Introduction

Evidence-based HIV behavioral interventions typically have been developed and disseminated to meet the prevention needs of specific target populations, defined by their cultural, gender, sexual orientation and risk factors. Substantial adaptation of these interventions is an expeditious and pragmatic strategy that builds upon lessons learned and has the potential to fill gaps in prevention programming and reach underserved population segments [1]. Adaptations created by community agencies as well as by program developers are increasingly common, given the relative advantages of building upon lessons learned and avoiding the need for reinvention of proven risk-reduction strategies [2, 3]. There are, however, few reports of the execution process and whether the resulting adaptations are efficacious when used with significantly different populations. The promise of this approach for addressing the diversity of at-risk populations has been supported by evidence of the successful transfer of several interventions, such as SISTA and Be Proud! Be Responsible! which have been adapted for different age or cultural groups and shown to be effective with these new target audiences [4, 5]. However, there is little evidence of successful transfer across population groups defined by sexual orientation, despite the pressing need to identify evidence-based programs for high-priority populations, including minority gay and bisexual men [6–8].

Here we report on how VOICES/VOCES (V/V), a widely disseminated intervention originally designed for African American and Latino heterosexual adults, was adapted for use with Latino men who have sex with men (MSM). V/V is a brief single-session group intervention that focuses on sexual safety, condom use and partner negotiation. In addition to demonstrating efficacy in an initial clinical trial, V/V has been shown to be effective in reducing sexual risk behaviors when delivered in diverse settings, including substance abuse treatment centers, shelters, and clinics [9]. Community-based organizations have extended its reach to multiple populations, including youth, MSM, Haitians, Native Americans, prison re-entry programs, migrant farm workers, the deaf community, adults with spinal cord injuries, and other populations that range in size and risk profiles. Taken together, these adaptations provide evidence for the intervention model’s robustness. However, they also raise an unanswered question about the efficacy of adapted models, especially when populations served are substantially different from those included in prior studies.

The No Excuses/Sin buscar excusas study addresses this question. As part of a CDC multisite initiative aimed at identifying promising interventions for minority MSM, we adapted V/V to meet the prevention needs of English and Spanish-speaking Latino MSM, and then assessed its impact on men’s sexual risk behaviors through a randomized field trial. The intent of this report is not to suggest that resource-intense efficacy trials are needed whenever adaptations for new populations are made. Indeed, such requirements can undermine many of the efficiencies and benefits of using adapted programs rather than inventing and testing new ones. Rather, we address a gap in the prevention and technology transfer literature by documenting the execution and evaluation of a successful program adaptation and, importantly, an adaptation that crosses lines of sexual orientation to serve the prevention needs of a high-priority population.

In the U.S., gay and bisexual men of all races and ethnicities are most disproportionately affected by the HIV epidemic; they are the only risk group in which numbers of new infections have continued to increase [10]. In 2011, the estimated rate of new diagnoses of HIV infection among Latino males was 43.4 per 1,000, about three times the rate of whites [12]. Among Latino adult and adolescent males, 79 % of new infections were attributed to male-to-male sexual contract [10]. Urban residents in the Northeast and MSM aged 25–34 account for the highest numbers of newly identified infections [10]. In New York City, where this study was was based, 61 % of the 2,644 males newly diagnosed with HIV in 2011 identified as MSM; 32 % of new infections among Latino MSM [11].

Multiple structural, social and behavioral factors contribute to these figures, from poverty, recent immigration and lack of health insurance to sexual network characteristics and individual beliefs and behaviors. In a study of 21 major cities, almost 20 % of MSM were HIV-positive; of those infected, 44 % were not aware of their status. Among HIV-infected Latino MSM under 30, 63 % were unaware of their status, compared to 40 % of young white MSM [12]. Because of the high prevalence of HIV among MSM, uninfected men face a greater risk of being exposed with each sexual encounter. In addition, many do not get recommended testing, which not only reduces the effectiveness of treatments, but also results in unknowing viral transmission and inaccurate assessments of the status of partners [13–15]. Other barriers, such as stigma, underestimation of risk, beliefs that HIV is no longer a serious health threat, and the interpersonal difficulties of consistent condom use, contribute to behavioral patterns that fuel ongoing disparities [16–19].

While some of these barriers are specific to the context of risks and relationships of MSM, the V/V model provides a foundation for making content modifications that address gender, cultural, and orientation relevancy. Largely, this is due to design features, based on Rogers’ theory of diffusion, which intentionally built in simplicity, transparency, transportability, and ease of use [20]. In addition to addressing the prevention needs of intended participants, V/V addressed the constrained time, limited resources, and delivery challenges of providers. Grounded in a social-cognitive theoretical model, the V/V session lasts about 45 min and begins with a soap-opera style video that models common risks, relationships and prevention messages in culturally relevant contexts. It also serves as a non-judgmental tool for initiating a 15–20 min discussion of sexual safety geared to participants’ experiences that is followed by condom education addressing common problems and barriers to use (10 min). Each participant then chooses a selection of condoms to take home, and the facilitator ends the session with a repeated message reinforcing the importance of consistent condom use, for oneself, one’s partners, family, and community.

In the early 1990s, a randomized controlled trial demonstrated V/V is efficacious in promoting behaviors that reduce risk of exposure to HIV infection and in lowering rates of new STI in heterosexual African American and English- and Spanish-speaking heterosexual adults attending a large public STI clinic [21–23]. A subsequent study examined use in four different types of agencies and produced the empirically informed V/V dissemination package that has been nationally distributed [24]. More recently, a study in New York and Puerto Rico replicated the original efficacy findings when the intervention was delivered in non-research settings by local agencies [25]. Building on this research base, the challenge here was twofold: first, to adapt V/V for use with culturally diverse, English- and Spanish-speaking gay and bisexual Latino men and, second, to conduct a test of its efficacy in reducing risk behaviors as well as promoting HIV testing, a critical component of prevention not addressed in the original V/V intervention.

Methods

The Intervention Model: Adapting V/V into No Excuses/Sin Buscar Excusas

The adaptation was accomplished through the collaborative efforts of the original V/V development team: a film company experienced producing videos for V/V; about 70 community participants who attended a series of focus groups; and an advisory board with expertise working with Latino MSM. Advisors included 10 representatives from health departments, project leaders from community agencies, and direct service HIV prevention counselors and prevention providers; several were based outside NYC, in the southwest and Puerto Rico. Board meetings were conducted bimonthly, both in-person and through conference calls. Materials for review were sent in advance; advisors provided input on materials before they were shared with focus groups and received summaries of focus group input. Procedures followed adaptation guidance developed for CDC Replication of Effective Programs packages, which underscores the importance of getting input from members of the target population and key stakeholders (33). The behavior change logic model for V/V provided the theoretical underpinnings for the adaptation; this was updated to include the addition of an intended outcome related to HIV testing, which has become an important prevention goal. The intervention content and activities were adapted to assure cultural and linguistic relevancy, while maintaining the core elements, or critical design features, of the original intervention.

The most resource-intense development was the creation of two new videos, one in English and one in Spanish (core element 1); substantial revisions to the original videos were needed to address the cultural, gender and sexual contexts and prevention needs of Latino MSM. Drafts of story lines were first created by the development team. Paralleling the V/V development process for other populations, attention was paid to character selection (e.g., age, orientation, cultural diversity); relationships (i.e., new, casual, and steady), contexts (e.g., online hook ups), key prevention messages (i.e., condom use, safety negotiation, and HIV testing), and modeling of realistic, desired behaviors as well as peer and community support. These were shared with advisors, refined, and then used by a professional bilingual script writer to prepare English and Spanish scripts. Drafts were reviewed by advisors and then vetted in focus groups. Focus group participants were recruited at gay-friendly community events, through community-based organizations and using websites, such as Craigslist. Eligibility requirements reflected characteristics of the target population, i.e., men who self-identified as gay/bisexual and Latino and had engaged in recent risk behaviors. Ten groups were led by staff familiar with the V/V intervention and Latino MSM communities. Men were asked whether they would prefer to participate in English or Spanish; an equal number of groups, each with 4–8 participants, were held in each language.

Advisory board and focus group meetings were iterative in nature; materials were shared, revised, and presented for new input. For example, in focus groups, volunteers were asked to read and act out lines, stimulating suggestions for making storylines and prevention messages more culturally and linguistically relevant. Questions posed included: What key messages will people get from each scene? How do Latino MSM negotiate condom use with a main partner or with a non-main partner? To update and augment the original intervention, participants considered challenges and concerns related to HIV testing and the use of serostatus information in decisions about partners and sexual behaviors. Such questions included: When and why do Latino MSM get tested? Why don’t they get tested? How are these issues addressed or not addressed in the script? Input helped hone what information should be included in videos, which are designed to last up to 10 years, as opposed to other intervention materials, such as handouts and facilitator guides, which can be more cheaply and regularly updated. For example, recommendations about testing or testing procedures may change quickly, but the importance of knowing one’s status, disclosure issues, and negotiations around status are more stable and suited to video modeling. Advisors were involved in decisions about casting and invited to attend script rehearsals and filming. Video rough cuts were shared to obtain input on editing, flow, sound quality, music, and other features designed to capture and maintain viewer attention.

Once video content was set, attention turned to adaptation of the discussion guide that facilitators use to structure their interactions with participants (core element 2). When disseminated, V/V has been delivered by diverse facilitators, including health educators, counselors, nurses, and community volunteers who have a wide range of training and experience. The brevity of the structured discussion can be challenging—many detours can derail content away from skills-building and key prevention messages. Thus, lessons from past use underscore the importance of clear instruction. Toward this end, a draft guide was shared with advisors and then piloted with focus groups. Particular attention was paid to identification of specific video trigger points designed to spur discussion. For example, in one scene, Roberto, a lead character, is shown using the Internet to hook up with a potential sex partner; after arranging a date, he reaches into a drawer and gets a small bag to take with him, thus raising questions about what’s in the bag (drugs? condoms?) and their relationship to risky sex. The guide frames questions to get participants to relate to the character and identify possible negative and positive consequences of his actions. Key prevention messages and talking points are underscored—e.g., Roberto was feeling lonely when he was surfing the Internet. Questions included: What strategies does he try that might keep him safer, instead of hooking up with a casual partner? How would this work for you? What do you think about Roberto saying he isn’t worried because he is always a top? And his partners are HIV-negative? Trying out these questions in focus groups helped refine tips for supporting skills-building within the time constraints of the intervention’s 45–60 min session.

Once the discussion guide was finalized, adaptations were made to support the expansion of core element 3 (condom selection, offer of an HIV test, and distribution of locally vetted referral information). For MSM, it was important that this final component be expanded to highlight the importance of HIV testing, knowledge of one’s serostatus, and using condoms outside of a committed mutually monogamous partnership in which both partners have followed testing guidelines and accurately disclosed their current status. Revisions were made to tailor the condom exercise, updating reasons why men may not want to use a condom and modeling “comebacks” for these excuses, consistent with the video title and key prevention message: No Excuses/Sin buscar excusas. An updated condom board highlights varieties of condoms available and how features can be matched to personal needs. Advisors reviewed information on condom problems (e.g., breakage, insufficient lubrication, and slipping) and potential solutions. They provided input on the facilitator script to encourage HIV testing and status disclosure. Finally, we gathered information on marketing and recruitment, including placement of flyers to reach out to diverse population segments, such as English and Spanish speakers, recent immigrants, men who have sex with men and women, and men who do not know their HIV status or who have not previously participated in HIV prevention programs. Having completed the adaptation, the next step was to obtain evidence of acceptability and short-term efficacy.

Study Recruitment and Randomization

Over 13 months from August 2008–2009, men were recruited through multiple channels, including street outreach, listservs, and marketing through the Hispanic AIDS Forum and Callen-Lorde Community Health Center, the study sites. Screening was conducted either face-to-face or over the phone by trained research staff. Because this study was part of a multi-site CDC initiative, there were requirements to adhere to a common research protocol, which defined eligibility criteria. While each project developed and tested a different intervention, this allowed pooling of baseline data for research purposes. Eligibility criteria included self-identification as Latino, sex with two or more partners in the last 3 months, unprotected anal intercourse with a male in the last 3 months, HIV status negative or unknown, and no exposure to HIV prevention education (research or non-research) in the last 6 months. The eligible age range was broad, including men 18–49 years of age, whereas V/V has been primarily used with those under 40. Eligible participants were invited to a study site to complete baseline audio-computer assisted surveys (A-CASI) and other study activities (depending upon their random assignment). Men were able to choose one of several dates, times, and locations, as well as whether they preferred to participate in English or Spanish. They were advised to make an appointment when they could spend up to 2–2.5 h at the site. If more than 2 weeks elapsed between screening and this appointment, men were rescreened.

After completing surveys, men were assigned at random to participate in either a No Excuses/Sin buscar excusas 1-session intervention (in the language of their choice) or a non-attention control condition. Those in the control condition were offered an HIV test after they completed their surveys, but did not participate in a group activity. Intervention participants were offered an HIV test at the conclusion of the 45-min session. Men were re-contacted 2 weeks prior to their 3-month follow up date to schedule a time for completing the final survey. Of 370 men enrolled, 346 completed follow-up surveys, for a retention rate of 93.5 % (93.2 and 93.9 % among intervention and control participants, respectively). Study procedures were approved by the Institutional Review Boards of Education Development Center and CDC.

Measures

At both baseline and 3-month follow up, men were asked about the number and gender of their sexual partners and sexual risk behaviors over the past 3 months. Primary behavioral outcomes include: (1) total number of unprotected anal intercourse (UAI) acts with last with last two male partners; (2) condom use at last intercourse with a male partner, yes (1) versus no (0); and (3) self-report of an HIV test during the 3-month follow up window. Several potential mediators drawn from previous research on VOICES efficacy were also considered: attitudes about condom use (five items, e.g., using a condom turns me off, Cronbach’s alpha >.80); and self-efficacy to engage in safer sex (four items, e.g., “I can choose safer sex with a man I have sex with regularly,” Cronbach’s alpha >.80). Responses to 5-point Likert scales were summed and dichotomized into positive attitudes and high self-efficacy (1) versus other (0). Almost all surveys were completed in their entirety, with few or no (<2 %) missing values on any item; for these rare cases, mean substitution was used.

Statistical Analyses

Preliminary analyses were conducted to assess whether there were significant differences by socio-demographic and risk behaviors between the two conditions at baseline using Chi square (percentages) and one-way ANOVA (means). Outcome analyses were conducted within an intent-to-treat model. All 190 participants assigned to the intervention completed the brief, single-session group; 36 sessions were held, with a range of 4–9 men per group (mean = 5). Repeated-measures ANOVA was performed for the outcome of number of unprotected sex acts using the same measure at baseline and 3-month follow-up. This procedure examines change in the outcome over time controlling for individual and intragroup differences. Logistic regression was used to examine the relationship of the intervention to the dichotomized follow-up outcome of condom use at last sex, adjusting for the effects of covariates. Covariates considered include language preference (Spanish vs. not), sexual orientation, age (40 and under vs. over 40, based on core V/V audience), education (less than high school; high school; some college; college graduate); and HIV status (negative vs. unknown). Across conditions, 24 men were not resurveyed at follow up, yielding an analytic sample of n = 346; there were no significant baseline differences in covariates or outcome measures between those retained and those lost to follow up. Given that younger adults (under 40) were the primary target audience for the parent program, we also considered whether effects differed by age by adding a cross-product term (age by intervention) to equations. When this interaction was significant at p < 0.05, analyses were repeated within age subgroups. Analyses were conducted using PASW Statistics 18.

Results

Table 1 presents baseline characteristics of the sample by condition. The average age of participants was 36.6; 56.7 % were 40 years of age or younger. Just over half (55.6 %) were born in the U.S. mainland, 42.2 % had a Spanish language preference. About a third of the men had never been tested for HIV; 24.6 % had been tested with the previous 12 months. Although all men reported they were HIV−/status unknown to screen into the study, seven disclosed they were HIV+ on the computer survey. Overall, 30 % self-identified as bisexual and 64 % as gay (same gender loving/homosexual). Gay identity was the only significant baseline difference across study arms; at p < 0.05, men in the intervention group were more likely to identify themselves as gay (70 %) or bisexual (26.8 %), versus those in the control condition (57.5 % gay and 34.1 % bisexual). In terms of sexual behaviors, participants reported having an average of 5.4 male partners and 6.5 total partners (male, female, and transgender) in the previous 3 months; 57.9 % reported having a main partner. They had engaged in an average of 7.4 acts of unprotected anal intercourse with their last two partners; 60.8 % reported not using condom use at last sex with a male partner). Only 65.4 % had ever been tested for HIV. There were no significant differences in baseline reports of sexual behaviors by treatment condition (all p > 0.20).

Table 1 Characteristics of 370 men enrolled in the No Excuses/Sin buscar excusas study in New York City by condition
Full size table

As shown on Table 2, at baseline, 71.5 % reported three or more total sex partners, and 60 % had three or more male sex partners. Potentially influencing power to detect differences by treatment assignment, men in both conditions reported significantly lower levels of risk at follow up. There was an overall reduction in the number of male partners with whom the respondent had engaged in UAI from 2.8 at baseline to 1.2 at follow up, and a decrease in the mean number of total sex partners to from 6.5 to 3.6. Similarly, the proportion of men who reported three or more partners decreased from 71.5 to 42.7 %.

Table 2 Reports of recent sexual behaviors at baseline and 3-month follow up among men enrolled in the No Excuses/Sin buscar excusas study, New York City
Full size table

Table 3 presents results of repeated measures ANOVA for the outcome of number of UAI acts (F = 4.10, p < .05). From baseline to follow-up, there is a sharper drop in mean number of UAI acts in the intervention group, with a decrease of 59 % over the two time points (from an average of 8.4 unprotected acts to 3.8), compared with a 39 % change among controls (6.2–3.8). These analyses adjust for covariates, none of which reach the level of significance. Table 4 provides results the logistic regression examining the dichotomized outcome of condom use at last sex. Men exposed to the intervention were more likely to report this protective behavior (AOR = 1.69; 95 % CI 1.02–2.81, p < .05).

Table 3 Change by condition in number of recent unprotected anal intercourse acts with last two partners from baseline to follow up (analytic sample n = 346)
Full size table
Table 4 Logistic regression of treatment condition on men’s reports of condom use at last sex, adjusted for baseline covariates (n = 346)
Full size table

We then considered whether these intervention effects differed for older and younger men by adding an age by intervention interaction term to the analytic models; this term was significant at p < 0.05 for condom use at last sex, no condom use at last two partners, and number of UAI acts. For each outcome, there is a significant or trend-level treatment effect for men 40 and under, but not for older men. Restricting analyses to the subsample of younger men (n = 208), 42.4 % of intervention participants, compared to 31.7 % of controls, said they used a condom the last time they had sex (p < 0.05). In terms of risk behavior, controls were more than twice as likely to report they had not used condoms with their last two partners (27.1 % versus 14.8 %, p < 0.05) and, at a trend level, they engaged in a higher number of UAI acts (AOR = 1.24; 95 % CI 0.75–2.06, p < 0.10). Younger men receiving the intervention also reported more positive condom attitudes than their control counterparts (AOR 2.07; 95 % CI 1.02–4.22, p < 0.05) and at p < 0.10, higher condom self-efficacy (AOR = 1.85, 95 % CI 0.92–3.60, p < 0.10). We did not find significant treatment effects for men 40 and over.

Overall, 25.4 % of men got tested for HIV at the time of their enrollment. There was no difference in test acceptance by condition at that time. At follow up, 88 men reported they got an HIV test post-study, including 50 men who had not been tested prior to participation. Of the 88 men who got a post-study test, 37.5 % were intervention participants and 28.2 % were controls. While this difference was not significant for the whole sample, it was among men 40 and under; 46.2 % of younger men who got tested were intervention participants, compared to 20 % of controls (p < 0.05).

Discussion

Despite significant biomedical advances in HIV prevention and treatment, behavioral prevention for high-priority populations will remain a critical element in the HIV/AIDS armamentarium. Here we provide evidence that a brief intervention designed for heterosexuals is sufficiently robust to be adapted for MSM, as well as a description of the steps that were taken to assure that the content would meet the prevention needs of this different audience. While only short-term effects were examined, compared to controls, men who participated in No Excuses/Sin buscar excusas reported significantly fewer unprotected sex acts and were more likely to use condoms at their last sex encounter. These findings are similar to evidence from prior V/V studies that also have demonstrated increased condom use and reductions in unprotected sex, and provide evidence that the adaptation to a different target population was successful. The question of whether this impact is sustained or results in reduced infections goes beyond the scope of this study, although in prior studies, V/V exposure has been related to decreases in the incidence of sexually transmitted infections over 12–18 months [21, 25].

Because of its relative ease of implementation, the V/V model has been used by many agencies, often with adaptations to extend to other audiences. This study provides support for the acceptability and efficacy of such adaptations, with the cautionary note that careful targeting and tailoring are needed. Notably, our adapted intervention worked better for men under the age of 40, which parallels the age group for whom the intervention was initially developed. Older men are at significantly lower risk of acquiring new infections, but there are also few interventions that serve their issues, needs and concerns. Our findings suggest that changing behaviors among older men may require additional community-informed adaptations, but we believe this tailoring to a subpopulation can accomplished without the need for an additional costly efficacy study. Indeed, evidence that V/V can be successfully transferred for populations that differ in sexual orientation provides support for reducing the research burden on other adaptations and, especially, the resource-intense requirements of a randomized trial. There is a clear tradeoff between the costs of requiring such “gold standard” evidence for adaptations, on the one hand, and the benefits of moving efficiently to deliver program adaptations to populations, such as Latino MSM, that continue to be underserved. By showing that an adaptation can be successful across boundaries defined by sexual orientation, our findings support lightening the “burden of proof” for carefully-developed adaptations. The efficacy trial component of our study required a six-figure cash investment over two years that was more than double the first year costs incurred for developing the adapted intervention materials. Beyond the monetary costs, however, the time required to complete such a study delays the delivery of culturally-tailored prevention programs and focuses attention on repeated evaluations of efficacy at the cost of greater attention to implementation and reaching those at greatest risk.

A brief video-based intervention, designed with the needs of both the intervention deliverers (i.e., community agencies and health services) and the end users (Latino MSM) has the potential to overcome many of the barriers identified through previous dissemination efforts. These include limited resources, insufficient personnel to provide intensive interventions or one-on-one prevention counseling, high staff turnover and inadequate training in prevention, cultural and linguistic barriers, inattention to the different needs of diverse audiences, and the lack of confidence among agency staff in the effectiveness of behavioral interventions [26]. Additionally, there is a pool of prevention providers who are familiar with V/V and could transfer their experiences to using No Excuses/Sin buscar excusas. Thus, this adaptation can be readily incorporated into ongoing prevention programming, consistent with the need for high-impact prevention that reaches high-priority populations, such as Latino MSM, with proven intervention approaches. The intervention might also be useful for promoting HIV testing as well as reducing sexual risk behaviors.

Several methodological issues merit discussion. First, in both conditions, there was a marked decrease in all risk behaviors from baseline to follow-up. These overall decreases may have limited power to detect treatment differences. While the sample of Latino MSM was relatively large for an initial efficacy trial and included a broad age range of participants, there was limited power to conduct subgroup analyses. The notable drop in reported risk behaviors over time also should be considered in calculations of statistical power for future trials that recruit high-risk participants. Second, we considered the applicability of adjusting for intra-class correlations (ICC), which have been shown to introduce bias in group-randomized trials [27]. However, this study has a number of features that set it apart from standard group-randomized experiments. Participants were recruited individually and randomly assigned either to the single-session intervention or to a non-group control, a design for which modeling the influence of ICCs is still limited. Further, unlike classroom-based interventions, multi-session groups, or studies of patients clustered in provider practices, the session itself provides only a short time for group engagement, and there were a relatively large number of groups with a small numbers of participants, factors that mitigate ICC bias [28].

Broader lessons from this study point to the limitations of even brief, one-session face-to-face group interventions for reaching underserved priority populations. First, it was relatively difficult to recruit men for the study. Research requirements added substantially to participant burden, more than doubling the time for a V/V session alone, and requiring a return visit to take a follow up survey. Although the protocol did not require participants to provide social security numbers or other verification, it is possible that undocumented Latinos would be wary about participating in a research study that required personal contact information for scheduling return visits. We do not know whether recruitment for the intervention on its own would be difficult, but in our work with agencies implementing V/V and other group interventions, we have witnessed how convening groups can be problematic, although addressable through staff training and technical assistance. Second, a substantial minority of men—18 %—had participated in a prior research or prevention program within the last year. Reaching out beyond the “choir” was hard not only at our study site, but also at other sites funded under the same initiative. This underscores the importance of greater attention to outreach and implementation issues. Online and mobile applications may provide a way to deliver the intervention directly to men without requiring special visits to local agencies; they may also be useful as tools for providing tailored, reinforcing prevention messages over time. In light of biomedical advances, next generation adaptations of the parent intervention, V/V and No Excuses/Sin buscar excusas should consider greater integration of behavioral messages with strategies for linking participants to medical care, for prophylactic medication as well as timely treatment of new infections. Thus, at the same time it is desirable to provide agencies with an evidence-based intervention for MSM, it also is imperative to explore and expand complementary strategies that may help achieve the critical population saturation needed to promote normative and sustained behavior changes that reduce the spread of infections.