Balance of Risk in COVID-19 Reveals the Extreme Cost of False Positives

Abstract

COVID-19 public health responses, including lockdowns and diagnostic testing strategies, have had consequences. Economic costs (see the CHD paper in this issue) could reach $16 trillion dollars, 90% of the US annual GDP. While harm to small businesses, unemployment, worsening poverty, death from cancer, increased suicides, social isolation, and restriction of freedom all increase the perceived need for drastic responses from the top, flawed measures are costly. A diagnostic assay[1] of tests for COVID-19 depends for its validity on its sensitivity and specificity assessed in terms of the true positive rate (TPR), false positive rate (FPR), true negative rate (TNR), and false negative rate (FNR) of the assays. In this pandemic, Real Time — Polymerase Chain Reaction (RT-PCR) testing has been relied on for drastic top-down responses (as in shutting down the economy of whole nations or the entire world). Here I focus on false positive results where RT-PCR testing suggests many infections by SARS-CoV-2 where there are none. I show by mathematical modeling how reporting positive results of RT-PCR testing, ones known to be false in a measurable percentage of instances, is at least 40 times more impactful (in a detrimental way) than increasing or decreasing the number of tests conducted. To balance the risks of errors in diagnosis, false positive results must be minimized by validating nucleotide sequences and estimates of viremia to avoid flagging individuals as contagious when they are not.