We propose that busy referral centres are soon going to become the front line in the fight against the pandemic across Africa. Severely ill patients with COVID-19 from the community, will start appearing at emergency departments in the coming weeks, and will spread the infection to other vulnerable patients and staff, whilst they wait for RT-PCR results that are already stretched to maximum capacity. Recent analysis by Public Health England has suggested rapid antibody tests for SARS-CoV-2 perform very well in severely ill patients, with IgM and/or IgG being readily detectable. We will conduct a multi-centre evaluation of two affordable and available antibody-based RDTs at five referral hospitals screening adult admissions with severe respiratory tract infection. We will determine their positive predictive value for RT-PCR-confirmed active infection, and model their potential for rapid triage to quickly separate COVID-19 cases from other admissions. Results will be published and updated online on a weekly basis, providing the global community with the latest possible information. We will also document the course of COVID-19 and outcomes in a HIV & TB endemic setting.

Overall Aim

To implement a cross-sectional diagnostic evaluation to compare two rapid antibody tests with gold-standard RT-PCR, in adults admitted with severe respiratory tract infection in a high HIV/TB burden setting.

Specific Objectives & Hypotheses

To evaluate the accuracy of two Ab rapid diagnostic tests (RDTs) in adult admissions presenting with severe respiratory tract infection, with RT-PCR as gold-standard - We hypothesize that Ab RDTs might be 80% sensitive in being able to predict RT-PCR positive cases of COVID-19.

To document the clinical course of COVID-19 in a high HIV/TB burden setting and identify factors associated with poor outcomes – We hypothesize that mortality at 1 month will be higher among those with COVID-19 and that co-morbidities such as HIV and TB will be differentially distributed between those with and without COVID-19.

To model triage screening algorithms for COVID-19 using clinical and diagnostics results – drawing on data from the proposed trial, and all other data available globally during the latter half of the project – We hypothesize that a diagnostic algorithm can be developed that utilizes Ab RDTs to provide rapid triage and reduce the burden on RT-PCR services.

For further information on this project see HerpeZ website https://www.herpez.org/rapidcov