Malaria and dengue pose notable threats to global public health. According to the World Health Organization (WHO, 2019), approximately 100 to 400 million infections with dengue are recorded per year, with half of the world’s population being at risk. Both diseases are predominantly found in tropical and sub-tropical countries like the Philippines, as well as other Southeast Asian regions. Severe complications and possibly death may result if cases are not properly managed.

The overall alarming increase in case numbers over the last two decades is partly explained by a change in national practices to record and report dengue to the Ministries of Health, and to the WHO. But it also represents government recognition of the burden, and therefore the pertinence to report dengue disease burden. (WHO report, 2020)

A 3-phased field-based clinical risk assessment and diagnostic testing research taking place at the EHA Clinics for a period of 3 years (2019 through 2022) is supported by the Pfizer through the release of the THINKMD global innovation grant for the development of febrile disease differentiation using supervised machine learning (ML) algorithms. 

The first phase (2019-2020) prioritized the validation and correlation between THINKMD mHealth malaria risk assessment compared to malaria Rapid Diagnostic Test confirmed cases; determining the specificity, sensitivity, positive predictive value, and negative predictive value for THINKMD mHealth malaria risk assessment using mRDT confirmed true and false positive cases and true and false negative cases of malaria and generating acute febrile illness (AFI) profiles, starting with malaria, using other correlations of THINKMD’s clinical risk and associated symptomology and confirmatory test results.

Phase II which was between 2020 and 2021 focused on understanding the effect of an embedded mRDT result/OpenRDT image into THINKMD’s platform on the algorithm’s malaria-risk assessment, including the correlation of PCR as the “gold standard” to compare mRDT/THINKMD malaria-risk assessment and perform a comparative analysis between THINKMD’s malaria risk assessment, a front health worker’s interpretation of an mRDT result.

The ongoing phase III aims to determine the specificity and sensitivity of THINKMD’s new dengue clinical risk assessment algorithms with correlative diagnostic dengue RDT (dRDT) by comparing THINKMD clinical dengue risk assessment to diagnostic dengue RDT for the same patient and perform a comparative analysis between the prevalence and incidence of malaria and dengue fever as generated by THINKMD’s clinical risk algorithms with established prevalence and incidence data.

A clinical risk assessment logic for the identification of dengue risk developed by the funding partner is being used for children and adults by the research coordinator from the EHA Clinics for malaria and dengue-focused risk assessments. Before the commencement of the study, the use of the THINKMD logic and application was demonstrated to get feedback on the influence of the tool on workflow and the realities of the workplaces.

The study recruits 500 participants including child, youth and adolescent; 2months-15 years presenting to the Clinics with fever (80%) and 20% who do not have fever.