The surveillance survey among MSM in Benin is a cross-sectional study with an analytical orientation, conducted from November 13 to 24, 2017 for 2017 study and from November 19 to December 11 for 2015 study.

The MSM in the 2015 study were recruited from cities in Benin, including Cotonou, Abomey-Calavi, Ouidah, Bohicon, and Parakou ( Table 1(a) ). The arrangements put in place to ensure the safety of respondents described in 2017 were observed in 2015.

Cn = Cotonou; Ac = Abomey-Calavi; Ou = Ouidah; Bh = Bohicon; Pk = Parakou; Pn = Porto-Novo; Sec: Secondary; Sup: Superior, NG= Not given.

The localities covered by the study in 2017 were the 19 cities which were covered by the situation analysis of MSM and injectable drug users (IDU) in the fight against STI, HIV and AIDS in Benin carried out in 2012 and the first Second Generation Surveillance Survey (ESDG) carried out in 2013.

With the help of the head of the two MSM networks (Benin Synergie Plus and Coalition AIDS Benin) and identity associations, collection sites were identified in the selected localities. In order to guarantee the safety of the respondents as well as the discretion of the operation, it was decided to install the collection sites in health centers and headquarters of certain identity associations, according to the constraints of each locality. This is the Cotonou 1 health center, the Coalition AIDS Benin (CAB) headquarters in Abomey-Calavi, the headquarters of the Adorable Club Association in Porto-Novo, the Ouidah zone hospital, the headquarters of the MSM Love Story association in Bohicon and the Zongo health center for the city of Parakou ( Table 1(b) ). The sites where the blood samples were stored were all part of the PSLS network of STI, HIV and AIDS care sites. They also had the advantage of having facilities to collect and store blood samples.

Eligibility criteria for MSM enrolment included being born as a biological male who reported having had at least one anal (receptive or insertive) or oral (fellation and/or rimming) sexual encounter with a male partner in the past 12 months and being at least 18 years old of age and voluntarily consent to participate in the survey and residing in Benin during the last six months preceding the survey regardless of nationality, whether or not to belong to an association that is a member of one of the two MSM networks (BESYP and Coalition Sida Benin and have a valid reference coupon.

Excluded of the study were MSM who did not meet one of the inclusion criteria at the time of the survey and who were in a state that did not allow them to give informed consent (under the influence of drugs or alcohol, mental illness or any other cause).

Second Generation Surveillance Survey (ESDG) among men who have sex with men in Benin used Respondent Driven Sampling (RDS) [7] method to recruit MSM both in 2015 and 2017.

The first MSM surveyed called “seeds” were identified in a reasoned way. In order to promote better representativeness of the sample, arrangements have been made to ensure a good diversity of seeds.

This diversity of basic characteristics made it possible to avoid an over-representation of certain groups of participants with more extensive networks, thus improving the representativeness of the final sample. It was only after that and after accepting to play the role of “seed” that MSM was chosen. A total of eight and nine MSM were selected as starting points (seeds) for recruiting other MSM in 2015 and 2017 respectively. These men were chosen so as to obtain a better representativeness of the MSM population in the country.

Each seed received three recruitment coupons which it then distributed to three other MSM called “recruited” that it knows and who know him. Each coupon distributed bears a unique number making it possible to know at all times who has been recruited and by whom and to follow the characteristics of the recruitment. These coupons were validated for a specific week in order to modulate the influx of people presenting with a coupon and wishing to participate in the study. Each recruit received in turn three recruiting coupons which they also distributed to three other MSM, thereby becoming a “recruiter”. The recruitment chain started from each seed continued until the expected sample was reached.

The sample size calculation took into account the prevalence of HIV, the level of knowledge about HIV and the level of use of condoms and lubricant gels during anal intercourse. Thus, the formula [11] used to calculate the sample size and was 364 in 2015 and 359 in 2017.

A structured questionnaire was used to collect the data both in 2015 and 2017. It included the following components:

Sociodemographic characteristics, knowledge, sexual behaviour/practices, opinions and attitudes towards STIs and HIV/AIDS, types and number of sexual partners, use of condoms and lubricating gels, sexually transmitted infections, HIV, voluntary HIV testing, social network, socio-cultural, health and legal contexts of MSM and Access to prevention services and tools, especially condoms, lubricants, and HIV information services.

Serum samples were first tested for HIV by enzyme linked immunosorbent assay (Vironostika® HIV Uni-Form II Ag/Ab (bioMérieux bv)). Confirmation was obtained with (Génie III HIV 1&2®, Bio-RAD). Samples that produced discordant results were further analyzed with western blot (INNO-LIA® HIV I/II Score).

All the data collected in 2015 and 2017 were checked through a written procedure manual and then double-entered using the EPI-DATA software. Further data cleaning and data analysis were done using SPSS 22.1. Data tabulation was done with RDSAT software version 7.1.38 5 [12] .

For each calculated indicator, the RDSAT software uses the so-called “bootstrapping” technique, a robust non-parametric method, to calculate the 95% confidence interval. Thus, all the indicators calculated and presented in this report have been weighted according to the RDSAT procedures with their 95% confidence intervals. In combination with the NETDRAW software, the RDSAT also provides a graphical representation of the recruitment structure.

Results were presented as proportions with a 95% confidence interval. Qualitative variables were compared using the chi-2 or Fisher’s exact tests.

The dependent and explanatory variables were analyzed by univariate and/or multivariate logistic regression to study the relationships between the determinants of HIV infection.

The national ethics committee for health research (CNERS) of Benin gave its approval of the study. The consent of each participant was requested after explaining the benefits, potential risks and other conditions for participating in the survey.

Informed consent covered both the behavioural component of the study and the biological component (anonymous HIV screening test). All those recruited to participate in this study were free to accept or refuse.

In 2015, the sizes of the social networks of MSM seeds ranged from 2 (Abomey-Calavi site) to 60 (second site of Cotonou). The age of MSM seeds ranged from 19 years to 32 years. The total number of participants in the 2015 study per seed ranged from 37 to 50 ( Table 1(a) ).

Table 1(b) shows some general characteristics for each MSM seed in the survey of 2017. The sizes of the social networks of MSM seeds ranged from 7 (Abomey-Calavi site) to 42 (Porto-Novo site). The age of MSM seeds ranged from 22 years to 33 years. One seed has not given its age

366 MSM were recruited in 2015 survey. Of these, two were eliminated to avoid bias in recruitment. Thus, a total of 364 MSM were retained, including (186/364, 50.5%) in Cotonou, (45/364, 12.4%) in Abomey-Calavi, (50/364, 13.7) in Ouidah, (40/364, 11%) in Bohicon and (45/364, 12.4%) in Parakou.

A total of 359 MSM were included in 2017 study with (118/359, 32.9%) in Cotonou, (60/359, 16.7%) at Porto-Novo, (48/359, 13.4%) at Ouidah, (46/359, 12.8%) at Abomey-Calavi, (45/359, 12.5%) at Bohicon and (42/359, 11.7%) at Parakou.

About half of the respondents (46.8%) were between 18 and 24 years old and younger than those recruited in the 2015 (80.6%) survey (p < 0.01). On the other hand, those aged 25 and over were more numerous in 2017 (19.4%) than in 2015 (53.2%) (p < 0.01%). Less than 10% of MSM were married in 2017 and more numerous (8.6%) than those of 2015 (1.9%) (p < 0.01) while single MSM were numerous in 2015 (94.7%) than in 2017 (73.5%) (p < 0.01%) ( Table 2 ). Almost all MSM were in school, and over 90% had a high school diploma or above in 2015 as well as in 2017. No difference was observed up to the secondary education level of the MSM recruited in 2015 and 2017 apart from the higher education level where there were more MSM recruited in 2017 (49.3%) than in 2015 (38.1%) (p < 0.01%).

Craftsmen represent the majority of people questioned (75.5%) in 2017 and more numerous than in 2015 (20.8%) (p < 0.01%) while the pupils/students questioned in 2015 were higher (74.2%) than those of 2017 (3.1%) (p < 0.01). On the other hand, more salaried employees were recruited in 2017 (21.4%) than in 2015 (5%) (p < 0.01).

The prevalence of traditional and Muslim MSM remained stable from 2015 to 2017 (9.9% vs 8.9%) (p = 0.74) and Muslim (21.4% vs 21.2%) (p = 0.99). The prevalence of Christian MSM increased from 55.3% in 2015 to 65.2% in 2017 (p < 0.01). MSM were predominantly Beninese and their prevalence did not change from 2015 (97.4%) to 2017 (98.3%) (p = 0.46). Only 2% were of foreign origin (Togo, Ghana) and the prevalence remained stable in 2017 (p = 0.73).

There were no significant differences in the orientation and sexual role of MSM from 2015 to 2017 ( Table 2 ).

STI: Sexually transmitted infections, *p-Value. Fisher’s exact test.

MSM prevalence who have both precise knowledge about preventing sexual transmission of HIV and who reject major misconceptions about HIV transmission had remained stable from 2015 (66.6%) to 2017 (66.0%) (p = 0.95). Similarly, the prevalence of MSM living with HIV (p = 0.81) and sexually transmitted infections (p = 0.84) remained unchanged. The prevalence’s of the average number of male (p = 0.98) and female (p = 0.69) sexual partners remained unchanged during the last 6 months of the survey in 2015 and 2017 ( Table 3 ). In contrast, the prevalence of MSM who reported consistently using a condom with lubricating gel every time they had anal sex with a male partner in the past 6 months increased from 2015 (21.1%) to 2017 (44.1%) (p < 0.01). Prevalence did not change among MSM who reported consistently using a condom every time they had vaginal intercourse with a female partner in the past 6 months from 2015 (49.8%) to 2017 (52.4%) (p = 0.52). Regarding the prevalence of MSM having access to screening services, no variation was observed (p = 0.48) ( Table 3 ). This non-variation in prevalence was also observed among MSM who reported using a condom the last time they had anal sex (p = 0.95).