Abstract
Estimating unreported cases of Covid-19 in a region or country presents
a significant challenge, specially when they do not coincide with
asymptomatic cases. This paper introduces a mathematical extension to
deterministic models of total case growth, designed to describe the
temporal evolution of unreported cases for each reported case of
Covid-19. The proposed expression accounts for the growth of unreported
cases in flat regions and their decline towards the end of an epidemic
wave. We apply this extension to both a Gompertz growth model and a
generalized Richards model, fitting these models to data from 30
countries/regions using Bayesian methods. The results show that fits
from the extended models outperform those from the standalone Gompertz
and generalized Richards models, as evidenced by superior values in the
Bayesian Information Criterion (BIC) and R 2 metrics. In conclusion, the
extended models offer a more accurate depiction of Covid-19 epidemic
wave progression and provide short-term predictions of epidemic trends
and criteria for identifying the end of an epidemic wave.