This study was conducted in Ibadan, the capital of Oyo State in Southwestern Nigeria. Oyo State has a high adult male literacy rate (80.5%) and low male unemployment rate (3.7%), with 59.4% of households classified as poor. In general, ~82.7% of urban residents in the state have access to satisfactory medical services.
The success of the informal ARFH educational program at reaching and retaining men led us to attempt to preserve the strength of the Nigerian program, while incorporating elements from evidence-based interventions that may have relevance for the Nigerian setting.
The curriculum was developed using a participatory process involving US and Nigerian researchers, ARFH nursing, education and research staff, men who had participated in the prior psychoeducational program and young men who were representatives from an advisory ARFH youth group. Senior investigators conducted two intensive 3-day interactive Intervention Development Workshops that introduced all attendees to manualized interventions and provided participants with hands-on experience in the development of group-delivered intervention activities. During the workshop, attendees were assigned to three- to five-person ‘Work Groups’ of ARFH staff and adult and teen men from the target communities, supervised by the senior investigators. Prior to convening the Intervention Development Workshop, the Nigerian and US investigators reviewed intervention manuals of evidence-based programs to identify pre-existing intervention modules that could be tailored to the local Nigerian context. These modules, coupled with elements of the Nigerian educational program, were used by the Work Groups as the basis for developing the intervention program. The goal was for the Nigerian teams to develop culturally relevant intervention activities and messages and to create an intervention the team could ‘own’. Each Work Group was charged with reviewing, adapting and augmenting pre-identified intervention materials pertinent to an assigned topic area (e.g. HIV knowledge, partner negotiation), with input from senior US and Nigerian investigators. By the end of the workshops, each Work Group had developed manualized exercises for its topic area, which were subsequently further refined by senior investigators. The intervention was further refined and finalized during the course of a Facilitator Training Workshop with the two men who would deliver the intervention.
The final intervention included seven modules designed to promote dual protection, delivered over the course of two 5-hour workshops, scheduled 1 week apart, with an additional 2-hour ‘check-in’ session 1- and 2-month post-intervention to review and support personal risk reduction goals. Modules covered confronting HIV-related stigma; promotion of knowledge of HIV/STIs and an accurate sense of personal susceptibility; understanding pregnancy and STI/HIV risks and prevention options; exploring risk reduction strategies (abstinence; male and female condom use, with and without use of an additional contraceptive); facilitating sexual negotiation (framed by the Work Group as capitalizing on Nigerian competence and pride in their innate skills as negotiators); challenging gender-based violence (including furthering norms within the group that challenge legally permissible wife beating) and setting and implementing personal sexual risk reduction goals. Sessions emphasized the communication, assertiveness and negotiation skills requisite for practicing safer sex and attempted to foster positive safer sex and gender-equitable attitudes. A combination of methodologies—didactic and interactive teaching, small group discussion, scripting behavior through vignettes and role plays, proverbs, songs, stories and games—was used to engage men and to facilitate skills development.
The comparison condition consisted of a half-day didactic, question-and-answer educational workshop on male and female and female sexuality, reproduction, contraception and HIV/STI transmission, symptoms, treatment and prevention delivered in a group.
Recruitment and procedures
The proof-of-concept trial was conducted between January and August 2005. A combination of strategies was used to recruit men. Female clients attending local family planning clinics were asked to deliver an invitation to their husbands/male partners, informing them of a reproductive health program for men and inviting their participation. Flyers were also distributed in diverse community venues (e.g. trade associations, markets, churches/mosques, government offices, transportation hub centers). In this quasi-experimental study, men were not randomized. Rather, condition was assigned based on the geographic area from which men were recruited; intervention and comparison areas were pre-matched to ensure comparable socioeconomic characteristics.
Following informed consent, all men who responded to the study advertisement underwent a 20- to 30-min eligibility screening for which they were paid the equivalent of US $2. Eligible men were then given a detailed description of the program, and those interested completed an additional consent procedure and were invited to attend either a single- or a multi-session workshop, based on recruitment area. Free transportation to the intervention sessions and lunch were provided.
All workshops were conducted at the ARFH site by two trained, carefully supervised male facilitators. Four two-session experimental and four one-session comparison interventions were delivered, with 25–30 men attending each session. The study was approved by the Institutional Review Boards of both ARFH and the New York State Psychiatric Institute/Columbia University, Department of Psychiatry. All study materials were translated into Yoruba by an experienced ARFH translator, with both assessment and intervention delivered in Yoruba.
Participants were interviewed at baseline and 3-month post-intervention by trained interviewers. Items, by domain, are described below.
Participants were asked to provide personal data regarding age, education, religious affiliation (Christian or Muslim), monthly income, socioeconomic status (lower/higher) based on recruitment sites and marital status (not married, monogamous or polygamous marriage).
HIV/STI risk perceptions, knowledge, testing experiences and stigma
Using a four-point Likert scale (grounded by 0 = not at risk and 3 = a great deal at risk), participants were asked to indicate ‘your view of your own risk’ for STIs and HIV in two separate questions. Fourteen items, developed by the Nigerian team and used in previous research, were used to assess participants’ ‘HIV knowledge’ based on a three-option response scale. The items covered the issues of HIV transmission, prevention and care. A ‘don't know’ response was regarded as lack of knowledge. Items were coded so that a higher score indicated higher HIV knowledge. HIV testing experiences were assessed by asking participants whether they knew where an HIV test could be obtained and whether they and their main partner had been tested for HIV. At follow-up, men were asked whether they were tested since their baseline interview and whether their main partner had ever been tested.
‘Social Exposure to HIV’ was assessed by asking participants whether they knew any HIV-positive person or anyone who had died of AIDS (1 = knew of someone, 0 = did not). ‘Stigma’ was assessed via seven items constructed by the Nigerian team. Using three response options (agree, disagree and not sure), men were asked how they felt about people living with HIV/AIDS, including items that assessed beliefs about how the virus was acquired, whether the virus was a punishment from God, personal comfort around infected persons and human rights of infected persons. A higher mean score indicates more stigma (α = 0.77 and 0.69 at baseline and follow-up, respectively).
Relationship context, skills and intentions
The Nigerian and US team jointly developed two measures to assess the context of men's primary relationship with regard to the climate in which condoms were negotiated and power structure. ‘Relational Response to Condom Use’ was assessed via three items using a four-point scale (grounded by 1 = strongly agree and 4 = strongly disagree). Participants indicated the extent to which they believed their main partner would be angry or suspect sexual infidelity were they to request condom use and whether the participant would be ‘outraged’ if his main partner asked him to use a condom. A higher score indicates a more negative response to condoms (α = 0.84 and 0.64 at baseline and follow-up, respectively). Seven items were used to assess ‘Interpersonal Power’ in main partnerships. Participants rated whether they, their main partner or both equally ‘had the most to say’ regarding choice of friends, leisure and household activities, timing of serious discussions and of sex, contraception, condom use and general power in the relationship. A higher score indicates greater male dominance (α = 0.91 and 0.83 at baseline and follow-up, respectively).
Two skills-related variables developed by the Nigerian and US teams were assessed. ‘Comfort with Sexual Communication’ in primary partnerships was evaluated via three items (grounded by 1 = strongly disagree and 4 = strongly agree): ‘It is easy for me to tell my wife/main partner what I do or don't like to do sexually’, ‘My wife/main partner really cares about what I think about our sexual relationship’ and ‘Talking about sex with my wife/main partner is usually fun for both of us’. A higher score indicates better sexual communication (α = 0.64 and 0.63 at baseline and follow-up, respectively). To assess ‘Safer Sex Self-Efficacy’, participants rated on a four-point scale (grounded by very unsure = 1 and very sure = 4) how certain they were that they could perform 11 safer sex-related behaviors (e.g. discuss HIV/STI prevention with their partners, assess partner risk behavior, convince a long-term or new partner to use a male or female condom; α = 0.88 and 0.82 at baseline and follow-up, respectively).
We also evaluated ‘Intentions for Condom Use’, with participants indicating the likelihood of using a condom every single time they had vaginal or anal sex in the next 6 months on a six-point response scale (from 1 = ‘very unlikely’ to 6 = ‘very likely’).
HIV risk behavior
Sexual behavior was assessed separately for main and ‘other’ partners. Participants reported the number of occasions of vaginal intercourse in the prior 3 months and the number of those occasions protected by condoms. Since dual protection was our central focus, condom use, with or without the use an additional contraceptive, was the central focus of evaluation. A single binary variable was created to reflect whether or not there were any occasions of condom-unprotected vaginal intercourse in the prior 3 months (yes = 1, no = 0). Condom use the last time the participant had vaginal intercourse with their main partner also was assessed and was specified a priori as the major study outcome. Men additionally were asked at baseline whether they had ever refused condom use with their main partner and, at follow-up, whether this had occurred since baseline.
Descriptive statistics and contingency tables are provided to indicate the proportion of men and mean values by condition for key variables. To assess differences by condition at baseline and follow-up and examine differences between participants with and without follow-up data, logistic regression and t-tests were used for categorical and continuous variables, respectively. Given the non-random design, all follow-up analyses controlled for baseline values. Although men did not differ significantly at baseline in education, marital status, religion, occupation or income, there was a significant difference in age by condition at baseline [mean = 35.7 for comparison and 39.8 for intervention; F = 8.41 (df = 1178), P < 0.01]. Therefore, confirmatory follow-up analyses examining differences by condition were conducted both with and without controlling for age.