Testing a Talkstory Intervention to create Supportive and Safe Violence-Free Communities for Women

Abstract

The purpose of this paper is to report on a community-designed and led talkstory intervention to increase awareness of intimate partner violence (IPV), decrease acceptability of IPV, and increase community leadership to address IPV. In collaboration with women engaged in prior IPV outreach and education in Hawai‘i, a talkstory intervention for IPV was developed, and a single-group, pre-post-test design was used to test it. The intervention included five talkstory sessions over seven months with community groups interested in violence prevention. Pre- and post-testing were conducted to determine changes in group means on three measures. Ninety-two individuals participated in the intervention, 77 (84%) of these completed the 1-month follow-up measure, and 59 (64%) of these also completed the 6-month follow-up measure. The findings included: (1) participants in the talkstory intervention groups decreased their acceptability of violence and increased their awareness, knowledge, and confidence to address IPV; (2) the community leaders in the intervention groups gained skills in facilitation; and (3) intervention groups continued to sponsor other IPV awareness-raising activities in their communities following completion of the study. Working with community leaders to design and facilitate the intervention not only provided IPV education within the context of the community, but also led to sustainable efforts to enhance the safety and wellbeing of women experiencing violence.

Introduction

The purpose of this article is to report the results of a community-designed intervention conducted in Hawai‘i to determine if it would increase awareness of intimate partner violence (IPV), decrease acceptability of IPV, and develop community leadership to prevent and address IPV. IPV is a complex issue affecting individuals around the world.^1,2 Experiencing IPV has been linked to negative acute and long-term physical, psychological, social, and economic outcomes.³ In the United States (US), nearly 1 in 4 adult women (23%) and approximately 1 in 7 men (14%) report having experienced severe physical violence (eg, being kicked, beaten, choked, or burned on purpose, having a weapon used against them, etc) from an intimate partner in their lifetime.⁴ Over 40% of female homicide victims in the US are killed by an intimate partner.⁴

Adverse health effects of IPV can include cardiovascular, gastrointestinal, reproductive, musculoskeletal, and nervous system conditions, as well as depression, posttraumatic stress disorder (PTSD), and engagement in risky health behaviors, such as substance abuse.⁴ In 2000, costs associated with nonfatal injuries and deaths due to IPV exceeded $70 billion.^5,6 Traditionally, interventions have been sponsored by the justice system and by psychological counseling providers; however, many women experiencing IPV are reluctant or unable to access these services.⁷

In Hawai‘i, at least 20% of women aged 19–64 years have been victims of IPV in their lifetime.⁸ The Hawai‘i State Department of Human Services contracts with seven non-profit entities to provide emergency support to IPV victims. In 2015, there were 9,081 IPV victims served, 16,900 hotline calls, and 3,473 victims and survivors provided IPV advocacy.⁹ Although this information does not describe specific cultural groups, previous studies have investigated ethnic variation in reports of IPV. In an earlier retrospective review of medical records at four community health centers on O‘ahu, 337 medical records from over a 5-year period were examined to identify documented cases of IPV. Native Hawaiians represented 32% of 31 documented IPV cases, despite the fact that Native Hawaiians comprised only 19% of the study sample.¹⁰ In qualitative research with Pilipino^* women, 16% of participants reported IPV and noted that this was an important concern for their community.¹¹ Between 2000 and 2012, 67 women in Hawai‘i were murdered as a result of IPV, and more than 70% of those murdered were Pilipino or Native Hawaiian women.¹²

Findings from research in Hawai‘i suggest that many individuals do not use conventional IPV resources due to language or cultural barriers and fear of discrimination from the legal, child-protection, and immigration systems.¹² Advocacy interventions, defined as interventions in which IPV victims receive help with safety planning and social services beyond the clinic, are gaining popularity and appear to yield more benefits than standard care.¹³ Interventions focused on developing individual capacity to access information on IPV have primarily focused on leaving the relationship, but significant service gaps exist for women who require support after leaving an abusive relationship.¹⁴ A review by Ellsberg et al. concluded that the most effective IPV interventions are participatory, engage multiple stakeholders, support critical discussion about gender relationships and the acceptability of violence, and support greater communication and shared decision making among family members, as well as non-violent behavior.¹⁵

Consideration of culture is important in intervention development in Hawai‘i because of the ethnic diversity of the population, estimated at 21.3% Native Hawaiian, 22.7% Caucasian, 16.3% Japanese, 17.2% Pilipino, 6.8% Chinese, and 15.7% others.¹⁶ Traditional cultural beliefs, practices, and norms regarding gender roles and decision-making patterns can function as protective or contributing factors to IPV and may be influenced or disrupted over time.¹⁷

Methods

Developing the Intervention

The intervention was developed by three authors, (LM, JS, CS) and other members of the Consortium for Health Safety and Support (CHSS), a group of community members, professionals from the Hawai‘i Domestic Violence Action Center (DVAC) and other legal, health and social services providers, and faculty from the University of Hawai‘i at Manoa (UHM). The Office of Womens' Health (OWH) initially funded 16 sites nationwide to assess community needs and develop research proposals that addressed the specific needs of women. During the first year of funding, the CHSS conducted a community needs assessment of the Leeward communities on O‘ahu (Ewa Beach, Kapolei, Makaha, Nanakuli, Waianae, and Waipahu) with high percentages of Native Hawaiian and Pilipino residents. “Talkstory” was identified by community members as an important strategy to address IPV. The members of the community believed that talkstory would increase community engagement and ownership and create safe spaces for discussions about IPV. Continued funding from OWH was received to implement and test the intervention in these communities.

Talkstory refers to an informal, laid-back conversation involving a “reciprocal exchange of thoughts, ideas, feelings about self, and other issues,”¹⁸ and is important to Hawai‘i residents across multiple ethnic groups. The use of talkstory demonstrates respect of the local culture and customs by generating dialogue that is not targeted to a single perspective or endpoint. Talkstory discussion groups have been used in previous studies to address primary care and prevention issues on the Leeward Coast (southwest side of the Island of O‘ahu.)¹⁹ It has been used as an approach to address disharmony between family and community members and to reach accepted solutions to remedy the discord and restore harmony to relationships.^18,19

Theoretical Orientation

Although IPV is a multi-faceted phenomenon and no single theoretical approach provides a complete explanation,²⁰ critical social theory (CST) was chosen by the CHSS to guide the intervention. The intent of CST is to “challenge conventional assumptions and social arrangements and to move beyond the “what is” to the “what could be.”²¹ The context for IPV is a belief system about the relationships between power, societal structures (primarily related to race, gender, and class), and resulting conditions of society. Dominant voices hold power over marginalized voices, creating both privilege and marginalization. Gender roles within culture can increase the tolerance of abuse and decrease the reporting of abuse.²⁰

To complement CST, a community-based participatory research (CBPR) approach was adopted. The purpose of CBPR is to increase shared leadership, community capacity, and intervention relevancy for community members.²² These strategies, while recognizing the power of community participants, also place an obligation on the community partners to take action that is consistent with the participants' voices. Thus, the intervention was designed to use a talkstory approach to help participants reflect critically on the traditional rules, practices, structures, and assumptions that have guided the perceptions of IPV and the resulting programs in communities.

The CHSS developed and pilot tested an initial curriculum with Native Hawaiian and Pilipino community groups, and refined it based on the pilot groups' suggestions to become the talkstory intervention.²² The intervention consisted of five sessions, each lasting approximately two hours. Three sessions were facilitated during the first month, followed by one session three months later and a final session six months later. Conceptually, the sessions were designed to encourage community members to discuss five topics: 1) their perceptions of IPV; (2) actions they took individually to prevent, interrupt, or stop IPV; (3) suggested actions that community groups (eg, churches, schools, canoe clubs) could take together; (4) resources to prevent, intervene, or interrupt IPV; and (5) resources still needed. Woven through the talkstory sessions was information on (1) understanding IPV; (2) gender role expectations/healthy relationships; (3) effects of IPV on the family and community; (4) support and safety within the community; (5) strategies and skills to create safe environments to address IPV; and (6) creating a community-owned network of safe support groups. This information was compiled into a document entitled, “Talkstory Toolkit.” In preparation for the implementation of the study, 20 facilitators were trained to become group leaders. They attended at least two, 2-hour sessions that included: (1) instruction in safety measures and methods of conducting groups; and (2) human subjects' protection. Trained community leaders usually co-facilitated with someone from DVAC and/or UHM. Table 1 describes the topics, research instruments, and schedule.

Table 1.

Talkstory Topics and Data Collection Schedule

	Week 1	Week 2	1Month	3 Months	6 Months
Talkstory Topics	Understanding Intimate Partner Violence (IPV)	Gender role expectations and healthy relationships	Effects of IPV on the family and community	Support and safety within the community	Strategies and skills to create safe environments to address IPV in communities
Demographic Questionnaire Resource Utilization Survey	xa x				x
Acceptability of Violence	x		x		x
Awareness, Knowledge and Confidence	x		x		x
Community Capacity	x		x		x

^ax= Data collected at these points

Ethical Considerations

The Institutional Review Board (IRB) from UHM granted approval (CHS #20030) for this study. All participants were informed about the implications of participating in the study and were required to provide their informed consent prior to participating. In addition, a Certificate of Confidentiality was requested from the National Institutes of Health and all participants were required to sign a Confidentiality Agreement to maintain ethical integrity. Safety procedures to be taken in the event of unforeseen interruptions of the session were also developed. With these procedures in place, no breaches in safety were experienced.

Study Participants and Setting

A single-group, pre-post-test design was used to test the talkstory intervention. To recruit women into interventions groups, community leaders who represented the partnering agencies on the Leeward Coast purposively identified and invited community residents 18 years of age or older willing to discuss IPV from among friends, neighbors, and respected informal leaders. Participants' IPV status was disclosed on the demographic form. One of the purposes of the study was to change attitudes and perceptions regarding IPV, and both those who had experienced violence and those who had not were included in the study. No participants were known to community leaders as perpetrators.

Although the selection of talkstory emerged from the community, retention of intervention participants throughout the seven-month intervention initially proved challenging. After discussion with group leaders and OWH, and with the approval of the IRB, a progressive increase in the incentives was offered to participants. The original incentive schedule of $20 per session was increased to $25 for the first session, $30 for the second, $35 for the third, $40 for the fourth, and $50 for the fifth session attended. An assistant was also hired to maintain contact with the participants between sessions.

Data Collection

All participants were asked to complete a demographic questionnaire and three paper assessment instruments: (1) Perceptions of the Acceptability of Violence; (2) Awareness, Knowledge, and Confidence regarding IPV; and (3) Perception of Community Capacity to Address IPV. The completion of the assessment instruments occurred at three points: baseline, 1 month, and 6 months.

In the “Perceptions of the Acceptability of Violence Tool” (three items), respondents marked their level of perceived acceptability of IPV on a 3-point scale (1=never, 2=sometimes, 3=always). This tool was developed by Torres, et al.²³ In her previous research, women marked their level of perceived acceptability of IPV in their communities, their families, and themselves on a 10-point scale, and item scores were categorized into tertiles—low, mid, and high. Following pretesting in our communities, we used a 3-point scale instead of a 10-point scale.

The “Awareness, Knowledge, and Confidence Tool” (nine items) solicited self-assessment of individual capacity to address IPV (1 for beginning capacity; 2 for developing; 3 for accomplished). Examples of items included “I'm aware of community needs for prevention of IPV” and “I'm confident in my ability to work with community agencies for the prevention of IPV”. The authors have used this tool with students and found that their self-rated competence increased along the 3-point scale after engaging in interdisciplinary team work on IPV.

The “Perception of the Capacity of the Community Tool” (six items) asked participants to assess the competence of the community, eg, percentage of the community aware of IPV (0=<25%, 1=25–49%, 2=50–74%, 3=≥75%). This tool has not been used in research and was tested for language clarity and understanding during the pilot study and refined accordingly. We expected this to allow us to see a difference from baseline to post-intervention measures.

Data Analysis

Demographic variables were analyzed using frequencies. Because of the large number of items in the outcome measures, we created “total scores” for the three outcome measures. The “Violence Acceptability” score was created by summing the three items from that scale (range 3=violence is not acceptable to 9=violence is acceptable). The “Awareness, Knowledge, Confidence” score was created by summing the nine items in this scale (range 9=very low awareness-knowledge-confidence to 27=high awareness-knowledge-confidence as perceived by the respondent). The “Community Capacity” score was created by summing the six items in this scale (range 6=perceived low community awareness about and resources to reduce IPV to 18=perceived high community awareness about and resources to reduce IPV). For each scale, a Cronbach's alpha was estimated. Values were .76, .85, and .86, respectively, suggesting that these measures had good internal reliability. Because responses were not normally distributed, we used the Wilcoxon signed-rank test, a non-parametric test, to compare measures at baseline with those at 1 and 6 months. SPSS v21 (IBM Corp, Armonk, NY) was used for data management and analysis.

At the final evaluation conference with participating group leaders, each group was asked to identify the number of individuals who expressed or documented that they had been abused and were given assistance because of the project.

Results

Participants included 92 Leeward Coast residents in ten intervention groups. As hoped, about 36% of the sample was Native Hawaiian, 33% Pilipino, and 2% Other Pacific Islander. Other participants were predominantly Asian (9%) or Caucasian (13%). The participants included 28 men and 64 women, and the mean age of participants was 39 years. About 18.5% had a high school degree only, 4% attended college but did not earn a degree, and 39% reported having a bachelors, masters, or higher degree. About 75% of participants reported being employed, and 86% of participants reported having health insurance. Almost all participants reported income (n=89); about 41% of the sample reported annual income of <$39,999, and 55% reported annual income of $40,000 or more (Table 2).

Table 2.

Characteristics of Intervention Participants Over Time^a

	Baseline n=92 n (%)	1-month follow-up n=77 n (%)	6-month follow-up n=59 n (%)
Ethnicity
Native Hawaiian Pilipino Other Pacific Islanderb Other Asianc Caucasian Otherd	33 (36) 30 (33) 2 (2) 8 (9) 12 (13) 7 (8)	27 (35) 24 (32) 2 (3) 8 (10) 10 (12) 5 (7)	15 (26) 23 (39) 0 6 (10) 10 (17) 4 (7)
Mean age (years)	38.3	39.7	37.6
Gender
Female Male	64 (67) 28 (31)	54 (70) 23 (30)	46 (78) 13 (22)
Education
High school or GEDe Some college Bachelor's degree Masters or higher	17 (19) 39 (42) 20 (22) 16 (17)	13 (17) 32 (42) 18 (23) 14 (18)	7 (12) 28 (48) 12 (20) 12 (20)
Employed (yes)	69 (75)	58 (75)	46 (78)
Insurance (yes)	79 (86)	67 (87)	50 (85)
Income
< $20,000 $20,000–$39,999 $40,000–$59,999 > $60,000 Missing	17 (19) 21 (23) 28 (30) 23 (25) 3 (3)	11 (14) 18 (23) 24 (31) 21 (27) 3 (4)	10 (17) 15 (26) 19 (32) 14 (24) 1 (2)

^aPercentages may not total 100% due to rounding.

^bIncluding Micronesian, Marshallese, Guamanian, Samoan.

^cIncluding Cambodian, Chinese, Japanese, Korean, Thai, Vietnamese.

^dIncluding African American, Hispanic, Native American.

^eGeneral Educational Development Test

Although 92 participants provided baseline data on the outcome measures, only 77 (84%) completed the 1-month post-test, and only 59 (64%) also completed the 6-month post-test. As shown in Table 2, larger proportions of women participated in the 1- and 6-month assessments compared to baseline; however, this difference was not statistically significant. Nor did the baseline and follow-up samples differ significantly on any other demographic measure, suggesting no systematic bias in attrition.

Over time, there was significant improvement on each of the three scales (Table 3). Specifically, post-test scores were lower for Violence Acceptability (from 4.03 to 3.74; P=.048 at 1 month, to 3.50; P<.001 at 6 months), higher for Awareness, Knowledge, Confidence (from 18.11 to 22.19; P<.001 at 1 month, to 25.38 P<.001 at 6 months), and higher for Community Capacity (from 10.66 to 11.68; P =.024 at 1 month, to 13.10; P<.001 at 6 months).

Table 3.

Comparison of Scores at Baseline and After 1 and 6 Months

	Baseline (n=92)	1-month follow-up (n=77)	Significance from baselinea	6-month follow-up (n=59)	Significance from baselinea
Total “Violence Acceptability” Scoreb
Mean (Standard Deviation) Range Skewness	4.03 (1.47) 3–9 1.22	3.74 (1.03) 3–7 1.36	0.048	3.50 (0.88) 3–6 1.48	<.001
Total “Awareness/Knowledge/ Confidence” Score
Mean (Standard Deviation) Range Skewness	18.12 (4.29) 9–27 0.23	22.19 (4.09) 14–38 0.50	<.001	25.38 (2.32) 18–27 1.40	<.001
Total “Community Capacity” Score
Mean (Standard Deviation) Range Skewness	10.66 (3.12) 6–21 1.12	11.68 (2.91) 6–18 0.66	0.024	13.10 (3.15) 8–21 0.51	<.001

^aSignificance by Wilcoxon Signed Rank Test.

^bA lower score is better.

Community leaders reported that many of the intervention group members did not realize that what women experienced in their homes was actually abuse from their partners. In previous research conducted by this group, community participants indicated they felt it was normal for women to be emotionally and physically beaten up.²⁴ Leaders reported that participants gained a better understanding of the dynamics of abuse. Based on intervention group discussion and self-report either on the demographic form or verbally in the talkstory group, community leaders estimated that 25–30 members of the 92 intervention participants, about one-third of total participants, were experiencing or at risk of IPV during the talkstory intervention. At least five women sought help during the period of the intervention, most often leaving the unsafe situation.

Other leaders described that their groups' concerns about IPV led them to make signs and engage in sign-waving events along the only major highway leading to the community, particularly during Mother's Day and Father's Day. One leader reported that 135 residents participated in sign-waving, with group leaders counting more than 2,000 “honks” of approval accompanied by the “shaka sign” from people driving past the participants. Community members also designed a T-shirt with the logo, “kNOw MORE,” to emphasize their hope for changing the social acceptance of IPV to a culture of no tolerance for IPV. A “kNOw MORE” march was held by community residents and involved more than 200 people. Following completion of the research study, four of the original community groups continued to include prevention and activities to address IPV in their regular programs to create different social norms regarding IPV. One group, for example, continues to provide “Healthy and Hapai” baby showers for new mothers and fathers, providing baby supplies as well as support and an opportunity for pregnant young families to reach out for help if needed during this vulnerable period.

Discussion

Although legal and social policies are enacted to address IPV and provide remedies for those experiencing violence, they are not enough to stop the problem. Effective prevention and response require a coordinated effort across many sectors, and community activism plays an important role. Michau, et al, urged that prevention programs be designed to change “attitudes, norms, and behaviors.”²⁵ They also maintain that community-based, rather than individually focused, programs are more likely to foster social change.

This talkstory intervention appears to have positively influenced attitudes in the community and begun to change behaviors regarding IPV. A particularly valuable outcome of the process was that it developed local leadership for IPV prevention. Natural leaders built their capacity regarding IPV through the training activities provided to them and through the process of facilitating their groups. At the end of the project, they described their plans to continue work in IPV-awareness raising and to engage the broader community in periodic sign waving and ongoing discussions.

The talkstory groups were conducted with mixed genders. This format was designed by members of the community because they believed that it would more fully engage the community. Michau, et al,²⁵ state that “evidence now shows that work with both women and men (in gender-specific and mixed groups, depending on the topic and the situation) is more likely to promote non-violent norms around masculinity and less passive norms around femininity than work that only engages men or women separately.”²⁵

One of the goals of Healthy People (HP) 2020—a set of nationwide health promotion and disease prevention goals developed by the US Department of Health and Human Services—under educational and community-based programs (ECBP) is to increase the number of community-based organizations providing population-based primary prevention services against violence (ECBP-10.2).²⁶ Throughout the study, a myriad of activities developed by community partners has helped address this goal in Hawai‘i. At the end of the intervention, participating community leaders were asked to report ways the intervention impacted perceptions of IPV on the Leeward Coast. Leaders estimated the number of participants in their groups experiencing and/or at risk of IPV, how the intervention helped these individuals, and activities the leader and/or group conducted to further raise awareness of IPV in their community. These activities help overcome silence around IPV on the Leeward Coast, where large percentages of Native Hawaiians and Pilipinos reside.

A second goal of HP 2020 under injury and violence prevention (IVP) is to reduce violence by current and former intimate partners (IVP-39).²⁷ This goal is highly desirable, and working towards a reduction in violence is undoubtedly important. Yet, success should not be measured in the short term. Lacayo²⁸ used Complexity Theory as a framework to analyze change fostered by a social program in Nicaragua. She describes societies as being composed of people who are “unpredictable and uncontrollable,” and the process of social change as being “a nonlinear, contradictory, messy, emergent, self-organizing, and long-term process.”²⁸ A study such as the talkstory intervention conducted at the community rather than the individual level promotes social awareness, which can lead to social change of attitudes and beliefs over time.

Conclusions

From this test of the talkstory intervention, it appears there was a positive impact. Working with community leaders to design and facilitate the intervention provided IPV education within the context of the culture of the Leeward Coast. It also led to sustainable efforts to enhance the safety and wellbeing of women of the community, and in ways that were safe and did not judge or isolate women experiencing violence. Participants in the talkstory interventions decreased their acceptability of violence and increased their awareness, knowledge, and confidence to address IPV, the community leaders gained skills in facilitation, and groups went on to sponsor other IPV-awareness-raising activities.

IPV is a complex issue involving families and communities and is best addressed with their full participation. Social change is dynamic and complex and is gradual rather than dramatic in most instances. The community members involved in this project have provided testimony regarding the changes in their neighborhoods. Since the project, they have gained a better understanding of the dynamics of abuse, are aware of ways to decrease it, and have greater capacity to address IPV in their communities.

Acronyms

CBPR

Community-Based Participatory Research

CHSS

Consortium for Health Safety and Support

CST

Critical Social Theory

DVAC

Domestic Violence Action Center

HP 2020

Healthy People 2020

IRB

Institutional Review Board

IPV

Intimate Partner Violence

OWH

Office of Womens' Health

Conflict of Interest

None of the authors identify any conflict of interest.