86084592388Health Educ ResHealth Educ ResHealth education research0268-11531465-364827923860648719510.1093/her/cyw045HHSPA1021219ArticleUsing a domestic and sexual violence prevention advocate to implement a dating violence prevention program with athletesJaimeM. C. D.1*StockingM.2FreireK.2PerkinsonL.3CiaravinoS.1MillerE.1Pediatrics, Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center, Pittsburgh 15213Division of Violence Prevention, National Center for Injury Prevention and Control Centers for Disease Control and Prevention (CDC), Atlanta 30341, USACDC Foundation, Atlanta 30303, USACorrespondence to: M.C.D. Jaime.maria.virata@chp.edu and mcdjaime1026@gmail.com19420191220162842019316679696

‘Coaching Boys into Men’ is an evidence-based dating violence prevention program for coaches to implement with male athletes. A common adaptation of this program is delivery by domestic violence and sexual violence prevention advocates instead of coaches. We explored how this implementer adaptation may influence athlete uptake of program messages and outcomes. Randomly, one school received the program delivered by an advocate while another school received the program delivered by coaches. Athletes completed baseline and follow-up surveys (n = 148), and a subset who received the advocate-led program participated in focus groups (four groups; n = 26). We compared changes in athlete attitudes and behaviors and conducted thematic analyses with qualitative data. We found no significant differences between athletes who received the program from the advocate versus their coaches. Athletes highlighted the advocate’s delivery and role as a non-judgmental adult ally as qualities that influenced their uptake of program messages. The acceptability of the advocate-led program may be related to the implementer type along with specific implementer characteristics and delivery methods. Using advocates together with coaches as implementers could increase the reach of this program. Further study of best practices for Coaching Boys into Men adaptation is needed to guide program dissemination and sustainability.

Introduction

Dating violence (defined as physical, sexual, and emotional abuse by a current or former dating partner) is highly prevalent among adolescents, with 1 in 10 teens reporting physical violence and 1 in 10 teens reporting sexual violence from a dating partner in the past year [1]. Dating violence during adolescence is associated with a myriad of adverse health outcomes, including mental health problems, substance abuse, and poor sexual health [2, 3]. The 2013 Youth Risk Behavior Survey documented that among high school students who dated, females were twice as likely as males to experience some form of teen dating violence and reported a higher prevalence than males for both physical and sexual dating violence [4].

Male perpetrated dating and sexual violence against women and girls have been associated with gender inequitable attitudes, which include endorsement of hyper-masculinity, homophobic attitudes, and condoning rape and abuse perpetration [5–16]. The culture of sports in the United States may also create a context in which such inequitable attitudes and practices are encouraged, in particular within some sports such as football and basketball [11, 12, 17–19]. Athletic programs represent a promising setting for implementing programs to prevent dating violence experienced by female adolescents. On the one hand, some studies show that male student athletes are overrepresented among violence perpetrated by males against females [10, 11], and that such behaviors are associated with gender-inequitable and rape supportive attitudes [12–16]. On the other hand, because male athletes often possess significant social status and are recognized leaders among their peers [10–12, 19, 20], they may have greater ‘potential’ to spread nonviolent, gender equitable attitudes in the school setting. Furthermore, protective factors for teen dating violence include connection with caring adults and prosocial peers [21, 22]; such positive social networks are often embedded in the athletic setting with adult coaches and student athletes.

‘Coaching Boys into Men’ (CBIM) is an evidence-based teen dating violence prevention program for male adolescent athletes (Table I). This program uniquely targets an important out-of-classroom setting through existing relationships between coaches and athletes. Rooted in social norms theory [5, 23] and theories of gender and power [24–26] CBIM addresses gender-based violence by leveraging the influential position of coaches to deliver messages to their athletes about respect, gender-equity and the responsibility to intervene when witnessing violence against women and girls.

Social norms theory postulates that individual behavior is informed by the perceptions and misperceptions of others’ attitudes and behaviors [27]. CBIM equips coaches and their athletes to change these perceptions and to promote positive attitudes and behaviors by becoming active bystanders, identifying and stopping disrespectful and abusive behaviors among their peers. Since athletes also learn from observing role models, coaches are particularly well-positioned to influence athletes’ attitudes and behaviors because of their role as consistent, non-parental mentors invested in the personal development of their athletes [28]. Observing coaches deliver CBIM messages and intervene when witnessing harmful behaviors builds athletes’ skills and confidence to intervene themselves. The original CBIM program is based, in part, on Bandura’s social cognitive theory which posits that behavior is learned through observation of others behaviors and the consequences of those behaviors which in turn increases individuals’ self-efficacy to modify one’s behavior and social context [29]. Thus, for CBIM as a prevention program, observational learning is a core component. This raises a key question about the extent to which changing the implementer from coach to a violence prevention advocate may shift opportunities for such observational learning.

CBIM also draws on theories of gender and power by targeting adolescent male’s attitudes and behaviors related to dating and sexual violence. This ‘gender transformative’ approach theorizes that altering gender norms related to dating and sexual violence perpetration, promoting bystander intervention and understanding the role of gender and power can reduce violence against women [24–26]. Conversations about masculinity, standing up against violence against women, and handling power responsibly are core messages embedded in this program. Athletes’ and coaches’ awareness of gender-based violence and engagement as active bystanders is expected to lead to larger social change to stop violence against women and girls [30–33].

A large scale cluster-randomized trial found the CBIM program significantly increases participating athletes’ recognition of abuse and positive bystander behaviors—intervention to stop peers’ dating violence-related behaviors [31]. One year after program delivery, intervention athletes had a relative reduction of abuse perpetration compared to controls and had lower rates of negative bystander behaviors (i.e. doing nothing or laughing and going along with peers’ abusive behaviors) [30]. Researchers further assessed the association between gender-equitable attitudes, bystander behavior, and recent abusive behaviors. Athletes with higher gender-equitable attitudes and greater intentions to intervene were less likely to perpetrate abuse. Those athletes who had more negative bystander behavior had a higher likelihood of abusive behaviors toward their female dating partners [16]. Abuse perpetration was more common among athletes participating in football and basketball, who also endorsed more gender inequitable attitudes [17].

CBIM was developed by Futures Without Violence (Futures), a national non-profit organization that provides training and technical assistance for advocates and other practitioners involved in domestic and sexual violence prevention [34]. CBIM is freely accessible through their website (www.coachescorner.org). Each of the 12 brief weekly discussions can be integrated into regular athletic activities (such as team practice) (Table II). According to CBIM administrator and staff at Futures, they primarily disseminate the program through domestic and sexual violence service agencies interested in implementing the program in their communities (B. Connors and Y. Gorbea, Futures without Violence, Personal Communication). Via this method of dissemination, Futures offers a train-the-trainer program for local domestic violence (DV) and sexual violence (SV) prevention community educators, commonly referred to as DV and SV advocates in the field and heretofore referred to as advocates. Advocates, in turn, partner with schools to provide a brief training for athletic coaches. CBIM program materials recommend that coaches maintain close contact with an advocate throughout program delivery, particularly to guide coaches on how to address sensitive topics and to provide services such as crisis counseling for athletes who disclose personal concerns.

Futures has collaborated with multiple communities across the United States. The communities and stakeholders who have implemented CBIM and continue to report evaluation of their implementation include 12 different agencies from 10 different cities in 9 different states. Communities range from large urban cities to small rural towns. The majority of program implementation takes place in high schools with a few communities implementing at the middle school and collegiate levels. CBIM has also been implemented in non-school settings including local YMCAs, community sports leagues and at least one juvenile detention center (B. Connors and Y. Gorbea, Futures without Violence, Personal Communication).

As CBIM becomes recognized as a promising prevention practice [35], the adaptation of the program’s implementer from an athletic coach to a DV/SV prevention advocate is increasingly common, with advocates sometimes co-delivering with coaches or delivering the program themselves [B. Connors and Y. Gorbea, Futures without Violence, Personal Communication; 36]. Advocate delivery and co-delivery with coaches can happen for a variety of reasons, including coaches feeling unprepared to deliver CBIM on their own. In some cases, the advocate’s supporting role in delivery facilitates subsequent independent delivery by coaches (B. Connors and Y. Gorbea, Futures without Violence, Personal Communication). There are important tradeoffs to consider in having advocates versus coaches deliver CBIM. Advocates often have greater knowledge of CBIM topics and more experience delivering prevention programming due to their professional background. Nonetheless, coaches’ close relationships with their athletes may be a necessary context for athletes to regard CBIM messages as credible. Given the role of advocates in CBIM’s dissemination and delivery in some settings, it is important to determine whether or not advocates can effectively deliver the program and how athletes receive CBIM messages. To date, however, there have been no studies that examine how changing the implementer from a coach to an advocate influences athlete receptivity to program messages and in turn, program outcomes. The aims of this study were to compare differences in violence prevention outcomes for athletes, specifically their recognition of what constitutes abusive behavior, gender attitudes, intentions to intervene, and bystander intervention behaviors, between advocate-led (adapted model) and coach-led (original model) CBIM groups and to explore athletes’ perspectives on receiving CBIM from an advocate. Therefore, this adaptation study focused on studying the differences in program implementation and outcomes based on changing the implementer (i.e. from coach to prevention advocate) and not on program evaluation. Further, the exploration of athletes’ perspectives concentrated on the potential effectiveness of an advocate as a CBIM implementer and did not emphasize the theoretical frameworks of the CBIM program.

MethodsRecruitment and study procedures

Two local high schools in the Pittsburgh region were identified to participate in the study based on their interest in implementing the CBIM program, their existing partnership with a local DV/SV prevention agency, and the presence of large athletic programs. The school districts did differ in overall demographic characteristics – school district with the coach-led delivery had far more African American students (65 compared with 22%) and a greater proportion of students on free/reduced lunch (98 versus 42%) [37, 38]. Prior to the start of the study, the Principal Investigator and the Program Director of the local agency met and received permission from school administrators, including superintendent and principals, to have their school’s athletic department participate in the study. Both school districts and athletic directors were amenable to using either coaches or the advocate to deliver the program. The research team randomly assigned one school to receive CBIM delivered by a male advocate (adapted model) and assigned the other school to receive CBIM delivered by male coaches (original model) to minimize biases from the advocate’s or school’s preferences for program delivery. The advocate was a Caucasian adult male currently involved in implementation of CBIM locally, employed by the local DV/SV prevention agency, and who worked in close proximity to the two schools involved in the study. At both schools, CBIM was delivered to the male football team during the fall sports season and the male basketball and wrestling teams during the winter sports season.

The University of Pittsburgh Institutional Review Board approved the study with a waiver of parental permission and waiver of written consent for survey and focus groups. Research staff and the advocate approached each school’s athletic director to coordinate recruitment and survey administration. At the beginning of each season, research staff distributed study information packets to male student athletes to take home to review with their parents/caregivers. The information packet included a parent letter with contact information for parents/caregivers to call if they did not want their child to participate in the study. Research staff did not receive any contact from parents/caregivers.

Prior to survey administration, research staff reviewed an assent information sheet and received verbal consent from each student athlete willing to participate in a survey at the start of each sports season (Time 1) and 3 months later at the end of each sports season (Time 2). Each participant who completed the anonymous paper survey, which took ~15–20 min, created an individual identification code by responding to a series of questions to which only he would know the answer. Participants received a $10 gift card for participating in each survey.

Once Time 1 surveys were collected, coaches delivering the program (original model) received CBIM training and a Coaches Kit (www.coaches-corner.org), which included a guide on how to deliver the program, scripted cards for weekly discussion, and a list of youth relevant resources such as crisis hotline numbers. The advocate delivering the CBIM program was trained by Futures prior to the start of the study. The advocate, in turn, trained the football, basketball and wrestling coaches delivering the program and connected with them biweekly to provide additional guidance on implementation. Having the same advocate provide implementation support across coaches controlled for variation in training and technical assistance at the coach-delivered school.

Student athletes who completed both baseline and follow up surveys in the advocate-delivered school were invited by their coaches or the advocate to participate in focus group discussions. Interested student athletes received an additional parent letter to review with their parents/caregiver. Similar to the surveys, the research staff did not receive any contact from parents/caregivers about not wanting their child to participate.

Before the start of each focus group discussion, the research staff reviewed the assent form and allowed each participant to decide if they wanted to take part in the discussion. The focus group facilitator reminded participants to not share personal information during the discussion and reviewed the definition of confidentiality. Two female research staff facilitated the focus group discussions. The same focus group questions and prompts guided all discussions. Discussions were ~30–40 min and included 4–10 student athletes. The focus groups were held in a private and quiet location after school on school campus. A total of 4 focus group discussions (n = 26) with student athletes who received the advocate-led program were conducted and audio recorded. Recordings were destroyed after transcripts were reviewed for accuracy and any identifying information removed. Participants received a $10 gift card for their participation.

During the course of this adaptation study, observations as well as feedback from coaches and advocates were collected. However, reviewing all this process evaluation data in depth was beyond the scope of this manuscript which focused on how changing the implementer influenced the athletes themselves.

InstrumentsQuantitative surveys

Demographic questions included grade level, race/ethnicity, parental level of education, and sport. Baseline and follow up (Time 2) surveys included the following athlete outcome measures: recognition of abusive behaviors, gender-equitable attitudes, intentions to intervene when witnessing abusive behaviors, positive and negative bystander intervention and abuse perpetration. A five-point Likert scale was used to measure recognition of abuse, gender attitudes, and intentions to intervene. For example, recognition of abuse included a series of questions about how abusive a behavior was, rated from not abusive (1) to extremely abusive (5); intentions to intervene had items about how likely one would be to stop a particular harmful behavior from happening, rated from very unlikely (1) to very likely (5), calculated as mean scores. The bystander intervention and abuse perpetration items queried behaviors witnessed or perpetrated within the past 3 months. Athletes were asked about perpetrating any of 10 abusive behaviors and witnessing 9 abusive behaviors by peers. A summary abuse perpetration score was created by adding together any ‘yes’ responses indicating perpetration. Bystander intervention was converted into two separate summary scores for any positive and negative intervention. Positive intervention included the following responses: ‘I told the person in public that acting like that was not okay’, ‘I told the person in private that acting like that was not okay’, ‘I talked to our coach about it privately’, or ‘I talked to another adult (not coach)’. Negative intervention included ‘I didn’t say anything’ or ‘I laughed or went along with it’. These measures were developed and used by researchers from the original CBIM cluster-randomized control clinical trial [31].

Focus group discussions

Discussion topics included athletes’ experiences receiving the CBIM program (e.g. how the implementer typically delivered program sessions, level of athlete participation, and athlete recollection of program messages and discussions); perception of implementer (e.g. comfort with, belief in and connection to advocate); and preferences of having their coach versus the advocate deliver the program.

AnalysisSurveys

Chi squared and Fisher’s exact tests were used to compare demographic characteristics between athletes who completed follow-up compared to those who did not, as well as between athletes who received CBIM from the advocate versus coaches. Changes in outcome scores between baseline and follow-up were calculated and tested for significant differences within coach-led and advocate-led groups using paired t-tests (and as needed Wilcoxon signed ranked test to account for non-normal distribution). Adjusted linear regression models were constructed to account for significant between-group differences by grade and race at baseline, to compare advocate-led versus coachled changes in key outcomes of interest. Using an adjusted difference of differences controls for the effect of grade and race and calculates an adjusted difference in the effect of the program delivery per group. Specifically, we assess the effect of the program on the mean change scores of recognition of abuse, gender attitudes, intentions to intervene, and bystander intervention, and the difference of those scores between advocate-led and coach-led athletes. These analyses included matched data only—participants who completed both baseline and follow up surveys. As missing data were minimal (<3% across variables of interest), floating sample sizes were used for analyses rather than imputing missing values. Alpha level for significance was set at P < 0.05.

Focus groups

Two members of the research team independently coded focus group transcripts with NVivo 10 soft-ware using codes developed a priori from focus group questions and codes developed inductively through an initial review of athlete responses. Coded transcripts were merged and coders discussed and reconciled areas of disagreement, added codes, and finalized the codebook. The qualitative approach focused on data reduction, data display and conclusion drawing and verification [39] Following the initial descriptive coding process, one team member created matrices to display key content organized by code. Thematic analysis was used to search for common patterns and themes within and across codes most relevant to study aims. Salient themes were discussed and validated in a series of co-analysis meetings between members of the research team. Quotes were used to illustrate themes. Since previous CBIM studies demonstrated the effectiveness of coach-led delivery [30, 31], qualitative data from coaches and athletes who were involved in the coach-led delivery were not included in this analysis. Instead the analysis focused on athletes’ perspectives regarding the potential effectiveness of an advocate as an implementer of CBIM as opposed to a coach.

Results

Of the 193 athletes who completed the baseline survey, 148 completed the follow up survey (77% retention). Those athletes lost to follow up did not significantly differ from those retained by grade, race/ethnicity, place of birth, or parental education. Attrition was similar across both advocate and coach groups. Non-completers were similar to those retained in terms of baseline intentions to intervene, knowledge of abuse, and bystander behavior. However, non-completers had significantly lower gender equitable attitudes (P = 0.036, t =–2.12) and higher abuse perpetration (P = 0.002, t = 3.15) scores at baseline compared with those who completed the study.

Demographic characteristics for the 148 athletes who completed both baseline and follow up surveys are shown in Table III. The majority of athletes (61%) played football and the remaining athletes played basketball (20%) or wrestling (19%). Most athletes reported being in 10th and 11th grade (32% and 25%, respectively), with grade level ranging from 7th to 12th grade. Because one of the coach-led teams pulled in middle school athletes for card delivery, seventh and eighth graders accounted for 4.7% of all athletes. The majority of the sample identified their race/ethnicity as Black or African American (54%). Parental education was a proxy for socioeconomic status—41% reported at least one parent was a college graduate. The two intervention arms differed significantly by grade (P = 0.011, Chi-square test = 16.53) and race/ethnicity (P < 0.01, Chi-square test = 32.21) with the coach-led group being younger and more identifying as Black or African American compared to the advocate-led group (Table III).

There were significant increases within the advocate-led group from baseline to follow up for recognition of abusive behavior (P = 0.008, Wilcoxon signed ranked test = 425.50) and the coach group for gender attitudes (P = 0.013, Wilcoxon signed ranked test = 437.50) (Table IV). As seen in prior evaluations of the CBIM program [31], the absolute values from baseline to follow up show similar positive changes. There were trends in the advocate group towards greater intentions to intervene and in the coach group towards a reduction in negative bystander behavior, but these within group changes were not statistically significant. The positive bystander and abuse perpetration scores showed no significant changes for either group. For comparisons in intervention effects between coach-led versus advocate-led delivery, there were no statistically significant differences in changes in scores for any outcome between the two groups.

Athlete focus group results

Of the student athletes who participated in focus groups, a majority were football players 50% (n = 13) and the remaining were wrestling 31% (n = 8) and basketball 19% (n = 5) players. Qualities that influenced athletes’ perceptions of the advocate’s delivery of CBIM and their experience as program participants coalesced under three main themes. These included: the advocate’s efforts to build relationships through athletic engagement, perceptions of the advocate as a non-parental adult ally, and the advocate’s delivery style and expertise.

Relationship building through athletic engagement

Prior to CBIM delivery, the advocate cultivated relationships with athletes through his participation in athletic activities including team weight-lifting sessions and practices. As one athlete reported, “He learned how we act towards each other. He’d get out there, and he’d wrestle with us.” Another athlete tied relationship building activities to his changing perception of the advocate as an outsider and his receptivity to program messages:

He was in the weight room, and I was like, ‘Who’s this bulky guy?... This is our weight room.’ And then towards the end of the season, I was like, ‘This is a cool guy.’ The longer he was there, the more I listened.

A few athletes also reported feeling connected to the advocate because of his sports background. One athlete explained, ‘He was even a wrestler in high school... so he knew like what we were going through in the season and everything’.

Athletes on all sports teams consistently described how their responsiveness to CBIM and their acceptance of the advocate increased over the course of program delivery as they built a trusting relationship with the advocate, explaining that, ‘the more weeks he came in, we bonded more’, and ‘he became one of those people you trust to talk to’. During CBIM delivery, the advocate continued to participate in athletic activities outside of program delivery to maintain relationships and connect with athletes during their regular team training. These activities increased athletes’ engagement with the advocate beyond program delivery times.

Athlete perception of advocate as an adultally

Athletes from multiple sports teams reported that they were comfortable talking with the advocate about CBIM topics because they viewed him as a non-judgmental adult ally. Some athletes suggested the advocate was easy to talk to because he approached them as equals instead of adopting an attitude of authority like a coach or teacher might.

He tried to relate to us... build like a friendship relationship with us more than like a parent or a coach or a teacher... I mean like the first day he came in and the first thing he said is, ‘Don’t call me sir. Don’t call me any of that. Just call me [advocate name].

Other athletes reported they felt comfortable sharing their experiences with the advocate because he did not, ‘care how you talked’ or impose judgment. As one athlete explained, ‘[the advocate would] laugh like along with us, but never was like, ‘Wow! You shouldn’t do that!’ or ‘That’s wrong!’... so it was easy’. Another athlete said that he may have felt judged by his coach if his coach was the one implementing CBIM. As he explained, ‘I think that we would have hesitated to respond how we did with the advocate just for the simple fact that it’s our coach. He sees us every day... he might have judged’. Finally, many athletes discussed how the advocate’s life experiences were similar to their own as male adolescents. Athletes explained how these common life experiences influenced their beliefs that the advocate could relate to their situation. One athlete stated, ‘He was a teenage boy... we all generally go through the same stuff, so he definitely understood what was going on’.

Delivery style and expertise

Athletes consistently reported viewing the advocate as a credible and effective messenger because of his skill in delivering the program. Some athletes explained how the advocate’s confident and natural delivery style contributed to his credibility and the ease of program discussions. Many athletes cited examples of the advocate’s delivery style. Various athletes explained that the advocate, ‘wouldn’t just read straight from the article’, ‘didn’t look confused or flustered’, and ‘talked very casually’. Other athletes pointed to the advocate’s use of real life examples to explain program content as influencing their sense that he was believable and trustworthy. One athlete reported, ‘I believe him because there’s been real good examples’. Another participant explained, ‘When he would give personal experiences it showed a sense of that he’s putting himself out there’. Several athletes reported that the advocate’s use of interactive teaching methods impacted their learning and made them feel meaningfully engaged in discussions. As one athlete described, ‘he would ask for examples of certain behaviors... and then he would collaborate on anything we said’. Last, a few athletes suggested that compared to their coaches, the advocate had superior content area knowledge. One athlete noted, ‘I don’t feel coach would know the information that the advocate knows’. Another said, ‘The advocate’s like well-versed in all the topics that he’s talking about, and our coach like would just be making it up as he went along’.

Discussion

This study explored how CBIM sessions delivered by an advocate rather than by athletic coaches might influence athletes’ attitudes and behaviors. The CBIM intervention effects did not vary between advocate and coach-led delivery. However, the athletes who received CBIM from the advocate had a significant increase in recognition of abusive behavior from baseline to follow-up. The athletes receiving the program from their coaches had improved changes in gender attitudes from baseline to follow up. In focus groups, athletes’ consistently reported that the advocate made efforts to build a relationship with them, was perceived as a non-judgmental ally, and had an interactive delivery style and content area expertise. Our findings support that an advocate can influence athletes’ uptake of program messages. In addition to the qualitative study of athletes’ perceptions, survey results indicate that the advocate and coach deliveries resulted in similar athlete outcomes. This is particularly relevant for advocates implementing the program as a way to build initial connections and trust with schools and athletic programs or for advocates co-implementing with coaches who need additional hands-on support.

Notably, while none of the differences in degree of change comparing advocate to coach delivered CBIM were statistically significant, the recognition of abusive behaviors among athletes who received the advocate delivered program had a significant increase from baseline to follow up. There were also trends for the advocate group towards greater intentions to intervene. Advocates’ typical expertise in violence prevention education likely gives them an advantage relative to coaches in delivering messages about abusive behaviors and bystander intervention. As a result, athletes may have greater uptake of these messages. Athletes whose coaches delivered the program had a significant increase in gender equitable attitudes and reported less negative bystander behavior at follow-up, possibly reflecting a coach’s ability to influence athletes’ attitudes and behaviors as intended by the original program design. These changes were in the expected direction based on the randomized controlled trial results that assessed program effectiveness when delivered by coaches (original model) [31]. The abuse perpetration scores showed no significant changes at this short term follow up, similar to findings from the randomized trial where abuse perpetration reduction was demonstrated only at 1-year follow-up [30].

Athletes’ reflections in group discussions suggest that their openness to the advocate as an implementer of CBIM hinged heavily on the additional time and effort the advocate invested cultivating relationships with them. From the outset, the advocate intentionally integrated himself into team activities such as workouts and practices to get to know athletes and gain their trust. Athletes also reported that their bond with the advocate became stronger through interaction during program sessions. The advocate’s efforts to build relationships with athletes prior to program delivery may have been critical in laying the foundation for athletes’ positive views of the advocate during delivery. Other researchers studying violence prevention programs have found that implementers add or modify program activities to build relationships with participants prior to program delivery (L Perkinson, KE Freire, ME Stocking, et al., in preparation) and improve participant engagement during delivery [31, 40–43]. Youth Empowerment Solutions (YES), is a youth violence prevention program designed to engage youth participants in leadership development and a community improvement project [44]. In a study of YES implementation in four cities, Perkinson et al. (in preparation) found that some implementers added activities to build trust and relationships with youth participants because they perceived a need for participants to feel more comfortable with the adult implementers and to view them as credible resources. For a program like CBIM, which assumes that there is an existing relationship between coaches and athletes, relationship-building may be critically important for other types of implementers or for coaches who lack a well-developed relationship with their athletes.

Athletes also identified two specific characteristics that made the advocate more relatable and influenced their openness to program messages during implementation. First, the advocate’s background as a former male athlete and a younger appearing adult provided common ground for connecting with athletes. Athletes’ perceived that the advocate was someone who understood their experiences both as teenage boys and as athletes. Second, athletes viewed the advocate as a non-judgmental ally and discussed how this view influenced their willingness to trust the advocate with sensitive information during program implementation. Some athletes also noted that the advocate did not present as an authority figure, but instead used the experiences he shared with athletes to connect with them as equals. Taken together, these advocate characteristics greatly influenced athletes’ willingness to participate in planned program discussions and interact with the advocate outside of program delivery. Because discussions between athletes and the implementer is a critical part of each CBIM session, athletes’ engagement in program discussions is an indicator of program implementation fidelity [45]. From a process evaluation perspective [46], the program ‘dose delivered’ is dependent not only on the advocate’s delivery of program content but his ability to engage athletes in conversation about the topics. Conversely, the program’s ‘dose received’ is dependent on athletes engaging in discussions on CBIM topics and relating CBIM messages to their own experiences. Additional analysis comparing the observations of advocate-led and coach-led program delivery and coaches and advocate interviews may provide further insight about implementer’s ability to engage athletes and the athletes’ level of engagement in the program.

In addition to possessing relatable characteristics and developing a strong relationship with athletes, athletes viewed the advocate as a credible source for CBIM messages, referencing his engaging delivery style and content area expertise. As athletes explained, the advocate put program messages into his own words, related complex concepts to real life examples, and elicited sharing through interactive teaching methods. Some athletes reported that the advocate seemed especially knowledgeable and comfortable discussing sensitive content. Other researchers have found that implementer characteristics, such as comfort with content, values [42, 47, 48], and previous experience implementing programs [40, 51] influence program delivery. In particular, implementers that endorse and have previous experience with interactive teaching methods tend to incorporate this approach (vs. more didactic) to increase student engagement [40], and research suggests that interactive teaching methods benefit student learning [43, 50]. Our observations are consistent with these findings. In this case, the advocate’s interactive style resulted in the intended effect of a high level of athlete engagement.

Findings from this study highlight specific characteristics that DV or SV prevention advocates may need to embody for athletes to consider them as credible sources for violence prevention messages. The advocate’s relationship and connection with athletes through sports was consistent with a coach’s natural role. Athletes may be more receptive to advocates who have athletic qualities, can easily integrate into the athletic setting, and are able to invest additional time and effort to build trusting relationships. The athletes’ feedback also provided insight that certain implementer characteristics may be important beyond implementer type (coach versus non-coach). Namely, being perceived as a non-judgmental ally in contrast to a strong disciplinary or authority figure and using an engaging and interactive delivery style may elicit greater athlete participation in program discussions and thus improve uptake of program messages. Given the prominent role of DV/SV advocates in the dissemination of CBIM and the increasingly common adaptation of the program’s implementer from an athletic coach to advocate, this has important implications for the spread and reach of CBIM.

Further investigation is needed to explore stakeholders’ perspectives on changing the implementer and using an advocate to deliver CBIM. Investigation of coaches’ perspectives on an advocate-delivered program and their acceptance of the advocate may provide additional guidance on how the program may be best implemented with advocates. Similarly, observations and feedback from coaches and advocates may help identify strategies used by the advocate in building relationships and trust with the coaches themselves. For instance, advocates could offer more intensive hands-on training or initially co-implement with coaches who feel uncomfortable or not yet ready to deliver the program on their own. Program spread and reach may increase if agencies work with athletic programs to implement these types of hybrid models. This could lead to further hypothesis generation about best practices for CBIM implementation and evaluations examining different roles advocates and coaches serve during program implementation. In addition, a longitudinal assessment of longer term influences on athletes (e.g., whether the advocate- delivered program has similar lasting effects as the coach-delivered approach) may help to delineate benefits of this implementer adaptation and how it can improve dissemination and sustainability of CBIM as an evidence-based practice.

Limitations

These findings are not without limitations. This was a small scale exploratory study to examine differences in program delivery comparing two different implementer-types qualitatively with only two groups, and thus was not powered to assess program efficacy to the same magnitude as the original randomized controlled trial. Since we only randomized two schools, between group differences may not have been detected. Further, when we compared the two schools, we did not account for the nesting of athletes within teams within schools or the use of multiple comparisons which likely resulted in inflation of p-values; this means that any trends emerging between groups were likely not to be statistically significant. The shifts in some knowledge, attitudes, and behaviors certainly are in similar directions to outcomes found in the RCT, suggesting some program effect in both groups.

As with other violence prevention studies, the surveys relied on self-report and likely introduced social desirability bias. Despite athletes creating their own self-generated anonymous code with survey administration, which was used to preserve anonymity and reduce likelihood of misreporting around sensitive questions, the anonymous code may not have been enough to eliminate such bias towards under reporting undesirable behaviors. Non-completers had significantly lower gender equitable attitudes and higher abuse perpetration scores at baseline compared to those who completed the study; it is unclear how much intervention effects would differ for this higher risk group. A small number of participants also participated in both fall and winter sports within the advocate-led (5%) and coach-led (4%) groups, which may have influenced the results. Lastly, even though the analysis controlled for racial differences and there were no statistically significant differences in parental education (a proxy for socioeconomic status) between the two schools, there were considerable differences in the racial and socioeconomic composition of the two school districts. These differences by school districts were not matched based on the risk factors related to violence perpetration.

A major strength of the study is the use of qualitative data to provide an interpretive context in addition to the survey results, allowing us to explore how athletes received messages from the advocate and perceived him as a credible implementer. The primary limitation of the qualitative findings is the potential for selection bias. Because a combination of self-selected and coach-selected athletes participated in focus group discussions, participants may have had more positive views of the advocate or program than non-participants. Participants, however, were representative of athletes across all three athletic teams, and themes presented herein reflect common perceptions among athletes across teams.

Additionally, despite athletes’ positive perceptions of the advocate in this study, findings cannot be generalized to other prevention advocates and youth athletes. The advocate in this study was male, Caucasian, athletic and shared common interests with athletes. Many DV/SV prevention advocates are female, racial background vary, and advocates may or may not have athletic or coaching experience to share and connect with athletes. More youth athletes identified as African American in the coach-led arm than the advocate-led arm. The race of the advocate in relation to the race of the athletes was not explored in this study and could also have an impact on the program’s efficacy. Therefore, the advocate in this study may not reflect the typical violence prevention practitioner working for local agencies and the athletes perceptions may not be representative of all youth athletes, specifically youth from different racial backgrounds.

The study focuses on athletes who participated in the advocate-led program, and does not include discussion among athletes in the coach-led program or the process evaluation including formal observations of advocate and coach-led deliveries. The previously completed randomized controlled trial already provides results of the effectiveness of coach-led delivery. This study took advantage of CBIM implementation in practice to learn more specifically about the advocate’s effectiveness in delivering CBIM as a non-coach implementer.

Finally, other healthy relationships and violence prevention program evaluations within classroom or school based settings have shown the importance of other contextual factors, such as content of program materials and curricula, varied methods to use to facilitate and instruct program content, and differences between who is delivering the program, an internal (teacher) or external (community stakeholder) person [51–53]. These additional contextual factors were not explored in this study since the adaptation was focused on the advocate as the CBIM implementer.

Conclusions

Part of the appeal of the CBIM program is that it can be implemented by coaches rather than relying on violence prevention educators and advocates, thereby increasing local capacity in communities to address teen dating and sexual violence prevention. Thus, in the longer term, program scalability and sustainability ultimately depend on coaches as primary implementers and require strategies to address coach barriers to implementation. That said, our findings demonstrate that CBIM was successfully implemented by an advocate. Using prevention advocates to initially deliver or co-deliver CBIM with coaches who may otherwise not adopt and independently deliver CBIM may increase the spread and reach of the program. To achieve widespread adoption of CBIM, implementing organizations such as DV and SV agencies could consider this option. Advocates can help to expand reach and ensure CBIM’s core messages are well received by athletes while working closely with coaches within the school sports setting to increase likelihood of coaches implementing the program themselves.

Acknowledgement

We wish to thank our partners Center for Victims, Pittsburgh, PA, a local domestic violence agency and Futures Without Violence, San Francisco, CA, the developers of Coaching Boys Into Men.

Funding

This work was supported by a grant from the Robert Wood Johnson Foundation [grant number 69547] to the CDC Foundation and implemented in partnership with the CDC Division of Violence Prevention. The CDC Division of Violence Prevention funded researchers from Children’s Hospital of Pittsburgh, University of Pittsburgh Medical Center.

Disclosure

The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Centers for Disease Control and Prevention.

Conflict of interest statement

None declared.

ReferencesKannL Youth risk behavior surveillance - United States, 2013. MMWR Suppl 2014; 63:1–168.Exner-CortensD, EckenrodeJ, RothmanE. Longitudinal associations between teen dating violence victimization and adverse health outcomes. Pediatrics 2013; 131: 71–8.23230075SilvermanJG, RajA, MucciLA Dating violence against adolescent girls and associated substance use, unhealthy weight control, sexual risk behavior, pregnancy, and suicidality. jama 2001; 286: 572–9.11476659VagiKJ, O’Malley OlsenE, BasileKC Teen dating violence (physical and sexual) among US high school students: Findings from the 2013 National Youth Risk Behavior Survey. JAMA Pediatr 2015; 169: 47–82.BerkowitzA, ed. Fostering men’s responsibility for preventing sexual assault In: ScheweP (ed). Preventing Violence in Relationships: Interventions Across the Life Span. Washington, DC: American Psychological Association, 2002.VermaR, PullerwitzJ, MahendtaK Challenging and changing gender attitudes among young men in Mumbai, India. Reprod Health Matters 2006; 14: 135–43.17101432ReedE, SilvermanJG, RajA Social and environmental contexts of adolescent and young adult perpetrators of intimate partner violence. Am J Mens Health 2008; 2:260–72.19477789McMahonS, PostmusJL, KoenickRA. Conceptualizing the engaging bystander approach to sexual violence prevention on college campuses. J Coll Stud Dev 2011; 52: 115–30.ReedE, SilvermanJG, RajA Male perpetration of teen dating violence: associations with neighborhood violence involvement, gender attitudes, and perceived peer and neighborhood norms. J Urban Health 2011; 88: 226–39.21311987CrossetTW, PtacekJ, McdonaldMA Male student-athletes and violence against women: a survey of campus judicial affairs offices. Violence Against Women 1996; 2: 163–79.12295457HumphreyS, KahnA. Fraternities, athletic teams, and rape: importance of identification with a risky group. J Interpers Violence 2000; 15: 1313–22.BoeringerSB. Influences of fraternity membership, athletics, and male living arrangements on sexual aggression. Violence Against Women 1996; 2: 134–47.12295455KossMP. Rape. Scope, impact, interventions, and public policy responses. Am Psychol 1993; 48: 1062–9.8256879MurnenSK, KohlmanMH. Athletic participation, fraternity membership, and sexual aggression among college men: a meta-analytic review. Sex Roles 2007; 57:145–57.SawyerRG, ThompsonEE, ChicorelliAM. Rape myth acceptance among intercollegiate student athletes: A preliminary examination. Am J Health Studies 2002; 18: 19–25.McCauleyHL, TancrediDJ, SilvermanJG Gender- equitable attitudes, bystander behavior, and recent abuse perpetration against heterosexual dating partners of male high school athletes. Am J Public Health 2013; 103: 1882–7.23947324McCauleyHL, JaimeMC, TancrediDJ Differences in adolescent relationship abuse perpetration and gender-in-equitable attitudes by sport among male high school athletes. J Adolesc Health 2014; 54: 742–4.24582876GageEA. Gender attitudes and sexual behaviors: comparing center and marginal athletes and nonathletes in a collegiate setting. Violence against Women 2008; 14: 1014–32.18703773BoeringerSB. Associations of rape-supportive attitudes with fraternal and athletic participation. Violence against Women 1999; 5: 81–9.ForbesGB, Adams-CurtisLE, PakalkaAH Dating aggression, sexual coercion, and aggression-supporting attitudes among college men as a function of participation in aggressive high school sports. Violence Against Women 2006; 12: 441–55.16617170CapaldiDM, KnobleNB, ShorttJW A systematic review of risk factors for intimate partner violence. Partner Abuse 2012; 3: 231–80.22754606WilkinsN, TsaoB, HertzM Connecting the dots: an overview of the links among multiple forms of violence. In: National Center for Injury Prevention and Control Centers for Disease Control and Prevention, Prevention Institute, eds. Atlanta GA, Oakland CA, 2014.FabianoPM, PerkinsHW, BerkowitzA Engaging men as social justice allies in ending violence against women: evidence for a social norms approach. J Am Coll Health 2003; 52: 105–12.14992295BarkerG, RicardoC NascimentoM Engaging Men and Boys in Changing Gender-Based Inequity in Health: Evidence from Programme Interventions. Geneva, Switzerland: World Health Organization, 2007.International Center for Research on Women and Instituto Promundo. Engaging Men and Boys to Achieve Gender Equality: How Can We Build on What We Have Learned? Washington, DC: International Center for Research on Women and Instituto Promundo, 2007.PulerwitzJ, SegundoM, NascimentoM “Promoting more gender-equitable norms and behaviors among young men as an HIV/AIDS prevention strategy,” Horizons Final Report. Washington, DC: Population Council, 2006.BerkowitzAD. An overview of the social norms approach In: LedermanLC and StewartCP (eds). Changing the Culture of College Drinking: A Socially Situated Health Communication Campaign. Creskill, NJ: Hampton Press, 2005: 193–214.JaimeMCD, McCauleyHL, TancrediDJ Athletic coaches as violence prevention advocates. J Interpers Violence 2014; 30: 1090–111.25015237BanduraA Social cognitive theory In: VastaR (ed). Annals ofChild Development, vol. 6 Greenwich, CT: Six Theories of Child Development, 1989.MillerE, TancrediDJ, McCauleyHL One-year follow-up of a coach-delivered dating violence prevention program. A cluster randomized controlled trial. Am J Prev Med 2013; 45:108–12.,23790995MillerE, TancrediDJ, McCauleyHL “Coaching boys into men”: A cluster-randomized controlled trial of a dating violence prevention program. J Adolesc Health 2012; 51: 431–8.23084163BanyardVL, MoynihanMM, PlanteEG. Sexual violence prevention through bystander education: an experimental evaluation. J Commun Psychol 2007; 35: 463–81.CokerAL, FisherBS, BushHM Evaluation of the green dot bystander intervention to reduce interpersonal violence among college students across three campuses. Violence Against Women 2015; 21: 1507–27.25125493Futures Without Violence. 2015 Available at: http://www.futureswithoutviolence.org/. Accessed: 16 June 2016.DeGueS Preventing Sexual Violence on College Campuses: Lessons from Research and Practice Part One: Evidence-based Strategies for the Primary Prevention of Sexual Violence Perpetration. Division of Violence Prevention, Centers for Disease Control and Prevention, 2014.Centers for Disease Control and Prevention. Understanding teen dating violence. 2014; Available at: http://www.cdc.gov/Features/datingviolence/. Accessed: 16 June 2016.Pennsylvania Department of Education. Enrollment Reports and Projections. 2014–2015 [cited 2016]; Available at: http://www.education.pa.gov/Data-and-Statistics/Pages/Enrollment%20Reports%20and%20Projections.aspx. Accessed: 05 June 2016.Pennsylvania Department of Education. Pennsylvania Department of Education Food and Nutrition Division Building Data Report (Lunch). 2015–2016. Available at: http://www.education.pa.gov/Documents/Teachers-Administrators/Food%20and%20Nutrition/Reports/2015-2016%20Building%20Data%20Report.pdf. Accessed: 05 June 2016.MilesMB, HubermanAM, SaldanaJ Qualitative Data Analysis: A Methods Sourcebook, 3rd edn Thousand Oaks: Sage Publications, 2014.OzerEJ, WanisMG, BazellN. Diffusion of school-based prevention programs in two urban districts: adaptations, rationales, and suggestions for change. Prevention Science 2010; 11: 42–55.19697133AaronsGA, MillerEA, GreenAE Adaptation happens: a qualitative case study of implementation of The Incredible Years evidence-based parent training programme in a residential substance abuse treatment programme. J Child Serv 2012; 7: 233–45.Miller-DayM, PettigrewJ, HechtML How prevention curricula are taught under real-world conditions. Health Educ 2013; 113: 32–44.PettigrewJ, Miller-DayM, ShinY Describing teacher-student interactions: a qualitative assessment of teacher implementation of the 7th grade keepin’ it REAL substance use intervention. Am J Commun Psychol 2013; 51: 43–56.ZimmermanMA, StewartSE, Morrel-SamuelsS Youth empowerment solutions for peaceful communities: combining theory and practice in a community-level violence prevention curriculum. Health Promot Pract 2011; 12: 425–39.21059871DaneAV, SchneiderBH. Program integrity in primary and early secondary prevention: are implementation effects out of control? Clin Psychol Rev 1998; 18:23–45.9455622StecklerAB, LinnanL IsraelB Process Evaluation for Public Health Interventions and Research. San Francisco, CA: Jossey-Bass, 2002.HillLG, MaucioneK, HoodBK. A focused approach to assessing program fidelity. Prevent Sci 2007; 8:25–34.MooreJE, BumbargerBK, CooperBR. Examining adaptations of evidence-based programs in natural contexts. J Prim Prevent 2013; 34: 147–61.DusenburyL, BranniganR, HansenWB Quality of implementation: developing measures crucial to understanding the diffusion of preventive interventions. Health Educ Res 2005; 20: 308–13.15522898ResnicowK, DavisM, SmithM How best to measure implementation of school health curricula: a comparison of three measures. Health Educ Res 1998; 13: 239–50.10181022TuttyL Healthy Relationships Preventing Teen Dating Violence: An Evaluation of the Teen Violence Prevention Program. 2011 Available at: http://www.canadianwomen.org/sites/canadianwomen.org/files/PDF%20-%20VP%20Resources%20-%20CWF%20H-ealthy%20Relationships%20-%20FULL%20REP0R-T%20-%20April%2029%202011.pdf. Accessed: 14 September 2016.FoxCL, HaleR, GaddD. Domestic abuse prevention education: listening to the views of young people. Sex Educ 2014; 14: 28–41.OllisD The role of teachers in delivering education about respectful relationships: exploring teacher and student perspectives. Health Educ Res 2014; 29: 702–13.24939857

CBIM program description

Program materialsProgram contentProgram deliveryTraining and tech-nical assistanceOutcomes for athletes
‘Coaches Kit’ isAbusive behaviorImplementerCoaches receive 1–3-Primary
 freely download-Coaches define dis-An athletic coach h training led by aIncreased awareness
 able from coaches- respectful and abu- who is an adult trained violence and recognition of
 corner.org and sive behavior, and mentor trusted and prevention abusive behaviors
 includes: promote healthy respected by advocate.
CBIM Playbook alternatives. athletes.Trained advocate is
Provides coach with available to assistIncreased gender-
 facts and informa- coaches with con- equitable attitudes
 tion about teenGender attitudesCard training cerns that ariseIncreased intention to
 dating violence, and norms lessons during program de- intervene
 ‘teachable moment’Each training lesson livery, including
 scenarios, and tips is at least 10– disclosures.
 and tactics on howCoaches promote 15 min.Futures Without
 to talk to athletes gender-equity andTraining lessons are Violence, CBIM’sSecondary
 and respond to positive, non-vio- delivered over at developer, hostsIncreased positive by-
 harmful behavior lent attitudes and least 6 weeks train-the-trainer stander intervention
 or language. norms about mas- (no > 2 lessons per sessions for vio-
 culinity and male weeks). lence prevention
CBIM Card Series sexuality.Training lessons are advocates and pro-Decreased abuse
Twelve training cardsBystander participatory and vides support to perpetration
 guide coaches to intervention discussion-based. implementing
 lead brief discus-Coaches promote andTeachable moments agencies.
 sions on topics model bystanderCoach addresses
 relevant to dating intervention skills opportunities to re-
 violence preven- to speak up and inforce and model
 tion. Each card in- intervene when positive language
 cludes an witnessing disres- and behavior with
 introduction, ask pectful and harmful athletes during and
 the player ques- behavior. outside card train-
 tions, and discus- ing lessons.
 sion and wrap-up
 points.
Other resources
Tip sheets with stra-
 tegies on how to
 get started with the
 program and share
 messages with
 broader
 community.
List of professional
 violence prevention
 agencies for
 referral.

CBIM Card Series Description

Training Card SeriesSummary of content
Prep Card 1 Respect, Integrity and Non-ViolenceOverview of the components of the toolkit: CBIM Playbook and Card Series. Discusses the use of the training cards, finding time within the existing practice schedule to work in discussions of each of the cards throughout the sports season.
Prep Card 2 Recruit AlliesGain school administrator and parent support by letting them know about the team’s engagement in CBIM, know the school’s disciplinary policies regarding sexual harassment and violence (including reporting requirements), and know local resources to refer athletes should disclosures arise.
Prep Card 3 Lead Brief Weekly CBIM TrainingsPlan to use one of the Training Cards at least once a week, discuss for 10–15 minutes each. Each card contains a ‘Warm Up’ script that the coach can read directly or adapt, questions to prompt discussion with the team, and some ‘Wrap Up’ comments that can also be read or adapted. Additional guidelines for adapting and delivering the program.
Training Card 1 Pre Season SpeechClarify ground rules and expectations for the sports season; introduction to CBIM (anticipate that coach will be discussing respect for women and girls throughout the sports season)
Training Card 2 Personal ResponsibilityRecognize the consequences of behaviors, how these reflect on themselves, team, and others; accept responsibility and hold themselves accountable for their actions.
Training Card 3 Insulting LanguageRecognize that language can be harmful in unexpected ways; refuse to use language that degrades women and girls; adapt language to show respect for others.
Training Card 4 Disrespectful Behavior Towards Women and GirlsRecognize how certain behavior towards women and girls can be hurtful, scary or even violating; refuse to engage in lewd or foul behavior such as catcalling, whistling, sexual innuendo, or harassment.
Training Card 5 Digital DisrespectRecognize that the actions taken online and on cellphones have real consequences; refuse to participate in abusive or controlling digital behavior including sending repeated and unwanted text messages, violating someone’s privacy, or pressuring someone for a naked, private, or embarrassing picture; create safe and respectful digital spaces for themselves and others.
Training Card 6 Understanding ConsentDiscuss and respect personal boundaries around intimate and sexual activities to create healthy,safe relationships and situation; refuse to use pressure, threats, or force in any physical or sexual encounter; actively oppose and prevent incidents of rape, sexual coercion and assault.
Halftime:Enlist Your Local Sports ReporterTell your school newspaper or local sports reporter about the CBIM program and your team’s focus on strengthening respect for women and girls; the community should know that your athletes not only perform well on the field, but are upstanding young men off the field, too; your athletes will be proud to see that the community is recognizing their commitment; this is a great way to involve the broader community and provide other opportunities for your athletes to model respectful behavior.
Training Card 7 Bragging about Sexual ReputationRecognize that bragging about sexual reputation doesn’t prove or improve one’s masculinity; telling stories about someone else’s sexual activity is disrespectful and harmful; refuse to speak disrespectfully about a girl’s sexual reputation.
Training Card 8 The Responsibility of Physical StrengthRecognize the power and responsibility of having physical strength; decide when it’s appropriate or inappropriate to use physical strength.
Training Card 9 When Aggression Crosses the LineRecognize different degrees of aggression on and off the field; identify when becoming aggressive and adjust behavior.
Training Card 10 There’s No Excuse for Relationship AbuseReject the use of pressure or intimidation in relationships and friendships; refer back to tools on how to monitor aggression and determine when it crosses the line; refuse to become violent or abusive in relationships.
Training Card 11 Modeling Respectful Behavior Towards Women and GirlsRecall the CBIM messages that coach discussed; treat others how one wants to be treated; recognize athlete role as leaders for younger athletes and friends; model respectful behavior towards women and girls; if one witnesses disrespectful or harmful behaviors, say something or speak to a safe adult.
Training Card 12 Signing the PledgeActively affirm individual and collective commitment to respect women and girls; review a team pledge: “I know that violence is neither a solution nor a sign of strength and I commit to treating everyone with honor and respect. I believe that strong men publicly and actively denounce violence against women and girls and that I can be a role model to others by making this pledge.”
Overtime Card Host a Fan Pledge DaySuggestions for increasing community awareness by having attendees at a home game sign the CBIM pledge; additional ideas for team to create posters, messages, other strategies to increase awareness about stopping violence against women.
Create a Team Campaign
Teachable Moment Card(i) Stop and identify the behavior; (ii) Call a time-out (with individual athlete or team depending on incident witnessed); (iii) Defense—correct the language and behavior explaining why
How to handle a teachable momentit was wrong; (iv) Offense—review positive alternatives with team.

Participant characteristics of athletes who completed pre- and post-surveys

% Total (n)a n = 148% Advocate-led (n)b n = 69% Coach-led (n)bn = 79
Sport
 Football60.8% (90)44.4% (40)55.6% (50)
 Basketball20.3% (30)53.3% (16)46.7% (14)
 Wrestling18.9% (28)46.4% (13)53.6% (15)
Grade**
 7th2.0% (3)0.0% (0)100% (3)
 8th2.7% (4)0.0% (0)100% (4)
 9th14.8% (22)77.3% (17)22.7% (5)
 10th31.8% (47)38.3% (18)61.7% (29)
 11th25.0% (37)43.2% (16)56.8% (21)
 12th20.3% (30)53.3% (16)46.7% (14)
Race**
 White/Caucasian27.0% (40)72.5% (29)27.5% (11)
 Black/African American54.1% (80)27.5% (22)72.5% (58)
 Hispanic/Latino0.7% (1)0.0% (0)100% (1)
 Asian0.0% (0)0.0% (0)0.0% (0)
 Pacific Islander0.7% (1)0.0% (0)100% (1)
 Native American0.7% (1)100% (1)0.0% (0)
 Multi-Racial9.5% (14)78.6% (11)21.4% (3)
 Other4.1% (6)66.7% (4)33.3% (2)
Highest parental education
 Some high school0.0% (0)0.0% (0)0.0% (0)
 High school graduate19.0% (28)35.7% (10)64.3% (18)
Some college/technical school14.2% (21)38.1% (8)61.9% (13)
 College graduate41.2% (61)52.5% (32)47.5% (29)
 Completed graduate school19.0% (28)53.6% (15)46.4% (13)
 Does not apply3.4% (5)40.0% (2)60.0% (3)

Note. *Does not add up to 100%, due to<4% of missing values.

P < 0.05;

P < 0.01.

column percent.

row percent.

Comparison of changes in outcome measures between advocate-led versus coach-led delivery

Advocate-led
Coach-led
Adjusted difference in differencesaEst. parameter (SE)b
Baseline Mean score (SD)Follow-upMean score (SD)DifferencesMean scorechange (SD)bBaselineMean score (SD)Follow-upMean score (SD)DifferencesMean score change (SD)b
Recognition of abuse3.30 (1.00)3.60 (0.76)0.30 (0.98)d*3.23 (1.00)3.33 (1.01)0.12 (,82)d–0.15 (0.15)
(n = 69)(n = 69)(n = 69)(n = 77)(n = 78)(n = 77)
Gender attitudes3.05 (0.53)3.12 (0.51)0.07 (0.37)42.88 (0.56)3.00 (0.60)0.11 (0.53 )4*0.08 (0.08)
(n = 69)(n = 68)(n = 68)(n = 79)(n = 79)(n = 79)
Intention to Intervene2.37 (1.06)2.54 (1.11)0.18 (1.15 )2.47 (1.09)2.48 (1.10)–0.02 (1.18)–0.20 (0.20)
(n = 69)(n = 69)(n = 69)(n = 78)(n = 78)(n = 77)
SummarySummarySummary scoreSummarySummarysummary score
 score score mean score mean score mean
 mean (SD) mean (SD) change (SD)c (SD) mean (SD) change (SD)c
Bystander
 intervention
 Positive0.70 (1.31)0.68 (1.47)–0.01 (1.38)0.70 (1.38)0.48 (1.40)–0.21 (1.55)–0.11 (0.25)
 intervention(n = 69)(n = 69)(n = 69)(n = 79)(n = 79)(n = 79)
 Negative1.77 (2.35)1.57 (2.12)–0.20 (1.77)2.11 (2.39)1.62 (2.45)–0.49 (2.60)–0.26 (0.39)
 Intervention(n = 69)(n = 69)(n = 69)(n = 79)(n = 79)(n = 79)
 Abuse perpetration0.16 (0.66)0.25 (0.70)0.09 (0.48)0.50 (0.95 )0.47 (0.89)–0.04 (1.11)–0.11 (0.15)
(n = 68)(n = 68)(n = 68)(n = 79)(n = 78)(n = 78)

Adjusted for grade and race/ethnicity.

Reflect the difference in changes in mean scores or in summary scores from baseline to follow up of coach led minus advocate led athletes and tested for significant differences.

Difference in baseline and follow-up scores, tested using paired t test, except otherwise noted.

Tested using Wilcoxon sign rank test to account for non-normal distribution of differences.

P < 0.05;

P < 0.01.