preprints.org > medicine and pharmacology > epidemiology and infectious diseases > 202005.0064.v1

Working Paper Brief Report Version 1 This version is not peer-reviewed

Version 1 : Received: 28 April 2020 / Approved: 5 May 2020 / Online: 5 May 2020 (06:04:41 CEST)

Several articles have reported that the low number of COVID-19 cases in Japan is attributed to the small number of diagnostic RT-PCR tests performed. The criticism is based on the low number of the tests performed, and they suspected there would be more potential cases in Japan. The use of pre-test probability among tested subjects is proposed in order to evaluate of the sufficiency of test availability instead of the number of the tests. The pre-test probability means the average probability, i.e., ‘suspicion level’, of having coronavirus among the tested subject in a country. The higher pre-test probability is regarded as less sufficient opportunity of the tests, and the test availability could be evaluated by the pre-test probability. Thus, potential problems of underestimation of COVID-19 cases by insufficient amount of the test would be clear by using pre-test probability. The comparison of the pre-test probability could be replaced with that of the positive rate of the test because of the linear relationship between them under the assumption of common sensitivity and specificity. Japan shows the third lowest rate (8.6%), and is considered that the considerably sufficient number of the tests have been performed. In conclusion, the positive rate of the test as a surrogate index of the pre-test probability is useful to evaluate the sufficiency of test amount instead of the number of the tests performed. In present, the potential problem of underestimation by insufficient test availability would be less serious in Japan.

COVID-19; RT-PCR test; pre-test probability; positive rate

Medicine and Pharmacology, Epidemiology and Infectious Diseases

* All users must log in before leaving a comment