The study was conducted in the five main large cities of Burkina Faso (Ouagadougou, Bobo-Dioulasso, Koudougou, Tenkodogo, and Ouahigouya). Due to the specificity of the hidden populations, these cities were selected according to several criteria: geographical disparities in HIV prevalence, cultural variation, and crowds of sexual activities in these cities.

A cross-sectional survey was carried out via the respondent-driven sampling method, which is one of the best methods to reach hidden populations, such as FSW (10, 11). The survey took place from June to August 2022.

The study population included female sex workers (FSWs) living in the five study towns. Female sex workers are defined as “girls and women over the age of 18 who receive money or other gifts in exchange for sexual services, either regularly or occasionally.” The inclusion criteria were as follows: being female, at least 15 years old, having sexual intercourse in exchange for money as the main income source for 12 months, having a valid coupon (the validity of the coupon was set at 15 days), and consenting to participate in the study. HIV-positive FSWs were excluded from this analysis. We included FSWs with less than 18 years old because a previous study showed an important proportion of them who started sex work as minors (12).

The sampling process and sample size calculation have been described elsewhere (8). In brief, the sample size was 1,424 FSWs to be recruited. The sample size was calculated using the Salganik (13) formula. The sampling followed the RDS methods. The recruitment started with two seeds in each city. Seeds were FSWs known as the leader of their community with a high network size. The seeds who meet the study eligibility were non-randomly selected with the contributions of NGOs and FSW associations. They were selected considering their influence on the FSW community. They are known as leaders in the FSW community of their city. After their inclusion in the study, they were trained on the process of recruitment before receiving three coupons to recruit other FSWs. Second, the following participant also received three coupons to recruit other participants until the sample size was reached. Each coupon was identified by unique numbers; thus, each person recruited can be identified by his coupon number. This allows us to know the order of recruitment and reduce any risk of double inclusion in the survey.

The study methods have been described elsewhere (8). In brief, following the RDS approach, data collection started by checking the eligibility of the participants. If a participant was eligible, he was invited to complete a face-to-face questionnaire administered by a trained investigator. The questionnaire was structured to include sociodemographic and behavioral data and was programmed on a digital tablet via the CSPro application. After receiving an explanation of what HIVST consists of, participants were invited to answer the following question: “If you were offered an HIV self-test, would you use it?” After the interviews, the participants underwent HIV testing via a rapid diagnosis test (RDT).

The outcome variable in our study was the willingness to use the HIVST kit. The outcome variable was treated as a binary variable: FSWs who agreed with self-testing (code = 1) versus those who did not agree (code = 0).

The independent variables include demographic and socioeconomic variables, behavioral factors, and HIV knowledge variables. These variables were age (<25 years and >=25 years) of FSW, current marital status, educational level, monthly income (Dollar), age at first sex (<18 and >=18), age at initiation of sex work (<18 and >=18), history of drug use (Yes/No), condom use at last sex with clients (Yes/No), condom use at last sex with partners (Yes/No), history of previous incarceration in last 12 months (Yes/No), current alcohol consumption (Yes/No) with refer to alcohol consumption within last 30 days, lifetime history of drug use (Yes/No), HIV knowledge (Sufficient/Non-sufficient) as defined by UNAIDS, number of biological children (neither, one, two or three, more than three), and membership in an association (Yes/No) also known as peer group membership (it seems to be a way for HIV prevention among female sex workers including HIVST delivery (14, 15)).

After the data were cleaned, we first excluded all participants who tested positive for HIV. We then describe the characteristics of the participants. We estimated the rate of FSW willingness to use HIVST via the unweighted and RDS weighted options. For each rate, a confidence interval was calculated. The rate of willingness to use HIVST was presented by the demographic and risk behavior characteristics of the participants. We conducted a modified Poisson regression using a generalized estimating equation with an exchangeable correlation structure as recommended (16) to identify factors associated with willingness to use HIVST among female sex workers. From this model, we derive the prevalence ratio and its confidence intervals. All analyses were performed using Stata 18. All the statistical tests were considered significant when the p-value was <5%.

The protocol of this study was approved by the National Ethical Committee for Health Research (Deliberation Number 2020-02-029 of 5 February 2020).

Among the 1,425 participants, 87 HIV-positive individuals were excluded. The final sample size for this analysis was 1,338 FSWs. The participants’ characteristics are presented in Table 1.

The rates of willingness to use HIVST among FSWs are presented in Table 1. Among the 1,338 FSWs, 1,082 agreed to use HIVST, for a rate of 89.5% (95% CI: [87.0; 91.6]). The proportion of FSWs who were willing to use HIVST for different subgroups of demographics and risk behaviors was high, ranging from 85.1 to 98.2%. This proportion was highest (98.2%) among FSWs who had more than three children and lowest (51.3%) among FSWs who had a history of previous incarceration in the last 12 months before the data collection. The proportion of FSWs who were willing to use the HIVST rate appeared to be affected by monthly income, but from an income of 700$ to 5,014$, the prevalence was the lowest at 85.8%.

Table 2 below displays the results of the modified Poisson regression on the willingness to use HIVST among FSWs in Burkina Faso. As reported in the table, the proportion of willingness to use HIVST was 10% greater among FSWs who were married than among those who were not married (single), with a significant p-value (p = 0.034). The proportion of willingness to use HIVST among FSWs whose first sex was less than 18 years (<18 years) was 14% higher than that among those whose first sex was at least 18 years old (p = 0.024). The proportion was significantly (p = 0.010) higher among those who used condoms during their last sex with a non-paying partner. Compared with those who have a current alcohol consumption, female sex workers who did not have a current alcohol consumption had a 6% higher willingness rate to use HIVST, with a significant p-value (p = 0.026). Those who are members of an association of FSWs have a 10% higher rate of willingness to use HIVST than those who are not members of any association of FSWs (p = 0.014).

This is the first analysis of HIVST among FSWs in Burkina Faso. Our findings suggest a high rate (89.5%) of willingness to use HIVST among FSWs in Burkina Faso. The willingness to use HIVST seems to be greater among FSWs who are married, those who initiated sex before 18 years of age, those who did not use condoms at last sex with non-paying partners, those who used drugs, those of did not consume alcohol, and those who are member of FSW association.

Compared to other studies, our study revealed a high rate of willingness to use HIVST among FSWs (89.5%) in Burkina Faso. Kim et al. (17) reported a lower rate (30.1%) of willingness to use HIVST among FSWs in Malaysia. While a qualitative study conducted in western Africa (Mali, Cote d’Ivoire, and Senegal) showed positive attitudes and willingness of FSW toward HIVST use (18), studies that focused on the willingness to use HIVST are sparse. However, many studies have assessed the acceptability and uptake of HIVST among key populations in SSA. Figueroa et al. (19) reported in their literature review that three of 23 studies on HIVST focused on FSW. Witzel et al. (20) noted that among 11 RCTs on HIVST compared with standard HIV testing, only three RCTs focused on HIVST among FSWs. The HIVST acceptability rate appears to be high in sub-Saharan Africa. Indeed, the HIVST rate reported in studies is close to 59.3% in Ethiopia (21). A systematic review revealed greater acceptability of HIVST among FSWs (19, 20). In particular, oral HIVST has a high acceptability rate, close to 90.5% in Morocco (22) and 81% in Botswana (23). Among other populations, studies have reported various levels of HIVST acceptance rates. In 2020, 70.4% of university students in Nigeria (24) and 94.3% in Senegal were first-time testers (25), and 84.5% were among the general population in Zimbabwe, 76.5% in Malawi, and 64.5% in South Africa. The STAR initiative in 13 countries has reported good acceptance of HIVST; the majority of participants found the test kits easy to use and interpret, and approximately two-thirds were willing to pay for HIVST (26). However, studies have shown some limits regarding the use of HIVST. The main limitation was the linkage HIV care service including confirmation test and ART initiation. Kra et al. (27, 28) reported that only 56% of the key population with HIVST positive was linked to a confirmation test. In 2021, Neuman et al. reported a limited effect of HIVST on ART initiation compared with that of standard testing services alone (29). From the ATLAS study, we learned that FSWs were reluctant to promote HIVST use among their casual clients (18). However, in another context, a recent community-based study showed a good linkage between confirmatory testing (89%) and ART initiation (85%), and 69% of participants were willing to pay for the HIVST kit (26). In Burkina Faso, a national strategy plan was drafted in 2022 to improve the application of a differentiated service approach known to significantly improve HIV cascades of care (30–32). From this plan, we noted that FSWs are among the main population for whom this approach is implemented. The approach included HIV testing, index testing, and HIVST as new strategies for HIV testing in addition to community facility-based HIV testing. Our findings suggest a good framework for scaling up HIVST services among FSWs in Burkina Faso. This approach might also accelerate the achievement of the first 95 since studies have shown that peer-assisted HIVST increases the number of newly diagnosed HIV-positive cases (33).

According to studies, various factors increase the likelihood of FSW’s willingness to use HIVST, depending on the setting. In our study, the age of initiation of sex work was not associated with FSW’s willingness to use HIVST. However, Nibret Eskezia (21) noted that spending more than 5 years of sex work increased the likelihood of HIVST acceptability. Age at first sexual intercourse was associated with HIVST. Nibret Eskezia et al. noted that FSWs aged >19 years at first sexual debut are 3.23 times more likely to accept HIVST. In our study, we did not assess knowledge about HIVST. Studies have shown good acceptability of HIVST among FSWs with good knowledge of HIVST (21). The association between risky sexual behavior and HIV testing is often difficult to interpret because of important desirability bias even if the data collector is well trained to avoid it. Indeed, our study reported that risky sexual behavior, such as not using a condom, was associated with a lower rate of FSW willingness to use HIVST. However, Stalter et al. noted that individuals who reported high-risk behaviors were more likely to repeat HIV tests (34). Manu et al. reported no significant association between risky sexual behaviors and HIV testing (35).

Studies have reported low rates of HIV testing among FSWs (36–38). Our previous study showed that only half (57.8%) of FSWs reported having been tested for HIV in the last 12 months suggesting an important gap in HIV testing in this population (8). There are many barriers to the use of standard HIV testing services. Among these barriers are stigma and discrimination regarding HIV-positive individuals and sex work (38). Many studies have highlighted that HIVST may address this issue, as it is safe, confidential, and increases the acceptability and frequency of HIV testing (20). Indeed, HIVST increased testing uptake by 36% compared with standard HIV testing among FSWs (20). A modeling study from the ATLAS project revealed that HIVST might increase the HIV diagnosis rate by 1.3% in Côte d’Ivoire, 10.6% in Senegal, and 3.6% in Mali by the end of 2028 (39). HIVST is projected to reduce disparities in access to HIV testing among key populations and their partners in Western Africa (39). Regarding the benefit of HIVST in HIV prevention, we recommended scaling up this intervention in Burkina Faso. This might help fill the gap in HIV testing among FSWs in the country. However, the linkage to HIV care seems to be low compared with that to standard care (20). Indeed, the systematic review revealed a lower linkage to HIV care via HIVST than to standard HIV care (20). Linkage to care seems to be the key aspect of HIVST, which needs to be addressed to improve the benefit of HIVST. Indeed, Kouassi et al. noted that only 56% of participants with a positive HIVST result followed up with confirmatory HIV tests. While compared to standard HIV testing, HIVST seems to better screen HIV-positive FSWs (33). This approach might help reach FSWs who are not able to use facility-based HIV testing services. The ATLAS HIVST strategy revealed that HIVST increased HIV testing coverage by reaching the underserved population across West Africa (40). Since studies such as that of Mavhu et al. (45) noted that HIVST is preferred over health provider-based HIV testing, public health policymakers in Burkina Faso must promote HIVST among FSWs to increase the HIV testing rate and reach the WHO recommendation regarding HIV testing among FSWs (7). The program level cost of HIVST strategies seems to be higher than HIV standard service (41); however, a part of the cost may be supported by the population since they are willing to pay for HIVST kit (26). Promoting HIVST needs to be combined with risk behavior control intervention among FSWs since evidence among MSM shows that risk behavior seems to increase after HIVST introduction (42). It is also necessary to take into account strategies that might support FSWs who tested HIV positive to disclose their status to their partners and invite them for testing (43).

Face-to-face interviews in the context of sex work stigma might cause social desirability. There is recall bias regarding some characteristics reported as independent variables in this study. The study was conducted in five cities of the country, and the results might not be generalizable to all FSWs in the country. Some factors, such as self-perceived risk known to be associated with HIV testing, were not collected. However, this is the first quantitative assessment of HIVST acceptability among FSWs in Burkina Faso. We also used the RDS approach, which is among the best approaches for studying hidden populations (11, 44).