Effectiveness and Cost-Effectiveness of a Rapid Diagnostic Test for Malaria

Overview

There is increasing evidence from African countries, including Burkina Faso, that at least in some settings/seasons the proportion of fevers attributable to malaria is low or very low: this means that the current strategy of treating all fever cases as malaria is only to the advantage of very few. Rapid, antigenic tests might be of help, particularly in peripheral health centres such as the "Centres de Santé et Promotion Sociale" (CSPS) that lack any laboratory facilities. Nevertheless two major problems could arise: * False negatives: as only the negative result would change the decision to treat, versus the current "presumptive" strategy, false negatives would not be treated for malaria. * False positives: they would be exposed to the risk to be left without treatment for the true cause of their fever instead. The main purpose of this study is to assess if the short term outcome of febrile patients treated after testing with the Rapid Diagnostic Test Paracheck® is at least equivalent (not inferior) to that of controls (presumptively treated without any test) in terms of clearance of fever and other major symptoms and signs. To do so, febrile patients will be randomly assigned to be submitted to the test before clinical decision, or to be managed the usual way with no test. A follow up will be carried out at Day 4th in order to determine the proportion of patients in both groups with persistence of fever and other main clinical symptoms.

Primary outcome: proportion of cases (fever patients randomised to be submitted to the rapid test before treatment) and controls (fever patients randomised not to be submitted to the rapid test) with persisting fever at follow-up. Secondary outcomes: * proportion of cases (fever patients randomised to be submitted to the rapid test before treatment) and controls (fever patients randomised not to be submitted to the rapid test) with persisting/new major clinical symptoms/signs and/or clinical complications at follow-up; * proportion of cases and controls referred to hospital; * prevalence of malaria infection among febrile and not febrile patients; * attributable fraction (AF) of fever episodes to malaria infection (as established by subsequent microscopic examination of blood slide and determination of the parasitemia level); * optimal cut-off level of parasitemia(treatment threshold) in order to define a case of malaria disease vs. a simple malaria infection (i.e., presence of malaria parasites in the blood but presumably not the cause of fever); * performances of Paracheck® (sensitivity, specificity and predictive values) on malaria disease in the field, compared with expert reading; * proportion of cases and controls with malaria disease that have not been treated with antimalarials/have been treated for other presumed causes of fever; * proportion of cases and controls without malaria disease that have been treated with antimalarials/ have not been treated for other presumed causes of fever; * proportion of cases and control receiving any other treatment, at enrolment and/or at follow up. Cost - effectiveness models will take into account all costs involved in diagnosis and treatment both for malaria and other conditions. Sample size. Sample size is determined for the main clinical outcome under study, that is, fever clearance. Expected freq of fever persistence at day 4th: 40 %; maximal accepted difference 10%; power of study 90%; alpha error 5%: sample size 814. We plan to enrol at least 500 subjects with fever per arm in both seasons, in order to account for loss to follow up. Expected total enrolment: 4000 (1000 fever cases and 1000 patients without fever in each study period). Only fever patients will be randomised to be submitted or not to the rapid test. Patients without fever will be consecutively enrolled and submitted to malaria slides and clinical questionnaire, but not to the rapid test. Study start: 24th April 2006. Second Phase: 2nd October 2006. In each study period at the end of the dry season and the end of the rainy season, for up to to three weeks in each study period, all patients \> 6 months, presenting for clinical consultation at one of the 10 peripheral health centres participating in the study, responding to inclusion criteria and consenting to participate/with parental consent to participate, will be enrolled. All will be submitted free of charge to a thick film and a thin film for malaria, and a brief questionnaire on clinical history. All slides will be stored for further reading. Those patients with fever at presentation (axillary temperature \>= 37.5°C), will be randomised (based on computer-generated random number lists) either to be submitted to a Paracheck® rapid test for malaria (cases), or to be managed the usual way, that is, on clinical grounds alone (controls). Only Paracheck® result will be used for decision on malaria treatment for cases. Treatment of controls, and treatment of any other condition for both cases and controls, will solely depend on the judgment of the clinical officer. No other formal algorithm will be followed than the national guidelines. Sensitisation of the study sites and information to the public has been carried out before the start of the intervention. A detailed information about the study purposes and implications will be provided to all patients attending the clinical consultation, and witnessed signed informed consent will be obtained. Inclusion criteria. All patients \> 6 months presenting to clinical consultation and consenting to participate/with parental consent to participate. Exclusion criteria. Refusal to participate. Severe clinical conditions with emergency treatment needed as judged by the clinical officer. Pregnancy not an exclusion criterion. Follow-up. This will be carried out ad Day 4th after recruitment for fever patients only. The following main outcomes will be determined at follow-up: 1. Duration/clearance of fever, time of fever clearance; 2. Duration/clearance of other clinical symptoms/signs registered at entry; 3. Appearance of new clinical symptoms/signs, including danger signs, according to a checklist; 4. Patients submitted to new / modified treatment 5. Patients needing referral to hospital 6. Deaths.