Program Coordinators at each medical center.

A national screening expansion of this scale requires rigorous, systematic evaluation. Although prior work exists evaluating IPV screening implementation for women veterans in primary care settings [23-25,34-37], no effort to date has systematically evaluated the implementation of IPV screening among all patients across the entire health care system. Data are needed to assess the implementation of the screening expansion, as well as the impact of screening new patient populations. In this protocol paper, we describe the recently funded Partnered Evaluation of Relationship Health Innovations and Services through Mixed Methods (PRISM) initiative. The aims of this work are to (1) evaluate the implementation of the RHS 3.0 IPV screening and response to national expansion and (2) identify the impact of IPV screening and potential gender differences. Specifically, we will assess implementation outcomes across Reach, Effectiveness, Adoption, Implementation, and Maintenance (RE-AIM) domains and examine potential differences in outcomes by patient characteristics (eg, gender, age, race, ethnicity, sexual orientation, and marital status). To examine the “no wrong door” approach, we will also identify clinical settings most and least likely to adopt IPV screening, as well as yield disclosures during screening. We will examine the impact and potential gender differences by assessing service use and clinical outcomes following positive screens and exploring experiences and perceptions of clinicians who screened and patients who disclosed IPV during screening encounters.

The PRISM initiative is guided by 2 robust implementation science frameworks—the RE-AIM (outcomes framework [38]) and the updated Consolidated Framework for Implementation Research (CFIR 2.0; determinants framework [39]). Although RE-AIM helps guide the organization of evaluation outcomes, it does not necessarily explain the conditions that influence variation in outcomes across the system, including among patient subgroups and clinical settings. The CFIR 2.0 is particularly helpful for structuring the exploration of contextual factors essential for the implementation of innovative programs, including at the VHA [40], making it ideally suited to guide the evaluation of multilevel factors that impact the implementation success of the VHA IPV screening and response national expansion. Integrating these 2 frameworks will support the assessment of implementation outcomes alongside understanding potential barriers and enablers of screening implementation, clinicians’ experiences with IPV screening and patients’ experiences with being screened (particularly among newly targeted patient populations, like men), and the implementation process overall.

Evaluation of the RHS 3.0 IPV screening expansion will include mixed methods data across VA health care facilities nationally. We will integrate quantitative and qualitative data sources, including EHR data and qualitative interviews (see Table 1 for a summary of data sources and outcomes). Quantitative data will be extracted from the VA’s Corporate Data Warehouse (CDW), a centralized data repository that aggregates clinical, administrative, and financial data from the VA EHR across all 139 VA medical centers and satellite clinics [41]. Our evaluation sample will include all VHA veteran patients with at least 1 outpatient VHA health care encounter during the evaluation observation period. Qualitative data sources will include semistructured interviews with veterans, clinicians, and VA leadership from the national IPV Assistance Program. The development of the semistructured interviews was guided by the CFIR 2.0 domains. Specifically, clinician interviews will focus on CFIR 2.0 constructs related to the RHS 3.0 screening and response protocol itself, the outer and inner settings (ie, their clinics and facilities), responders’ experiences and perceptions of the screening and response (including perceived impacts for patients), and the implementation process. Veteran interviews will also focus on their perceptions of being screened, including experiences with the screening process itself and resulting outcomes (eg, services offered or received).

This evaluation is a quality improvement (QI) initiative jointly supported by the VA Care Management and Social Work Service’s IPV Assistance Program; the VA Quality Enhancement Research Initiative (QUERI); and the IPV Center for Implementation, Research, and Evaluation (IPV-CIRE) at VA Connecticut Health care System. The PRISM Initiative was designed for internal purposes in support of VA QI as an internal operations evaluation designated as nonresearch by VA, thus not requiring institutional review board approval [42]. Empirical research conducted with data collected from this QI initiative was approved by the VA Connecticut Health Care System institutional review board (protocol #1792152).

Quantitative data grounded in the RE-AIM outcomes will be extracted from the CDW and analyzed using a longitudinal observational design with repeated measurement periods at baseline (T0), year 1 (T1), and year 2 (T2) [38]. We operationalized reach as the proportion and representativeness of veteran patients who were administered IPV screening, and those administered the RHS 3.0 specifically, calculated as veterans screened out of those eligible for screening. Eligibility for screening was defined as having at least 1 VHA health care encounter during the observation period. To examine representativeness and potential disparities between those screened and not screened, we will use generalized linear mixed models (GLMM) to assess differences by patient characteristics (gender, age, race, ethnicity, sexual orientation, marital status, rurality, and housing instability status) over each of the 3 time points while specifying facilities as random effects to account for nesting [43,44]. We will also examine the change in the probability of being screened over time through GLMMs specifying facilities and patients as random effects. Additionally, we will examine whether the effect of time varies significantly across facilities. In post hoc analyses, we will investigate whether the change in the probability of being screened over time differs by patient characteristics.

Effectiveness, for this evaluation, focuses on determining the extent to which the RHS 3.0 expansion itself is effective. We defined the effectiveness of the expansion as whether there is an increase in the proportion of IPV cases detected, referrals offered, and universal education provided following the implementation of the RHS 3.0. Using the quantitative CDW data, we will calculate changes in the proportion of these factors following positive screens through the same GLMM process described above. Effectiveness will also be examined through the exploration of qualitative data regarding clinicians’ and veterans’ experiences with and perceptions of the RHS 3.0 and its potential unintended consequences. Adoption was operationalized as differences in screening uptake by facility and clinical setting [38]. We will combine stop codes, identifiers used by the VHA to track which clinic group and location provided a service, into meaningful categories of services (eg, primary care mental health, social work, etc) to examine descriptive statistics (frequencies and range) of screening uptake across medical center facilities and clinical settings. We will also assess adoption through an exploration of the barriers and facilitators related to IPV screening expansion via qualitative data.

Implementation fidelity was defined as the extent to which IPV screening and response procedures are implemented as intended (ie, per IPV Assistance Program guidelines) [20,38]. For example, we will identify the percentage of veterans screened out of those eligible, completion of the secondary screen based on positive primary screen responses, referrals offered, and universal IPV education provided. We will also explore contextual factors related to fidelity reported during qualitative interviews. Finally, to examine maintenance of the RHS 3.0 expansion, we will identify whether the proportion of veterans screened (and proportion screened across patient subgroups), percent of facilities using the RHS 3.0, and number of unique clinicians using the RHS 3.0 remains at, above, or below T0 levels during T1 and T2 [38].

In addition to examining the implementation of the IPV screening expansion, the PRISM initiative will evaluate the impact of expanded IPV screening on patients, clinicians, and the health care system as a whole. Unique from RE-AIM’s effectiveness domain, which targets whether the screening expansion itself is effective, determining the impact of expanded screening will involve examining linkages between screening and connection and engagement to VHA services and resources. Leveraging available CDW data, we will examine patients’ connection to, and use of, health care and social support services following IPV screening, including services for physical and mental health conditions and resources for social services or essential needs (see Table 1 for detailed description). Examination of CDW data will also enable the identification of responses to IPV cases identified as high risk for lethality or serious injury, including a number of safety plans completed or same-day consults placed. Additional analyses will include a calculation of the frequency of services received in the 60 days following a positive screen and the same day. We will determine associations between positive screens and health care or resource use by using GLMMs with facilities as random effects, controlling for service use in the 60 days prior to the screen. We will also examine rates of high-risk cases detected from secondary screener results and the proportion of safety plans completed among high-risk cases.