With the increasing variety of RDTs on the market, policy makers must identify the most appropriate test, and circumstances where presumptive treatment remains the preferred strategy. This choice is likely to vary widely not only in response to test characteristics but also to characteristics of the population where RDTs are to be deployed. An economic model was developed as a decision aid, adaptable to different scenarios of ACT and RDT costs, and test accuracies. The model also enables the user to vary estimates for other factors, such as the potential harm of treatment, including risk of adverse events and drug resistance, and the probability that clinicians will adhere to test results. In a recent trial the accuracy of two RDTs (detecting either pLDH or HRP2 antigens) was evaluated in 7 sites across Uganda. The data on costs and accuracies were entered into the model to illustrate its use and results. Output was then obtained at increasing levels of comprehensiveness, starting with direct expenditure on diagnostics and treatment alone, and then introducing patient health outcomes, compromised adherence with test results, and the broader societal costs associated with overprescription of antimalarials. Results suggest that given current RDT and ACT prices, use of the HRP2 RDT would be justifiable across most prevalences and age groups. This however depends to a great extent on whether factors such as the harm associated with use of antimalarials and the probability clinicians adhere to results is included in the analysis. Excluding the harm of treatment, presumptive treatment is justified for younger children, and the benefit in the use of RDTs for older patients is also limited. Results also indicate to the need to ensure that clinicians adhere to negative test result if RDTs are to remain an efficient use of resources. Results are expected to vary widely by location and over time as prices and effectiveness of RDTs and ACTs change, therefore the model was designed for easy incorporation of local and up to date parameter estimates for identification to support local decision making.