This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Research Protocols, is properly cited. The complete bibliographic information, a link to the original publication on http://www.researchprotocols.org, as well as this copyright and license information must be included.
Sub-Saharan African Migrants (SAM) are the second largest group affected by HIV/AIDS in Belgium and the rest of Western Europe. Increasing evidence shows that, more than previously thought, SAM are acquiring HIV in their host countries. This calls for a renewed focus on primary prevention. Yet, knowledge on the magnitude of the HIV epidemic among SAM (HIV prevalence estimates and proportions of undiagnosed HIV infections) and underlying drivers are scarce and limit the development of such interventions.
By applying a community-based participatory and mixed-methods approach, the TOGETHER project aims to deepen our understanding of HIV transmission dynamics, as well as inform future primary prevention interventions for this target group.
The TOGETHER project consists of a cross-sectional study to assess HIV prevalence and risk factors among SAM visiting community settings in Antwerp city, Belgium, and links an anonymous electronic self-reported questionnaire to oral fluid samples. Three formative studies informed this method: (1) a social mapping of community settings using an adaptation of the PLACE method; (2) a multiple case study aiming to identify factors that increase risk and vulnerability for HIV infection by triangulating data from life history interviews, lifelines, and patient files; and (3) an acceptability and feasibility study of oral fluid sampling in community settings using participant observations.
Results have been obtained from 4 interlinked studies and will be described in future research.
Combining empirically tested and innovative epidemiological and social science methods, this project provides the first HIV prevalence estimates for a representative sample of SAM residing in a West European city. By triangulating qualitative and quantitative insights, the project will generate an in-depth understanding of the factors that increase risk and vulnerability for HIV infection among SAM. Based on this knowledge, the project will identify priority subgroups within SAM communities and places for HIV prevention. Adopting a community-based participatory approach throughout the full research process should increase community ownership, investment, and mobilization for HIV prevention.
“Know your epidemic, know your response,” has become the directive of the Joint United Nations Programme on HIV and AIDS (UNAIDS) for intensifying HIV prevention [
In Belgium, 27% (n=230) of the newly reported HIV diagnoses in 2013 were in individuals of sub-Saharan African origin [
Reported characteristics of newly diagnosed SAM in Belgium are in line with the generalized epidemic in sub-Saharan Africa. In 2013, the majority (64%) were women, heterosexual contact was the main transmission mode (89%), and most (78%) were diagnosed between 20 and 45 years [
Late diagnosis and delayed initiation of care not only affect disease prognosis [
Belgian surveillance data show that 10.4% of all SAM diagnosed in 2013 report having acquired HIV in Belgium. Yet, this is based on the physician’s assessment at diagnosis and data are missing for 30% of cases [
HIV prevention comprises the continuum of primary prevention, promotion of HIV testing and counselling, and “positive health, dignity, and prevention,” which includes the prevention of HIV transmission. Yet, in Belgium [
The prevention of new HIV infections, or primary prevention, among SAM has not been made priority because traditionally the HIV epidemic in the African diaspora in Europe was understood to be imported [
Second, estimates of the proportions of SAM with undiagnosed HIV are lacking. In Europe, one third of persons living with HIV are assumed to be unaware of their HIV status [
Third, previous research mainly focused on SAM’s individual knowledge, attitudes, and practices related to sexuality and HIV-preventive behavior, thus underestimating the social, cultural, religious, and migration-related contexts that increase vulnerability with respect to HIV [
Although small in numbers, SAM communities are characterized by a high degree of heterogeneity due to diverse ethnic and cultural backgrounds, migration patterns and residence statuses, educational and socioeconomic backgrounds, and religious beliefs [
Many SAM live in socioeconomically vulnerable and legally unstable conditions. Together with prevalent HIV-related stigma and culturally grounded taboos on sexuality, this translates into little demand for HIV prevention [
The TOGETHER study’s overall aim was to increase the communities’, researchers’, and policymakers’ in-depth understanding of the dynamics of the HIV epidemic among SAM, to improve primary prevention interventions. This translated into the following objectives:
To assess the HIV prevalence and proportion of undiagnosed HIV infections among SAM socializing in community settings in Antwerp city.
To identify individual, community-level, and structural risk factors for HIV infection among SAM.
To identify priority settings and groups for future primary HIV prevention interventions.
To increase community ownership, involvement, and mobilization for HIV prevention.
To develop policy recommendations to improve HIV prevention for the target group of SAM.
To assess the feasibility and acceptability of community-based participatory research on HIV prevalence in SAM communities and adopted research tools.
To meet these objectives, the TOGETHER Project applied mixed methods and a community-based participatory research approach (CBPR) [
For the different study components, separate study protocols were developed and ethical approval was obtained from the Institutional Review Board of the Institute of Tropical Medicine and the ethical committee of the University Hospital Antwerp.
To account for the heterogeneity of Antwerp’s SAM and to ensure that the study methods and tools were acceptable for all subgroups (objective 6), we chose a community-based participatory approach [
Outline of the TOGETHER project.
This study ran from June 2012 until June 2013 and had the triple objective of (1) determining the sampling frame for the HIV prevalence study, (2) identifying priority settings for future HIV prevention, and (3) increasing communities’ ownership of HIV prevention.
To account for the heterogeneity of the SAM communities and ensure inclusion of hidden subpopulations, for example, SAM of undocumented status or MSM, we chose to employ the systematic approach of the PLACE Method (Prioritising Local AIDS Control Efforts) [
In step 1, “identifying a priority prevention area,” Antwerp city was selected based on demographic and epidemiological aspects: 22% of all SAM in Flanders live in Antwerp city [
The second formative study contributed to the second overall project objective (ie, assessing individual, community-level, and structural risk factors). This study’s first phase took place between April 2013 and December 2013. The 3 specific objectives were: (1) informing the development of a structured questionnaire for the HIV prevalence study, (2) qualitatively contextualizing the findings of the HIV prevalence study, and (3) informing the development of future HIV prevention interventions. Study participants were SAM living with HIV, who were unique cases to retrospectively identify multi-level risk and vulnerability factors.
We conducted an empirical inquiry investigating a phenomenon within its real-life context, using multiple sources of evidence [
In the first study phase, a convenience sample of SAM living with HIV was recruited through physicians and nurses of the HIV clinic and facilitators of an HIV support group for SAM. To arrive at a representative sample of the patient population of sub-Saharan African origin, in the second phase, with a start date of April 2016, we switched to purposive sampling.
All participants in the first study phase were consenting adults who received their HIV diagnoses between 6 months and 10 years ago, who were assessed by health care providers as being psychologically stable and were followed up by a social nurse at the time of the interview. The latter was important to assure linkage to psychosocial care in case the interviews evoked emotional upheaval. The same rationale led to significant attention being paid to the informed consent procedure. Prior to the first interview, the study’s rationale, objectives, procedure, and confidentiality measures, and participants’ rights, benefits, and disadvantages were discussed extensively with the participant, before the informed consent form was signed. During this process, we asked for explicit approval to consult the participants’ patient files. For follow-up interviews, the procedure was repeated and verbal informed consent was obtained. As a token of appreciation, participants received an incentive of €25 for each interview.
To ensure participants’ anonymity and confidentiality, all data were coded and stored in a password-protected folder. All data were uploaded to NVivo 10 and a first
For the HIV prevalence study, we opted for collecting oral fluid samples to determine HIV status, since reluctance toward blood taking is known to limit HIV testing uptake among SAM [
This intervention of the Institute of Tropical Medicine offered free oral fluid HIV tests (Oracol device, Malvern Medical Developments, Worcester, UK) in community settings of two target groups (MSM [
The primary objective of this cross-sectional study, which ran from December 2013 to August 2014, was to determine HIV prevalence among SAM socializing in community settings in Antwerp city (objective 1 of the TOGETHER Project). Secondary objectives were: (1) Identifying the individual, community-level, and structural risk factors for HIV infection among SAM; and (2) Identifying priority settings for future HIV prevention interventions (project objectives 2 and 3, respectively).
HIV prevalence was the primary outcome measure. The sample size was calculated using an anticipated HIV prevalence of 4%, a required precision of 2% for the 95% confidence intervals, and a cluster sampling design effect of 2. This resulted in a required sample size of 714 SAM.
A 2-stage time location sampling (TLS) was adopted. TLS takes advantage of the fact that some hard-to-reach populations tend to gather or congregate at certain types of locations [
To be eligible, potential study participants had to self-identify as belonging to the SAM communities, be 18 years or older, agree to answer the behavioral questionnaire, donate an oral fluid sample, and be willing and able to provide written informed consent. Prior participation in the prevalence study was an exclusion criterion.
The study combined biological and behavioral measures. To measure the study’s primary outcome, HIV prevalence, oral fluid samples were collected and tested for HIV antibodies in the AIDS Reference Laboratory. These samples were linked through a unique code to an anonymous behavioral questionnaire. This instrument included questions on participants’ socio-demographic and economic background, migration and mobility background, health-seeking behavior, HIV-testing behavior, sexual and relational history (last year and lifetime), attitudes towards condom use, actual condom use, and level of assistance needed to complete the questionnaire. The structured electronic questionnaire was developed based on the findings of formative study 2, consultation of available questionnaires from comparable studies [
Detailed study procedures were described in the Standard Operating Procedures (SOP) developed in collaboration with the CRs, CAB, and the AIDS reference laboratory. The SOP were refined after 2 pilots. At pre-arranged moments, a study team visited the selected sites and randomly selected 14 of the SAM present. When approaching potential participants, the CRs identified themselves and introduced the study’s objectives and methodology, stressing the anonymous and voluntary nature of participation. To avoid self-exclusion of HIV-positive individuals, they explicitly mentioned that everybody was invited to participate, regardless of HIV status. Interested individuals were invited to a quiet area in the setting, if available. After discussing and signing the informed consent form, participants were asked to complete a structured anonymous electronic questionnaire on a tablet through SurveyToGo. To build confidence and ensure data rigor, participants first received a short tutorial on how to use the tablet. Here, special attention was given to the demonstration of how anonymity was guaranteed. Questionnaires were available in French, English, and Dutch, which are reference languages for most SAM. The preferred interview method was self-completion; however, assistance was offered if needed. By ethnically matching the CRs to the settings, translation of questions to a local language was possible, if preferred. The CRs were trained to offer assistance with sensitivity and respect to confidentiality.
After completing the questionnaire, the CR demonstrated the procedure of oral fluid collection using the collection device. Next, participants were asked to self-collect the sample. The sample was then linked with the informed consent form, questionnaire, and a letter explaining how to collect the results through a
Finally, participants were asked to provide information on their frequency of
Within 7 days of collecting the sample, the AIDS reference laboratory of the Institute of Tropical Medicine performed the analysis according to a validated algorithm using oral fluid specimens [
Data from the questionnaires, attendance forms, laboratory data (HIV status), and HIV test result collection form were linked via the unique code, merged, and stored in an SPSS Statistics 22 (IBM) database. After data cleaning, statistical analysis was carried out. An analysis plan to take into account cluster sampling and a weighting factor [
Results have been obtained from 4 interlinked studies and will be described in future research.
The TOGETHER Project was conceptualized to respond to the growing need for investing in targeted primary prevention interventions to reduce new HIV infections among SAM. Increasing evidence indicates that more SAM than previously assumed acquire HIV after migration to West European host countries [
Our HIV prevalence study adopted methods successfully implemented in previous bio-behavioral surveys and cross-sectional studies [
Study methods and tools were continuously refined based on the input of a CAB, CRs team, and the findings of two additional formative studies. To the best of our knowledge, this was the first time that such a carefully designed CBPR design had been adopted in Europe to assess HIV prevalence among a representative sample of SAM.
Due to practical limitations, the study setting was limited to Antwerp city. While Antwerp is the major city where SAM reside in Flanders, this implies that, although the HIV prevalence study will generate a sound estimation of HIV prevalence among SAM, the results cannot be generalized to Belgium or other West European countries. Different compositions of SAM communities in other cities have to be considered. For example, in Brussels 47% of SAM are of Congolese origin [
Adopting the CBPR approach as outlined is known to be challenging in terms of assuring skills of lay researchers, and safeguarding continuous motivation and data quality [
Ability to ensure community mobilization for HIV prevention may be challenged by time constraints. While some subgroups are stable communities, others are known to evolve rapidly, community venues are often unstable, and leadership fluctuates [
With respect to the multiple-case study, it should be mentioned that our approach to identifying factors that increase SAM’s vulnerability to HIV infection was innovative (ie, triangulation of data from life history interviews, lifelines with patient files). In particular, the use of lifelines was, according to our literature search, new to HIV research. These qualitative insights were valuable in the analysis and contextualization of the HIV prevalence study’s findings. Together, they will form a solid basis for policy recommendations and the development of future HIV prevention interventions, once all study results are available.
community advisory board
community-based participatory research
community researchers
greater involvement of people living with HIV/AIDS
men having sex with men
Prioritising Local AIDS Control Efforts
sub-Saharan African migrants
time location sampling
The authors wish to thank the team of community researchers for their insights and continuous engagement: Sophiah Atieno, John Che Akangwa, Sandra Karorero, John Mugabi, Morgan Ndungu, Eveline Pilime, Electa Tamasang, Daniel Tantoh, and Jean Senga. Janvier Muhizi, the project’s study assistant is highly appreciated for his all-around support. We also acknowledge the members of the community advisory board for their input in the studies’ design: Laura Albers (Helpcenter-ITM), Rans Aubin (Ghana Welfare), Florent Batoum Bahiengraga (ACASIA vzw), Erika Delporte (ARC, ITM), Levis Kadia (Bilenge vzw), Tine Vermoesen (ARL, ITM), and Hans Willems (City of Antwerp). In addition, we wish to thank Dominique Van Beckhoven and Andre Sasse from WIV for their input on the epidemiological background. Veronica van Wijk and Sabien Salomez are thanked for their language editing and help with reference manager. Finally, we wish to thank the Scientific Fund for Research on AIDS, managed by the King Baudouin Foundation, for financially supporting the project.
None declared.