Recovering From Intimate Partner Violence Through Strengths and Empowerment: Findings From a Randomized Clinical Trial

J Clin Psychiatry 2022;83(1):21m14041

To cite: Iverson KM, Danitz SB, Shayani DR, et al. Recovering from Intimate Partner Violence through Strengths and Empowerment: findings from a randomized clinical trial. J Clin Psychiatry. 2022;83(1):21m14041.
To share: https://doi.org/10.4088/JCP.21m14041

© Copyright 2021 Physicians Postgraduate Press, Inc.

^aWomen’s Health Sciences Division of the National Center for PTSD, VA Boston Healthcare System, Boston, Massachusetts
^bDepartment of Psychiatry, Boston University School of Medicine, Boston, Massachusetts
^cDissemination and Training Division of the National Center for PTSD, VA Palo Alto Healthcare System, Palo Alto, California
^dDepartment of Psychiatry and Behavioral Sciences, Stanford University, Stanford, California
^eVA Center for the Study of Healthcare Innovation, Implementation and Policy, VA Greater Los Angeles Healthcare System, Los Angeles, California
^fDepartment of Psychiatry and Biobehavioral Sciences, University of California Los Angeles Geffen School of Medicine, Los Angeles, California
^gWestern New England University, Department of Psychology, Springfield, Massachusetts
^hWomen’s Health Center, VA Boston Healthcare System, Boston, Massachusetts
ⁱDepartment of General Internal Medicine, Boston University School of Medicine, Boston, Massachusetts
^jVA Center for Health Equity Research and Promotion, Corporal Michael J. Crescenz VA Medical Center, Philadelphia, Pennsylvania
^kSchool of Social Work, Temple University, Philadelphia, Pennsylvania
^lDr Mahoney is now affiliated with the University of Colorado at Colorado Springs and the Lyda Hill Institute for Human Resilience, Colorado Springs, Colorado
^mDr Gerber is now affiliated with the Department of Medicine, Albany Medical College, Albany, New York
*Corresponding author: Katherine M. Iverson, PhD, Women’s Health Sciences Division of the National Center for PTSD, VA Boston Healthcare System, 150 South Huntington St, Boston, MA 02130 (Katherine.Iverson@va.gov).

Intimate partner violence (IPV), including physical, psychological, and sexual violence from past or current intimate partners, is a population health problem.^1–3 Although individuals of all gender identities experience IPV, women are disproportionately affected in terms of health impacts.^2,3 At least 1 in 4 women in the United States (US) experiences IPV during their lifetime.^2–4 IPV is associated with numerous emotional and psychiatric issues, including low self-efficacy, empowerment, and quality of life; depression; posttraumatic stress disorder (PTSD); and suicidality.^1,5–8 Consequently, women who experience IPV present frequently to health care services.^9,10

Health care visits provide opportunities for IPV screening and referral to interventions.^11–13 Screening is effective in identifying IPV,¹⁴ but screening alone may not lead to positive health outcomes; individualized and structured interventions (eg, manualized protocols) are needed to improve health.^15,16 IPV advocacy services are typically provided in community-based settings and include education, empowerment counseling, linkages to resources, and safety planning.¹⁷ Yet, evidence suggests the effectiveness of advocacy is equivocal,¹⁸ perhaps due to the lack of structure and rigorous evaluation of these services. Additionally, some people may not want to go outside the health care setting for services. Psychotherapies targeting mental health conditions, particularly IPV-related PTSD, may also be effective for IPV survivors,^19–21 although not all survivors have a mental health condition or want psychotherapy. Further, enhancing quality of life, particularly as it relates to survivor’s ability to engage in the things that matter to them (ie, valued living), is an important treatment target.^8,22 Experiencing IPV (which often includes coercive control and psychological abuse) is inherently disempowering and can lead individuals to distrust their ability to make good decisions. As long recognized in the advocacy literature,^23–26 interventions that facilitate empowerment are important for this population. Principally, survivor-defined practice recognizes the complexities of survivors’ lives and emphasizes survivors’ personal goals and priorities beyond victimization and safety to include social and psychological well-being.^27,28

Recovering from IPV through Strengths and Empowerment (RISE) is a clinician-administered brief counseling intervention designed for delivery in health care settings that builds on survivor-defined practice to provide a bridge between community-based advocacy services and specialized psychotherapies for IPV survivors. RISE is rooted in overlapping principles of trauma-informed care^28,29 (eg, shared power, facilitates trust and emotional healing, collaborative treatment planning) and empowerment (eg, strengths-focused, fosters survivor voice and choice).^23,30 RISE is an individualized, variable-length (1–6 sessions), modular, structured manualized protocol that targets key concerns of IPV survivors and incorporates Motivational Interviewing (MI).³¹ MI emphasizes strengths and self-efficacy using a collaborative approach focusing on changes that individuals can control by themselves (eg, enhancing self-care and coping, social support, and health). Only a handful of studies have examined MI specifically for this population.²⁰ Findings from 2 studies with general health care–seeking IPV survivors demonstrate MI’s effectiveness in reducing depressive symptoms (though not self-efficacy or quality of life).^32,33 RISE, with its use of MI in addition to survivor-centered practice and individualized skills building, is theorized to improve self-efficacy and empowerment,³⁴ outcomes associated with increases in psychosocial health and safety among individuals who experience IPV.^5,35–39 Through use of MI, RISE is also theorized to facilitate changes that are consistent with survivors’ goals and values (ie, valued living) and increase confidence in managing health needs (ie, patient activation), which may help reduce psychological distress and IPV over time.³⁴ Feasibility, acceptability, and the potential helpfulness of RISE were demonstrated through pilot work in the Veterans Health Administration (VHA).³⁴

We examined the effectiveness of RISE compared to an advocacy-based enhanced care as usual (ECAU) condition consisting of education, safety planning, resources, and referrals in a sample of women VHA patients. Women Veterans are an important population for intervention testing as they experience high risk for IPV,^40–43 with 18.5% of VHA primary care patients experiencing past-year IPV.⁴⁴ VHA is implementing IPV screening programs and is in need of effective IPV interventions.⁴⁵ We hypothesized that RISE would improve self-efficacy, empowerment, patient activation, and valued living, as well as depressive symptoms and IPV, more than ECAU.

METHOD

Participants & Procedure

Participants were recruited for a clinical trial (registered in ClinicalTrials.gov; NCT03261700) between October 2018 and September 2020 from an urban area in the northeast US by self-referrals, clinician referrals, and recruitment letters mailed to VHA primary care patients.⁴⁶ Women were included if they reported (1) experiencing past-year IPV (ie, 1 or more incidents of physical, psychological, or sexual IPV on the Conflict Tactics Scale-Revised [CTS-2]),⁴⁷ (2) having received past-year VHA care, (3) being at least 18 years old, and (4) willingness to have sessions audio-recorded. Exclusion criteria were minimal to maximize external validity and included current bipolar or psychotic symptoms/disorder and homicidal and/or suicidal ideation with intent.

Figure 1 provides a CONSORT diagram of participant flow through the study. The study was approved by the VA Boston Healthcare System’s Institutional Review Board. Trained bachelor’s-level research assistants and doctoral-level psychology staff conducted screening, consent, and assessment procedures. Women completed a phone screening, and study staff reviewed health care records to determine eligibility (ie, past-year VHA care). Those eligible were scheduled for an in-person enrollment session; ineligible women were referred to alternative services. During the enrollment session, participants completed informed consent and pretreatment measures. Participants were then randomly assigned to RISE or ECAU, which they received immediately following their pretreatment assessment. Most participants (n = 50) received their intervention in person, as intended in the original design; 10 women received telehealth sessions due to COVID-19 (n = 6 enrolled during COVID-19). Participants were reassessed at 10 and 14 weeks post-enrollment. RISE participants also completed primary outcome measures before each session. Participants were compensated for pretreatment, 10-week, and 14-week assessments ($25, $50, and $75, respectively). Sixty participants enrolled and were randomized, of which 59 were included in intent-to-treat analyses (n = 1 dropped due to missing data at both post-enrollment assessments).

Measures

Pretreatment characteristics. Sociodemographic and mental health characteristics were assessed via self-report. Probable mental health conditions were determined using clinical cutoffs on the Center for Epidemiologic Studies Depression Scale (CES-D; 16+),⁴⁸ Depression Anxiety Stress Scale-Anxiety Subscale (8+),⁴⁹ and PTSD Checklist-5 (33+).^50,51

Primary outcomes. Self-efficacy was measured with the 10-item General Self-Efficacy Scale (GSES),⁵² which assesses optimistic self-beliefs to cope with difficult life demands.⁵³ Respondents rate statements such as “I am confident that I could deal efficiently with unexpected events” on a 4-point Likert scale. Items are summed, with higher scores indicating greater self-efficacy. The GSES has demonstrated good construct validity and internal consistency⁵³; Cronbach α was 0.88 in this study. Empowerment was assessed using the 28-item Personal Progress Scale-Revised (PPS-R).³⁸ Participants rate their agreement with statements such as “I am feeling in control of my life” on a 7-point scale (“almost never” to “almost always”). Items are summed; higher scores indicate greater personal empowerment. The PPS-R has been used to evaluate other empowerment-based IPV interventions^19,54,55 and has demonstrated good construct validity and internal reliability³⁸; Cronbach α in this study was 0.89. Patient activation, or confidence managing one’s health and health care, was assessed with the 13-item Patient Activation Measure (PAM-13).⁵⁶ An example item is “Taking an active role in my own health care is the most important thing that affects my health.” Responses are on a 4-point scale (1 = strongly disagree to 4 = strongly agree), and total scores are computed. Higher scores indicate higher patient activation. The PAM-13 has demonstrated good internal reliability and construct validity⁵⁶; Cronbach α was 0.88 in this study. Valued living was measured by the Valued Living Questionnaire (VLQ),⁵⁷ as values consistency is theorized to be enhanced during RISE sessions (eg, goal-setting related to the parent they want to be, if parenting is an important valued domain for the participant). This 2-part measure uses a 10-point scale to assess (a) the importance of 10 valued domains (eg, parenting, spirituality, community life, and physical self-care) and (b) the concordance between one’s actions and values in these domains. Scales are multiplied for a composite score; higher scores indicate greater valued living. The VLQ has demonstrated construct validity and internal reliability⁵⁷; Cronbach α was 0.84 in this study.

Secondary outcomes. Depressive symptoms were measured with the 20-item CES-D.⁴⁸ Participants rate frequency of past-week symptoms on a 5-point scale (0 = rarely/none of the time to 4 = most/all of the time). Higher scores indicate greater symptoms. The CES-D has demonstrated construct validity and internal reliability⁴⁸; Cronbach α was 0.90. Past-year IPV was assessed with CTS-2⁴⁷ subscales of physical (eg, shoving), sexual (eg, forced sex), and psychological (eg, threats) IPV. Count scores represented the number of different IPV acts endorsed, with higher scores indicating greater levels of IPV.¹⁹ The measure has demonstrated construct validity⁴⁷; Cronbach α was 0.95. Satisfaction was assessed with the valid and reliable Client Satisfaction Questionnaire (CSQ-8),⁵⁸ with higher scores indicating greater satisfaction (Cronbach α = .93).

Intervention: RISE

RISE is a variable-length psychosocial intervention³⁴ that includes 6 optional modules administered over 10 weeks: (A) Safety Planning, (B) Education on Health Effects of IPV and Warning Signs, (C) Improving Coping and Self-Care, (D) Enhancing Social Support, (E) Making Difficult Decisions, and (F) Connecting with Resources and Moving Forward. Session 1 covers the philosophy and structure of RISE and the participant’s experiences with IPV and introduces self-efficacy and goal setting. Participants review and discuss an agenda-setting handout and select a module that best suits their current needs and circumstances. Modules include handouts, exercises, and goal setting. Clinicians utilize MI³¹ to help women navigate ambivalence and articulate the importance and confidence of their goals. Modules do not need to be delivered sequentially nor do they all need to be covered, and women can choose to repeat modules. At the end of each session, women choose whether to schedule another session, and they are connected with VHA and community resources if desired. Empowerment is embedded throughout by facilitating personalized goal-setting related to personal values, amplifying self-efficacy in each session, and giving choice and control (eg, asking permission to share information and giving  women the ability to select modules and to determine whether to schedule a subsequent session). RISE providers integrate validation throughout sessions and maintain a nonjudgmental, survivor-centered stance (as opposed to being overtly directive in ways that are inconsistent with the survivor’s values). RISE’s theory, content, and structure are detailed previously.³⁴

Intervention: ECAU

The ECAU intervention was an advocacy-based approach that incorporates best practices for addressing IPV in VHA.¹⁵ Participants received a single 60-minute session with a provider. Participants are given VHA’s IPV educational brochure to guide information provision and discussion of the different forms of IPV and the effects of IPV on physical, mental, and social health. Participants are offered safety planning and information on VHA and community resources. ECAU includes supportive statements and validation throughout (eg, “I’m sorry this is happening to you”).

Resources/referrals. Community and VHA resources were identified via handouts for both interventions, and additional resources were discussed according to participants’ stated preferences. Community referrals involved providing contact information and offering to contact the resource together during session. Community referral included Vet Center counseling, IPV programs (eg, shelters), and a community program that offers individual and family services for Veterans. VHA referrals typically involved active referrals (ie, placing consults, warm handoffs) that spanned mental health, case management, housing, legal, and employment services.

Interventionists

Interventionists were clinical providers, including 1 social worker and 3 psychologists, employed at VHA with relevant clinical experience and expertise in trauma-informed care; all were women. Providers delivered both interventions within primary care and mental health clinics. Providers completed VHA IPV trainings and attended an in-person day-long RISE training, followed by weekly consultation with the RISE developer throughout the study. Sessions were audio-recorded, and 40% of the RISE sessions were rated for fidelity to core components by the RISE developer, which guided provider feedback. Sessions were rated for adherence (0 = not at all, 1 = at least partially done; a percentage was calculated for overall adherence) and competence (quality of delivering 9 key aspects of RISE, including nonjudgmental stance, amplifying self-efficacy, collaborative goal setting, MI skills, and module comprehensiveness) on a 7-point scale (0–2 = poor, 3 = adequate, 4 = good, 5–6 = excellent). Average ratings for adherence (89%) and competence (mean = 4.5; SD = 0.8) indicated good fidelity.

Statistical Analyses

All analyses were conducted on the intent-to-treat sample (n = 59). We calculated descriptive statistics both for sociodemographic characteristics and for the primary outcome variables at each assessment time for both RISE (pretreatment, sessions 1–6, 10-week assessment, 14-week assessment) and ECAU (pretreatment, 10-week assessment, 14-week assessment). Secondary outcomes (ie, depressive symptoms and IPV experiences) were only measured at pretreatment, 10-week follow-up, and 14-week follow-up). Independent samples t tests and χ² tests examined differences between conditions on sociodemographic variables, as well as pretreatment primary and secondary outcome variables. Multilevel growth curve modeling using the mixed procedure of SPSS⁵⁹ was conducted to evaluate changes in all outcomes from pretreatment until 14-week follow-up, both within each condition and across both conditions. This approach is ideal for analyzing clinical trial data, including those with small samples,^60,61 and uses all available data irrespective of whether participants completed all timepoints.⁶² Unconditional change models were initially compared to identify whether to model linear or nonlinear change. A smaller −2 log likelihood (ie, deviance) value combined with other fit statistics (ie, Akaike Information Criterion, Bayesian Information Criterion) were used to determine the best fitting model. Effect sizes (d) for changes within conditions and differences in change between conditions were calculated using Feingold’s⁶³ procedures for effect size estimates from growth curve analyses, which are comparable to those from traditional repeated measures designs.

RESULTS

Table 1 displays participant characteristics at pretreatment; there were no significant differences between RISE and ECAU conditions on any variable. On average, participants were in their late thirties; 45% identified as a racial minority, and over half (56.7%) reported annual household incomes above $45,000. Probable mental health diagnoses were common (depression: 78.3%; anxiety: 100%; PTSD: 68.3%). Just under half the sample (45.0%) reported continued involvement with the violent partner.

As indicated in Table 1, there were no significant differences between RISE and ECAU on any of the primary or secondary outcome variables at pretreatment. For RISE, 25 women (83.3%) completed at least 2 sessions, 21 (70.0%) completed at least 3 sessions, 17 (56.7%) completed at least 4 sessions, 14 (46.7%) completed at least 5 sessions, and 9 (30.0%) completed all 6 sessions. Reasons for stopping RISE among the 8 participants who completed only 1 or 2 RISE sessions were variable (n = 1 got what she needed, n = 2 were ambivalent/uninterested in more sessions, and n = 5 wanted more sessions but life stressors, physical health problems, or partner interference were barriers). The percentage of participants selecting each module were as follows: Improving Coping and Self-Care (73%; n = 22), Education on Health Effects of IPV and Warning Signs (60%; n = 18), Enhancing Social Support (53%; n = 16), Connecting with Resources and Moving Forward (53%; n = 16), Making Difficult Decisions (50%; n = 15), and Safety Planning (33%; n = 10). Seven participants received community referrals (RISE: n = 5, ECAU: n = 2), and 30 participants received 1 or more VHA referrals (RISE: n = 14; ECAU: n = 16).*

Table 2 displays mean scores for primary outcome measures over time.

Effect of Time

Unconditional growth models suggested that it was appropriate to proceed with modeling variables other than time or non-linear time on all outcomes. Linear versus non-linear change was determined for each outcome before conducting subsequent analyses.

Final Models

Table 3 displays results for analyses of intervention condition as a Level 2 predictor of change in primary and secondary outcomes; effect sizes are reported in Table 2. For self-efficacy (GSES) and empowerment (PPS-R), the linear change model produced smaller deviance models and fit statistics than non-linear models and was thus used for these multilevel models. Supporting our hypothesis, there was a significant time-by-condition effect in that RISE participants demonstrated higher increases in empowerment (d = 3.46) and self-efficacy (d = 1.09) than ECAU participants (Table 3). Both of these were large effect sizes.⁶⁴

For patient activation (PAM-13), the non-linear (ie, cubic transformation) change model produced smaller deviance models and fit statistics than the linear model and was applied. Contrary to our hypotheses, there was no significant time-by-condition effect for patient activation (d = 0.63), nor were there main effects of time or condition; these results indicate there was no significant difference within or across conditions. For valued living (VLQ), the linear change model was used based on deviance models and fit statistics. There also was not a significant time-by-condition effect for valued living (d = 0.23), nor was there a significant main effect of condition. However, there was a significant main effect of time, indicating both conditions demonstrated significant increases in valued living.

With regard to secondary outcomes, there were not significant time-by-condition effects, nor were there main effects of condition for either depressive symptoms or IPV, which were both modeled using linear change (Table 3). However, the main effect of time was significant for both outcomes, indicating that depressive symptoms and IPV significantly declined for both conditions (Table 4).

Satisfaction (CSQ-8) was significantly higher for RISE participants (mean = 29.30, SD = 2.78) than ECAU participants (mean = 25.07, SD = 4.85) at 10-week follow-up (t₅₄ = −3.93, P < .001); the effect size was large (d = 1.23).⁶⁴

DISCUSSION

There is a need for brief and accessible health care–based counseling interventions that can address the most salient concerns of women who experience IPV.¹⁵ Results from this RCT suggest that RISE is effective in enhancing the psychosocial health of women VHA patients who experience past-year IPV. Women who received RISE, relative to those who received enhanced care as usual (ECAU), demonstrated large increases in empowerment and self-efficacy following the intervention. These findings support that RISE’s focus on strengths (eg, discussing self-efficacy each session, including ways in which progress on goals and skills from the modules impact positive coping), amplification of survivor choice (eg, survivors choose modules, number of sessions, and goals), and incorporation of MI (eg, goal identification and navigating ambivalence) foster empowerment and self-efficacy. RISE participants also experienced significant improvements in other key psychosocial health outcomes (valued living, depression symptoms, and IPV) over time, although these outcomes improved at similar rates for women who received ECAU.

Despite RISE’s better effectiveness overall, the advocacy-based ECAU condition was a robust intervention in its own right, especially considering the efficiency of a one-time session. The similar effectiveness observed for valued living, depression, and IPV may reflect the interventions’ shared strategies of providing knowledge about IPV and resources and talking through potential solutions based on women’s unique needs and preferences.^17,22 Comparable referrals across conditions and additional service receipt may have also contributed to improvements across conditions.

These findings are timely, especially given heightened attention to IPV stemming from the COVID-19 pandemic^65–67 as well as the critical need for effective trauma-informed interventions to complement IPV screening efforts within VHA. The current findings with a demographically diverse sample of VHA patients suggest the promise of implementing RISE in this context. RISE may be particularly useful to IPV Assistance Program Coordinators (typically social workers, psychologists, and psychiatrists) who are located at all VHA hospitals and are responsible for implementing IPV screening but currently lack a standardized intervention protocol.⁴⁵ RISE, with its brevity, modular approach, and focus on self-efficacy and empowerment, provides a foundation for a stepped-care approach to IPV care. Whereas RISE may be sufficient to meet the needs of some women, others may benefit from intensive psychotherapies to address mental health needs prevalent in this sample. Such treatments can further reduce distress and sometimes reduce risk for future IPV.^55,68 RISE complements and extends advocacy interventions through its delivery by clinicians within the health care setting, incorporation of MI, and serving as a bridge to services for other health needs.

Study limitations include the modest sample size, relatively brief interval for follow-up assessments (10 and 14 weeks post-enrollment), and lack of a time- and attention-matched control condition. This study was powered to detect significant between-group effects on empowerment and self-efficacy, as IPV intervention experts and advocates have long viewed these constructs as a critical foundation for recovery from abuse.^30,32,54,69 It is possible this study was underpowered to detect differences in other outcomes. Although no significant effects were observed for patient activation, the effect size for the time-by-condition interaction was medium to large, indicating the analyses may have been underpowered given the small sample size, suggesting the need for research with larger samples. Future research will also benefit from collecting follow-up assessments over longer durations of time to evaluate the maintenance of gains associated with RISE. Some secondary outcomes may continue to improve over time. Additionally, some outcomes (including satisfaction) may be impacted by intervention dose (number of sessions received). Future research should also examine the effectiveness of RISE with broader populations, including patients of various gender identities, as well as evaluate the feasibility and potential benefits of RISE in other contexts beyond VHA. These efforts can advance the goal of ensuring that all individuals who experience IPV have access to effective interventions.

Submitted: April 17, 2021; accepted June 29, 2021.
Published online: November 23, 2021.
Potential conflicts of interest: None.
Funding/support: This work was supported by the Department of Veterans Affairs (VA) Office of Research and Development Health Services Research & Development Services (IIR 16-062; Dr Iverson).
Role of the sponsor: The VA had no role in the design, analysis, interpretation, or publication of this study.
Disclaimer: This work does not represent the official position or views of the VA.
Acknowledgments: The authors are grateful to Stephanie Kaplan, PhD; Laura Brennan, PhD; Jessica Chmielewski, MSW; and Alessandra Grillo, BS, of the  VA Boston Healthcare System, Boston, Massachusetts, for their assistance with treatment delivery and project implementation for this study. None of these individuals reported a conflict of interest regarding their participation in the implementation of this study.

Clinical Points

• Health care systems are increasingly adopting intimate partner violence (IPV) screening practices, but there remains limited guidance on effective health care–based intervention following disclosure; RISE fills this gap.
• RISE is a brief, survivor-centered counseling intervention that is variable-length, modular-based, and rooted in principles of empowerment, trauma-informed care, and motivational interviewing.
• RISE is an effective intervention for enhancing psychosocial well-being, especially empowerment and self-efficacy, among women experiencing IPV.

*The numbers of participants who received community and VHA referrals are provided for descriptive purposes. The information provided includes the referrals placed or services that were contacted together in sessions (and some participants were referred to more than 1 service). These numbers do not necessarily reflect participant follow-up with these services. Participants in both conditions were provided resource and service information. In addition to facilitated referrals, participants could reach out to services on their own terms (as opposed to the interventionist facilitating referrals).

Editor’s Note: We encourage authors to submit papers for con­sideration as a part of our Focus on Women’s Mental Health section. Please contact Marlene P. Freeman, MD, at mfreeman@psychiatrist.com.