Access to prompt and effective malaria treatment in the Kilombero Valley, Tanzania

Abstract

Malaria is the most important parasitic infection in humans, causing an estimated one million deaths annually. Most cases occur in young children in sub-Saharan Africa, supporting the vicious circle of disease and poverty. Current control strategies have so far failed to reduce the disease in most parts of sub-Saharan Africa. Insecticide-treated mosquito nets (ITN) are effective in preventing malaria episodes and efficacious drugs (such as artemisinin-based combination therapies or ACTs) exist to cure malaria. However, a major problem is the delivery of quality health services, including life-saving drugs, to the ones in need. A variety of inter-linked factors influences patients’ access to prompt and effective treatment. While growing resistance against commonly used antimalarials such as chloroquine or sulphadoxine-pyrimethamine (SP) is being addressed with the introduction of ACTs, obstacles to effective malaria treatment have been identified at the levels of the households (the demand side), the health system (the supply side), and in health policy. The present thesis aimed at contributing to a better understanding of factors influencing access to malaria treatment in a positive or a negative way. The insights gained should inform the development of targeted interventions to improve access to malaria treatment and help to develop a general access framework. The research was carried out as part of the ACCESS Programme, which aims to understand and improve access to effective malaria treatment in the districts of Kilombero and Ulanga, in south-eastern Tanzania. The ACCESS strategy is based on a set of integrated interventions, including (1) social marketing for improved care seeking at community level, (2) strengthening the quality of case-management in health facilities, and (3) strengthening the commercial drug retail sector. The interventions are accompanied by a comprehensive set of monitoring and evaluation activities. Quantitative, semi-quantitative and qualitative methods were used for data collection in the area of the local Demographic Surveillance System (DSS) and the nearby semi-urban centre of Ifakara. Between 2004 and 2006, community-based surveys were conducted to investigate treatment-seeking behaviour and estimate communityeffectiveness of malaria treatment. A shop census and mystery shoppers (simulated clients) were used to monitor drug availability and the performance of shopkeepers in the retail sector. The DSS served as sampling frame for the community-based studies and provided demographic indicators, including morbidity and mortality data. The investigation of treatment-seeking and illness perception revealed a better overlap of local and biomedical illness concepts than reported in earlier studies from the same area. This is likely to reflect the intensive social marketing and health education campaigns carried out during the past decade. Modern medicine was clearly preferred by most patients and 87.5% (95% CI 78.2-93.8) of the fever cases in children and 80.7% (68.1-90.0) in adults were treated with one of the recommended antimalarials (at the time SP, amodiaquine or quinine). However, an estimation of community-effectiveness revealed that only 22.5% (13.9-33.2) of the children and 10.5% (4.0-21.5) of the adults received prompt and appropriate antimalarial treatment, despite high health facility usage rates. Quality of case-management was not satisfactory and the exemption mechanism for under-fives was not functional. Consequently, the commercial drug retail sector played an important complementary role in the provision of malaria treatment. In order to increase treatment effectiveness and maintain the high efficacy of the recently introduced ACT, both treatment sources should be strengthened and their quality should be improved. The seasonal movement of families to distant farming sites did not increase the risk of family members contracting malaria. In the fields, 97.9% (95.2-100) of all people were protected with mosquito nets but since few households stocked antimalarials at home, treatment had to be sought from distant health facilities or drug stores. Of the episodes that happened in the fields, 88.2% (72.6-96.7) were finally treated with an antimalarial, indicating that households made a considerable effort to obtain malaria treatment. It appeared that during the farming season, difficulties to mobilize resources coupled with the long distance to treatment sources led to delays in treatment-seeking. In this context, a comprehensive approach should be considered to improve access to treatment while at the same time assuring rational use of medicines and protecting fragile livelihoods. Investigations in the retail sector found that antimalarial availability had decreased by almost 50% in commercial shops following the policy change from chloroquine to SP as first-line treatment in 2001. This decline was noted mainly in general shops, which were not tolerated any more to sell SP (while they could generally sell chloroquine prior to the policy-change). In 2004, five out of 25 studied villages with a total population of 13,506 (18%) had neither a health facility, nor a shop as source of malaria treatment. While there was no immediately apparent impact on overall antimalarial use, the decline may have disproportionately affected the poorest and most remote groups in the community. In the light of the policy change to ACT these issues need to be addressed urgently if the benefits of these efficacious drugs are to be extended to the whole population. The assessment of shop keepers knowledge and behaviour revealed that drug store keepers had better knowledge of malaria and its treatment than their peers in general shops. In drug stores, mystery shoppers were more likely to receive an appropriate treatment (OR=9.6, 95% CI 1.5-60.5), even though at a higher price. As a distribution channel for ACTs, complementary to health facilities, upgraded drug stores may be the most realistic option. However, shopkeepers in drug stores need to be trained on the provision of correct malaria treatment. At the same time, the role of general shops as first contact points for malaria patients needs to be re-considered. Taking the importance of shops into account, interventions to increase the availability of ACTs in the retail sector are urgently required within the existing legal framework. The insights gained in the ACCESS studies helped to design a generic access framework embedded into the context of livelihood insecurity. This framework links social science and public health research with broader approaches to poverty alleviation. Apart from offering an analytical frame for further scientific research, it suggests access policies and interventions that reach beyond health services. In conclusion, the findings of this thesis underline the need for a comprehensive approach to analyze and improve access to treatment. In this setting, health systems factors appear to be major obstacles to treatment, while local disease perceptions did not appear to have a big influence on treatment access. There is an urgent need to improve quality of care at all levels and new avenues have to be explored to achieve equitable coverage with essential health interventions. Health policies need to be formulated and implemented in a way that they effectively improve the quality of services for all population groups. Considering the close link of disease and poverty, any health intervention is unlikely to succeed without taking the demand side into consideration. A comprehensive approach should therefore not only include measures that enable patients to access providers of good quality care, but also contribute to the strengthening of household economies. In order to achieve a decline in malaria morbidity and mortality in Africa, a concerted effort of all stakeholders is required to translate efficacious tools into effective, equitable and sustainable interventions.