Evaluation of Myanmar Artemisinin Monotherapy Replacement Malaria Project (July 2017)

from Department for International Development
Published on 28 Jul 2017 View Original


This report presents the summary of findings and final assessment of an independent, external evaluation of the DFID co-funded project “Replacement of malaria monotherapy drugs in the private sector to support the containment of drug resistant malaria in eastern Burma” or, in short, the “artemisinin monotherapy replacement” project (AMTR). The AMTR project was launched in 2012 while the evaluation project started with some delay in 2013.

The methodology applied by the evaluation team consisted of a combination of qualitative (key informant interviews, field visits, document review) and quantitative data collections (analysis of routine and survey data), as well as an in-depth Value for Money analysis and construction of a casemanagement model to explore project impact and project future developments under a defined set of assumptions and scenarios.

The results can be summarised as follows:

Theory of change and logframe

One of the identified weaknesses of the project was that the theory of change was not actively used as a management tool to guide the project until the final project year. This did not negatively impact on outcomes but would possibly have facilitated a more rapid adjustment of activities to the rapidly changing environment of a declining malaria incidence. The logframe initially had some unrealistic indicators and needed multiple revisions in light of the changing epidemiology and socio-economic context, but was ultimately a useful tool for assessing progress and short-comings.

Key evaluation questions

Has the replacement of oral artemisinin monotherapy been achieved?
It can be said that the AMTR project did successfully replace oral artemisinin monotherapy (oAMT) through the formal distribution channels with quality assured artemisinin combination therapy (QAACT), but some oAMT returned – mainly through informal channels outside of the project’s influence.
At the same time, QA-ACT showed overall declining trends due to the changing malaria epidemiology.

One concern that can be raised is that of sustainability of the social marketing approach under the assumption that some kind of subsidy or market support for QA-ACT will be needed in the future. An approach that delivers a subsidy directly to distributors and avoids brand ownership by the social marketing organisation could possibly have included more distributors, and hence larger coverage, and would be more sustainable in the future.

Has the proportion of fever cases correctly treated increased?

The evidence suggests that correct treatment of fever cases was quite favourable for those testing negative and those not tested, in part due to the declining malaria incidence and increasing awareness among providers that fevers are much less likely to be malaria. However, for correct treatment of true positive cases, the outcome showed only moderate improvement and the proportion of cases treated in such a way that resistance spread could be encouraged was still high at 40%. This was mostly a function of low testing rates in the face of declining malaria incidence.

Has the need for prior testing been established and are RDT used?

There were significant delays in RDT implementation and roll-out that were due to external factors but resulted essentially in testing rates too low to significantly improve the major intended outcome of increased adequate treatment of Plasmodium falciparum cases seen in the private sector. Further increases seem possible in the future with a focus on those outlets that are best suited for testing, but the issue of cost, or rather no-cost, to the consumer will have to be addressed creatively if RDTs are to be channelled through normal private sector supply chains, which in turn will be essential for sustainability.

What influence did the BCC and promotional activities of AMTR have?

There is evidence of success in promoting awareness and knowledge among both consumers and providers. However, overall, this has been less successful than was hoped for, and falls short of the targets in the logframe. The challenge here seems to be that the type of messages that are most adequate in a situation of rapidly decreasing malaria incidence can no longer focus on malaria as a danger (e.g. test your fever because it may be malaria), but rather acknowledge the increasingly rare character of the disease whilst emphasising that excluding malaria is essential for both individual health and the benefit of the entire society.

What was the contribution of the AMTR project to resistance containment?

This evaluation has highlighted that the proportion of adequately treated P.falciparum malaria cases that are seen by private sector providers has increased since the start of the project. This certainly can be seen as a contribution towards resistance containment, even if it was not as high as hoped for and although there were still 40% of true malaria cases that were highly inadequately treated with a potential to worsen resistance (reduced from 60% at the start). However, with the available evidence, it is not possible to say whether this contribution was in fact significant in terms of overall resistance development in Myanmar.

What is the long-term perspective for private sector support in malaria beyond AMTR?

Based on the observations and modelling results, the evaluation makes an attempt to look forward into the potential role of private sector health care providers and malaria commodity markets in Myanmar’s attempt to reach malaria elimination by 2030 and ultimately contain spread of resistance.

The major conclusion here is that elimination will not be possible without some kind of intervention in private sector case-management. Increasing testing of suspicious fever cases will be critical and, in conjunction with optimal consumer and provider behaviour, the correctly treated malaria cases in the private sector can be increased to 50-60%. Cases with highly inadequate treatments which would enhance resistance selection pressure and spread could be reduced to 9-12%.