1. Introduction

1.1. Background

1.2. Overview of HIV/AIDS in Ethiopia

1.3 Introduction to behavioral surveillance survey (BSS)

1.4 Objectives of BSS in Ethiopia

1.1. Background

Ethiopia is situated in the Horn of Africa where it is bordered by Djibouti, Eritrea, Sudan, Kenya and Somalia. The population of Ethiopia is estimated to be over 65 million. Less than 14% of the population lives in urban areas (CSA 1998). The majority of the population lives in the highlands of Ethiopia where subsistence farming predominates. Most inhabitants of the lowland areas are pastoralists. Agriculture accounts for 54% of the gross domestic product. In 1998, per capita gross national product was only US$ 100 (PRB 2000).

Christianity and Islam are the major religions; 51% of the population is Orthodox Christian, 33% Muslim and 10% Protestant. The remaining 6% follows a diversity of other faiths. Ethiopia is home for over 80 ethnic groups (CSA 1998).

Between 1974 and 1991, Ethiopia operated a central command economy, under the socialist banner of the Derg Regime. However, since the overthrow of this regime, the country has moved towards a market-oriented economy. A federal system of government exists with ten regional states plus the Dire Dawa Administrative Council (AC). Figure 1.1 presents a map of Ethiopia, which shows the various administrative regions. Each regional state is subdivided into zones and districts. The smallest administrative units of government are called kebeles; these are also known as urban dwellers associations (UDA) in urban areas and peasant associations (PAs) in rural areas.

The majority of the population resides in the rural areas and has little access to any type of modern health institution. According to Ministry of Health (MoH) figures, potential health service coverage is only 51.2% (MoH 2001). Approximately, 75% of the population suffers from some type of communicable disease and malnutrition, both of which are potentially preventable (TGE 1995). Life expectancy is 55.4 years for women and 53.4 years for men. Between 1984 and 1994, life expectancy did not improve for either gender (CSA 2001).

At present, the government’s health policy takes into account population dynamics, food availability, vulnerability, acceptable living conditions, availability and accessibility of health services and other requisites essential for improvement of health status (TGE 1993).

In 2000, a demographic and health survey (CSA 2001) was conducted in Ethiopia with the objective of providing up-to-date and reliable data on fertility and family planning behavior, child mortality, children's nutritional status, the utilization of maternal and child health services and knowledge about HIV/AIDS. This survey revealed a very low educational level amongst the majority of Ethiopians; about 62% of males and 77% of females had no formal education and <3% of males and 1.2% of females completed primary education. Only 18% of households had access to piped drinking water and the majority of Ethiopian households (82%) did not have a toilet. The practice of female circumcision was widespread in the country; about 80% of women were circumcised. About 25% of the Ethiopian women who died in the seven years preceding the survey had died from pregnancy-related causes. The maternal mortality rate was 871 deaths per 100,000 live births for the period 1994-2000. Neonatal mortality rate was 49 per 1000, infant mortality rate was 97 per 1000 and under-five mortality rate was 166 per 1000 live births (CSA 2001).

Figure 1.1 Map of Ethiopia showing administrative regions.

1.2. Overview of HIV/AIDS in Ethiopia

Over 6.6% of Ethiopia's adult population is thought to be HIV positive; accordingly, it is estimated that there are 2.2 million people infected with the virus (MoH 2002). As is the case elsewhere in Africa, transmission is almost exclusively through heterosexual contact. A large proportion of new HIV infection is occurring in young people (<25 years old).

In Ethiopia, it is widely believed that the HIV epidemic began around 18 years ago. The first evidence of HIV infection in Ethiopia was discovered in serological samples collected in 1984 (Tsega et al. 1988) and was followed by the first reported cases of AIDS in 1986 (Lester et al. 1988).

The national response to the epidemic was initiated promptly with the establishment of a task force on HIV/AIDS in 1985. Subsequently, in 1987, the Department of AIDS Control was established in the Ministry of Health and a national program to prevent and control HIV/AIDS was launched. These activities were followed by a number of sero-surveys across the country, to map the extent of the epidemic (Eshete and Sahlu 1996). Soon after the initial activities, major social and political changes occurred, including: the devolution of political power to the regional states; the creation of a federal system of government; economic liberalization; and the growth of the private and non-governmental sectors. These changes offered the regional states, non-governmental organizations (NGOs) and civil society a potentially conducive environment to tackle the spreading HIV epidemic.

In Ethiopia, where poverty, lack of formal education, and natural and man-made disasters are widespread, it is almost inevitable that the HIV/AIDS epidemic continues to grow relentlessly and expand despite efforts to curb its spread. Indicators suggest that a large segment of society is affected by the epidemic.

In response to this widespread problem, in August 1998, the government formulated a national policy on HIV/AIDS (GFRE 1998) and, in April 2000, it established the National AIDS Prevention and Control Council (GFRE 2000). Moreover, the Ministry of Health and regional health bureaus, collaborating with all stakeholders, have drafted multi-sectoral strategic five-year plans (MoH 1999a, 1999b and 1999c).

Kebede et al. (2000) reviewed the most up-to-date sero-surveys conducted amongst various population groups in Ethiopia; in 1999, the prevalence of HIV was 6.4% amongst blood donors in Addis Ababa. In the same year, the prevalence of HIV was 15% amongst women receiving antenatal care in Addis Ababa (MoH 2000). Furthermore, in 1999, HIV prevalence data from visa applicants (another ‘self-selected’ group of individuals) showed that 9.1% were HIV positive (Tegbaru et al. 1999). Data on area of residence for this group are not available but most are expected to have been from Addis Ababa. In combination, these data indicate that the HIV/AIDS epidemic has affected a large segment of Addis Ababa’s population.

Less data are available for areas outside of Addis Ababa. However, amongst demobilized soldiers who consented to VCT, the HIV prevalence was 6.6% (Yigeremu Abebe, personal communication 2001). Although data from rural areas are scarce, recent evidence indicates that about 3.7% of the adult population of rural areas is HIV positive (MoH 2002). Similarly, amongst 62,000 rural and 10,000 urban army recruits (studied between 1999 and 2000), the prevalence of HIV was 3.8 and 7.2% respectively (Yigeremu Abebe, personal communication 2001). Table 1.1 illustrates trends in the prevalence of HIV-1 infections amongst antenatal clinic attendees at various sero-surveillance sites in Ethiopia (MoH 2002).

An increase in HIV-related deaths is seriously limiting improvements in life expectancy in Ethiopia. Predictions of life expectancy made with or without the presence of AIDS in the population were markedly different e.g. 46 instead of 53 years in 2001, and 50 instead of the expected 59 years in 2014 (MoH 2002).

Table 1.1 Percentage of pregnant women testing HIV positive by sentinel site (urban and rural) 1989-2001.

Site	Year
	1989	1992–93	1995	1996	1997	1998	1999–2000	2001
Addis Ababa	4.6	11.2	21.2	17.8	17.5	-	15.1	15.6
Metu	-	10.7	-	-	-	-	4.0	10.5
Gambella	-	-	-	-	12.7	-	19.0	14.6
Dire Dawa	-	12.3	-	-	-	-	13.6	15.2
Awassa	-	-	-	-	-	14.4	11.5	10.0
Attat	-	-	-	-	-	0.8	4.0	1.5
Dilla	-	-	-	-	-	14.5	11.7	9.8
Gambo	-	-	-	-	-	-	0.7	1.1
Hossana	-	-	-	-	-	3.6	4.8	5.9
Aira	-	-	-	-	-	-	2.0	2.6
Soddo	-	-	-	-	-	9.2	10.7	11.6
Shashemene	-	-	-	-	-	-	14.3	13.1
Estie	-	-	-	-	-	-	7.3	10.7
Bahir Dar (HC)	-	13.0	-	-	-	-	20.8	23.4
Dire Dawa (HC)	-	12.3	-	-	-	-	13.6	15.2
Nazareth	-	-	-	-	-	-	-	18.7
Jijiga	-	-	-	-	-	-	-	19.0
Mekele	-	-	-	-	-	-	-	17.2
Maichew	-	-	-	-	-	-	-	16.8
Adigrat	-	-	-	-	-	-	-	16.2
Borena Dadim	-	-	-	-	-	-	-	1.7
Borena Gosa	-	-	-	-	-	-	-	1.7
Ambo Toke	-	-	-	-	-	-	-	4.6
Jimma	-	-	-	-	-	-	-	8.6
Nekemet	-	-	-	-	-	-	-	9.1
Ginir	-	-	-	-	-	-	-	3.1
Asaita	-	-	-	-	-	-	-	12.4
Dire Dawa (Hsp)	-	-	-	-	-	-	-	8.5
Bahir Dar (Hsp)	-	-	-	-	-	-	-	19.9
Gondar (HC)	-	-	-	-	-	-	-	15.1
Pawi								8.5

Note: Hsp = hospital; HC = Health center. Source: Ministry of Health 2002.

Compulsory reporting of AIDS cases started in 1986 with a report of two cases (Negassa 1990a and 1990b). By 2001, a total of 107,575 cases had been reported to the Ministry of Health (MoH 2002). At present, data on the number of reported AIDS cases disaggregated by age and gender, area of residence and from presumed high-risk groups are available only for the years up to and including 1994 (see MoH 1995). More recent data are not available because of under diagnosis, under reporting and delayed reporting. Nevertheless, AIDS case surveillance, with all its limitations, continues to be practiced in Ethiopia as it is used to estimate the number of HIV infections, related deaths and other impacts of the epidemic. For instance, AIDS case surveillance data have been used as the bases of several estimates and projections for Addis Ababa (Khodakevich et al. 1990; Addis Ababa Regional Health Bureau 1999; Mekonnen et al. 1999) and the entire country (MoH 1998; UNAIDS 1998; CSA 2001).

Several surveys have indicated high levels of awareness about HIV/AIDS in Ethiopia (Gebresellassie 1988) but only a few studies have attempted to examine actual changes in behavior. Serial data on the proportion of students who use condoms are available only for Addis Ababa and Gondar. A recent review of the literature (Kebede et al. 2000) indicated that youth and other high-risk groups in Addis Ababa and Gondar are changing their behavior and adopting safer sexual practices. In Addis Ababa, the proportion of high school students reporting condom use increased from 6.6% in 1990 to 27.7% in 1993. About 34% of college students in Addis Ababa reported condom use in 1993. In Gondar, the proportion of college students reporting condom use increased from 24 to 45.9% between 1990 and 1996. Moreover, in Gondar, the proportion of students reporting sex with a non-regular partner decreased from 47% in 1990 to 9.3% in 1996; in Addis Ababa, it decreased from 36.2% in 1990 to 10.3% in 1993.

Two large-scale nationwide studies on sexual behavior were conducted in 1987-88 and 1993 (Mehret et al. 1996; NAPCC 2000). Although the two studies used different methods to investigate condom use, differences between the results (condom use by 3.4 and 47% of participants, respectively) were too great to be dismissed on methodological grounds.

There are several limitations to the reported behavioral data. These limitations include a lack of standardization of methods used in the various studies; for example, the studies did not use similar or standardized questions. Furthermore, data often represented a small group of individuals and covered limited geographic areas. Consequently, it is difficult to compare results and reach substantiated conclusions; this in turn affects monitoring and evaluation of trends in HIV/AIDS/STI related behaviors. Nevertheless, in other countries where behavioral surveillance surveys (BSS) have been used to overcome similar limitations, experience has shown that tracking of behaviors is essential to the strengthening of national HIV/AIDS prevention and control efforts.

1.3 Introduction to behavioral surveillance survey (BSS)

BSS is a monitoring and evaluation tool designed to track trends in HIV/AIDS-related knowledge, attitudes and behaviors in subpopulations at particular risk of infection, such as female sex workers (FSW), injecting drug users (IDU), mobile men and youth. BSS consists of repeated cross-sectional surveys conducted systematically to monitor changes in HIV/STI risk behaviors based on HIV and sexually transmitted infection (STI) surveillance methods. The key benefit of this methodology is its standardized approaches to questionnaire development, sampling frame construction, and survey implementation and analysis. BSS findings serve many purposes: they yield evidence for impact of projects; provide indicators of project successes and highlight persistent problem areas; identify priority populations for interventions; identify specific behaviors in need of change; function as a policy and advocacy tool; and supply comparative data concerning risk behaviors.

National BSS have been conducted by Family Health International (FHI) in more than 25 countries and their use is growing. Since 1999, they have been used in Asia and Africa, where they have proved beneficial in understanding the pandemic from regional and country-specific perspectives. In several countries, multiple rounds of BSS have already been conducted and the trend data are being used to formulate new programs and to adapt existing ones. The first round of BSS in Ethiopia was initiated following a series of consultations, negotiations and meetings.

1.4 Objectives of BSS in Ethiopia

• To provide baseline information that will assist in the development of a system for tracking behavioral trend data for high risk and vulnerable target groups, i.e. those groups which influence the epidemic in Ethiopia;

• To provide information on behavioral trends amongst key target groups in some of the catchment areas where intervention projects are operating;

• To strengthen the second generation surveillance system in the country;

• To provide information that can guide program planning;

• To open and develop a dialogue on HIV/AIDS amongst the population from the policy to the community level;

• To provide evidence for the relative success of HIV prevention efforts taking place in selected sites; and

• To obtain data in a standardized format, which will enable comparison with other BSS carried out in other countries.