BACKGROUND: Malaria continues to be a serious public health problem particularly in Africa. Many people infected with malaria do not access effective treatment due to high price. At the same time many individuals receiving malaria drugs do not suffer from malaria because of the common practice of presumptive diagnosis. A global subsidy on artemisinin-based combination therapy (ACT) has recently been suggested to increase access to the most effective malaria treatment.

METHODS: Following the recommendation by World Health Organization that parasitological testing should be performed before treatment and ACT prescribed to confirmed cases only, it is investigated in this paper if a subsidy on malaria rapid diagnostic tests (RDTs) should be incorporated. A model is developed consisting of a representative individual with fever suspected to be malaria, seeking care at a specialized drug shop where RDTs, ACT medicines, and cheap, less effective anti-malarials are sold. Assuming that the individual has certain beliefs of the accuracy of the RDT and the probability that the fever is malaria, the model predicts the diagnosis-treatment behaviour of the individual. Subsidies on RDTs and ACT are introduced to incentivize appropriate behaviour: choose an RDT before treatment and purchase ACT only if the test is positive.

RESULTS: Solving the model numerically suggests that a combined subsidy on both RDT and ACT is cost minimizing and improves diagnosis-treatment behaviour of individuals. For certain beliefs, such as low trust in RDT accuracy and strong belief that a fever is malaria, subsidization is not sufficient to incentivize appropriate behaviour.

CONCLUSIONS: A combined subsidy on both RDT and ACT rather than a single subsidy is likely required to improve diagnosis-treatment behaviour among individuals seeking care for malaria in the private sector.