UGANDA

Monitoting and Evaluation of HIV/AIDS programs

David Serwadda

Background

Uganda has almost 21 million inhabitants and has a GNP per capita of US $240 per year. Uganda has been in the frontline of the AIDS epidemic in many ways: it was one of the first countries to report AIDS cases; one of the quickest to respond to the epi demic; one of the first to observe a decline in HIV prevalence; and one of the first to document dramatic increases in adult mortality associated with HIV/AIDS. By 1988 Uganda was facing a generalized epidemic with an estimated one million people infected by HIV. Current estimates are as high as 10% national HIV prevalence.

National response

As early as 1986 the president of Uganda, Yoweri Museveni, openly admitted that AIDS was a national health problem. The national AIDS control programme was established in 1987. One of its first activities was a national HIV sero-survey in 1988, followe d by the establishment of sentinel surveillance and active surveillance of AIDS patients within hospitals in 1989. In 1990, the multi-sectoral Uganda AIDS commission was established to strengthen and broaden the response to the epidemic. In 1994, the STD control programme was merged with the AIDS control programme. In 1996, the UNAIDS theme group was formed. The national AIDS programme includes a condom social marketing component (SOMARC) and voluntary counseling and testing (VCT) component. In 1990, the first VCT facility was opened and at present there are many VCT facilities. The NGO AIDS Support Organization TASO has been on the forefront with care and support for people living with HIV/AIDS. Evaluations of TASO services have been carried out by WHO in 1994 and by Danida/USAID in 1998.

Monitoring and evaluation

The HIV sentinel surveillance in antenatal clinics started with 8 sites in 1989 and gradually expanded to 20 sites. Programme staff collects forms and blood samples every four months and testing is done at a central location to assure quality. Currentl y, efforts are made to focus on younger age groups in antenatal clinics (under 20 years). Donors have contributed significantly to the development of monitoring and evaluation by insisting on baseline indicators for any project they fund (especially USAID ) and by financial and technical assistance to the AIDS control programme (particularly WHO). National policy is that the Ministry of Planning and Economic Affairs monitors all HIV/AIDS projects that are funded through the Ministry of Finance. Several pro jects however are funded directly to the Ministry of Health with separate project monitoring. This led to some fragmentation of M&E activities.

Population-based surveys on knowledge, attitudes and sexual behaviour were conducted in two districts in 1989, and in four districts in 1995. Both surveys were funded by WHO. Similar surveys were carried out in other districts. The DHS survey in 1995 a lso included AIDS-related questions. The protocols for the WHO/GPA prevention indicators survey were only used in a survey in Hoima district. This is the only survey that yields exact data on the prevention indicators. Other surveys provide data on knowle dge, multiple sexual partners and condoms use, but not with the same denominators as in the prevention indicators.

STD surveillance was supposed to be carried out in 13 of the 20 HIV surveillance sites, but data were not analyzed or reported. An evaluation in 1997 recommended the use of the national health information and management system, but it was decided to st rengthen STD data collection and analysis based on the 13 surveillance sites. In addition, an NGO (AMREF) is monitoring STD drug resistance. Substantial information on the determinants, course and consequences of the HIV/AIDS epidemic is obtained throug h longitudinal studies. These include the Rakai cohort study, which initially ran from 1989-1992, and the Medical Research Council cohort studies in Masaka district, which started in 1989 and is still ongoing. In 1994 an STD intervention trial was set up in Rakai district. The impact of the longitudinal studies on policy making has not been optimal.

Monitoring of HIV prevalence has led to a national response with strong commitment and openness at the highest level of government. The use of more detailed M&E data, from for instance the cohort studies, for decision making has been limited, which is partly due to inadequate dissemination and partly due to lack of resources to implement the recommendations following the evaluation results. The decline of HIV prevalence in adolescents has however led to more funding for monitoring HIV trends in thi s age group.

Input, output and context

Information, Education and Communication (IEC) on HIV/STD/AIDS is covered by more than 100 different organizations, which complicates input and output monitoring. More than 5,000 health educators have been trained and four training courses have been de veloped, more than 10 posters have been produced, and health education messages are disseminated through mass media. Data on knowledge and attitudes have been collected in all sexual behaviour surveys and the 1995 DHS. All showed high levels of awareness and knowledge of HIV/AIDS. There is no central place that keeps track of all condoms imported and distributed. Statistics on social marketing of condoms are available and show an increase to 10 million in 1997. The total number of condoms distributed in 1 997 may however be in the order of 25 million.

With regard to STD control data are available on the number of workshops, training manuals, numbers of health workers trained in STD syndromic management, and distribution of STD drugs. VCT centers report on the numbers of persons tested to the nationa l level, and the largest facility (AIDS Information Centre) had tested about 350,000 clients by 1997. Screening procedures for donors have been introduced in 1989 and are reportedly still in place. There is however no system to assess which units have bee n screened.

Proximate determinants

Comparative analysis of the two sexual behaviour surveys in the four districts between 1989 and 1995 showed a delay in first intercourse and an increase in condom use. In general, however, there are relatively few data on sexual behaviour changes from population-based surveys.

The cohort studies also documented changes in sexual behaviour, notably a reduction of the number of sexual partners and, to a lesser extent, an increase in condom use. The reported changes in sexual behaviour were accompanied by declines in HIV incide nce, particularly among young people. A follow-up study of 3,000 VCT clients three and six months after testing showed significant changes in sexual behaviour. In 1996, a facility survey was carried out in 10 districts to assess the quality of STD service s (prevention indicators 6 and 7).

HIV and STD

The cohort studies are a major source of HIV incidence and prevalence data in Uganda. An HIV surveillance system using both rural and urban antenatal clinics was established within the national programme in 1989 and has reported a decline in HIV prev alence. One of the limitations in the current system is the use of a broad age group of 15-24. Plans are underway to address this by oversampling younger groups. The STD surveillance system collects data from 13 sites.

AIDS impact mitigation

The cohort studies provide extensive data on the mortality and fertility impact of the AIDS epidemic. Two to three fold increases in adult mortality have been observed. In addition, the impact of the epidemic on orphanhood and households in general has been documented. Trends in adult and child mortality have also been described in a survey in six districts and in the DHS 1996.

Conclusion

In Uganda there is evidence that monitoring and evaluation of the HIV/AIDS epidemic has affected policymaking. This mainly pertains to HIV surveillance, although trends in sexual behaviour were partly supported by data from sexual behaviour surveys and cohort studies. Results from the cohort studies gave further credibility to the surveillance data trends, although in general it appears that the link between the research studies and policymaking and programme planning at the national level has been we ak.

As there are many organizations involved in HIV/AIDS/STD programme implementation there is a need for standardization of indicators and data collection procedures. The national AIDS programme should set these guidelines.