Vandepitte, Judith M.*; Malele, Faustin MD†; Kivuvu, Dieu-Merçi MD†; Edidi, Samuel‡; Muwonga, Jérémie‡; Lepira, François§; Abdellati, Said*; Kabamba, Joelle MD§; Van Overloop, Catherine MD†; Buvé, Anne MD*
THE CLINIC OF MATONGE IN Kinshasa started up in 1985 as a research centre for HIV/sexually transmitted infection (STI). It was a collaboration among the Congolese government, the Centers for Disease Control and Prevention of Atlanta, and the Institute of Tropical Medicine, Antwerp. The center consisted of an STI clinic where diagnosis, based on laboratory results, treatment, and condoms were offered exclusively to female sex workers (FSWs) and their partners. The clinic succeeded to survive during the political instabilities in the early 1990s, and in 1993, Médecins Sans Frontières (MSF) took over the running of the clinic, which continues to function today. Syndromic management of STIs, using the World Health Organization-standardized algorithms, intensive health education, and condom promotion, are the pillars of the program. In the meantime, also peer education and outreach activities were organized in the community encouraging sex workers to visit the Matonge clinic.
In 1988, a cross-sectional survey was conducted among FSWs in Kinshasa. In this study, participants were actively recruited at their place of work (hotel-, home-, or street-based sex workers) and sent to the Matonge clinic for diagnosis and treatment. Of the women attending the clinic, 35% were HIV-seropositive and 75% had at least one STI: 16% had serologic evidence of active syphilis, 23% gonorrhea, 13% Chlamydia trachomatis infections, 22% trichomoniasis, and 5% were diagnosed with genital ulcer disease.1
Political instability and war since 1993 have driven the Democratic Republic of Congo into a deep socioeconomic crisis. Young girls in Kinshasa as well as married, abandoned, or widowed women are pushed into prostitution as a survival mechanism. In addition to these categories of sex workers, new “categories” have emerged. The so-called Phaseures are homeless girls or women selling sex to survive. This group consists of street children, orphans, girls who have abandoned their families, and abandoned women. The so-called Masquées are women involved in clandestine and/or occasional sex work. This very heterogeneous group comprises women who need additional money to survive but who do not want to be recognized openly as commercial sex workers. In 2001, among women presenting at the Matonge clinic, 39% were Masquées, 28% were home-based sex workers, 18% were hotel-based, 3% were street-based, and 12% were Phaseures.
In this changed context and after more than 10 years of interventions targeting sex workers in Kinshasa, we set out to assess the prevalence of HIV infection and other STIs among sex workers attending the clinic in Matonge for the first time in 2002.
Study Population and Data Collection
All women involved in sex work presenting for the first time at the STI clinic in Matonge between September and December 2002 were eligible to participate in the study. Study procedures were explained to all eligible women and they were offered pre- and posttest counseling for HIV before informed consent was obtained to participate in the study.
A standardized questionnaire was developed for the data collection. A trained nurse interviewed the participants on sociodemographic characteristics, type of sex work, client recruitment, and condom use. In addition, information on current gynecologic symptoms or complaints was recorded.
After the interview, the study participants were requested to give a blood sample that was tested for HIV, syphilis, and herpes simplex virus type 2 (HSV-2). A physician carried out a clinical examination, including a pelvic examination. Vaginal secretions were collected on swabs for the diagnosis of gonorrhea, C. trachomatis infection, and trichomoniasis.
The patients were treated on the spot following the guidelines for the syndromic management of STIs. They also received health education and free condoms and were invited to return after 1 week for additional treatment, if indicated.
At the pelvic examination, 2 vaginal secretion specimens were collected. The first one was inoculated onto a culture medium for Trichomonas vaginalis (InPouch TV; Biomed Diagnostics, San Jose, CA). The InPouch culture medium was incubated for 5 days if no growth was observed after 3 days of incubation at 36°C. The second swab was stored at −20°C until shipment to Antwerp for further processing. There, the swab was tested for Neisseria gonorrhoeae and C. trachomatis using the Strand Displacement Amplification Assay (SDA; Probetec; Becton Dickinson, Sparks, MD). Samples testing positive on the SDA were confirmed with the Amplicor CT/NG polymerase chain reaction (PCR) test (Roche Diagnostic Systems Inc., Branchburg, NJ).
Syphilis serology was assessed with the rapid plasma reagin (RPR) test, Macro-Vue RPR Card Test (Becton Dickinson, Le Pont de Claix, France) and the Treponema pallidum particle agglutination (TPPA Serodia; Fujirebio, Tokyo, Japan). Women were considered to have active or untreated syphilis if they tested positive on the RPR and the TPPA.
Testing for HIV was done in Kinshasa using a dual testing strategy (Enzygnost HIV1/2 plus; Dade Behring, Marburg, Germany; HIV-Determine; Abbott Laboratories, Tokyo, Japan). Positive results were reconfirmed in the HIV/STI laboratory of the Institute of Tropical Medicine in Antwerp (ITM), Belgium, using the Vironostica HIV Uni-form II plus O ELISA test (BioMérieux, Boxtel, The Netherlands).
Finally, type-specific HSV-2 antibodies were determined using a commercially available enzyme-linked immunosorbent assay (Kalon Biologic Ltd., Aldershot, U.K.).
Trichomonas culture, syphilis, and HIV testing were done at the Laboratoire National de Référence SIDA et IST (LNRS). Aliquots of serum samples were shipped to ITM, Antwerp, for HSV-2 serology and external quality control of the HIV and syphilis serology.
Data were entered in Epi-info 4.06 (Centers for Disease Control and Prevention, Atlanta, GA) and analyzed using Stata 8.0 (Stata Inc., TX). The study population was described by summary statistics (percentages, means, and medians). The prevalence of HIV and other STIs were compared in the different types of sex workers using the Fisher exact test. Risk factors for HIV infection were examined first by univariate analysis. Continuous data such as age, time in prostitution, age at first sex, numbers of clients, and cost per sex act were categorized according to the mean or median. Odds ratios (with 95% confidence intervals) were used to measure the magnitude of the association with HIV infection. All variables for which a P value <0.10 was obtained in the univariate analysis were considered as potential confounders and included in the final model for multivariate logistic regression.
Ethical approval for the study was obtained from the Ethics Committee of the Institute of Tropical Medicine, the Programme National de Lutte contre le Sida (PNLS) and Médecins Sans Frontières.
Sociodemographic Characteristics, Sex Work, and Condom Use
Between September and December 2002, 585 women presented spontaneously and for the first time to the clinic in Matonge and 502 (81%) agreed to participate. Table 1 describes the characteristics of the study population in terms of sociodemographic characteristics, type of sex work, and condom use.
Seventeen women (3.4%) recruited their clients in hotels (hotel-based), 146 (29.1%) at home (home-based), and 10 (2.0%) on the street (street-based FSW); 40 (8.0%) were homeless people, living and working on the street (Phaseures) and 289 (57.5%) were women involved in clandestine or occasional sex work (Masquées). The mean age of the participants was 23 years (range: 10–46 years). The Phaseures were on average younger than the other women and the home-based FSWs were older (mean age: 18 and 29 years, respectively). Nearly 30% of the women had never been to school or had only reached the primary school level. Sixty percent reported having a steady partner, ranging from 40% among the street-based sex workers to 82% among those working in hotels. The median duration of sex work was 3 years, but 27% of the participants had worked as FSWs for ≤1 year. The mean age at sexual debut was 12 years for the Phaseures and 15 years for the other categories of FSWs. Overall, the women recruited an average number of 16 clients a week. Only 27% declared consistent condom use (always) with clients, 66% not consistent condom use (not always), and 7% never used a condom. However, when recalled for the last working day, 74% of the women reported condom use with all clients, 14% with some of the clients, and 12% did not use a condom with any of them.
Any differences between Masquées and the other categories of women who do openly acknowledge working as sex workers were further explored. Masquées were younger, were involved in sex work for a shorter period, and recruited by average more clients than the other FSWs (Table 2). The rate of consistent condom use was similar in both groups.
Prevalence of HIV and Other Sexually Transmitted Infection
Table 3 presents the prevalence of HIV and other STI for the different categories of sex workers. Overall, 12.4% (95% confidence interval [CI]: 9.62–15.58) of the women were HIV-seropositive. The home-based FSWs were most affected (24.0%, 95% CI: 17.30–31.73) followed by the street-based FSWs (20.0%, 95% CI: 2.52–55.61), the hotel-based (11.8%, 95% CI: 1.46–36.44), the Phaseures (10.0%, 95% CI: 2.79–23.66), and finally the Masquées (6.6%, 95% CI: 4.00–10.08).
Sixty-two percent of the study participants had at least one STI. N. gonorrhoeae and C. trachomatis were detected in, respectively, 7.8% and 8.4% of the women, and 2.2% of women were dually infected. The Phaseures had the highest rates of gonorrhea and chlamydia, 22.5% and 17.5%, respectively. The overall prevalence of T. vaginalis was 8.6%, but reached 20.5% and 30.0% in Phaseures and street-based FSWs, respectively. Positive RPR and TPPA, suggesting active untreated syphilis, were found in 2.6% of the study population. HSV-2 infection was the most prevalent STI, ranging from 50% in the Masquées and street workers up to approximately 75% in hotel- and home-based sex workers.
Factors Associated With HIV Infection
Older age, type of sex work (“other” vs. Masquées), longer time in prostitution, fewer numbers of clients per week, consistent condom use during the last day, positive HSV-2 serology, and active syphilis were independently associated with HIV infection in univariate analysis. After multivariate analysis, HIV infection remained statistically significantly associated with increasing age, fewer number of clients per week, consistent condom use in last working day, and HSV-2 infection (Table 4).
Among FSWs attending for the first time the Matonge STI clinic in Kinshasa, in 2002, 12.4% were found to be HIV-infected. The study population consisted of different categories of FSWs: women recruiting clients in hotels, at home, or on the street; homeless women involved in sex work; and clandestine FSWs. A similar cross-sectional study conducted in 1988 among hotel-, home-, and street-based sex workers in Kinshasa found an HIV prevalence of 35%.1 The HIV prevalence among the same 3 categories of FSWs in the current study was 22.5%, suggesting a decline in HIV prevalence. Also, the prevalence of other STIs decreased in the 3 overlapping categories: the prevalence of syphilis dropped from 16% in 1988 to 3.5% in 2002, gonorrhea from 23% to 8.1%, chlamydial infection from 13% to 5.2%, and trichomoniasis from 22% to 9.2%. In 2002, more sensitive laboratory tests were used for the diagnosis of STI compared with 1988. In the current study, gonorrhea and chlamydial infections were detected using PCR versus culture in 1988, and T. vaginalis was isolated by culture versus microscopy in 1988. So the actual decline in prevalence of other STIs may be even more pronounced.
Decreasing trends in prevalence of HIV and other STIs have also been observed in other settings where STI/HIV prevention programs were implemented for FSWs. In Abidjan, Côte d'Ivoire, the HIV prevalence dropped from 89% to 32% between 1992 and 19982; in Cotonou, Benin, from 53% in 1993 to 41% in 19993; and in Bobo-Dioulasso, Burkina Faso, there was a decrease from 57% to 41% from 1994 to 2000.4 All 3 studies also reported a significant decline in the STI prevalence. It is believed that changes in sociodemographic characteristics and shifts of the nationalities of the women may have played a role but that it was mainly the change in sexual behavior and the improved STI case management that had contributed to this success.
However, the decline in HIV and other STIs in the overlapping categories of FSWs in Kinshasa should be interpreted with caution; the studies in 1988 and in 2002 were both cross-sectional, conducted 14 years apart, and were not specifically designed to assess trends over time. The study populations had been selected in different ways. In 1988, sex workers were approached at their place of work and invited to the clinic in Matonge, whereas in 2002, we enrolled sex workers presenting spontaneously and for the first time at the clinic. Although the activities in the Matonge clinic are well accepted throughout the sex worker communities in Kinshasa, the actual coverage of the clinic is not known, so some selection bias may have occurred. Nevertheless, the results of the current study confirm that the HIV prevalence in Kinshasa does not increase dramatically as one would expect in a context of profound political and social instability. Also, Mulanga-Kabeye et al found a stabilization of HIV prevalence in selected population groups, including FSWs, in Kinshasa.5
Sexual behavior changes are difficult to quantify and their impact on the HIV/STI epidemic even more. Condom promotion and distribution is one of the major interventions offered by the MSF FSW Project in Kinshasa. Condom use is difficult to evaluate, because it relies on self-reported behavior.6 In the current study, only 27% of the women reported consistent condom use with clients, 66% answered sometimes, and 7% never. However, when they were asked to recall condom use during the last working day, 74% answered always, 14% sometimes, and 12% never. In contrast in 1988, 8% of hotel-based, 6% of home-based, and 20% of street-based sex workers reported regular condom use.1
The current study has demonstrated furthermore that the sex worker population may have changed considerably over time. At the beginning of the project, the clinic could easily categorize all women according to the place where clients were recruited, i.e., in hotels, at home, or on the street. By 2002, 2 new categories had appeared: the Phaseures and the so-called Masquées, the latter making up 58% of the clinic population in 2002. Although the Masquées did not want to be considered FSWs, they were not significantly different from the other self-acknowledging sex workers in terms of sexual behavior; they recruited even slightly more clients per day or per week and had a comparable pattern of condom use. Almost 7% of them were HIV-infected and 50.7% had HSV-2 antibodies; the prevalence of the other STI was comparable in both groups. The univariate analysis of risk factors for HIV infection showed that the Masquées were nearly 4 times less at risk than the other FSWs. However, this association disappeared in the multivariate analysis.
The multivariate analysis showed furthermore that FSWs with fewer clients per week were more at risk of being HIV-infected than those with more clients per week. A similar negative association between HIV infection and numbers of partners was found among men living in rural Tanzania.7 A possible explanation is that HIV-infected sex workers may have to reduce their workload because of HIV-related illness.
HSV-2 antibodies were detected in 58.5% of our study population and this was also the main risk factor for HIV infection (adjusted odds ratio: 9.26, P = 0.00). Home- and hotel-based FSWs had the highest HSV-2 seroprevalence (74.5% and 76.5%, respectively). The age-specific HSV-2 seroprevalence increased gradually: from 39.3% in women ≤19 years to 82.7% in those of 30 years and more. These findings are consistent with results from other studies conducted among FSWs in African settings. In Lagos, Nigeria, 59% of FSWs had HSV-2 antibodies,8 73% in Nairobi, Kenya,9 91% in Cotonou, Benin, 84% in Yaoundé, Cameroun, 94% in Kisumu, Kenya, and 87% in Ndola, Zambia.10
In conclusion, the Matonge clinic in Kinshasa, which opened in 1985, was one of the first initiatives targeting FSWs in sub-Saharan Africa. Many efforts have gone into improving the sexually transmitted disease case management and making the population aware of the threat of HIV and how it can be prevented. The clinic managed to continue functioning during the war of the early 1990s. Now, more than 10 years later, the prevalence of HIV and other STIs in the FSW population seems, at least, to have stabilized, if not declined.