TANZANIA

Monitoting and Evaluation of HIV/AIDS programs

Gernard Msamanga and Rowland Swai

Background

Tanzania has an estimated 30 million people and a GNP of $120 per capita. Health expenditure has remained less than US $4 per capita per year. The first AIDS case was identified in 1983 in Kagera Region near Uganda. Since then the epidemic has spread to all 20 regions of mainland Tanzania and to a lesser extent to Zanzibar. Even though there are large differences between regions, national prevalence is estimated to be between 5% and 10%, with more than 1 million HIV infected persons.

National response

An AIDS Task Force was established in 1985, while the national AIDS control programme was launched in 1988. The main intervention strategy was promotion of safer sex in the general population, youth and mobile populations in high transmission areas. During the single party system the national party was heavily involved in the mobilization of communities against AIDS, but this faded in the early 1990s with the introduction of a multi-party system. Two mid-term plans have been completed, the third has just started. The second plan envisaged a multi-sectoral response, but this has not been successful. The implementation of the mid-term plans has been plagued by lack of funds. For the second mid-term plan donors pledged US$ 41 million for five years. The UNAIDS theme group was formed late 1995.

Monitoring and evaluation

The national AIDS programme has a monitoring and evaluation unit. This is a small unit and it has suffered from frequent staff turnover, partly because many staff left for better paid jobs with international organizations or in other jobs. The annual reports are the main output and these reports have been disseminated widely, providing essential monitoring information. A short overview of research findings is also included in these reports. The national AIDS programme is also the main review body for all research proposals in the country.

At the end of each mid-term plan a review team consisting of a mix of international organization and national representatives evaluates all aspects of the programme. This has shown to be a valuable instrument for the AIDS programme and the implementati on of activities.

HIV surveillance is done among blood donors and antenatal women. Blood donors have become less useful as a source of HIV monitoring because of donor screening and self-selection. In 1990 a protocol for HIV and syphilis surveillance in 24 antenatal clinics in 11 regions was developed. Due to lack of funds only 8 sites were reporting information by 1996. Most antenatal clinics with HIV surveillance are located in one region with strong donor support.

Several national surveys have collected data on knowledge, attitude and sexual behaviour. These include WHO/GPA survey in 1991, DHS surveys in 1991 and 1996 and a KAP survey FP and AIDS in 1994. In addition, a number of population-based sero-surveys and cohort studies have been carried out.

Input, output and context

Information about financial support for the AIDS programme is incomplete. Funds are mostly not channeled through the national programme and involve many different donors and NGOs. Condom distribution is monitored and can be broken down by region of distribution. The two main sources are free condoms distributed through the national AIDS programme and social marketing condoms. The distribution of 'free' condoms peaked at 45 million in 1993, the total for 1997 is about 30 million condoms, including on e-t hird social marketing condoms. The STD programme is part of the AIDS programme. It is now in the process of expanding improved STD services.

The blood transfusion programme developed national guidelines based on research findings. Reporting of blood donations by hospitals ranges from 23% to 87%, but there is no system to evaluate whether all units have been properly screened for HIV. Shortages of HIV test kits have been reported in recent years.

Proximate determinants

The population-based surveys and cohort studies suggest a modest but significant change in sexual behaviour in many parts of Tanzania. The main change appears to be reduction of the number of partners, while there is also an increase in condom use with short-term partners. The WHO/GPA prevention indicators have not been used as such, as most surveys asked slightly different questions.

HIV and STD

The cohort studies have provided HIV incidence for several populations in Tanzania. In one case a decline in incidence was attributable to an STD intervention. In another case a dramatic decline in HIV prevalence was observed in serial population-based surveys (Kagera). HIV prevalence data from the region bordering Zambia show very high figures, while it is difficult to assess trends in most other parts of the country due to lack of data. Reporting of syphilis (diagnosed through RPR tes ts) has the same problems.

AIDS impact mitigation

By the end of 1997 slightly over 100,000 AIDS cases had been reported, but this is considered to be less than 20% of the actual number. The issue of human rights and legal issues was only introduced in 1997. Policy guidelines have been developed, but have not yet been passed by parliament.

Conclusion

Tanzania is facing a large AIDS epidemic, but has only limited resources to mount a response. This has affected the monitoring and evaluation component, where HIV surveillance had been very modest from the start but could not be sustained. Resources and staff for the monitoring and evaluation component of the AIDS programme are few.

The national AIDS epidemiological reports, mid-term reviews and population-based studies (including cohort studies) are the most important monitoring instruments.