Electronic Algorithms Based on Host Biomarkers Point of Care Tests to Decide on Admission and Antibiotic Prescription in Tanzanian Febrile Children

Health professionals in developing countries have limited ability to identify children at risk of dying and those in need of antibiotics. The main reasons are limited clinical skills and time, unavailability of diagnostic tests (laboratory or x-ray) and non-adherence to practice guidelines. Child mortality is therefore higher than it should be. Etiological diagnostic tests (detecting microorganisms) may not always help since the distinction between infection and disease and between mild or severe disease is not straightforward. Overprescription of antibiotics is therefore widespread and leads to the development of drug resistance. To address these challenges, decision charts for the management of febrile illness will be developed and include i) few clinical parameters simple to assess, and ii) POCTs results based on specific host markers that can discriminate between mild and severe disease, pneumonia and upper respiratory tract infections, and unspecific fevers of bacterial and of viral origin. This algorithm combining clinical and bedside laboratory tests will be built on an electronic support (android tablet). The first objective of the study is to assess the safety of new electronic decision trees that integrate simple clinical assessment and POCTs results (oxygen saturation and a combination of specific biomarkers of inflammation) as a triage tool to decide on admitting febrile children; the second objective is to assess the usefulness and safety of new electronic decision trees that integrate simple clinical assessment and POCT results (a combination of specific biomarkers of inflammation) as decision-making tool to prescribe antibiotics to non-severe febrile children. The development of such a tool will decrease mortality due to delayed admission, At the same time, it will decrease irrational use of antibiotics, and hence drug pressure and emergence of drug resistance, which represents one of the most important public health threat our world is facing today. This project has the potential of huge applicability since it is specifically designed for end-users with limited medical skills and low resources, as it is the case in most areas of developing countries.