Abstract
Epidemiological data suggest that Native Hawaiian, Pacific Islander, and Asian American (NHPIA) adolescent females in Hawai‘i are at elevated risk for sexually transmitted infections (STIs). These data also indicate that teen girls in Hawai‘i are more likely to engage in high-risk sexual behaviors than boys. Despite this compelling evidence, there is a paucity of published research on effective STI prevention protocols that target NHPIA female adolescents. In light of this need, the purpose of this study was to adapt an evidence-based, group-level STI behavioral prevention intervention targeting local NHPIA adolescent girls in Hawai‘i for implementation by community-based organizations (CBOs). This article describes the adaptation of an evidence-based STI prevention intervention in partnership with a CBO, using a locally modified ADAPT-ITT model as a basis for the adaptation process. This research consisted of two phases: (1) development of an initial test version of an adapted intervention and (2) the development of the final version of the adapted intervention, Girl Power Hawai‘i. The results provide the empirical and foundational research necessary for a tailored intervention that can be readily implemented by CBOs for local teen girls in Hawai‘i.
Keywords
Introduction
Epidemiological data suggest that Native Hawaiian, Pacific Islander, and Asian American (NHPIA) adolescent females in Hawai‘i are at elevated risk for STIs. A high percentage of adolescent STI cases in Hawai‘i are among NHPIA females (Hawai‘i Department of Health, 2008). An epidemiological study by Sasaki and Kameoka (2009) found that NHPIA adolescent girls engage in higher rates of risky sexual behaviors than girls of other ethnicities in Hawai‘i, and this may contribute to increasing rates of morbidity and premature mortality resulting from sexually transmitted infections (STIs).
Despite this compelling evidence, there is a paucity of published research on effective STI prevention protocols that target NHPIA female adolescents. As a result, the prevention community is underprepared to respond to this growing need for empirically supported STI risk–reduction interventions addressing the unique behavioral, cultural, and developmental issues confronting this at-risk population.
In light of this need, the purpose of this study was to adapt an evidence-based, group-level STI behavioral prevention intervention targeting local NHPIA adolescent girls in Hawai‘i for implementation by community-based organizations (CBOs). An important focus of this adaptation was gender- and age-appropriateness, as these dimensions are strong influences on the effectiveness and acceptability of evidence-based STI prevention interventions (e.g., DiClemente et al., 2013).
Another important aspect of this study was the priority multiethnic population of at-risk girls in Hawai‘i. Hawai‘i is home to a uniquely diverse resident population that is often referred to as local. According to the 2016 census data, NHPIA individuals comprise the largest percentage of the population (47.9%; Native Hawaiian and Other Pacific Islanders alone, 10.2%; Asians alone, 37.7%). Furthermore, far more residents claim multiethnic background (two or more races, 23.1%) than in any other state (U.S. Census Bureau, n.d.), and most Native Hawaiians are of mixed ethnicity (Ishibashi, 2004). Given this diversity, this adaptation was developed to address the needs of the majority, local population of NHPIA teen girls in Hawai‘i.
Background
STIs and Related Risk Behaviors Among Teen Girls in Hawai‘i
Epidemiological data over the last decade demonstrate that female girls nationwide are engaging in higher rates of high-risk sexual behavior and are at elevated risk for STIs in comparison with their male counterparts (Centers for Disease Control and Prevention [CDC], 2014). These same trends are observed among Hawai‘i youth. Since 2002, Hawai‘i has ranked consistently in the top 10 states in chlamydia rates (McGrath, Katz, Lee, & Rochat, 2011), particularly among those aged 15 to 24 years. In 2004, 63% of reported cases of chlamydia and 50% of gonorrhea cases in Hawai‘i were among people in this age group (Advocates for Youth, 2008). As in the case nationwide, young females have been at higher risk for STIs compared with males in Hawai‘i, with 78% of chlamydia and gonorrhea cases combined occurring among 15- to 24-year-old females in 2004 (Advocates for Youth, 2008).
High rates of STIs among adolescents in Hawai‘i may be attributed to the prevalence of high-risk sexual behaviors, as this group is more likely to engage in risky sexual behavior than teens nationwide. In 2013, condom use among adolescents in Hawai‘i was among the lowest nationwide with 54.1% denying use during last sexual intercourse compared with the national average of 40.9% (CDC, 2014). In Hawai‘i, teen girls are more likely to engage in high-risk sexual behaviors than boys, just as they are nationwide. According to Youth Risk Behavior Survey data, 58.5% of 9th through 12th grade girls in Hawai‘i reported not using a condom during their last sexual intercourse, compared with 46.5% of boys (CDC, 2014). These data suggest that the sexual health of teen girls in Hawai‘i is a growing problem, and there is a clear need for a locally adapted STI risk-reduction intervention for this vulnerable population.
The ADAPT-ITT Model
A variety of evidence-based interventions (EBIs) have been developed for preventing STIs among teen girls. Adapting a STI prevention EBI specifically for use in Hawai‘i would begin addressing a significant gap in service. The ADAPT-ITT model developed by Wingood and DiClemente (2008) has been used successfully to adapt evidenced-based STI prevention interventions for adolescent girls. ADAPT-ITT is an acronym for eight sequential phases of an adaptation methodology: assessment, decision, adaptation, production, topical experts, integration, training, and testing. ADAPT-ITT underscores the scientific importance of maintaining fidelity to core elements responsible for successful behavior change in the original intervention and incorporates community-based participatory strategies.
This article describes the adaptation of an evidence-based STI prevention intervention in partnership with a CBO, using a locally modified ADAPT-ITT model as a basis for the adaptation process. This research consisted of two phases: (1) development of an initial test version of an adapted intervention and (2) the development of the final version of the adapted intervention. The results provide the empirical and foundational research necessary for a tailored intervention that can be readily implemented by CBOs for local teen girls in Hawai‘i.
Method
The following phases describe the ADAPT-ITT methodology that was revised and used for this study (Kameoka & Takishima-Lacasa, 2013). The methodology outlined in Table 1 accommodate the unique attributes of the partnering CBO, the targeted local Hawai‘i teen girl population, and the logistics of conducting community-based research in Hawai‘i.
Adaptation Procedures
NOTE: The development of these procedures was guided by ADAPT-ITT model (Wingood & DiClemente, 2008). Study materials are italicized. CBO = community-based organization; CAB = Community Advisory Board; IRB = institutional review board. Grayed cells indicate phases that were not included in this preliminary investigation.
Priority population are directly involved in these phases of adaptation.
Phase 1: Selection of the Intervention
In Phase 1, a behavioral intervention that would be most appropriate for adaptation and implementation among local adolescent girls in Hawai‘i was identified. A comprehensive review of the adolescent STI risk reduction literature published in the past decade was conducted to identify evidence-based group behavioral interventions targeting teen girls. This review identified two interventions based on the AIDS risk reduction model (ARRM; Catania, Kegeles, & Coates, 1990), one intervention based on the information–motivation–behavioral skills model (IMB; Fisher & Fisher, 2002), and five interventions based on Bandura’s (1986) social cognitive theory (SCT). All eight interventions were grounded in a strong theoretical foundation and were uniquely customized to address the sociocultural, gender, and developmental characteristics of a target population.
The AARM-based interventions, Project IMAGE (Champion & Collins, 2012) and SHERO (Harper, Bangi, Sanchez, Doll, & Pedraza, 2009), were systematically developed, empirically tested, and shown to be efficacious sexual risk–reduction interventions for use among ethnic minority teen girls. Both interventions, however, involved highly restrictive limitations (e.g., intervention and content targeted exclusively for a particular group, staffing requirements, and length of the intervention) that made them relatively poor candidates for local adaptation. Also, despite extensive testing and some promising findings, the IMB-based intervention, HIP Teens (Morrison-Beedy et al., 2013) did not demonstrate strong evidence of efficacy and was not adaptation-ready.
The SCT-based interventions included Be Proud! Be Responsible! Be Protective! (BBB; Koniak-Griffin et al., 2003); Prime Time (Sieving et al., 2014); Sisters Saving Sisters (SSS; Jemmott, Jemmott, Braverman, & Fong, 2005); Sisters Informing, Healing, Living, & Empowering (SiHLE; DiClemente & Wingood, 2004); and HORIZONS (DiClemente et al., 2009). While outcome studies of BBB, Prime Time, and SSS provided promising evidence of the effectiveness of those interventions among the targeted populations, SiHLE and HORIZONS demonstrated the strongest efficacy and theoretical grounding in both SCT and the Theory of Gender and Power (Connell, 1987). Based on these assessments, the ethnically neutral version of HORIZONS, Girl Power, provided the most viable starting point for local adaptation in Hawai‘i.
Phase 2: Develop Adaptation Guide
In Phase 2, an Adaptation Guide that documents modifications made to the original Girl Power content was created. Girl Power consisted of two 3-hour group workshop sessions covering 22 activity modules. Modifications were based on a systematic line-by-line review of the intervention’s 22 activity modules and an assessment of cultural relevance of the intervention materials. The Adaptation Guide documented decision-making rationale for each modification and ensured fidelity to the original intervention’s core elements. Project meetings with the CBO’s Community Advisory Board (CAB) members were conducted to discuss the intervention’s content, adaptation process, and project procedures. These discussions informed the development of the Adaptation Guide.
Phase 3: Focus Groups
Onsite staff from the partnering CBO recruited girls between 13 and 18 years old in their after-school programs in two communities on the island of Oahu. Thirteen local teen girls participated in one of the four focus groups. A majority of the participants (n = 10, 76.9%) identified as mixed NHPIA ethnicity, two girls self-identified as Chinese, and one as Filipino. Participants ranged from 13 to 18 years in age (M = 15.69 years; SD = 1.5) and were enrolled in 9th to 12th grades (mode = 11th grade).
Focus groups questions were developed in consultation with the CAB. These questions were developed to obtain locally relevant information that informed the initial adaptation of Girl Power intervention content and were designed to generate discussions specifically about (1) local youth culture and (2) norms for sexual behavior among Hawai‘i teens. Standard focus group methodology was used to conduct all sessions (e.g., Côté-Arsenault & Morrison-Beedy, 2005).
In each of the focus group session, two methods of data collection were used. First, detailed field notes were recorded using note-taking protocol and template that documented participant responses, nonverbal behaviors, important themes, and discussion highlights. Second, an audio-recorded debriefing session involving the study investigator, research assistants, and CAB members was conducted immediately after each focus group session to identify discussion themes. The field notes and debriefing session transcripts for each focus group session were combined, resulting in four final focus group transcripts.
Phase 4: Develop Test Version of the Intervention
A thematic and comparative analysis methodology (Braun & Clarke, 2006) was used to analyze the transcripts and to identify relevant themes that, in turn, guided modifications of Girl Power. Standard qualitative data analytic methods (e.g., Strauss & Corbin, 1994) including unitization and two rounds of independent coding were used to identify common themes. Intercoder reliability was calculated for each theme identified and percent agreement (Popping, 2010) ranged from 67% to 100%, with an average of 95.8%.
Themes derived from the focus groups were incorporated into the Adaptation Guide and were used to guide revisions of Girl Power (e.g., words, phrases, graphics, music, activities), delivery style, and intervention materials. These revisions yielded the Test Version of the locally adapted intervention, Girl Power Hawai‘i, as named by focus group participants.
Phase 5: Theater Tests
Theater testing is a pretesting technique used in marketing to obtain electronically attitudes about audio-visual materials from an audience. The Test Version of Girl Power Hawai‘i was theater tested with a group of local teen girls and key community stakeholders. Participant recruitment procedures were similar to those used for the focus groups. Nine local teen girls from various high schools on Oahu and six community stakeholders participated in two theater tests. Seven of the participants identified as mixed NHPIA ethnicity, and two girls self-identified as Filipino. Participants ranged from 14 to 16 years in age (M = 15.33 years; SD = 0.87) and were enrolled in 9th to 12th grades (mode = 11th grade).
The research team demonstrated Girl Power Hawai‘i activities to participating audience by role-playing each revised intervention activity. Theater test questions were designed to obtain participants’ reactions to demonstrated intervention activities. A single question about factors that may influence condom use among local teen girls was also asked. These questions were administered using presentation software and an iClicker electronic response system that allowed participants to answer questions immediately and anonymously with a remote clicker. Participants’ responses to each question were instantly viewable on the presentation screen and were used to facilitate discussion of participants’ reactions to the demonstrated intervention activities. Data collection procedures were identical to those used for the focus groups and resulted in two transcriptions of theater test data.
Phase 6: Final Version of the Intervention
The thematic and comparative analytic methodology used to analyze focus group data was also used to analyze the theater test transcripts. Percent agreement for the second round coding ranged from 50% to 100%, with an average of 95.9%. The resulting theater test themes guided further revisions of the Adaptation Guide, which informed the production of the locally adapted intervention. Readability testing was conducted to ensure a Flesch–Kincaid third grade reading level for all intervention content (Wingood & DiClemente, 2008).
Results
Focus Group Themes
Themes Relating to Specific Intervention Content
Participant responses to questions pertaining to relevance of the original intervention content focused on (1) role models, slang terms, popular music contained in the original intervention and (2) STI knowledge. Responses were generally congruent with what would be expected among teen girls in the United States. For example, while the original intervention material on positive female role models included Oprah, participants referred to individuals such as Beyonce and family members as significant role models for girls in Hawai‘i. Therefore, references to only significant individuals identified were incorporated in the role-model activity.
In response to questions related to the intervention’s use of slang and music, participants emphasized the use of pidgin slang (e.g., “irraz”) and local Hawaiian musicians (e.g., Kolohe Kai) unique to Hawai‘i. As a result, these references unique to Hawai‘i’s youth were included in the adapted intervention along with other slang and musicians common in Hawai‘i as discussed in the focus groups.
Contrary to expectations, focus group discussions commonly suggested that STI knowledge was very poor among all participants. Although most participants have received some form of sex education in school, most appeared confused about STIs and were unable to provide accurate information about STI prevention and transmission.
Themes Relating to the Culture of Local Teen Girls in Hawai‘i
Participant responses to questions pertaining to the shared experiences of local youth suggest that social norms, and not ethnically based cultural factors, appear to influence their values, goals, and identity as “local.” For example, in response to the question “What does being ‘local’ mean to girls in Hawai‘i?” most of the girls described a broader local identity that involves knowledge attained only by being born or raised in Hawai‘i, regardless of ethnicity. Participants noted factors such as local foods (e.g., “We eat rice in everything”), local language (e.g., “The moment a person speaks, you instantly know if they’re local or not”), local traditions (e.g., “Hug and kiss, nonlocals wouldn’t understand”), and local attitudes (“[Local] culture is very chill”). Participants commonly emphasized the importance of community and connection, best summarized in the response: “You know the people in the community, the people around you.”
The shared experiences described by focus group participants were similar to what most American teen girls would endorse. For example, they reported enjoying dressing up, wearing makeup, shopping, dancing, and spending time with friends. Most participants emphasized the importance of family and doing well in school to be happy and successful in life, and that getting attention from others makes them feel good about themselves. A majority of girls reported they value trust, loyalty, and caring in their friendships. When asked why girls their age have romantic relationships and sex, participants discussed “support,” “security,” “feeling loved,” and “needing attention.” Most participants suggested the importance of balancing relationships with other priorities (e.g., school). Participants also noted “feeling loved” and “needing attention” as factors in sexual decision making, and most girls cited peer pressure as a major influence. When prompted to consider religious or family influences, most girls agreed that those factors do not have a significant impact on whether or not girls have sex.
Theater Test Themes
Theater test questions pertained to perceived relevance of the intervention activities demonstrated during theater testing of Girl Power Hawai‘i. Participant responses were used to guide further modifications of program materials in the adapted intervention. For example, participants suggested that various images portrayed in the activities were outdated or lack relevance and suggested more appropriate images that were incorporated in the adaptation.
During the theater test, participants were asked to discuss factors that may influence condom use among their peers. Four themes emerged from the data, including peer, family, religious influences, and access. Participants discussed peer pressure and the desire to “fit in” as significant influences. Most participants agreed that friends influence condom use, for example, by telling each other that sex feels better without condoms. Participants agreed that family-related factors, such as communication with parents regarding sex, affect some but not all girls. While several participants discussed family influences, they disagreed on how girls are influenced. For example, while a participant indicated that families encourage girls to use protection, others expressed that girls will get pregnant to be “rebellious” or “to get back at parents.” There was consensus among participants that sexual decision making of girls from specific ethnocultural groups, such as Filipino Catholics, is most affected by religion. While a participant stated, “Some [religions] want you to have babies,” many participants noted that if you are Christian, you are taught “no sex before marriage.” Finally, other participants noted that the cost of condoms and difficulty accessing free condoms significantly influenced condom use.
Discussion
This project meets an unmet need to develop an evidence-based STI prevention intervention by systematically adapting the Girl Power intervention by optimizing its relevance and acceptability among local teen girls in Hawai‘i. STI prevention researchers agree that the relevance and acceptability of an EBI significantly influence effective dissemination and implementation of the intervention (e.g., Wingood & DiClemente, 2008). Clearly, tailoring the relevance of intervention content and delivery methods increases the effectiveness and utilization of the intervention in a new priority population. Close consultation with community stakeholders in the adaptation process increases the likelihood that it will be adopted by disseminating organizations (Damschroder et al., 2009). Indeed, successful adaptation leads to enhanced community support, increased client participation, and improved program satisfaction and outcomes (Solomon, Card, & Malow, 2006).
The adaptation of Girl Power Hawai‘i was based directly on feedback from local teen girls in Hawai‘i who participated in focus group and theater test discussions about popular culture, social norms, and shared experiences among their peers. During the adaptation process, careful attention was paid to incorporating into the intervention elements identified by participants and stakeholders as uniquely relevant to the local culture.
Since CBOs typically deliver the majority of STI/HIV prevention services to ethnic minority and high-risk communities (Harper et al., 2009), community collaboration with the partnering CBO was a critical aspect of the adaptation methodology. Members of the CBO’s CAB were integrally involved throughout the adaptation process, from the planning of study procedures to coordinating data collection, and to participating in theater testing. This partnership with the CBO enhances the likelihood that the adapted intervention will be implemented and sustained by community organizations serving youth in Hawai‘i.
While program content was modified to enhance relevance among local teen girls and appropriateness for implementation in Hawai‘i, a fundamental guiding principle of this adaptation was that the core elements underlying the original behavioral intervention were not compromised. The development and use of the Adaptation Guide ensured that core elements of the original intervention (DiClemente et al., 2009) were preserved in the local adaptation, especially behavioral change strategies targeting high-risk behaviors and attitudes regarding safe sex self-efficacy (e.g., assertive communication and empowerment exercises). This adaptation study, however, must be considered in light of limitations with respect to characteristics of the study’s participants. Due to sensitivity of topics related to sexual risk behaviors, participants and their parents’ anonymity and confidentiality of information discussed in focus groups and theater testing was assured. This assurance precluded the ability to obtain data on personal sexual activity of participating girls and, thus, the sexual risk status of the sample was not clearly determined. Participant responses, however, suggest that focus group participants may be sexually naïve or, alternatively, unwilling to share personal experiences. For example, a majority of participants indicated they responded based on their perception of behaviors of their sexually active peers, rather than from personal experience. Also, in general, participants showed difficulty discussing sociocultural factors that may influence condom use among their peers. This difficulty may be due to limited insight on the issue, shyness, or a lack of sexual experience.
Despite these limitations, the adapted intervention adaptation developed in this study provides a strong and necessary foundation for an implementation study of Girl Power Hawai‘i in communities or venues serviced by CBOs. Also, the locally customized (Kameoka & Takishima-Lacasa, 2013) “ADAPT-ITT” methodology used in this study provides a framework that may guide the design of future community-based health promotion intervention adaptations in Hawai‘i and in other Pacific Island regions. This framework involves both qualitative (e.g., focus groups) and quantitative (e.g., controlled effectiveness study yielding outcome data) methods, uses multiple and novel adaptation strategies (e.g., theater testing), and provides a strategy for ensuring fidelity to the original EBI. The use of a dissemination and implementation model as described in this study increases the rigor of research efforts and encourages researchers to build on previous findings (Bartholomew et al., 2011; Sales, Milhausen, & DiClemente, 2006; Van Achterberg, Schoonhoven, & Grol, 2008).
Conclusions
This research represents an important contribution to the adolescent sexual health intervention literature. Girl Power Hawai‘i was developed to address a major gap in services targeting the needs of local NHPIA adolescent girls at risk for STI and related behavioral health issues in Hawai‘i. This adaptation provided the critical first steps in ensuring the relevance and acceptability of Girl Power Hawai‘i among local teen girls, thus paving the way for a community-based randomized controlled study to demonstrate the intervention’s effectiveness in local at-risk communities. Dissemination of the locally customized ADAPT-ITT model as a template for building community collaborations may likely help maximize the value and importance of those partnerships for local adaptation of EBIs in Hawai‘i and the Pacific.
