Estimating Methods of the Undetected Infections in the COVID-19 Outbreak: A Systematic Review

Abstract

Introduction: The accurate number of COVID-19 cases is essential knowledge to control an epidemic. Currently, one of the most important obstacles in estimating the exact number of COVID-19 patients is the absence of typical clinical symptoms in a large number of people, called asymptomatic infections. In this systematic review, we included and evaluated the studies mainly focusing on the prediction of undetected COVID-19 incidence and mortality rates as well as the reproduction numbers, utilizing various mathematical models.

Methods: This systematic review aims to investigate the estimating methods of undetected infections in the COVID-19 outbreak. Databases of PubMed, Web of Science, Scopus, Cochrane, and Embase, were searched for a combination of keywords. Applying the inclusion/exclusion criteria, all retrieved English literature by April 7, 2022, were reviewed for data extraction through a two-step screening process; first, titles/abstracts, and then full-text. This study is consistent with the PRISMA checklist.

Results: In this study, 61 documents were retrieved using a systematic search strategy. After an initial review of retrieved articles, 6 articles were excluded and the remaining 55 articles met the inclusion criteria and were included in the final review. Most of the studies used mathematical models to estimate the number of underreported asymptomatic infected cases, assessing incidence and prevalence rates more precisely. The spread of COVID-19 has been investigated using various mathematical models. The output statistics were compared with official statistics obtained from different countries. Although the number of reported patients was lower than the estimated numbers, it appeared that the mathematical calculations could be a useful measure to predict pandemics and proper planning.

Conclusion: In conclusion, our study demonstrates the effectiveness of mathematical models in unraveling the true burden of the COVID-19 pandemic in terms of more precise, and accurate infection and mortality rates, and reproduction numbers, thus, statistical mathematical modeling could be an effective tool for measuring the detrimental global burden of pandemic infections. Additionally, they could be a really useful method for future pandemics and would assist the healthcare and public health systems with more accurate and valid information.

Keywords: COVID-19; Undetected infections; clinical symptoms; network-based sampling; sampling; snowball sampling.