|
Programs |
|||
|
|
||
II Objectives Primary objective:
Secondary objectives:
III Methods MAPPING To evaluate the need for inclusion of MSM in the FHI/IMPACT HIV/STI prevention program, Family Health International (FHI) / IMPACT undertook a mapping exercise of MSM between February and May 1999 in Phnom Penh. The objectives of this mapping were to determine the locations where MSM congregate in Phnom Penh and to estimate the number of MSM who frequent these locations. Due to the hidden nature of this population and difficulty identifying them, a snowball approach was used for the mapping. In snowball sampling, key informants in a sub-population identify other members of their community, or in this case other locations where MSM congregate. The people in each cluster are contacted, and they in turn identify further contacts. The process goes on until an adequate mapping is achieved and/or the number of sites exhausted. One fieldworker, who is a member of the Phnom Penh MSM community, was involved in the mapping exercise. The FHI/IMPACT Behavior Change Intervention (BCI) and Evaluation officer supervised the fieldworker. At the research sites MSM were asked what other locations they knew and/or frequent to meet partners. The fieldworker then went to the newly identified locations and asked the same questions to new contact persons. Communication was established before asking questions. This was important in order to enable information exchange with this stigmatized group. As a result of the effort to first establish a relationship for communication, some MSM offered to personally introduce the fieldworker to new locations. In general five to ten persons were approached and interviewed per location. The fieldworker did not tape or record the collected data on site in order to ease the discussion and avoid mistrust. All collected data was entered into a database at the FHI office the next day. The exercise showed that there were MSM networks in Phnom Penh, however they were discrete enough that they were not obvious to the general population. The study team identified twenty-seven locations where MSM meet/gather including parks, karaoke bars, brothels, discotheques, massage parlors, cinemas and streets were identified during this exercise. The venues most frequented by MSM were parks (10) throughout the city followed by karaoke bars (7) and discotheques/nightclubs (4). Because of the floating nature of the population, it was not possible during this mapping to measure the exact number of MSM in Phnom Penh but it was possible to estimate the number of MSM at each location. An average of 19 MSM per location was recorded in a given evening with a range of five to 50. In addition, field workers estimated that fifty percent of the MSM identified might be selling sex services to clients – both to other males and, to a lesser extent, to females. Because the mapping research raised additional questions related to the vulnerability of MSM, it was decided to develop further research on male-to-male sex behaviors in Phnom Penh. In April and May 2000, directly prior to the implementation of this cross-sectional study, a second mapping exercise was conducted. It was important to update the mapping since MSM meeting locations can change from month to month due to police crackdowns or pressures from the community living in those areas. To ensure that accurate data was available to design the sample frame, the methodology used for the second mapping was the same as used for the first mapping. This second mapping identified 16 locations compared to 27 that were identified in the 1999 mapping exercise. The findings of the mapping exercise were then used to develop the sampling framework as detailed below. STUDY DESIGN The protocol for this study was approved by Family Health International's Protection of Human Subjects Committee (PHSC) and the Ministry of Health in Cambodia A cross-sectional study design was employed at 16 selected locations, as defined by the mapping exercise, in Phnom Penh, Cambodia. The study population for the survey was men found in these locations who reported having sex with other men. The sample size was 200, and in order to reach the most representative sample of this floating population, a two-stage time-location sampling methodology was used. A total of 87 time-location clusters, made up of 16 locations, were defined. Forty-one clusters were randomly selected using equal probability sampling at the first stage. At the second stage, a fixed number of five individuals per cluster were randomly selected. Two drop-in centers were established for the study where interviews and specimen collection were conducted. Two study teams of four men were organized, each consisting of one supervisor, one medical assistant, one interviewer, one outreach worker and one "count worker" responsible for counting all MSM at each location to calculate sampling probabilities for each cluster. All of the men in the teams were Cambodian and MSM. Prior to the collection of data, field teams were trained in interviewing skills, sampling methodology, desensitization on sexual practices among MSM, and instructions on administering the questionnaire. Medical assistants practiced appropriate specimen collection for two weeks at an STI clinic. One-week of pilot testing was organized to field test the sampling approach and the tools used for this study. Questionnaires and core indicators from the Guidelines for Repeated Behavioral Surveys in Population at Risk of HIV8 were adapted for use in this study. The questionnaire was written in English, and then translated into Khmer. Inclusion criteria for recruitment into the study included self-reported male-to-male sex behavior, inclusive of non-penetrative sex, during the previous 12 months and being age 18 years or older. No effort was made to categorize MSM into sub-groups, such as male sex workers, transvestites, self-identified gays or others as not enough was known about the extent to which these categories were appropriate. Potential participants were approached and told about the study. If they agreed to participate, they were accompanied to one of the drop-in centers established for the study. Witnessed oral informed consent was administered at the study site and no identifiers were recorded. Participants were able to stop the interview and specimen collection process at any time. After completing the interview and administering the questionnaire, participants underwent a medical and physical examination. Anal specimens were obtained from all participants regardless of symptoms, and urethral specimens were taken from those presenting with urethral discharge. Specimens were inoculated onto Modified Thayer Martin Media in the clinic and placed in CO2 extinction jars. All men were asked to provide a urine sample, and blood was collected for HIV and syphilis serology. Rectal and urethral swabs and urine specimens were collected for the detection of Neisseria gonorrhoeae and Chlamydia trachomatis. Individuals with symptoms suggestive of an STI were treated syndromically without charge. Education and counseling were offered to each participant. Shower facilities were also offered to all participants. A small incentive consisting of a t-shirt and condoms were provided to the participants after completion of the interview, medical examination and specimen collection. A card with an ID number was given to the participants and they were invited to come back to obtain their STI results one week later. The results of the HIV test were not given at the drop-in center because of the absence of trained counselors in the team. Instead, each participant was offered a referral and transportation to a local anonymous testing site for free voluntary counseling and testing. LABORATORY METHODS Serologic testing for HIV and syphilis, Neisseria gonorrhoeae cultures and specimen handling were performed at the Pasteur Institute of Cambodia. Initial HIV screening was performed with a direct particle agglutination test (Serodia-HIV1, Fujirebio INC, Tokyo, Japan). If positive, a confirmatory ELISA (Vidas HIV DUO, bioMérieux Sa. Marcy l'Etoile, France) was performed. If the initial screening test was negative, the specimen was retested using a third generation ELISA (Geenscreen HIV 1/2, Biorad, France). The negative results of this test confirmed the HIV negative sero-status. Discrepant results were tested using the Vidas ELISA system. Sera were tested for syphilis using quantitative Rapid Plasma Reagin (RPR) (Becton, Dickinson, Cockeyville MD, USA). RPR reactive sera were confirmed using particle agglutination (Treponema Pallidum Passive Particle Agglutination Serodia-TP-PA, Fujirebio INC, Tokyo Japan). 8 Family Health International, 2000. |
|||
 |
|||
