Syphilis and HIV prevalence and associated factors to their co-infection, hepatitis B and hepatitis C viruses prevalence among female sex workers in Rwanda

Background Human Immunodeficiency Virus (HIV), syphilis, Hepatitis B Virus (HBV) and Hepatitis C Virus (HCV) are sexually transmitted infections (STIs) and share modes of transmission. These infections are generally more prevalent among female sex workers (FSWs). Methods This is a cross-sectional study conducted among female sex workers (FSWs) in Rwanda in 2015. Venue-Day-Time (VDT) sampling method was used in recruiting participants. HIV, syphilis, HBV, and HCV testing were performed. Descriptive analyses and logistic regression models were computed. Results In total, 1978 FSWs were recruited. The majority (58.5%) was aged between 20 and 29 years old. Up to 63.9% of FSWs were single, 62.3% attained primary school, and 68.0% had no additional occupation beside sex work. Almost all FSWs (81.2%) had children. The majority of FSWs (68.4%) were venue-based, and most (53.5%) had spent less than five years in sex work. The overall prevalence of syphilis was 51.1%; it was 2.5% for HBV, 1.4% for HCV, 42.9% for HIV and 27.4% for syphilis/HIV co-infection. The prevalence of syphilis, HIV, and syphilis + HIV co-infection was increasing with age and decreasing with the level of education. A positive association with syphilis/HIV co-infection was found in: 25 years and older (aOR = 1.82 [95% CI:1.33–2.50]), having had a genital sore in the last 12 months (aOR = 1.34 [95% CI:1.05–1.71]), and having HBsAg-positive test (aOR = 2.09 [1.08–4.08]). Conclusion The prevalence of HIV and syphilis infections and HIV/syphilis co-infection are very high among FSWs in Rwanda. A strong, specific prevention program for FSWs and to avert HIV infection and other STIs transmission to their clients is needed.


Background
Human Immunodeficiency Virus (HIV), syphilis, Hepatitis B Virus (HBV) and Hepatitis C Virus (HCV) are sexually transmitted diseases (STIs) and have shared modes of transmission. Female sex workers (FSWs) are at high risk to contract STIs due to the work and their behavior [1,2]. The World Health Organization (WHO) reported that the global prevalence of HIV among FSWs was 11.8% with significant variation by region, with the highest prevalence found in sub-Saharan Africa with an aggregated prevalence of 36.9% [3]. The prevalence of HIV among FSWs is 10-20-fold higher than the general population in many African countries [4].
A meta-analysis study conducted in African countries found that the HIV prevalence among FSWs varied between 19% and 60% [5]. Despite the control of HIV among the general population in Rwanda (3.0%) [6], STIs are still a concern among FSWs. In 2011 a 24.0% prevalence of HIV was estimated among FSWs in Kigali, Rwanda [7] which was three times higher than the prevalence of HIV in the general population in the same city (7.3%) [8] in the same period. WHO reported that 10 countries had greater than 10% syphilis prevalence among sex workers, while some countries reported a syphilis prevalence of more than 20% [1]. A low prevalence of syphilis among FSWs was reported in few African countries. In Somaliland, the prevalence of syphilis was 3.1% [9], and 3.3% among FSWs in Kisumu, Kenya [10]. A high prevalence of syphilis among FSWs was reported in other countriesfor example, it was 21.0% in Kampala, Uganda [11] and 52.4% in Addis Ababa, Ethiopia [12] in two separate cross-sectional studies.
Few data about HCV and HBV among FSWs have been published. WHO estimated the worldwide number of people living with chronic HCV between 130 and 150 million [13] while this number was estimated at 240 million of people chronically infected with hepatitis B [13]. In China, the prevalence of HCV was estimated at 0.8% in one province [14] and 1.1% in another province [15], it was 2.8% in India [16].
The prevalence of HBV in India was estimated at 7.6% [17] while it was 17.1% in Nigeria [17].
The interaction between syphilis and HIV is not well documented [18]. Syphilis as an ulcerative infection, increases HIV transmission. Syphilis in clinically immunosuppressed HIV-infected patients is reported to be associated with greater organ involvement [19]. In Nepal the proportion of syphilis and HIV co-infection was 31.0% [20]. Factors such as STIs symptoms, lower level of education were described being positively associated with syphilis and HIV infections [21].
Currently, few data are published on HBV and HCV among FSWs in Rwanda. A study conducted in 2013 found that the prevalence of HBV was 1.6% among blood donors, while the prevalence of HCV was 2.9% among health workers [22]. A meta-analysis of studies from sub-Saharan Africa reported an HBV prevalence of 4.1% and an HCV prevalence of 2.1% among HIVinfected persons [23].
The aim of this study is to describe the burden of syphilis, HIV, HBV, HCV and HIV/syphilis co-infection and associated factors among FSWs in Rwanda.

Study setting and population
This study was a cross-sectional behavioral and biological survey among FSWs in Rwanda in 2015. Venue-Day-Time (VDT) sampling method was used to identify specific places and times where FSWs await their subsequent sexual clients [24]. The time frame was determined by the peak hours of the presence of FSWs at hot spots.
Before the data collection process, a rapid assessment to estimate the number of expected FSWs was conducted. The number of FSWs at each site was estimated by key informants selected in the same target population. The survey was conducted in all hot spots with 5 or more FSWs per day. The number of participants recruited in each hot spot was determined using probability proportional to size of each selected site.

Population and inclusion criteria
Self-reported FSW aged 15 years or more found in a FSW hot spot area identified by a FSW key informant was eligible to participate. Excluded from the survey were: FSWs less than 15 years old, not self-identifying as a female sex worker, declining to consent to participate in the survey.

Sampling
A two-stage sampling method using VDT sampling, referring to the specific peak time and place that FSWs were present at their hot spot was used. The primary sampling unit was the time and place where FSWs were present at the hotspot, and the secondary sampling unit was FSWs in the selected time and place. A "take-all" approach sampling was used. Once the expected sample size at the site was obtained, data collection was closed at that site. For primary sampling unit (PSU) selection, 180 hotspots were selected from the sampling frame in the country.

Sample size calculation
The sample size calculation used the Z score value of 1.96 with an alpha level of 0.05 (95% confidence), the design effect of 3.2, a relative 10% precision around a point estimate (HIV prevalence) of 51% among female sex workers in 2010. A minimum sample size of 1980 FSWs was calculated.

Data collection methods
FSWs were contacted by data collectors to take part in the survey. If eligible, a verbal informed consent was administered with no identifying information. After verbal consent was received, pre-test counselling was conducted by a nurse counsellor. The laboratory technician drew blood by venipuncture. A Personal Digital Assistant (PDA) questionnaire was administered by survey staff. Rapid diagnostic test results were returned to the participant and post-test counselling was conducted in few minutes. Participants whose test was positive for HIV, syphilis, HBV or HCV were referred to the health centre of the patient's choice for clinical care. Blood samples were transported to the National Reference Laboratory (NRL) for confirmatory HIV testing for survey purposes.

HIV testing
HIV rapid testing was performed by a trained laboratory technician using the nationally approved algorithm of three serial rapid tests (currently approved protocol: Shanghai-Kehua, Determine, and Uni-Gold). For the survey purpose, ELISA-based testing was used (Vironostika® HIV Uni-form II Ag/Ab, 4th Generation) for all samples. For samples testing positive with ELISA, a confirmatory test used a second EIA called Murex HIV Ag/Ab Combination 4th Generation Murex®.

Syphilis testing
Screening for syphilis was conducted on site using the SD BIOLINE SYPHILIS 3.0 rapid test to detect syphilis anti-bodies. A reacting sample was retested using the RPR test at NRL. A sample that was positive for both tests was recorded as positive; a non-reacting sample to the second test was recorded as negative.

Hepatitis B testing
Screening for HBV was performed using the SD BIO-LINE HBV rapid test to detect HBV surface antigen (HBsAg). A reacting sample was recorded as positive.

Hepatitis C testing
Screening for HCV was performed using the SD BIO-LINE HCV rapid test. The test detected HCV antibody (anti-HCV). A reacting sample was recorded as positive.
All FSWs with positive tests for any test were transferred to the nearest health facility or the health facility of their choice in order to provide quick results for clinical care.

Data analysis
STATA 13 software was used for data analysis. Mean, proportion, and 95% confidence interval were computed for descriptive analysis of variables. Prevalence in different socio-demographic characteristics was estimated for HIV, syphilis, HBV, HCV infections and for HIV and syphilis co-infection.
HIV and syphilis co-infection associated factors significant at the p < 0.05 level in bivariate analysis were included in a multivariable logistic regression model. Multivariable analysis was computed to determine the effect of independent variables to HIV and syphilis coinfection. Variables were retained in the final model when achieving p < 0.05 significance. In bivariable logistic regression, age 25 years or older, being ever married, having more than five years of duration of sex work, having children, having had genital sores in the last 12 months, and having HBsAg positive test were positively associated with syphilis and HIV coinfection, whereas secondary or high education level was negatively associated with syphilis and HIV co-infection (Table 3).

Discussion
FSWs are among key drivers of HIV and other STIs in many countries. The current study was conducted among self-identified FSWs at different hot spots in the country to characterize the scope of this public health problem in Rwanda.
Two important findings resulted from this survey. Syphilis and HIV prevalence among FSWs in Rwanda was very high and increased with age and the duration of sex work. FSWs with any positive test were transferred to the nearest health facility or the health facility of choice for appropriate clinical care.
Overall HBV and HCV prevalence among FSWs in Rwanda was low (<5% each). Older FSWs, secondary or higher education, STIs symptoms, and HBV infection are independent predictors of HIV and syphilis coinfection.
Compared to the prevalence of the general population (3.0%) [6], the HIV prevalence among FSWs (42.9%) of our study was more than 14 times higher; and it is among the highest reported to date in Africa [3]. In the Rwandan context, sex work is illegal and often considered a shameful activity, some among FSWs operate in hidden unsecured places due to poor socio-economic conditions. This high HIV prevalence may be attributed to inconsistency in condom use, lack of negotiating power with clients over condom use, sexual violence and access to HIV care and treatment. In the current study, a proportion of 81.2% among FSWs had children with a risk of vertical transmission. In sub-Saharan Africa, in Togo FSWs had higher HIV prevalence (76.7%) [25]; For unknown reasons, other countries had lower HIV prevalence. It was 32.2% in Kenya, 15.7% in Uganda, 3.5% in Democratic Republic of Congo [25].
The overall prevalence of syphilis of our study was (51.1%) while it was 1.0% in the general population of Rwanda [26]. Although the current prevalence of syphilis in our study could be current or past infection, it is still very high. Due to the high prevalence of HIV in the current study (42.9%), it is not surprising to find a similar prevalence of syphilis. The prevalence of syphilis among FSWs in our study was higher compared to the prevalence of syphilis in many other countries. For instance it was 3.3% in Kisumu, Kenya [10], 21.0% in Kampala, Uganda [3]. However, the prevalence of syphilis among FSWs in Rwanda was similar to the prevalence of syphilis among FSWs in in Addis Ababa, Ethiopia 52.4% [12].
HIV and syphilis have similar modes of transmission. The burden of HIV and syphilis among FSWs is considerable in Rwanda; it could contribute to high morbidity and mortality rates in this key population. The prevalence of syphilis and HIV co-infection was high (27.4%). In terms of proportion of co-infection, in Nepal, the proportion of syphilis infection among HIV-positive FSWs was 31.0% [18], it was 69.2% in our study. In Addis Ababa the proportion of HIV among syphilis-positive FSWs recruited in STIs clinic was 85.1% [12].
Older age was identified as an associated factor of high HIV and syphilis co-infection; the same factor was identified in prevalence of syphilis in a Chinese study [21].
Secondary or higher education was negatively associated with higher prevalence of HIV and syphilis coinfection (aOR = 0.50). The assumption behind this finding may be the fact that FSWs with secondary and higher education could have additional occupation to the sex work, thus reduce the risk of exposure. In addition, they may have possibility to access prevention programmes, and could have power for condom use negotiation. They could also access to STIs treatment compared to illiterate FSWs. In China, researchers also found that years of education was a protective factor of STIs [21].
A recent history of genital sores was identified as an independent variable positively associated with high HIV and syphilis co-infection (aOR = 1.34 [95% CI: 1.05-1.71]). Ulcerative STIs facilitate HIV and other blood borne diseases transmission and acquisition. The same observation regarding HIV infection risk factors was found in India [21].
The prevalence of HBV among FSWs in our study was lower (2.4%). There is no previous estimated prevalence  of HBV in the general population or in any other key population including FSWs in Rwanda. HBV prevalence was 8.8% in Brazil [10] whereas in sub-Saharan African countries, for instance in Nigeria, it was 17.1% [17] and it was 13.3% in Nairobi, Kenya [25].
The prevalence of HCV was 1.1% in our study. HCV is not a high burden among FSWs in Rwanda compared to HIV (42.9%) and syphilis (51.1%). Few data on this area are available. However, a study on HCV and HBV conducted among health workers in a tertiary hospital reveled that the prevalence of HBsAg-positive and anti-HCV-positive was respectively 2.9% and 1.3% [27]. Another study conducted among HIV-positive patients in Kigali, Rwanda, showed that HBsAg-positive and anti-HCV-positive was respectively 5.2% and 5.7% [28]. HCV prevalence was similar with the prevalence found in Nairobi, Kenya 0.8% [18], but much lower than the estimated prevalence of HCV in Brazil (23.1%) [10].
Several limitations were encountered in our study. First home-based FSWs and those operating inside hotels were underrepresented in the study because the recruitment was focused on streets and around venues where FSW are found such as hotels, cabarets and night clubs. Second, venues with FSWs less than five per day were excluded by the sampling methodology. Third, specific sexual practices and behaviors and other risk factors associated with STIs among sex workers were not explored in depth. Fourth, the syphilis prevalence reported in this study was first performed using SD Bioline rapid test; confirmed with RPR test. Due to budget constraints a confirmatory test using TPHA test was not performed.

Conclusion
Based on our findings, we conclude that HIV and syphilis prevalence are very high among FSWs in Rwanda. A strong specific prevention program to avert HIV infection and other STIs transmission to their clients; and proper treatment should be reinforced. Current programs sponsored by the Ministry of Health can help to fight against stigma and discrimination against sex workersspecifically, promoting outreach approach to test and treat all HIV FSWs; testing and treating other STIs among FSWs using rapid tests in their community and improve prevention program for HIV and other STIs reduction. Strengthening HIV and STIs prevention program including condom distribution in the community, training peer educators to be involved in program implementation and linking FSW's community to care providers in confidential way. Organizing a regular campaign for HIV and STIs testing and treatment in the community. Supporting income generating activities and regular mentorship in order to improve the poor socio-economic status of most of FSWs.