Malaria is a common, but decreasing, cause of fever in endemic areas. The use of rapid diagnostic tests could improve treatment of malaria at the local community level. Deployment of these tests is, however, a considerable cost. The aim of the study is to evaluate their effect on improving treatment of fever when used by Community Health Workers in Afghanistan. In phase I of the study, the hypothesis is that an RDT diagnosis deployed with standard training and support will improve the accuracy of treatment applied to fever by community health workers when compared to a diagnosis that is based on symptoms alone. In Phase II of the study, the hypothesis is that the accuracy of treatment can be improved by additional training and supportive interventions given to community health workers compared to those who have only had standard training.

This is a cluster randomised trial conducted in 22 clinics (clusters) in two provinces of Afghanistan. Clinics in Afghanistan have a number of community health workers (CHWs) who are residents of villages near to the clinic. The clinic provides them with their supplies of drugs and other items (e.g. family planning interventions) and the CHW provides a range of services in the community including treatment of acute episodes of fever. Currently, the CHW has no way of diagnosing malaria, except based on the presenting signs and symptoms of the patient. With this method is it likely that many patients who do not have malaria, but have a fever, are treated with antimalarial drugs and some patients with malaria are likely not to be treated with an antimalarial drug.

The use of Rapid Diagnostic Tests (RDT) for malaria could potentially provide the CHW with an easy and safe way of diagnosing malaria based on the presence or absence of parasites in the blood. However, use of tests is not straightforward. They need to be applied correctly, following a strict protocol, to be accurate. There is now much interest in widespread deployment of these tools which are recommended by the World Health Organisation for all fever patients in malaria endemic areas. However, evidence for their effectiveness when used by CHWs is patchy and there are no randomised trials that have been conducted in Asia. Since they represent a considerable cost, examination of their effectiveness is required.

In Asia (unlike most of Africa), malaria is caused by two species - Plasmodium vivax accounts for the majority of cases and P. falciparum for the minority. Treatments differ between the species because of differing drug resistance patterns. Therefore accurate diagnosis is important for providing accurate treatment for the different types of malaria and also for treatment of cases of fever who do not have malaria with alternative treatments. In areas where health services are restricted by short opening hours, insecurity, or inaccessibility due to distance or terrain, the CHW should be able to use RDTs and apply the correct treatment for both a positive and a negative test at the village level.

This study will randomise 22 clinics (and their constituent CHWs) to receive either an RDT for use by the CHW in diagnosing malaria in patients with fever, or no intervention (the current situation) where CHWs base their treatment decisions on symptoms alone (i.e. there is no parasite based diagnosis). The RDT group will be given the standard national training for use of RDTs and management of malaria and the standard supplies. The non-RDT arm will have training on management of malaria and a short introduction to RDTs. The study will measure the accuracy of treatment applied by the CHW in the RDT vs. the non-RDT arms, and examine the accuracy of the RDT itself when it is applied in the field under operational conditions. The accuracy will be measured by collecting filter paper blood spots from the patients which will be tested by PCR for malaria. A variety of comparisons will be made between the two groups, but the main aim is to measure and difference in accurate treatment of the patients according to the reference (PCR) diagnosis.

In phase II of the study the non-RDT arm will be discontinued and the clinics will be re-randomised to different interventions. One group of clinics will have the standard RDT roll out continued, as previously, and the second group will get additional training and a range of supportive interventions to assist in their practice. The second phase will use the same methods of assessing the study outcomes. The aim of the second phase is to see whether additional training (and cost) results in improved practice by CHWs.

Inclusion Criteria:

• Any patient where the CHW* or clinician considers malaria in the diagnosis - either prescribing an antimalarial or would request a malaria test if available or referring for diagnosis of malaria elsewhere.

  * *any community health worker who consults with patients and prescribes treatment, and is administratively attached to one of the study clinics.

• Patient, or parent/guardian, gives informed consent to the study.

Exclusion Criteria:

• Patients with a diagnostic result from another facility.
• Patients referred on for diagnosis in the private sector.
• Patients who have signs of severe or complicated disease and are referred prior to giving any diagnosis using the interventions.