An analysis of HIV-related risk behaviors of men having sex with men (MSM), using respondent driven sampling (RDS), in Albania.
The question of how to access hard to reach populations has been troubling, both domestically and internationally. When discussing the epidemiology of the HIV virus, it is assumed that certain high risk populations are fundamental in the proliferation of the virus, particularly at the beginning of a country's epidemic. The problem however, has been how to reach these populations, particularly in countries where activities such as intravenous drug use, commercial sex and same-sex encounters are so highly stigmatized. The use of a Respondent Driven Sampling (RDS) appears to offer a solution to that dilemma, allowing researchers to access hard to reach populations through their social networks. (1,2) This paper presents results utilizing RDS amongst Men Having Sex with Men (MSM) in Tirana, Albania to identify behaviors relating to the potential spread of HIV and other sexually transmitted infections (STI). This paper reports the results of the second representative study of MSM behaviors in Albania (the first one was conducted in 2005), and allows a comparison of results between two studies. The methodology used will be discussed, along with a comparison of key indicators and a discussion of results and study limitations.
Respondent Driven Sampling (RDS) was used to sample target group members. The methodology behind RDS2, in brief, is a referral based system whereby participants refer only two to three other members of the target group that they know to participate and receive an incentive for both participating and successfully recruiting others. Data are analyzed using the recruiting linkages and yield a representative sample of the population in the target area. The surveillance in Tirana was part of a larger survey investigating biological prevalence of HIV and other sexually transmitted infections and behaviors and knowledge associated with the spread of these infections. Injecting Drug Users, and the Roma population were also surveyed during the same time period (July-August, 2008).
One hundred and eighty nine MSM were surveyed in one location, offices of a MSM NGO conducting outreach and prevention programs. Data collection included face to face interviews and encompassed a 5 week period with hours of data collection from 9:00-16:00 Monday-Saturday.
In accordance with RDS methodology, the initial participants of the study, known as "seeds" (3,4) were members of the target group.
Twelve seeds were selected at the beginning of the study, and because of diversity of the seeds, each seed received three coupons. As of wave 1, participants received two recruitment coupons. Fewer coupons help ensure that recruitment chains are long and linear as opposed to short and wide. Coupons were numbered in such a way that the original recruiter was recognized and that the chains and number of waves could be tracked.
Seeds were selected from diverse geographic areas in Tirana, married and unmarried, of different self-identified types (homosexual, gay, bisexual) and of different socioeconomic level.
Participants received cash incentives of the equivalent of 10 USD for participating and an additional 5 USD for each successful recruit. The study teams consisted of 2 interviewers, a study supervisor, a trained VCT counselor, physician and laboratory technician. Data collection included biological samples for HIV and other STI prevalence and a questionnaire relating to behavior.
The questionnaire topics included socio demographic indicators, drug and alcohol use, sexual history, behavior and condom use, knowledge of STI symptoms and transmission routes and previous HIV testing. The survey participation lasted approximately 1.5 hours and refusal rates were low. Less than 10% of those approached to participate refused to accept the coupon from participants. The study was approved by the Medical Ethical Committee of Albania. Data were entered in SPSS version 12, and all the data were double entered for quality control. Ranges were reviewed in SPSS to ensure cleanliness and the data were analyzed using RDS Analysis Tool version 6.0.1.
MSM did report a high level of alcohol use; estimates indicate that 42.2% of the MSM population consumes alcohol daily.
In terms of drug use, an estimated 64.8% of MSM have tried drugs (injected or non-injected). Population estimates indicate that more than half of MSM inject drugs (59.2%), with heroin being the most frequently injected drug among MSM in Albania
The MSM population is sexually active, with almost 90% estimated to have had sex in the past 6 months. The overwhelming MSM respondents (89.7%) had anal sex in the past 6 months. The questionnaire included questions related to the type of anal intercourse in which the respondent was involved: insertive or receptive. It is estimated that MSM who are involved in sexual activities seem to be almost equally insertive and receptive partner. More than half of MSM who are the insertive partner had more than four partners.
Sex with commercial partners is frequent, with 74.2 % of MSM did report to have had anal sex with a commercial partner in the 6 months preceding the survey. Out of MSM who have had a commercial sex partner, the majority (77.1%) used a condom with that partner during the last anal sex. On the other hand, almost 58% (only 16% for the first Bio BSS)7 used a condom consistently during every anal sex act with a commercial sex partner.
More than half of the MSM population in Tirana (55.7%) did report to have had sex with a non-commercial sex partner in the past 6 months. A non-commercial sex partner is defined as a person with whom the respondents have had sexual relationships without paying or being paid for it. Multiple partnerships are frequent, with an estimated 34.2%. of MSM who have had sex with a non-commercial partner in the past 6 months having had 5 or more such partners. About 60% used a condom at last sex, with 44% reported consistent condom use with noncommercial partners. Reasons for not using condoms for both types of partners are diverse with more frequent response "not liking them".
Sex with female partners seems to be common; almost 50% in the Tirana MSM network have had sex with a female. It is estimated that 84% of those had sex with a female in the past 6 months, and the majority 71.3% having 1-3 female partners.
One section of the questionnaire used with MSM was dedicated to exploring their knowledge, opinions, and experiences regarding HIV/AIDS and STIs. 66% has heard about HIV/AIDS and 72.3% had heard about diseases that can be transmitted through sexual intercourse. MSM have more formal knowledge about HIV/AIDS than about the above-mentioned STIs. The majority know of personal preventive strategies. An estimated 77.2% also know that HIV can be transmitted by using previously used needles, and 73.4% know that a pregnant mother can transmit the virus to her unborn child. The 96% of MSM (10% in the first Bio BSS)7, recognized that ARV drugs could reduce the risk of mother-to-child transmission of HIV. When it comes to knowledge about modes of transmission, some MSM believe, incorrectly, that there is a risk of transmission when there really is not. For example, almost one-third (30.4%) of MSM know that HIV is not transmitted through mosquito bites and 45.2% know that HIV is not transmitted by sharing a meal with someone living with HIV.
The majority of MSM are estimated to know that confidential HIV testing is available in Tirana with 78.3%. All those who know about the test have had and HIV test. The sample characteristics indicate that of those MSM who had been tested, two in three had taken the HIV test voluntarily, and that the majority of those tested had received their test results. Most of the MSM tested have taken the test in the last 12 months (71.2%).
Regarding the biological data, it is estimated that the prevalence of HIV infection among MSM in Tirana is 1.8%. Among the MSM sample, the rate of syphilis was 2.6 %, and the rate of hepatitis C was 3.5%.
RDS network size and recruitment information
RDS data also provide information on recruiting patterns among respondents. To determine if recruiters and their recruits are similar, RDS calculates a homophily (H) measure of self bias. For example, if homophily equals one (H = 1), then all network ties are formed within a group (e.g., illiterate recruit only illierate). If homophily equals zero (H = 0), then all network ties are formed randomly. And if homophily equals minus one (H = -1), then all network ties are formed out of a group (e.g., illiterate recruit only those with university); a score of H = -1 is also referred to as heterophily. (2,3,5) Acceptable homophily scores range between H = -0.3 and H= +0.3. Recruitment patterns and homophily scores were examined with respect to age and education. These factors are important because they provide a better understanding of the demographic characteristics of this hidden population.
MSM were asked specifically about the number of MSM they knew and who likewise knew them.
The RDS software, with this information, was used to calculate adjusted network sizes for MSM by age and by education. Overall, the adjusted average personal network size for MSM did not vary according to age. The adjusted network size for MSM for those [less than or equal to] 24 years of age was 5.2 persons, while that for those ages [greater than or equal to] 25 and older was 4.9 persons. In terms of education, the adjusted network size was larger among MSM who had completed university. (7,8)
The network structure of age recruiting relations reflects no homophily nor heterophily in respect of age. The homophily indices were -0.04 for [less than or equal to] 24 years old and -0.065 for those [greater than or equal to] 25. As can be seen the first category age group ([less than or equal to] 24) 58% of the time recruited the members of the other category. They recruited amongst themselves 42% of the time. While the other category ([greater than or equal to] 25) 47% of the time recruited the other category, and 53% among themselves.
The homophily indices for MSM recruitment for education show that MSM with university have recruited only peers from other groups (H= -1.0) resulting in score indicating heterophily. Overall, MSM from all level of education tended to recruit participants with secondary education (37.5% of those with no education, 33.3% with primary, 50% with secondary, and 50% of those with university) recruited peers with secondary education.
The research conducted in Tirana with MSM is very important, given the fact that a representative sample has been undertaken to investigate behaviors relating to the spread of HIV and other sexually transmitted infections. This study has provided policy makers and those involved in prevention of HIV and STIs with information to better plan and target interventions. It has provides a reference point for which planning and resource acquisition can occur. It has long been hoped that the conservative nature of many of the countries in the Balkans has provided a measure of protection against the HIV virus. (6) Based on the results of this study, it would appear that the HIV epidemic is in it's early stages in Albania, however reported behaviors indicate that an increase in prevalence is likely to occur without increased funding and efforts to control the virus. (8)
MSM network is young, with almost 44% of the population below the age of 25 years old. Less MSM than in the first study (almost 16% in comparison with about one-third in the first Bio-Bss)7 are mobile. Almost 43% drink alcohol every day, and the majority (61.3%) have tried drugs (mainly marijuana, cocaine, heroin, ecstasy, and valium).
Among MSM, 26.1% have not attended school and 38.1% have ever been married to a female. These socio-demographic factors are vital for developing appropriate behavior change activities for this population. In terms of risk behaviors, high percentages of MSM have ever injected drugs (almost 60%) (mainly heroin) and 48.6% have injected heroine last month. Almost 90% of MSM have engaged in anal and oral sex with multiple partners in the past 6 months. The majority of MSM (74.2%) also reported having engaged in sex with male commercial partners, 85% of whom had two or more such partners. Despite these high-risk behaviors, and the fact that most MSM know they can obtain condoms at a pharmacy, consistent condom use was much higher than the figures reported during the first round of the Bio BSS, (57.8% with commercial partners and 44% with noncommercial partners). In addition, 40.7% of MSM reported using lubricants. In regard to risk behaviors with males, MSM respondents also reported having recent sex with females. In the 6 months preceding the Bio-BSS, almost half of MSM reported having sex with multiple female partners (MSM who have had two or more female partners). The 42.3% used condoms consistently with their female partners (one in four MSM for the first round of the Bio BSS).
Overall, the combination of MSM risk factors-engaging in unprotected sex with males and females, and injecting drugs-makes it imperative that this group be targeted as part of any HIV prevention strategy in Albania. HIV prevalence rates were similar to the first Bio BSS in terms of absolute number of MSM HIV positive, while the population estimates show an increase of up to 1.8%. Even this round of the Bio-BSS did not include information about acute STIs, which could provide more in-depth insight into the associations between behaviors and infections. Nonetheless, MSM are engaged in several risk behaviors with different populations, making them a possible link between drug-using populations and female populations in Tirana.
Additionally, continued movement across borders and neighboring countries where prevalence amongst high risk groups is higher (i.e. to Greece or other and Eastern European countries) provide access to other populations where prevalence is higher- thus creating a migration effect. (9)
Although homosexuality was decriminalized in 1995, MSM remain deeply stigmatized and discriminated against. As a result, they felt considerable anguish about their sexuality and kept it hidden, and to all appearances lead conventional heterosexual lives, including marriage. MSM reported being reluctant to contact the public health service about STIs for fear of their sexual orientation being discovered, only going when the symptoms were severe. (9,10) Many MSM reported extreme frustration at their social and economic circumstances, while nearly one third reported having suicidal thoughts.
It is possible that limitations to the study could affect the results. Data collection took place over a 5 week period in middle June-July. Had data collection been undertaken at other points of time, perhaps results would differ.
Furthermore, the data is representative only of the capital city- had data been collected in other cities or in rural areas, results could also differ.
Finally, the body of literature regarding the use of Respondent Driven Sampling amongst hard to reach populations in developing countries is still limited. Little is known about the effect of incentives on populations in poor countries, or if random recruiting (a fundamental assumption underlying the RDS methodology) is valid in countries where this group is highly stigmatized and where a proportion of the population is poor. A greater body of research employing these methods in developing countries is needed.
Because MSM participants were recruited through RDS, it is likely that the study sampled from a wide range of MSM thereby increasing the sample representativeness. Study methodology likely drew a more diverse sample of MSM compared to traditional recruitment by outreach workers or through needle exchange programs. It greatly enhances our understanding of MSM and their network characteristics, and also provide evidence for the effectiveness of using RDS in recruiting hard-to-reach populations in environments where stigmatization and criminalization against their behaviors are commonplace.
It is important to note that unprotected anal sex is one of the most risky behaviors for HIV/AIDS transmission. The risk of becoming infected with HIV is significant for both partners involved in unprotected anal sex, but the receptive partner has a much higher risk of becoming infected than the insertive partner. When developing a strategy to address the transmission of HIV, it is essential to take into account that men have sex with other men for a wide range of reasons and under very different circumstances. Because anal sex also takes place within heterosexual relations, it is essential to understand that what is risky is having unprotected anal sex rather than having sex with other men.
The authors are grateful to the Institute of Public Opinion and the Institute of Public Health for giving the opportunity to further explore the Biological and Behavioral Surveillance Study 2008.
Conflict of interest: None declared.
(1.) Heckathorn DD. Respondent driven sampling II: deriving valid population estimates from chain-referral samples of hidden populations. Soc Probl. 2002;49(1):11-34.
(2.) Heckathorn DD. Respondent-driven sampling: a new approach to the study of hidden populations. Soc Probl. 1997;44(2):174-199.
(3.) Salganik M, Heckathorn DD. Sampling and estimation in hidden populations using respondent-driven sampling. Sociol Method. 2004;34:193-239.
(4.) Johnston, Lisa G., Rasheda Khanam, Masud Reza, Sharful I. Khan, Sarah Banu, Md. Shan Alam, Mahmudur Rahman, and Tasnim Azim. 2007.http://www.respondent drivensampling.org/reports/AIDSbeh_2007.pdf" The Effectiveness of RDS for Recruiting MSM in Dhaka, Bangladesh." AIDS & Behavior. 2006 Special issue on RDS: Journal of Urban Health, Vol. 83, No. 7
(5.) Heckathorn, Douglas DD. http://www.respondentdrivensampling.org/reports/ Heckathorn2007Preprint.pdf Extensions of Respondent-Driven Sampling: Analyzing Continuous Variables and Controlling for Differential Recruitment." Sociological Methodology, 2007
(6.) Rhodes T, Simic M. Transition and the HIV risk environment. BMJ, 2004, 331:220-223
(7.) Behavioral and Biological Surveillance Study, Institute of Public Health, Family Health International, 2005
(8.) HIV/AIDS epidemiological situation in 2008. Tirana, Albanian Institute of Public Health
(9.) Rapid assessment of and response to HIV/AIDS in Albania among Roma and young men who have sex with men. Geneva, UNICEF, Institute of Public Opinion, 2004.
(10.) Rapid assessment and response on HIV/AIDS among young MSM. Tirana, Institute of Public Opinion and UNICEF, 2004.
Corresponding author: Elda Sharra, MD, MPH, Rr. "Aleksander Moisiu", No. 80, Tirana, Albania; E-mail: email@example.com
Elda Sharra , Roland Bani 
 Global Fund Program, Tirana, Albania;  Institute of Public Health, Tirana, Albania
Table 1. Demographic information Characteristic Estimated Population Proportion % (95% CI) Age Median 26.5 [less than or equal to] 24 43.9 (35.4-52.3) [greater than or equal to] 25 56.1 (47.7-64.6) Highest education completed Illiterate 26.1 (18.9-35.7) Primary School 16.8 (11-23.5) Secondary School 35.1 (26.7-43.3) Higher 12 classes 16.4 (9.3-23) University 2.8 (0.9-4.8) Religion Muslim 68.6 (62.7-83.4) Catholic 6.5 (NC) Orthodox Christian 9.6 (5.8-21.3) No religion 3.2 (0-4.9) No answer 6.2 (0-8.3) Married/Currently Living with Female Sex Partner Ever married to a female 38.1 (29.5-46.9) Currently married or living with 22.4 (13.7-27.7) female sexual partner Spent 1 month or more away 15.8 (8.6-24.1) from Tirana in past year Table 2.MSM Adjusted Network Size, by Selected Characteristics Network Size Characteristic (Adjusted) Age < 24 5.2 > 25 4.9 Education None 4.4 Primary 5.5 Secondary 5.8 Higher 3.9 University 7.8
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||ORIGINAL ARTICLE; human immunodeficiency virus|
|Author:||Sharra, Elda; Bani, Roland|
|Publication:||Archives: The International Journal of Medicine|
|Date:||Apr 1, 2009|
|Previous Article:||A practical method to diagnose muscle degradation in normonourished patients with chronic heart failure.|
|Next Article:||Study of the relationship between perceived maternal parenting and self-conscious affects.|