[Meta-analysis of diagnostic tests for the detection of COVID-19].

Ochoa-Díaz, Diana; Mendoza-Olazarán, Soraya; Zarate, Xristo; Casillas-Vega, Néstor

Background COVID-19 disease emerged at the end of 2019 and it started to be considered a pandemic at the beginning of 2020. The reference methodology for the COVID-19 diagnosis is the RT-qPCR. Objective It was assessed the sensitivity and specificity of molecular test kits found on the list of tests of the Instituto de Diagnóstico y Referencia Epidemiológicos (InDRE, Institute for Epidemiologic Diagnosis and Reference). Material and methods By using the PubMed and the Google Scholar platforms, a search was carried out for articles, out of which 14 were selected, retrieving the number of samples analyzed and their status (positive or negative), according to the molecular reference method, the name of the test kit used, as well as the characteristics and results obtained, expressed as true positives, false negatives, true negatives and false positives. Sensitivity and specificity of 8 kits were calculated, with compilations of the OpenEpi, version 3.01, statistical calculators, by using the Wilson score with a 95% confidence interval. Results The sensitivity and specificity results for the kits were greater than 90%. Conclusions Tests evaluated allow the detection of presence or absence of virus, but they can be affected by conditions during the stages of the method.