Executive Summary
Introduction and methods
Behavioral surveillance surveys (BSS) are an integral part of surveillance systems that enable monitoring of the HIV epidemic. Surveillance systems for assessing and monitoring of trends in HIV and sexually transmitted infection (STI) prevalence, and their interactions, are key to informing an effective and efficient response to the epidemic. United Nations Program on HIV/AIDS (UNAIDS), World Health Organization (WHO), Family Health International (FHI) and others have worked together to develop a new framework for HIV surveillance. This framework, known as Second Generation HIV Surveillance, stresses the need to design a surveillance system that is appropriate to the stage of the epidemic in a country, focusing surveillance resources on population groups in which HIV infection is most likely to be concentrated. In particular, it emphasizes the importance of using behavioral data to inform and explain trends recorded in HIV infection amongst a population, and advocates for the increased use of behavioral data in planning and evaluation of appropriate responses to HIV.
The BSS methodology is a monitoring and evaluation tool designed to track trends in HIV/AIDS-related knowledge, attitudes and behaviors in sub-populations at particular risk of HIV infection, such as female sex workers, injecting drug users, migrant men and youth. However, groups included in a BSS depend on the stage of the epidemic in the country and on the distribution of risk in population groups.
The BSS was introduced to Ethiopia in 2001 to compliment the sero-prevalence and other HIV surveillance systems instituted nationally. Ethiopia is currently experiencing a ‘generalized’ HIV epidemic, defined as an epidemic where the HIV sero-prevalence level amongst sexually active adults in the general population is greater than one percent. Furthermore, it is recognized that HIV sero-prevalence is considerably above this level amongst specific high-risk population groups such as sex workers and mobile populations.
In Ethiopia, the fieldwork for the BSS Round One was conducted between December 2001 and May 2002. The basic objectives of the BSS were to: (i) identify risk behaviors which need to be addressed through interventions amongst specific sub-populations; (ii) identify priorities for the planning of prevention programs and for appropriate distribution of limited resources; (iii) establish a baseline to allow monitoring of trends and patterns in risk behaviors; (iv) provide information to explain changes in HIV prevalence; (v) provide key information for advocacy and policy making; and (vi) provide information to evaluate program impact.
Ten different population groups were studied in this first round of BSS. These groups included in-school youth (ISY; 15-19 year olds, currently enrolled in school), out-of-school youth (OSY; 15-24 year olds, currently not enrolled in school), uniformed services (ground forces and the air force), transport workers (truckers, intercity bus drivers, minibus drivers and their assistants), farmers, pastoralists, factory workers and female sex workers (FSWs).
These population groups were chosen as targets for the study because of their potentially high risk for HIV infection (FSWs, uniformed services, youth and drivers) or low risk for HIV infection (farmers, pastoralists and factory workers). Target group size was determined using probability-sampling techniques. Data were collected using quantitative and qualitative methodologies. Generic BSS questions were adapted to the Ethiopian context. However, considerable standardization was retained to allow for the tracking of behavioral indicators over time and comparison with BSS data from other countries. Questions addressed socio-demographic characteristics, sexual behavior, condom use (knowledge and practice), alcohol and drug use, STIs, availability of and access to HIV testing, HIV-related knowledge and beliefs (including misconceptions), and exposure to HIV interventions. Trained interview teams administered the questionnaires. Data were analyzed for each target group and region, as appropriate.
The quantitative findings of the survey were enriched by a qualitative study, which was carried out between August and September 2002, after the dissemination of the preliminary findings of the BSS. The purpose of the qualitative research was to elaborate on the gaps identified and to further explore some of the major results. A total of 24 focus group discussions and 23 in-depth interviews were conducted amongst the target groups.
Summary of findings
Almost all of the target groups had heard of HIV/AIDS and of ‘diseases that can be transmitted through sexual intercourse’. However, people knew little about details of the diseases. Level of knowledge relating to HIV prevention was highest amongst the ground forces and lowest in the female farmers of Butajira (79 and 1.8%, respectively). Comprehensive knowledge was also lowest amongst the female farmers of Butajira; less than 1% of this group had comprehensive knowledge. Almost all of the target populations had heard of male condoms and reported that they were readily accessible and inexpensive. Consistent condom use was low amongst most population groups. Nevertheless, the majority of FSWs consistently used condoms with their paying clients. In both adult and youth populations, use of alcohol and drugs (mostly khat or shisha) in the previous 12 months was associated with increased levels of unprotected sex (i.e. lower levels of condom use).
Notable findings specific to each target group include the following:
Youth
Amongst the 20,434 youth (6210 ISY and 14,224 OSY), average age was 18 years. Alcohol and khat use were higher amongst OSY than ISY, and higher amongst males than females. Knowledge of HIV/AIDS preventive methods was higher amongst ISY than OSY. General awareness about HIV, STIs and condoms was widespread amongst the ISY. More than two-thirds of the youth said that they knew someone with HIV or someone who had died of AIDS. Almost all of the youth expressed one or more negative attitude about people living with HIV/AIDS (PLWHA).
Amongst the 15- to 19-year-olds, more of the OSY than ISY had ever had sex. The OSY tended to start having sex at an earlier age than the ISY. Amongst the younger male OSY, those in the Gambella region and the Southern Nations and Nationalities Peoples Region (SNNPR), respectively, had the highest (54%) and lowest (24%) reported levels of premarital sex. Males reported that their first sexual partners were approximately the same age as themselves; in contrast, females reported that their first partners were older than themselves by around 5-10 years. Considerable proportions of sexually active OSY in Tigray (61.1%), Beneshangul-Gumuz (55.2%) and Gambella (48.9%) reported that they had had more than one sexual partner in the previous 12 months. Although condoms were readily accessible, only about 50% of sexually active ISY and OSY had ever used them with non-commercial partners. Consistent condom use in the past 12 months with non-commercial partners was low, 39 and 73.6% for the OSY and ISY, respectively. Amongst the youth, about 17% had had unprotected sex with one or more sexual partners in the previous 12 months. Levels of unprotected sex were highest amongst OSY in the Gambella, Beneshangul-Gumuz and Afar regions. The most common reason given for not using condoms was that individuals trusted their sexual partners. Less than 7% of the youth had ever had an HIV test.
Adults
Amongst the 5765 adults (uniformed services, factory workers, transport workers, pastoralists and farmers) interviewed, the average age was 30.4 years. Almost 66% of these individuals had attended school; however, years of schooling varied between the groups. With the exception of the uniformed services, male adult respondents commonly used alcohol and khat. Awareness of HIV/AIDS, STIs and condoms was relatively high in all groups (>90%).
Knowledge of the three major preventive methods was highest amongst the uniformed services (all male) and lowest amongst female farmers (1.8%). Levels of misconceptions about HIV/AIDS were lowest amongst long distance truck drivers and their assistants (>60% had no misconceptions) and highest amongst the male and female farmers of Butajira (only 17.2 and 15.6%, respectively, had no misconceptions).
Almost a third of the adult groups who were sexually active in the past 12 months, reported more than one sexual partner in the last 12 months. About 41% of the ground forces reported more than one sexual partner in the past year as compared with only 7% of the factory workers.
Extra marital sex was reported by 33% of the adult respondents in the past 12 months. The ground forces reported the highest level of extramarital sex (60%) while factory workers reported the lowest level (8%).
Amongst adult respondents who had commercial sex, 95% had used condoms during their last (most recent) commercial sex. Around 66% had used condoms with non-regular partners in the previous 12 months. The practice of unprotected sex (with any type of partner) in the previous 12 months was most common among the ground forces and minibus drivers (18%) and least common amongst factory workers. The main reason given for not using a condom was trust of a partner. Very small proportions of the adult groups had taken HIV tests; overall, only 10% had ever been tested. The highest levels of HIV testing were amongst the uniformed services (air force 27% and ground forces 15.6%).
Female sex workers (FSWs)
Interviews were conducted with a total of 2487 FSWs, from six regions. On average FSWs were 22.2 years old. About 74% of the respondents had attended school; however, levels of literacy varied between the regions. The proportion of illiterate FSWs was high in Bahir Dar compared with the other urban centers. The majority (about 63.5%) of FSWs were born in a town/city. Amongst these FSWs, 73% were working in an area other than their place of birth. Around 20% of the FSWs reported that they were engaged in income-generating activities, additional to their sex work.
About 28% of the FSWs were supporting at least one other person (adult or child). About 33% of the respondents reported drinking alcohol everyday; about 50% used khat regularly. Amongst the urban centers, the highest levels of alcohol and drug use by FSWs were found in Awassa and Nazareth.
Amongst the FSWs, general awareness about HIV, STIs and condoms was universal. A high proportion of FSWs (86%) knew that consistent condom use could prevent HIV infection. Literate FSWs were about 3-times less likely to have misconceptions about HIV/AIDS than their illiterate counterparts. There were some regional differences in levels of knowledge about the preventive methods; knowledge was lowest amongst FSWs in the Gambella town and highest amongst those in Dire Dawa.
Around 8% of the FSWs were <15 years old when they first became involved in commercial sex. Median numbers of sex partners reported by FSWs were three paying and one non-paying partner in the 7-day period preceding the interview. Overall, 92% used condoms during their last commercial sex (i.e. with paying partners), 91% used condoms consistently with paying clients in the 30 days preceding the interview and 70.5% used condoms consistently with their non-paying partners in the previous 12 months. Only 7.7% of the FSWs reported that they had ever had an HIV test.
Between group comparisons
Across the entire adult population studied, 13% reported that they had had unprotected sex at some time in the previous 12 months; in the youth, the corresponding figure was 17.4%. Interestingly, considerable proportions (around 73 and 80% of the adult and youth respondents, respectively) who had unprotected sex in the previous 12 months perceived themselves to be at no or low risk for HIV infection. Notably, less than 10% of the adult and youth respondents had had an HIV test in the previous 12 months.
Conclusions
In Ethiopia, the first round of BSS was successful in documenting HIV/AIDS-related knowledge, attitudes and behaviors in sub-populations including youth, adult groups and FSWs. A key benefit of the methodology was the standardized approach to questionnaire development, sampling frame construction and survey implementation and analysis. This BSS represents an important achievement that can be built upon by future surveillance efforts in Ethiopia. It is important that the results of BSS are recognized as a valuable national resource. Based on the findings the following conclusions were made:
- About half of the respondents did not have knowledge of the three preventive methods. The knowledge gap amongst the rural population was high.
- Misconceptions about HIV/AIDS existed amongst all population groups; however, level of misconceptions was highest amongst the farmers and pastoralists.
- There was a high level of stigma and discrimination amongst all target groups. Level of stigma was highest amongst the rural population.
- Risk perception relating to HIV infection was low amongst all target groups.
- Premarital sex was more common amongst the OSY than ISY.
- The level of HIV testing was very low despite a high level of willingness to undergo voluntary counseling and testing.
- Apart from the exposure to mass media, involvement in community interventions was low.
Recommendations
- At all levels, involvement of leaders and sustained political commitment are needed to support behavioral change, and decrease stigmatization and discrimination relating to HIV/AIDS.
- Parents, religious organizations, governmental and non-governmental organizations need to be mobilized and coordinated to influence public opinion relating to HIV/AIDS, and associated high-risk behaviors.
- Current information, education and communication (IEC) activities need to be evaluated with a view to adapting the country strategy for behavioral change communication (BCC) in order to enhance behavioral change and address widespread misconceptions. In the design and execution of BCC programs, the heterogeneous characteristics of the population and age/gender differences should be taken into account.
- The youth require special attention in relation to HIV/AIDS interventions. In particular, anti-AIDS clubs should be strengthened amongst the ISY and need to be expanded to OSY. Emphasis should be given to discouraging drug and alcohol use amongst the youth. Moreover, there is a need to promote the expansion of recreational facilities and libraries, and focus on the provision of employment opportunities for the youth.
- HIV/AIDS intervention programs need to extend to the rural communities (i.e. to pastoralists and farmers) and special strategies should be designed to increase access of these communities to the mass media; moreover, alternative methods that are appropriate to rural areas, such as outreach services, should be used.
- Increased awareness and expansion of VCT services are required to promote and sustain behavioral change. In particular, VCT services should be more accessible to the community, the uniformed services and mobile population groups.
