A male promotores network for Latinos: Process evaluation from a community-based participatory project

INTRODUCTION

This article presents a process evaluation of “De la Mano con la Salud” (Lend a hand for Health), a male lay health advisor (LHA) network in an emerging Latino community, where Latinos are a small yet rapidly growing segment of the population (Cunningham, 2006). In these locales, Latinos are frequently misinformed and few services are culturally tailored to them (Cunningham, 2006; Documet, et al., 2013). This adds to the scarce literature on process evaluation of male LHA.

BACKGROUND

LHA, also frequently referred to as community health workers or promotores when working with Spanish speakers, are trusted community members trained to provide information and support to their peers (Fisher, et al., 2014; Rhodes, Foley, Zometa, & Bloom, 2007). LHA, or promotores, have a long history of delivering health promotion and disease prevention both in Latin America and among Latinos in the United States (Ramos, Hernandez, Ferreira-Pinto, Ortiz, & Somerville, 2006) to address health concerns ranging from cancer control to nutrition (Ayala, Vaz, Earp, Elder, & Cherrington, 2010; Rhodes, et al., 2007). Despite being ideal for emerging communities, few LHA interventions in these areas have been described in the literature.

Latino immigrant men have significant health needs: they face negative public opinion (Chavez, 2008; Hainmueller & Hiscox, 2010), are at heightened risk of injury, alcohol abuse, depression, and sexually transmitted infections (Sanchez, et al., 2012; Vasquez, Gonzalez-Guarda, & De Santis, 2011; Walter, Bourgois, & Loinaz, 2004; Weine & Kashuba, 2012). Furthermore, the immigration experience exposes them to loneliness and anxiety (Caplan, et al., 2013; Walter, et al., 2004). Individuals who are poorer, less educated and lacking health insurance are particularly vulnerable. Mexicans and Central Americans are overrepresented in this group (Hacker, et al., 2011; Motel & Patten, 2012).

LHA interventions are based on the ability of peers to build rapport and thus effectively deliver social support in a culturally appropriate manner (Fisher, et al., 2014). While there have been many successful interventions with Latino female LHA, few have focused on male LHA working with male participants (Ayala, et al., 2010; Rhodes, et al., 2007). Activating interpersonal networks may be useful for Latino immigrant men (McQuiston & Flaskerud, 2003; Rhodes, et al., 2006; Shumaker & Hill, 1991; Zimmerman, Ramirez-Valles, Suarez, de la Rosa, & Castro, 1997), yet few LHA programs focus specifically on them (Rhodes, Hergenrather, Bloom, Leichliter, & Montaño, 2009). Furthermore, qualitative research suggests that Latino men are not inclined to serve as LHA (Ayala, et al., 2010; Villa-Torres, Fleming, & Barrington, 2014). There are a few exceptions worth noting. In Connecticut, LHA organized groups of men who had sex with men for training on sexual identity, AIDS prevention and referrals (Singer & Marxuach-Rodriguez, 1996). In North Carolina, LHA conducted volunteer outreach through soccer teams, providing condoms and HIV prevention education (Rhodes, et al., 2009). Another intervention trained men to facilitate participatory, non-directive groups with farm workers around intimate partner violence in four states (Nelson, et al., 2010). Finally, in Florida LHA encouraged eyewear protection use among citrus migrant workers (Monaghan, et al., 2011).

Our project aimed at connecting Latino men to services and the Latino community. It was developed using a community-based participatory approach (Rhodes, et al., 2007) in which the community was an integral part of activities and decision-making, including topic selection, project design, data collection, analysis and dissemination of results (Israel, et al., 2003). We conducted a process evaluation with a formative focus; such evaluations are scant in the male LHA literature. Process evaluation provides useful feedback to replicate or disseminate a program and reveals patterns of modifications in response to contextual factors (Patton, 1997; Saunders, Evans, & Joshi, 2005).

We focus on three questions crucial for the implementation of future LHA interventions with men in emerging Latino communities: Is it feasible 1) to maintain the interest of a diverse coalition in guiding the project?; 2) to hire, train and retain male LHA that match the demographics of the local Latino population?; and 3) for male LHA to recruit and assist underserved participants?

METHODS

Setting and Community Participation

The 19,070 Latinos in Allegheny County’s Latino represented less than 2% of the total population in 2011 yet Latinos had increased 71% from 2000 to 2010 (U.S. Census Bureau, 2010). Therefore it fits in the definition of emerging Latino community, where Latinos have difficulty finding services that are tailored to their culture and language (Cunningham, 2006) Slightly over half of Latinos in Allegheny county were male, and Mexicans represented the largest group (36%) (U.S. Census Bureau, 2010).

This project was the result of the efforts of the Latino Engagement Group for Salud (LEGS), a coalition of community members, researchers and health and social service providers. In 2010, the coalition conducted a participatory health assessment in Allegheny County, focusing on Latino immigrant men. The majority of participants had not completed high school and did not speak English. They expressed feeling isolated and lonely, and said they often reacted to this situation by working overtime and drinking alcohol (Documet, et al., 2013). As a response, the LEGS coalition designed and piloted De la Mano con la Salud to enhance participants’ social support, community connections, and healthcare access. The University of Pittsburgh IRB approved the study.

The principal investigator (PI) and the project coordinator comprised the management team. Project staff also included promotores (LEGS’s name of choice for LHA), trainers, and research assistants. Coalition members and the management team held LEGS meetings every two months. An external evaluator was hired to ensure that the project was participatory.

Originally Planned Project

The LEGS coalition decided to have a bilingual lead promotor, called super-promotor, to enable monolingual promotores to be part of the project and have a community member, rather than a researcher, as the promotores’ direct supervisor. The originally planned project is depicted in the logic model (Figure 1). A member of the LEGS coalition was selected a priori to be the super-promotor. He and ten other Latino immigrant men would be trained. The top five would be selected to become promotores. The other five would remain as back-ups, available to substitute promotores in case any stepped down.

Figure 1.

Logic Model for De la Mano con la Salud

Promotores would recruit participants from their own networks and at places where Latinos congregated. At initial face-to-face appointments promotores would use the Wheel of Life¹, a tool frequently used by life coaches and proposed by LEGS partners to elicit participants’ priorities. Promotores would listen to participants’ problems and ideas and together develop a plan of action. Then, according to the plan of action, promotores, with the support of LEGS partner organizations, would connect them with health and social services and offer connections to the community (e.g., libraries, ESL classes, churches with services in Spanish and exercise opportunities). Promotores would periodically follow-up on participants’ well-being and plan of action progress in person or via mobile phone. Often, this led to additional assistance for new needs. Promotores were to meet weekly with the management team to discuss their work and emerging difficulties, and to file project paperwork. Project staff would administer baseline and follow-up questionnaires. Finally, overarching issues and challenges were to be discussed in LEGS bimonthly meetings to receive feedback and suggestions for change if necessary.

Data and analysis

The data for this process evaluation included program operations documents, and also managers’ and employees’ narrative accounts (Saunders, et al., 2005). Specifically, we used the following:

• Meeting notes. LEGS agenda and minutes from 2008–2013, and notes taken by the PI during LEGS and promotores’ meetings.

• Management team journal. Details on staff hiring, contracts, trainings and other project matters.

• Project and coalition paperwork. LEGS memorandums of agreement (MOA), personnel recruitment forms, meetings’ sign-in sheets, curriculum materials, e-mails and text messages between team members, training logs, and project calendar.

• Project data. Participants’ demographics characteristics (produced in SPSS Version 21 from baseline questionnaires).

• Debriefings. Each promotor was debriefed by the PI or the project coordinator twice (six months after participant started recruitment and at the end of the project).

• External evaluator reports. Sent email feedback six times per year, plus two annual reports.

A logic model is a frequent starting point for process evaluation, because it lists the planned activities and their expected outcomes (Saunders, et al., 2005). The logic model in this study is based on the premise that peers who share cultural background with participants have a unique ability to deliver social support, especially in a population such as Latino men, who often face discrimination, social isolation and a hard economic situation (Fisher, et al., 2014). We compared the actual progress of the program to the evaluation plan (Table 1) and logic model, both outlined before the project started. For textual data we used qualitative analysis. Three investigators read all data sources in their original language (English or Spanish); each wrote summaries organized under our three focus questions. They also identified emerging challenges and subsequent solutions implemented and other factors that were repeatedly addressed in the documents. The investigators produced a joint account of the results that was shared with LEGS members to ensure it accurately captured the project’s history.

Table 1.

Process Evaluation De la Mano con la Salud

Activity	Associated Questions	Outputs	Data Sources
Maintaining a diverse steering coalition
LEGS membership	How many people signed (by role: community, researchers, organizations)	12 MOAs signed	MOAs
LEGS meetings	How many people attended on average? How many from the community? Were meetings participatory? What topics were touched?	Participatory meetings conducted every two months	Sign-in sheets PI notebooks Evaluator feedback	Journal Minutes
LEGS publicized	When and where was LEGS announced/publicized?	Work announced through: 2 media sources 3 email distribution lists	Journal Email (Management)
Hiring, training and retaining male LHA
Curriculum preparation	What topics were included? Who worked on the curriculum?	Curriculum with trainers responsible for each topic	Evaluator feedback Project coordinator debriefing	Curriculum Journal
Trainees hiring	How many men applied? Did their demographic characteristics match the priority population? Were they hired on time? Were there difficulties in hiring them?	20 men apply for job Hire 10 men to be trained as promotores	Personnel recruitment form	Journal
Promotores training	Were sessions as planned? Who conducted the training? Did trainees attend all session? If not, how was this dealt with? Was training useful to promotores?	5 4-hour training sessions conducted 10 promotores and 1 super-promotor trained	Evaluator feedback Project calendar	Journal Promotores debriefing
Select and hire promotores	How were trainees evaluated? Was it a “good” exam? Was the process fair?	5 promotores selected through exam 5 promotores and 1 super-promotor hired	Personnel recruitment form Evaluator feedback	Journal Promotores debriefing
Monthly trainings	What was the number and length of monthly trainings? What topics and what presenters? How were topics selected? Were they useful to promotores?	16 1-hour monthly sessions conducted	Monthly trainings log Evaluator feedback	Journal Promotores debriefing
Recruiting and assisting underserved participants
Enrollment	What venues were used? How many men were reached and enrolled? How long did it take? What were the demographic and baseline characteristics?	Over 500 LIM reached 125 clients enrolled (25 per promotor)	Promotores debriefing Demographic data	Journal
Face-to-face meetings	Did promotores and participants meet face-to-face? How often? Were the elicitation tools completed? Were the elicitation tools useful to promotores? To clients? Was the meeting useful to promotores? To clients?	125 face-to-face meetings and elicitation tools forms completed	In-depth interviews Promotores debriefing	Journal Wheel of life
Follow-up	How often did promotores follow-up on participants? Phone? Face-to-face? Was the follow-up useful to promotores? To clients?	125 calls each month 25 meetings each month	-Promotores debriefing	In-depth interviews
Promotores meetings	Were regular promotores meetings held? What was the attendance? What topics were covered? Was meeting useful to promotores?	32 promotores meetings held (originally 64)	Project calendar Text messages PI notebooks	Journal Promotores debriefing

RESULTS

Maintaining the interest of a diverse coalition in guiding the project

The LEGS coalition met every two months during project implementation, with attendance ranging from 7 to 19 participants. LEGS was publicized in a local Spanish language newspaper (four times), through email distribution lists (three times) and in community events (three times). LEGS members consisted of community members, university researchers, and service providers, including a Federally Qualified Health Center and a social service and advocacy organization. Twenty members (13 community members, 4 researchers and 3 service providers, with some having a dual position) committed to attend at least three of six yearly meetings and signed a Memorandum of Agreement (MOA). Members provided qualitative input on project planning and in decision-making.

While defining promotores’ responsibilities, the coalition determined that it would be inappropriate to require promotores, who were University employees, to volunteer their time to LEGS. Thus, participation in LEGS meetings was mandatory for the super-promotor and voluntary for other promotores. In spite of this, all promotores chose to become LEGS members and participated in decision-making.

Meetings included project updates and discussion regarding project strategy and funding. The coalition made programmatic decisions central to the project before and during implementation. For example, the LEGS designed the recruitment plan, required promotores only to be literate, and decided to support a soccer tournament.

Hiring, training and retaining male LHA

We advertised the 10-hours per week temporary promotor position through the University of Pittsburgh employment website, Latino churches, grocery stores, email listserves, and newspapers, and through LEGS members’ personal contacts by word of mouth, posters, Facebook and emails. The LEGS coalition expected applicants to be literate, native Spanish-speakers, Allegheny County residents for at least six months, good communicators, knowledgeable about and trusted by the local Latino community, and experienced in helping others.

Fourteen men applied for the position; two were LEGS coalition members. All were native Spanish speakers. Three applicants required help from the project coordinator in completing the computerized application. Several men who expressed interest did not complete an application; some cited lack of time or work authorization; others gave no reason. Four applicants had less than college education; ten applicants had college education or more. After consultation with LEGS members, eleven applicants, including the super-promotor, were selected for training according to the timeline (Table 2).

Table 2.

Applicants Selected to Receive Training as Promotores

Trainee	Country of Origin	Highest Education	Spanish/English Proficiency	Initially Selected as Promotor	Months as Promotor
1a	Guatemala	Less than high school	Yes/Yes	Yes	17
2	Mexico	Associate’s degree	Yes/Yes	Yes	17
3	Colombia	Graduate degree	Yes/Yes	Yes	17
4	Mexico	Less than high school	Yes/No	No	16
5	Guatemala	Bachelor’s degree	Yes/Yes	No	16
6	Puerto Rico	Graduate degree	Yes/Yes	No	16
7	Venezuela	Bachelor’s degree	Yes/Yes	Yes	13
8b	Mexico	High School	Yes/Yes	Yes	3
9	Panama	Graduate degree	Yes/Yes	Yes	3
10	Ecuador	Bachelor’s degree	Yes/Yes	No	0
11	Venezuela	Bachelor’s degree	Yes/Yes	No	0

Note. Bolded applicants represent the final group of seven promotores who worked with participants for over a year.

^aSelected super-promotor due to his success as promotor

^bA-priori selected super-promotor

Two trainers with experience in life coaching developed the curriculum and met weekly with the management team to refine the materials and schedule. The resulting 20-hour promotores training in Spanish was offered on five Saturdays by the trainers and project managers, and included the topics chosen by the LEGS coalition: (1) workplace safety; (2) U.S. health system; (3) case management; (4) emotional health and recreation; and (5) values, nutrition and diabetes. Overarching topics included ethics, available resources and communication. Three trainees missed sessions, and two watched the filmed session as a make-up. At the request of promotores and trainers for more practical training, an extra session was organized to role-play first contact with a participant.

LEGS coalition representatives observed a role-play exercise and rated trainees on their performance on the intake, consent process, communication skills, and referrals. They selected the five highest-scoring trainees as promotores; the remaining became back-ups. Back-ups expressed great disappointment despite being invited to all meetings and subsequent trainings and being paid for attendance. Therefore, we paid back-ups for outreach activities. Because some of them were extremely successful, the management team asked all trained men to enroll participants, paying hours according to performance. Eventually all but two back-ups who resigned early-on became promotores. One of the original promotores recruited only two participants and left the program, citing lack of time (see Table 2).

The original plan regarding the super-promotor, who had been selected a priori and therefore not evaluated, also changed. The LEGS coalition, in consultation with the external evaluator decided to replace him after two months, when promotores reported that his leadership proved inadequate even after extra coaching. One of the promotores who had demonstrated outstanding recruitment and leadership skills was named super-promotor. Talking about his leadership style he said, “I understand the promotores a little more because I did the same job and I’m still doing it. […] I think the basis for communicating with people is respect and humility on my part; and of course, sincerity.” Promotor 1

During the intervention promotores attended 11 additional one-hour trainings. Originally intended as refresher trainings, at promotores’ request these morphed into sessions on topics they collectively deemed necessary: immigration, housing, sexually transmitted infections, U.S. health system, stress management when working with participants, dental health and relationships with the police.

The evaluator wrote, “the promotores are clearly engaged beyond the training and with the [Latino immigrant men]. The proof of this is their request for more training in areas of interest that the [participants] have mentioned.” Promotores considered that their training was ongoing: “This has not stopped, the training continues.” (Promotor 5) Promotor 1 said that in the initial training he learned, “to listen. Before, one interrupted a conversation or gave answers to the person before he finished saying what he had to say. Then one […] did not know how to help the other person. And of course, confidentiality, learning to keep the secret of the private life of another person. In truth, I did not think about it, that it could be harmful to someone. But it’s true.” The subsequent trainings were also useful.

Recruiting and assisting underserved participants

Success in recruitment was heterogeneous. As Trainee 11 explained,

During the first stretch of one or two months […] it became clear that there were some promotores who were very successful. They had fifteen or twenty [participants], and some promotores had three or four. So… the general comment was that they didn’t have the skills to hook [participants] directly. That’s when [Promotor 1] said, well, I do it this way. And the other who was quite successful back then was [Promotor 8]. And the other who was very successful was [Promotor 5]. So they started to offer themselves as mentors for the others.”

According to informal reports from promotores, the barrier was the upper class appearance and demeanor of some promotores. All promotores were successful in retaining participants. Seven promotores maintained 26 participants on average (range 19 to 40) for over one year.

Some promotores reached out to their acquaintances (“we are from the same town […], we know each other from childhood” Promotor 4; others had to build trust with strangers, about which Trainee 11 said “the practical side of how to establish a contact, how to hook a [participant] I think that has to be [learned] on the street.”

By the fourth month, promotores had reached almost half of the enrollment planned over 10 months, so we changed the recruitment goal from 125 to 200 to enable more robust analysis with increased statistical power. Overall, promotores approached approximately 500 Latino immigrant men; 182 consented to the project.

At initial in-person interviews with participants, promotores administered consent and used the elicitation tool, which they considered helpful for building rapport with participants. Approximately two thirds of project participants required assistance on a variety of health issues such as referrals to medical, dental and mental health services; transportation to appointments; and interpretation. Promotores worked closely with LEGS partners and other organizations that provided health services and assistance navigating the health system.

Promotores embraced a holistic understanding of health. As expressed by Promotor 1: “What I like most about the program is that it goes beyond health.” As a result, support often included assistance finding jobs and English lessons, dealing with housing problems, immigration issues, legal processes, and handling bills. For this, promotores tapped non-health organizations in the areas of housing, assistance to crime victims, general social services, literacy, as well as immigration and criminal law. Additionally, promotores devoted time to unpaid endeavors that supported Latino men such as organizing a soccer tournament, a picnic, and the County’s first Spanish Alcoholics Anonymous chapter.

When dealing with situations where they had little or no control, like cases requiring an unaffordable medical solution or obtaining an immigrant visa, promotores expressed frustration, as exemplified by Promotor 7, “Sometimes people have expectations that we cannot meet. […]. The hardest part is telling people that you cannot help them.” Promotores also discussed difficulties setting limits to their involvement with participants. Common points of tension were participants’ expectations of promotores’ immediate attention and constant availability, as Promotor 1 explained, “Sometimes [participants] want everything done for them, now! They don’t want to put anything from their part. […] [S]ometimes it is hard to know about a limit, how far you can go to help a person.”

Collecting data presented additional challenges. The high speed of recruitment and project staff’s lack of an established relationship with participants hindered data collection as planned. After debating whether it represented a conflict of interest and consulting with the IRB the management team and the promotores decided that promotores would collect all data thereafter, which was done successfully for follow-up questionnaires. Baseline data were already affected and exist only for 145 participants. Demographic data of participants are presented in Table 3. The use of data from the intake forms was problematic, since the LEGS coalition originally decided that these documents were not to be made available to researchers for confidentiality. Later, LEGS and promotores decided, with IRB approval, that it was appropriate to use these documents for research. The management team was unable to collect all of these forms from promotores.

Table 3.

Demographic Characteristics of Project Participants (n=145) ^a

Characteristic	Frequency	%
Place of origin
Mexico	88	60.7
Central America	48	33.1
South America	3	2.1
Other	3	2.1
Missing	3	2.0
Age
18–25	27	18.6
26–35	52	35.9
36–45	37	25.5
Over 45	21	14.5
Missing	8	5.5
Years in the U.S.
Less than 5	61	42.1
More than 5	79	54.5
Missing	5	3.4
Education
No high-school completed	63	43.5
High-school or GED	59	40.7
Beyond high school	18	12.4
Missing	5	3.4
Living situation
Spouse or partner	32	22.0
Another family	23	15.9
One or more Latino immigrant men	69	47.5
Alone	13	9.0
Other	4	2.8
Missing	4	2.8

^aBaseline data was only collected for 145 men, despite 182 having been consented.

Promotores followed up with participants on a need basis. They called or met with participants who required intensive assistance more often than prescribed by the original plan. Conversely, they contacted participants who had “simple” problems (e.g., dental care) less often. Interactions often occurred at unexpected times, as need arose. Each promotor had his own follow-up and reporting strategy; some promotores kept a journal, others kept lists of activities on paper or in mobile phones, others resorted to memory. Promotores reported confidence in their follow-up systems, but we could not gather accurate information on the number of phone and face-to-face meetings that promotores had with each participant.

Communication among promotores and between promotores and the management team remained ongoing through phone conversations and meetings. These meetings, originally every Sunday to accommodate schedules, became biweekly following promotores’ request. Everybody attended them regularly; promotores reported absences ahead of time to the super-promotor. Promotor 5 said,

The meetings […] give us the opportunity to bring to the table what has happened to us during the two weeks […]. It provides an opportunity to ask questions about certain aspects, for instance if we are dealing with some problems with a [participant], […] listen to the reaction and opinion of the other promotores. They provide you a bigger help and you know you are not on your own in this regard. There is a group of people helping you.

The ability to work cooperatively that emerged during the training matured in these meetings, especially after the replacement of the original super-promotor.

DISCUSSION

While project implementation was conducted according to timeline, some of the original plans were altered to adapt to the local context. Taking into account contextual factors is an important part of the iterative nature of process evaluation (Saunders, et al., 2005; Wholey, Hatry, & Newcomer, 2004), which allows correcting when necessary. We were able, with the support of the LEGS coalition, to successfully change the original plan as challenges developed. We also identified modifications for future implementations.

Hiring, training and retaining male LHA were successful despite Latino men reported reluctance to become promotores, because they see this unpaid or low-paid position as more appropriate for women (Villa-Torres, et al., 2014). Furthermore, promotores, became LEGS members and volunteered their time in other community activities. Advertising in community venues and using word of mouth aided in reaching applicants for the promotor position. Assisting men in submitting their applications enabled us to hire men who had not finished high school or were Spanish monolingual. Our success in recruiting male promotores was tied to the LEGS coalition’s guidance, which included deciding to pay promotores a competitive wage. These successes resulted in half the promotores reflecting the population of interest.

We added four practice hours to address the concern of insufficient “hands-on” time. Future trainings need to build in practical time on outreach, communication, and paperwork. Selecting only some trainees created tension, which disappeared once we distributed paid hours according to performance in the field. This introduced flexibility, helped maintain team spirit, encouraged participation from all members and allowed planning for potential dropout as suggested by Earp et al. (1997). Likewise, selecting a super-promotor before the project started was premature. Being open to changes in leadership enabled the management team to appoint a natural leader as super-promotor, whose guidance facilitated the promotores’ sense of project ownership. Discussing challenges, giving and receiving support from each other during biweekly meetings, and collectively defining training themes also strengthened team spirit. The management team revised decisions periodically with promotores’ input. Regular meetings and addressing LHA’s concerns have been documented elsewhere (Kim, Flaskerud, Koniak-Griffin, & Dixon, 2005; Nelson, et al., 2010).

The significant number of participants having a high-school degree or less demonstrates our achievement reaching an underserved population. Some promotores’ challenges during outreach and enrollment were similar to those reported in a LHA intervention with Latino gay men (Singer & Marxuach-Rodriguez, 1996). Our community-savvy promotores offered a mentoring plan so that struggling promotores learned how to engage potential participants. This organically grown solution was unanimously accepted. After recruitment, all promotores reported success building trust.

As in other male LHA interventions (Singer & Marxuach-Rodriguez, 1996), participants expected to receive broader services than those within the project’s original scope, reflecting a holistic understanding of health. Enforcing boundaries was a challenging aspect of the promotores’ job. Fostering participants’ independence became a priority for them in the final months.

We switched data collection from the management team to promotores, who were in a better position to collect data because participants trusted them. We were unable to record exactly the number of promotor-participant contacts. A uniform strategy to document follow-up that includes collecting elicitation materials will help evaluate the dose of future interventions.

CONCLUSIONS

Lessons learned from this project can help other groups replicate our successes and avoid the mistakes we made. Broad participation enabled us to repeatedly adjust to emerging challenges and the local context. These adjustments and offering payment led to successful recruitment of promotores. Team building required frequent contact among promotores and the management team, fostering egalitarian participation, peer learning, and incorporation of the skills and needs of all involved. Hence, participation, flexibility and team building should be included in future LHA interventions. We also learned that the selection of promotores and distribution of hours should be made based on performance in the field and that promotores, as trusted peers, are in an ideal position to collect sensitive data. Based on our process evaluation, we suggest the following recommendations for CBPR interventions:

• Tailor promotores’ recruitment strategies to the local reality

• Cultivate a climate of horizontal communication, where everybody’s point of view has equal value.

• Assess promotes’ skills in the field. Role plays are not enough to gauge potential for success.

• Design the project with formal and informal opportunities for team building, including work-related and social interaction among LHA and project managers.

• To the extent possible, plan for promotores to also collect evaluation and research data.

Further research needs to formally test the effectiveness of male-to-male LHA interventions in emerging Latino communities.