Publications on Libya’s HIV epidemic mostly examined the victims of the tragic nosocomial HIV outbreak in the 1990s and the related dispute about the detention of foreign medical workers. The dispute resolution in 2003 included an agreement with the European Union on humanitarian cooperation and the development of Libya’s first National HIV Strategy. As part of this we conducted Libya’s first bio-behavioural survey among men having sex with men (MSM) and female sex workers (FSW).
Using respondent-driven sampling, we conducted a cross-sectional study to estimate the prevalence of HIV, hepatitis B virus (HBV), hepatitis C virus (HCV), and related risk factors among 227 MSM and 69 FSW in Tripoli (FSW recruitment ended prematurely due to the political events in 2011).
For MSM we estimated an HIV prevalence of 3.1%, HBV prevalence of 2.9%, and HCV prevalence of 7.3%, and for FSW an HIV prevalence of 15.7%, HBV prevalence of 0%, and HCV prevalence of 5.2%. We detected high levels of risk behaviours, poor HIV-related knowledge, high stigma and lack of prevention programmes. These results must be interpreted in the context of the political situation which prohibited reaching an ideal sample size for FSW.
There is urgent need to implement an effective National HIV Strategy informed by the results of this research. The risk of transmission within different risk groups and to the general population may be high given the recent military events that led to increased violence, migration, and the disruption of essential HIV-related services.
Citation: Valadez JJ, Berendes S, Jeffery C, Thomson J, Ben Othman H, Danon L, et al. (2013) Filling the Knowledge Gap: Measuring HIV Prevalence and Risk Factors among Men Who Have Sex with Men and Female Sex Workers in Tripoli, Libya. PLoS ONE 8(6): e66701. doi:10.1371/journal.pone.0066701
Editor: Michael Alan Polis, National Institute of Allergy and Infectious Diseases, United States of America
Received: December 20, 2012; Accepted: May 9, 2013; Published: June 19, 2013
Copyright: © 2013 Valadez et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: This study was funded by the European Commission under the terms of the project “Supporting the development and implementation of a National HIV-AIDS Strategy in Libya”. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: Dr. Hussain Ben Othman works for the Libyan National Centre for Diseases Control (NCDC), Ministry of Health, which is also responsible for the planning and implementation of the National HIV-Strategy informed by the results of this paper. The other authors have declared no competing interests exist. This does not alter the authors’ adherence to all the PLOS ONE policies on sharing data and materials.
An assessment of the HIV prevalence, related risk behaviour and knowledge among men who have sex with men (MSM) and female sex workers (FSW) in Libya is urgently needed to inform near-term policy making while the window of opportunity to act is still open , . History has shown that timely collection of necessary information for the design of effective HIV prevention strategies is crucial, in order to confine the epidemic at an early stage . This is particularly important in Libya which has not had a National HIV Strategy, and may face an expansion of the HIV epidemic due to increased migration, violence, and the disruption of services and structures related to recent warfare and socio-political transition , , .
Male-to-male sexual contact and transactional heterosexual intercourse have been main modes of HIV transmission in the early stages of the HIV epidemic in many countries , . Recently, new emergence and resurgence of HIV infection among MSM were observed in many parts of the world , , . Similarly, FSW continue to be disproportionally affected by HIV globally in almost all settings . Male to male sexual intercourse, which exists in diverse cultures and throughout history, is highly stigmatized in many parts of the world, and often criminalized, especially in low and middle income countries , . Likewise, commercial sex is illegal in most nations and severely punished in many . In Libya, both MSM and sex workers have been highly stigmatized, facing corporal punishment and sentences of up to several years in prison if arrested and sentenced . This behaviour creates incentives for these groups to remain hidden, resulting in an apparent lack of knowledge on HIV prevention and frequent inability to access HIV related services and tools. It might also be responsible for the striking lack of HIV-related data among these populations , , , .
Libya’s HIV epidemic has been classified as possibly concentrated among people who inject drugs (PWID) , which we recently confirmed . (A concentrated epidemic is defined as HIV prevalence below 1% in the general population, as measured in pregnant women in urban areas, and consistently above 5% in at least one subpopulation, such as PWID ). However, no data exist on the HIV prevalence among MSM and FSW . Sound evidence on the prevalence of Hepatitis B virus (HBV) and Hepatitis C Virus (HCV) infection among these groups is also missing .
So far the few international publications that exist on the HIV epidemic in Libya are mostly restricted to those examining the victims of the tragic outbreak of nosocomial infections in the 1990s . The related detention of six foreign medical workers and their trials in Libya had led to international political tensions, which were only resolved in 2003 after the detainees were released and an agreement was made with the European Union on humanitarian cooperation , . This included the development of Libya’s first National HIV Strategy, as part of which we conducted a Bio-Behavioral Surveillance Survey (BBSS) among most vulnerable populations (MVPs). We used the method of respondent-driven sampling (RDS), which has been successfully applied in many settings to sample hidden populations , , .
The study reported here provides baseline information on the level of HIV infection and related socio-demographic, behavioural and other characteristics among MSM and FSW in Tripoli to inform the design and future evaluation of Libya’s first National HIV Strategy.
Study Design and Participants
We conducted a cross-sectional BBSS among MSM and FSW in Tripoli, Libya. Given the hidden nature of these populations, we used the peer-driven chain-referral method of RDS to access study participants through their social networks , .
MSM were eligible to join the study if they were born male, and had anal sex with another man in the last six months. FSW were eligible if they were born female, and had earned all or part of their income through the exchange of money for vaginal or anal sex with more than one client in the last six months. In addition, participants of both groups had to be ≥15 years old, and have a valid referral coupon.
The study began on 21-Sep-2010 at a single site, with opening times differing for MSM and FSW, as they preferred not to share the same waiting room. Sampling started with initial participants, referred to as seeds, which were identified during an initial mapping exercise, carried out by an international consultant on social mapping to explore the situation, sensitize the populations of interest and identify their most active representatives as potential seeds. The criteria we used to select the seeds included clear understanding of the study goal and enthusiasm for it; good communication skills, a large personal network size and enjoying great respect among peers were assets. Because some of the seeds failed to successfully initiate recruitment chains due to the highly stigmatizing environment, we had to recruit a total of 14 MSM and 13 FSW seeds. The intensive outreach work continued even after the start of the survey. We made every effort to include seeds of diverse sociodemographic background and serostatus, as well as places of gathering. After undergoing the survey process themselves, the seeds initiated the recruitment chains by distributing up to three recruitment coupons to their peers and referring them to the study site. Eligible participants underwent a face-to-face interview, received pre-test counselling, provided blood for lab tests, had the opportunity to be seen by a doctor for sexually transmitted infections (STI), and were given the option to obtain HIV rapid test results and post-test counselling during the same, and HBV and HCV results during a second visit. They were then given a primary compensation payment to cover part of their time and transportation for the completed visit, as well as up to three coupons for peer recruitment. A secondary compensation was provided at a later stage for each successfully recruited peer.
MSM recruitment continued until a sample size of 227 was reached. This corresponds to the sample size required when using the formula proposed by Salganik et al 2006  with the following parameter settings: a standard error of 0.03, a design effect of 2.5, which is higher and thus more conservative than the recommended value of 2.0, and an assumed HIV prevalence of 9%. We chose this last estimate based on the highest known prevalence in neighbouring countries . We applied these same principles to FSW. With the same formula and highest known regional prevalence of 13% , we required a sample size of 314 FSW. However, FSW recruitment progressed much more slowly than for MSM, and was interrupted by the political events of February 2011 before the target sample size could be met .
The interviews were guided by a structured questionnaire that included questions on socio-demographics, sexual and other risk behaviours, HIV-related knowledge, access to services for HIV prevention and care, and other HIV-related factors. Questions were based on internationally recognized and standardized indicators and instruments ,  that were adapted to the local context. We also included RDS-specific questions on personal network size, and reciprocity of the relationship between recruiter and recruit , , , . The questionnaires were translated into Libyan colloquial Arabic and French, back-translated into English for verification, and refined during pre-testing.
Venous blood samples taken by qualified nurses were immediately tested for HIV by laboratory technicians who applied a three serial rapid testing strategy  using Determine HIV-1/2 Kit (Inverness Medical Innovations) for initial testing, and subsequently Uni-Gold HIV-Kit (Trinity Biotech) and Bioline HIV-1/2-Kit (Standard Diagnostics) for confirmation of reactive results.
HBV and HCV tests were performed daily at a central reference laboratory located at the National Centre for Disease Control (NCDC) in Tripoli. After detecting HBV surface antigen (HBsAg) and HCV antibodies by automated microparticle enzyme immunoassay (MEIA, AxSym, Abbot Diagnostics), positive results were followed up with enzyme linked fluorescent assay to determine HBsAg and anti-HBV core antigen (VIDAS, Biomérieux) and with line-immunoassay to confirm anti-HCV (INNO-LIA HCV-AbIII, Innogenetics).
After double data entry using EpiInfo v.3.5.1 and data pre-processing in MS access and excel v.2012, we developed programmes in MATLAB v.2010b for statistical analysis. All seeds were excluded from the analysis. We followed current RDS methodology which assumes that RDS chains of sufficient length are Markovian, , . We constructed the transition matrix A from the referral chains, where each element aij represents the proportion of the sample with characteristic i that recruit individuals with characteristic j. Using A and assuming that the characteristic of the recruit is only dependent on the characteristic of the recruiter (i.e. Markovian) we were able to estimate the expected sample proportions at equilibrium. To confirm that the sample was not biased by the selection of seeds, which are not necessarily randomly chosen, we computed whether equilibrium had been reached for all relevant indicators. We define equilibrium to be the point at which the sample proportion for the specific indicator changes minimally with continued recruitment, and is sufficiently close to the proportion expected from a simple Markov process . We distinguish equilibrium at 1%, where the mean absolute discrepancy between the sample proportion and the expected value from the recruitment matrix is <1%, from equilibrium at 2%, which results in a lower confidence. If the sample proportion is >2% different from the expected values, we conclude that the sampling process has not reached equilibrium for that indicator.
We used three methods to analyse the RDS chains (RDSI/DS, RDSII and RDSII/DA , ) and ensure that the representative population proportion estimates are robust. We used three methods to analyse the RDS chains (RDSI/DS, RDSII and RDSII/DA , ) to ensure that the representative population proportion estimates are robust. The first (denoted RDSI/DS), involves deriving a matrix of smoothed adjusted transition probabilities by adjusting the transition matrix A to account for differences in recruitment success. Using this matrix, the personal network size and the populations proportion estimates of each characteristic, pi, we establish a set of equations to solve for pi . With the second method (denoted RDSII), the sample proportions of individuals in each characteristic group are weighted by the ratio between the average network sizes from the total population and from the characteristic group. This ignores the variation in recruitment success rate between individuals with different characteristics . The third estimator (denoted RDSII/DA), accounts for biases that are introduced with differences in the success of recruitment of different groups by weighting the equilibrium proportion, rather than the sample proportion, by the same ratio of network sizes .
Confidence intervals were estimated using a re-sampling procedure with 10,000 reiterations  for all three methods. Further details on the computation of population proportion estimates are given elsewhere .
For the MSM data, we identified statistically significant correlates of HIV infection by calculating odds ratios using Fisher’s central exact test. The analysis was conducted in R v.2.13.1 and the packages epitools v.05-7 and exact2x2 v.1.1-1.0.
The study, including the consent procedure, was approved by the Liverpool School of Tropical Medicine Research Ethics Committee and the Libyan National Ethics Review Committee, and endorsed by the Libyan Ministry of Security, who ordered police to refrain from raids against MSM and FSW at and near the study site. Because respondents were reluctant to provide their signature for fear of being identified, we obtained oral rather than written informed consent from all study participants (having a staff member sign a certificate confirming that consent was fully informed and given freely and voluntarily). The informed consent form was administered by the screener, and the whole process was monitored by the supervisor. Informed consent forms, questionnaires and laboratory results were linked through code numbers rather than personal identifying information to ensure complete anonymity. The ethics committees also waived the need for informed consent from the next of kin, caretakers, or guardians on behalf of respondents aged 15–17, because the right for confidentiality in such a sensitive survey prevailed over the need for consent from their adult family members. This was the principle proffered by the Libyan NCDC.
Study participants who tested positive for HIV, HBV or HCV were referred to appropriate public health facilities for free clinical evaluation and treatment if needed.
We recruited a total of 227 eligible MSM through up to 15 recruitment waves and 69 FSW through 10 waves. For the MSM data, equilibrium was reached for all but one of the reported indicators and the three methods for estimating population proportions produced similar results (Table S1–S6 in File S1). However, given the interruption of the FSW study due to the political turmoil, the sample size was too small for equilibrium to be reached for some (11 of 54) indicators, including HIV status. Recruiter-recruit relationships were reciprocal with 95.6% of recruited MSM and 85.5% of recruited FSW peers indicating during screening that they would also have given a coupon to the same person who gave them their coupon.
During screening we also enquired about indicators for a potential overlap of networks in terms of being acquainted with a large number of persons from other most vulnerable groups. We found that 12% of the 227 MSM were acquainted with more than five PWID, and 13% of MSM knew more than five FSW. In turn, 17% of the 69 FSW in the sample indicated being acquainted with more than five MSM, and 13% of FSW knew more than five PWID.
Table 1 shows population estimates of sociodemographic characteristics for both groups.
HIV, HBV and HCV Prevalence
For MSM we estimated an HIV prevalence of 3.1%, HBV prevalence of 2.9%, and HCV prevalence of 7.3% (Table 1). None of the MSM were HIV/HBV co-infected, but 2.1% were HIV/HCV co-infected. Among FSW, the estimated HIV-prevalence amounted to 15.7%. The estimated RDS-adjusted HBV prevalence was 0% and there were no cases of HIV/HBV co-infection. In contrast, the HCV prevalence was 5.2%, with an HIV/HCV co-infection rate of 3.7%.
For MSM the median age of first anal intercourse with a man was 17 years. An estimated 58.1% of MSM had their first sexual intercourse with males before reaching 18 years, and 13.8% had a forced sexual debut (Table 2). 44.2% of respondents reported having had >3 anal sex partners during the past 6 months, and 16.9% had group sex at some point in the past. Only 21% of MSM used a condom during last anal intercourse. More than half of the respondents (54.3%) reported ever having used lubricants, of which only 0.6% were the recommended water-based lubricants. 26.5% of MSM had anal intercourse with a commercial partner in the past six months, but only 19.4% of these used a condom at last commercial intercourse.
More than two thirds (68.5%) of MSM also had sexual intercourse with a female partner in the past, and more than half (53%) had intercourse with a woman in the last six months. The median reported age of first intercourse with a woman was 18 years. 38.5% of MSM reported risky sex with both men and women during the past six months. Almost one third (29.5%) of respondents had used non-injecting drugs in the past six months, such as Hashish or Tranquilizers, and 4% reported having ever injected drugs. Only 5.7% of MSM felt they were at high risk for HIV infection.
Table 3 provides information about FSW commercial sex history. 18.2% of FSW reported having been forced to have sex during the past year. 42.8% of FSW had 10–50 sex partners during the past six months and 30.5% had 50 or more. 63.4% of FSW reported consistent condom use during sex with one-time clients, and 56.8% during sex with regular clients. 27.8% of FSW experienced STI symptoms in the past year. Only two respondents had ever injected drugs.
Access to Services, Knowledge and Attitudes
Access to HIV prevention services and HIV-related knowledge was limited (Table 4). Only 0.9% of MSM and 0% of FSW have been exposed to prevention programmes related to both HIV testing and condoms. Although most MSM (94.6%) and FSW (81.7%) knew how to obtain condoms, 0% of MSM and 1.1% of FSW had a condom to hand on the survey day, and only 12.1% of MSM and 49.2% of FSW knew how to use condoms correctly. 45.6% of MSM and 38.6% of FSW had been tested for HIV during the past year and knew the result. Only 16.8% of MSM and 18.6% of FSW succeeded in both, correctly identifying ways to prevent sexual HIV transmission and rejecting major misconceptions about HIV transmission. STI-related knowledge was very poor with only 1.3% of MSM and 5.4% of FSW correctly identifying at least two common STI symptoms in both men and women. Only 13.1% of MSM and 26.2% of FSW reported no stigmatizing attitudes against people living with HIV (PLHIV).
Factors Associated with HIV Seropositivity among MSM
Using Fisher’s central exact test we identified three factors associated with HIV sero-positivity among MSM (Table 2). The odds of HIV infection were 802 times higher among individuals who ever injected drugs, but the confidence interval was very wide (CI 72.7, 4.5e15). This is because some of the numbers were small, with only 1 of the 11 MSM who ever injected drugs testing HIV negative, and only 2 of the 12 HIV positive MSM in the sample indicating that they had never injected drugs. Odds ratios (OR) for associations with HIV sero-positivity could not be computed for the factors “needle sharing at last injection” and “non-injecting drug use in the past six months”, because of “zero cells”, i.e. none of those MSM who shared needles at last injection were HIV negative, and none of those who had never used non-injecting drugs were HIV positive. In addition, MSM who had been at least 18 years old at their first sexual intercourse were more likely to be HIV positive (OR 21.3, CI 3.0, 929.0), but this factor might be confounded by the fact that none of those MSM who had sex before the age of 18 had ever injected drugs. Further, those MSM who perceived themselves at high risk of HIV infection were more likely to be HIV positive (OR 115.0, CI 13.5, 5387.0).
Numbers were too small to fit a model that adjusts for all potential confounders, but the following additional univariate correlations (not shown in table) might assist in the interpretation of some of the results. Respondents who had ever injected drugs were more likely to perceive being at high or medium risk of HIV infection (OR 29.3, CI 4.0, 1292.6), more likely to have taken an HIV test and obtained their result in the 12 months preceding the survey (OR 12.4, CI 1.7, 545.5) and more likely to report using a condom during their last anal sex (OR 4.1, CI 1.0, 17.6) as compared to non-injectors. MSM who used a condom during last anal sex were also more likely to have taken an HIV test in the past year and obtained the result (OR 2.5, CI 1.3, 4.9). MSM who were ≥35 years old were more likely to be HIV positive (OR 82.2, CI 16.4, 570.5), but were also more likely to have ever injected drugs (OR 309.0, CI 35.6, 13424.2).
An HIV prevalence of 3.1% among MSM in Tripoli indicates an HIV epidemic that is approaching the 5%-threshold of a concentrated epidemic . Although the HIV prevalence among FSW is 15.7%, no definite classification is possible due to the small sample size. It is possible that the HIV epidemic among FSW in Libya is already or will soon become concentrated. So far, only a concentrated epidemic among PWID had been confirmed by our recent study in Libya .
There are no previous studies among MSM and FSW in Libya, nor studies among the general population stratified for sexually active males and females, to which our data could be compared , . However, looking at recent reports from other North-African countries, the HIV prevalence among MSM in our study seems comparably slightly lower , , , ,  and the HIV prevalence estimate among FSW comparably higher , , , . However, the prevalence varies for different cities in these countries, and more details on the age distributions would be needed for more informative comparisons as age is often associated with HIV-status.
Our study further detected a relatively moderate HBV prevalence of 2.9% among MSM and a 0% among FSW. In contrast, HCV prevalence was relatively high among MSM (7.3%) and FSW (5.2%). It is interesting that our study did not show any cases of HIV/HBV co-infection, but revealed HIV/HCV co-infection of 2.1% for MSM and 3.7% for FSW. This pattern might point to parenteral routes of HIV transmission given that HCV (in comparison to HBV) is more easily transmitted via unsafe injections than by mucosal exposure during sexual intercourse . However, these results should be interpreted with caution since the confidence intervals include the value zero, indicating the eventuality of no existing co-infection.
Sexual risk factors among MSM included a high number of anal sex partners and low use of condoms and the recommended water-based lubricants. In addition, a substantial proportion of MSM had group sex and commercial sex. Sexual risk factors among FSW included a high number of sex partners, and a condom use rate that was higher than among MSM, but still insufficient to ensure consistent protection. This may explain why a substantial proportion of FSW experienced STI symptoms during the past year. Not only a high rate of partner change, and paid sex, but also a history of STIs have been identified as key risk factors of HIV transmission in both early and advanced epidemics .
Additional risk factors in our study include MSM taking illicit non-injecting drugs. People who use non-injecting drugs have recently been identified as an overlooked population with a substantial burden of HIV due to their high-risk behaviours and overlapping social and sexual networks with PWID . Although only 4% of MSM and two FSW in our study reported ever having injected drugs, IDU is likely to more or less directly fuel the spread of HIV among MSM and FSW. Both FSW and 10 out of the 11 MSM who admitted having used injecting drugs were HIV positive.
Sample size restrictions did not allow us to run reliable correlation analyses among FSW, but the analyses among MSM showed that IDU was the highest risk factor for HIV transmission, although due to small cell counts the confidence interval is very wide and the result has to be interpreted with caution. Given the recently detected HIV prevalence of 87% among PWID in Tripoli , there is a risk for an expansion of the epidemic not only to and within different most vulnerable populations (MVPs), but also to the general population via heterosexual transmission to clients and partners of FSW, PWID and MSM.
There are a few possible explanations for the relatively low prevalence of HIV among MSM detected in our study despite the high level of sexual risk behaviours and the lack of HIV-related knowledge. While the HIV epidemic is not new among PWID, who have one of the highest HIV prevalence rates in the world , it is most likely new among MSM. Furthermore, although a majority of HIV positive respondents reported ever injecting drugs overall there was only a small number of MSM who indicated having ever injected drugs (only 4 %). It is also very likely that the networks of MSM who inject drugs are isolated from the networks of heterosexual drug injectors due to stigma. This, however, does not mean that the situation is stable and could not change. The HIV epidemic among PWID may penetrate MSM networks at any time, leading to a sharp increase in the prevalence of HIV among MSM. Moreover, given that only a fifth of respondents reported having used a condom at last sex, and given that close to 40% of MSM had unprotected sex with both men and women in this time period, the epidemic may quickly transfer to other population groups. An overlap and bridging between different MVPs has been described elsewhere in the Middle East and North Africa (MENA) , . For example, in Iran and Pakistan, newly emerging epidemics among MSM have been linked to well established epidemics among PWID , .
A further risk factor for an expansion of the epidemic is the criminalization of FSW and MSM that pose profound structural barriers to HIV prevention and care in the MENA region , . It drives the neediest populations away from services, increases their fear of disclosing risks to partners and providers, and renders them vulnerable to blackmail and abuse . In Libya, sex work can result in a penalty of up to five years imprisonment for sex worker and client, and corporal punishment (100 lashes by flogging) of the sex worker. Legislation can apply the same law to punish MSM with imprisonment .
Fear of stigma and persecution was also a reason for the relatively slow recruitment of FSW. Slow FSW recruitment has also been reported elsewhere , , . Given the outbreak of political turmoil in Libya, the FSW study prematurely ended and the target sample size could not be met. Hence, the FSW results, especially for those indicators, for which equilibrium could not be reached, have to be interpreted with caution, as they may not be representative for the whole FSW population.
Another study limitation is that, as in all cross-sectional studies, the relationships between cause and effect cannot be determined. Further, given that behavioural data were self-reported, social desirability bias might have led to over-reporting of preventive behaviour, such as condom use, and under-reporting of certain risk factors, such as non-injecting and injecting drug use, or being the receptive rather than the insertive sex partner. While 98.0% of MSM reported having insertive anal sex in the six months preceding the survey, only 6.3% admitted practicing receptive anal sex. Our preliminary social mapping of MSM revealed that practicing receptive sex was perceived as particularly stigmatizing among peers. Hence, it is possible that respondents tended to give a culturally more acceptable answer, claiming they practiced only insertive anal sex. Given that trafficking is frequent in the region, and that many FSW fear disclosing forced sex , it is also likely that some FSW (e.g. those held in connection houses or who met clients at their sex partner’s house) might have underreported forced sex as a reason for engaging in sex work. New studies are needed to explore the different types of FSW and MSM sub-groups in Libya, with results disaggregated for groups that might be particularly vulnerable to HIV, such as transgender people .
Despite the challenges and limitations described above, we showed that it is possible to engage MSM and FSW in Libya under difficult (pre-revolutionary) conditions and in the absence of pre-established links with hidden MVPs via NGOs. The results of the study show that an immediate response is crucial, in order to prevent the epidemic from spreading within different risk groups and to the general population. The risk of transmission is further fuelled by the recent political military events that led to increased migration, sexual and gender-based violence, the disruption of infection control and blood safety systems, and a nation-wide stock-out of antiretroviral drugs .
Comprehensive programmes for the prevention, treatment and care of HIV/AIDS and other STIs need to be rapidly and simultaneously scaled up among all MVPs and take account of the potential overlap of high-risk sexual and drug-using networks. Following international recommendations , ,  legal and policy barriers to the scale-up of effective programmes should be removed, and attempts made to establish meaningful collaborations with newly founded civil society groups representing MVPs and PLHIV. Thereby, experiences from other MENA countries could be taken into account. For example, Djibouti has a history of protecting the rights of PLHIV, Iran has led in the development of a harm-reduction programme, and NGOs in Morocco have successfully worked with MSM and FSW for HIV and STI prevention .
A likely escalation of the epidemic in Libya can be averted if vigilant surveillance and evidence- and human-rights based programmes are implemented while the window of opportunity to act is still open .
Table S1 – Socio-demographic characteristics and prevalence of HIV and other infections among MSM in Tripoli, Libya, 2010. Table S2 – Socio-demographic characteristics and prevalence of HIV and other infections among FSW in Tripoli, Libya, 2010. Table S3 – Sexual behaviour and risk factors for HIV infection among MSM in Tripoli, Libya, 2010. Table S4 – Sexual behaviour and risk factors for HIV infection among FSW in Tripoli, Libya, 2010. Table S5 – Access to services, knowledge and attitudes related to HIV among MSM in Tripoli, Libya, 2010. Table S6 – Access to services, knowledge and attitudes related to HIV among FSW in Tripoli, Libya, 2010.
The authors gratefully acknowledge Ms. Amal el Karouaoui, Dr. Lisa G. Johnston, Prof. Mehdi Karkuri, Dr. Abdalla Toufik, Dr. Simon D.W. Frost, Mr. William Vargas, Mr. James Smedley, and Prof. Patricia Cane for their consultancy work. We are also very grateful to Prof. Abdulhafid Abudher, then director of the NCDC, Dr. Omar El Ahmar, director of the NCDC laboratory and his team, as well as to Dr. Nabil Abumaer, then head of Department of HIV/AIDS Treatment-Management. We would also like to express our sincere gratitude to Dr. Fannir Abdallah, Mr. Aminou Mamane, Ms. Sarah Moxon, Mr. Abdullah Aljadi, Mr. Owais Al Wahedi, Ms. Nelly Manasaryan and all members of the study team in Tripoli, the data management teams at the NCDC and at LSTM. Last, but not least, we wish to thank all those who participated in this survey for their time and trust.
Disclaimer: This publication has been produced with the financial support of the European Commission. The contents of this publication are the sole responsibility of the authors and can in no way be taken to reflect the views of the European Union.
Conceived and designed the experiments: JJV LM SB. Performed the experiments: LM AAT RS. Analyzed the data: JT LD SB CJ. Wrote the paper: SB. Obtained funding: JJV. Led study preparations: LM. Negotiated the study with the Libyan government: JJV LM. Contributed to the study preparations: JJV AAT HO SB CJ. Provided technical assistance to the data collection: HO JJV SB CJ. Supported the data analysis and all authors contributed to the interpretation of results: JJV LM RS. Contributed to the writing of the manuscript, and read and approved the final manuscript: JJV SB CJ JT HO LD AAT RS LM.
- 1. Abu-Raddad L, Ayodeji Akala F, Semini I, Riedner G, Wilson D, et al.. (2010) Characterizing the HIV/ AIDS Epidemic in the Middle East and North Africa: Time for Strategic Action. Washington, DC: The World Bank.
- 2. UNAIDS (2009) NAIDS Action Framework: Universal access for men who have sex with men and transgender people. Geneva: Joint United Nations programme on HIV/AIDS (UNAIDS).
- 3. National AIDS Program, Libyan Arab Jamahiriya (2012) UNGASS Country Progress Report Libya, Reporting Period: January 2010-December 2011.: Ministry of Health, Libya.
- 4. Friedman SR, Bolyard M, Mateu-Gelabert P, Goltzman P, Pawlowicz MP, et al. (2007) Some data-driven reflections on priorities in AIDS network research. AIDS Behav 11: 641–651. doi: 10.1007/s10461-006-9166-7
- 5. Rachlis B, Brouwer KC, Mills EJ, Hayes M, Kerr T, et al. (2007) Migration and transmission of blood-borne infections among injection drug users: understanding the epidemiologic bridge. Drug Alcohol Depend 90: 107–119. doi: 10.1016/j.drugalcdep.2007.03.014
- 6. Baral S, Beyrer C, Muessig K, Poteat T, Wirtz AL, et al.. (2012) Burden of HIV among female sex workers in low-income and middle-income countries: a systematic review and meta-analysis. Lancet Infect Dis.
- 7. van Griensven F, de Lind van Wijngaarden JW, Baral S, Grulich A (2009) The global epidemic of HIV infection among men who have sex with men. Curr Opin HIV AIDS 4: 300–307. doi: 10.1097/coh.0b013e32832c3bb3
- 8. Baral S, Sifakis F, Cleghorn F, Beyrer C (2007) Elevated risk for HIV infection among men who have sex with men in low- and middle-income countries 2000–2006: a systematic review. PLoS Med 4: e339. doi: 10.1371/journal.pmed.0040339
- 9. Mumtaz G, Hilmi N, McFarland W, Kaplan RL, Akala FA, et al. (2010) Are HIV epidemics among men who have sex with men emerging in the Middle East and North Africa?: a systematic review and data synthesis. PLoS Med 8: e1000444. doi: 10.1371/journal.pmed.1000444
- 10. UNAIDS (2009) UNAIDS Guidance Note on HIV and Sex Work. Geneva: Joint United Nations Programme on HIV/AIDS (UNAIDS).
- 11. El Karouaoui A (2009) Project to Support the Development and Implementation of the National HIV/AIDS Strategy in Libya: Mapping for populations most at risk to HIV/AIDS, Tripoli- 2009. Liverpool: Liverpool School of Tropical Medicine (LSTM).
- 12. Abu-Raddad LJ, Hilmi N, Mumtaz G, Benkirane M, Akala FA, et al. (2010) Epidemiology of HIV infection in the Middle East and North Africa. AIDS 24 Suppl 2S5–23. doi: 10.1097/01.aids.0000386729.56683.33
- 13. Mirzoyan L, Berendes S, Jeffery C, Moir J, Bin Othman H, et al. (2013) New evidence on the HIV epidemic in Libya: why countries must implement prevention programmes among injecting drug users. J Acquir Immune Defic Syndr. 62(5): 577–583. doi: 10.1097/qai.0b013e318284714a
- 14. UNAIDS/WHO, Working Group on Global HIV/AIDS and STI Surveillance (2000) Guidelines for second generation HIV surveillance. Geneva: World Health Organization and Joint United Nations Programme on HIV/AIDS.
- 15. Kamal SM, Mahmoud S, Hafez T, El-Fouly R (2010) Viral Hepatitis A to E in South Mediterranean Countries. Mediterr J Hematol Infect Dis 2: e2010001. doi: 10.4084/mjhid.2010.001
- 16. Yerly S, Quadri R, Negro F, Barbe KP, Cheseaux JJ, et al. (2001) Nosocomial outbreak of multiple bloodborne viral infections. J Infect Dis 184: 369–372. doi: 10.1086/322036
- 17. Roberts RJ (2006) An open letter to Colonel Muammar al-Gaddafi. Nature 444: 146. doi: 10.1038/444146a
- 18. Nature Publishing Group (2007) Free at last. Nature 448: 511. doi: 10.1038/448511a
- 19. Malekinejad M, Johnston LG, Kendall C, Kerr LR, Rifkin MR, et al. (2008) Using respondent-driven sampling methodology for HIV biological and behavioral surveillance in international settings: a systematic review. AIDS Behav 12: S105–130. doi: 10.1007/s10461-008-9421-1
- 20. Heckathorn DD (2002) Respondent-driven sampling II: deriving valid population estimates from chain-referral samples of hidden populations. Soc Probl 49: 11–34. doi: 10.1525/sp.2002.49.1.11
- 21. Salganik MJ, Heckathorn DD (2004) Sampling and estimation in hidden populations using respondent-driven sampling. Sociological Methodology 3: 193–239. doi: 10.1111/j.0081-1750.2004.00152.x
- 22. Salganik MJ (2006) Variance estimation, design effects, and sample size calculations for respondent-driven sampling. J Urban Health 83: i98–112. doi: 10.1007/s11524-006-9106-x
- 23. UNAIDS/WHO Working Group on Global HIV/AIDS and STI Surveillance (2008) Epidemiological Fact Sheet on HIV and AIDS, Core data on epidemiology and response: Sudan, 2008 update, September 2008.
- 24. Ministère de la Santé de la Population et de la Réforme Hospitalière Ddlp, Comité National de lutte contre les IST/VIH/sida (2008) Rapport National de l’Algérie sur le suivi en janvier 2008 de la Déclaration d’engagement de la Session extraordinaire de l’Assemblée générale des Nations Unies sur le VIH/sida (UNGASS, 2001), Période couverte : janvier 2006 – décembre 2007. Geneva, Switzerland.
- 25. UNAIDS (2010) 09 United Nations General Assembly Special Session on HIV/AIDS - Monitoring the Declaration of Commitment on HIV/AIDS: guidelines on construction of core indicators : 2010 reporting. Geneva: UNAIDS.
- 26. Family Health International (2000) Behavioral Surveillance Surveys: Guidelines for Repeated Behavioral Surveys in Populations at Risk of HIV. Arlington, VA: Family Health International.
- 27. Frost SD, Brouwer KC, Firestone Cruz MA, Ramos R, Ramos ME, et al. (2006) Respondent-driven sampling of injection drug users in two U.S.-Mexico border cities: recruitment dynamics and impact on estimates of HIV and syphilis prevalence. J Urban Health 83: i83–97. doi: 10.1007/s11524-006-9104-z
- 28. Johnston LG (2007) Conducting respondent driven sampling (RDS) studies in diverse settings: A training manual for planning RDS studies. Centers for Disease Control and Prevention, Atlanta, GA and Family Health International, Arlington, VA.
- 29. WHO EMRO (2008) Inclusion of HIV rapid tests in HIV testing strategies in countries of the Eastern Mediterranean Region.
- 30. Volz E, Heckathorn DD (2008) Probability based estimation theory for respondent driven sampling. Journal of Official Statistics 24: 79.
- 31. Ministère de la Santé de l’Algérie (2012) Rapport d’activité sur la riposte nationale au sida - Algérie 2012: Suivi de progrès sur la déclaration politique sur le VIH/SIDA de 2011.
- 32. National AIDS Program E (2012) Global AIDS Response Progress Report 2012, Arab Republic of Egypt.
- 33. Royaume du Maroc (2012) Mise on œuvre de la déclaration politique sur le VIH/sida, rapport national 2012 - Période considérée: Janvier 2010-décembre 2011.
- 34. République Tunisienne, Ministère de la Santé Publique (2012) 2010–2011 Rapport d’activité sur la riposte au SIDA – Tunisie, Mars 2012. République Tunisienne, Ministère de la Santé Publique, Direction des Soins de Santé de Base, Programme National de Lutte contre le sida et les MST.
- 35. Abdelrahim MS (2010) HIV prevalence and risk behaviors of female sex workers in Khartoum, north Sudan. AIDS 24 Suppl 2S55–60. doi: 10.1097/01.aids.0000386734.79553.9a
- 36. Lavanchy D (2009) The global burden of hepatitis C. Liver Int. 29 Suppl 174–81. doi: 10.1111/j.1478-3231.2008.01934.x
- 37. Chen L, Jha P, Stirling B, Sgaier SK, Daid T, et al. (2007) Sexual risk factors for HIV infection in early and advanced HIV epidemics in sub-Saharan Africa: systematic overview of 68 epidemiological studies. PLoS One 2: e1001. doi: 10.1371/journal.pone.0001001
- 38. Strathdee SA, Stockman JK (2010) Epidemiology of HIV among injecting and non-injecting drug users: current trends and implications for interventions. Curr HIV/AIDS Rep 7: 99–106. doi: 10.1007/s11904-010-0043-7
- 39. Pakistan National AIDS Control Program (2011) HIV second generation surveillance in Pakistan - National report round IV. National AIDS Control Program.: Canada-Pakistan HIV/AIDS Surveillance Project.
- 40. UNAIDS (2011) Middle East and North Africa Regional Report on AIDS - 2011. Joint United Nations Programme on HIV/AIDS.
- 41. Beyrer C (2010) Global prevention of HIV infection for neglected populations: men who have sex with men. Clin Infect Dis 50 Suppl 3S108–113. doi: 10.1086/651481
- 42. Simic M, Johnston LG, Platt L, Baros S, Andjelkovic V, et al. (2006) Exploring barriers to ‘respondent driven sampling’ in sex worker and drug-injecting sex worker populations in Eastern Europe. J Urban Health 83: i6–15. doi: 10.1007/s11524-006-9098-6
- 43. Uuskula A, Johnston LG, Raag M, Trummal A, Talu A, et al. (2010) Evaluating recruitment among female sex workers and injecting drug users at risk for HIV using respondent-driven sampling in Estonia. J Urban Health 87: 304–317. doi: 10.1007/s11524-009-9427-7