Problem definition and substance of interventions

Matters related to sexual behaviour are rarely publicly discussed in Tanzania as is also the case in most of the rest of sub-Saharan Africa. This stands out as the main reason why the core issue of HIV control, the sexual behaviour patterns, have not been sufficiently addressed. Informal discussions with national colleagues at the National AIDS Control Programme and with other informants revealed a wealth of knowledge and considerable understanding of the link between sexual behaviour and HIV in Tanzania. Information on sexual behaviour clearly indicates large variations within the country, but although many have great knowledge of their own and some other ethnic group, no one, as elsewhere, seems to have an overview and a clear idea of the magnitude of different sexual patterns in the different population groups. Some of the information collected through these informal discussions have recently been confirmed by published research studies (Luke, 2003), but most of these are qualitative and also lack quantitative information. At the formal level within the Ministry of Health sexual behaviour is also discussed e.g. in the situation analysis of the MTP-III, but then in much more general and superficial terms.

The way in which HIV control was planned also contributes to making the analysis more superficial. When the Medium Term Plan III (MTP-III) (NACP, 1998) was outlined by the NACP, the regular “logic framework” based format was not used, but planning followed a UNAIDS procedure. This included several workshops and the participation of representatives from all sections of society including all sectors of government. The third medium-term plan was thus carefully prepared. It covered the necessary interventions against heterosexual transmission, such as reducing transmission among commercial sex workers and men with multiple partners (box on page 50). However, although the determinants were analysed in a special workshop they were never subject to any in-depth analysis of how and why sexual transmission takes place. It is probably more likely that such an analysis would have been undertaken if the planning had been done by local experts and if a problem tree had been created as prescribed by the “logic framework” approach. A deeper analysis was in the UNAIDS approach replaced by a ranking of determinants of HIV transmission identified in a preceding workshop. The ranking was carried

out by multi-sectoral groups of 5-6 people mainly consisting of civil servants from different ministries and of NGO representatives under the leadership of staff from the national level. Many of the participants had limited knowledge of HIV and its control.

The intervention that aimed at the reduction of multiple sexual partners was given low priority in the ranking (box on page 50). Although sex work was ranked higher among the priorities of the plan this might not have been well motivated. The identified interventions against heterosexual transmission tended to be based on a stereotype image created by international experts and included in the planning templates, such as those of the “sex worker”, “the poor woman as victim”, and the “sugar daddy” images. These images do sometimes, but not always fit in well with local contexts. Information early on conveyed an understanding of relatively widespread transactional sex, which at the 1998 stage of the epidemic probably would have been a much more important focus for prevention than sex work.

According to the epidemiological reports there was a great variation in HIV prevalence in blood donor data over the different parts of the country in 1996. The prevalence varied between 5% and 20% (Ministry of Health, National AIDS Control Programme, 1997). The reasons for this were not discussed in the situation analysis maybe at least partly because it was thought the differences were due to the time factor and would gradually even out over time. I have no records of either myself or any of my national colleagues using these prevalence levels to discuss sexual patterns and improve the understanding of sexual behaviour in the country in order to identify targets for prevention. No questions about the reasons for the uneven distribution of infection were thus posed and the question of the role of ethnicity as regards sexual patterns was never publicly discussed.

Table 8. Main findings and conclusions on management related issues of HIV control in Tanzania – papers III IV, V and VI

Issue Paper III

Management aspects of HIV control

Paper IV HIV control in Sub-Saharan Africa

Paper V Problems of allocation of large sums of money

Paper VI Is HIV control loosing its focus Problem

definition

&

scope of interven-tions

Little public debate on sexual behaviour as a determinant of HIV transmission

Limited analysis of sexual behaviour in different contexts; insufficient focus on prevention;

First need for action against determinants that drive the epidemic; then social norm change; ART to reduce stigma and open up discussion Planning

&

implemen-tation

Main purpose of plans is to attract funding - this leads to unrealistic plans; large structures and few qualified staff at district level and below;

all sectors involved and no prioritisation

Need for realistic plans; to adapt international templates and to separate short and long-term goals;

Prioritisation to avoid spreading human resources too thin; focus on key sectors;

need to retain the regional level and to use NGOs to increase capacity and grease the government machinery

Use of experienced NGOs to oil the government system

Plans based on international templates;

Need for locally formulated plans, which consider local resource limitations

Lack of penetrative analysis, which considers local social context;

current plans lack strategic choices;

Prioritisation of cost-effective interventions directed towards local factors that fuel the epidemic;

Initial focus on key sectors: health, local government, education and defence

Human &

financial resources

Shortage of financial resources and qualified staff hamper Implementation at al levels;

Need for long-term financing which spans 20-30 years

Drastic increase of funding in 2002/3.

Over-financing will not lead to major increase in service provision due to shortage of staff

Main problem not money, but human resources; risk of crowding out effects if additional health staff not employed;

Money welcomed, but over-financing risks distorting existing systems;

Need for money at measured pace over long time;

Need for balance between monetary input and human resources

Shortage of qualified staff is a main problem

Major obstacles to effective HIV control at community level

Low awareness of risk of HIV;

low quality of communication, lack of community involvement and little or no sharing of information with the community members; lack of political commitment;

resistance from religious leaders.

Continued risky sexual

behaviour; no focus on social norm change; few interventions with proven effectiveness;

Need to encourage open discussion on sexual matters in the community