Novel coronavirus disease is spread worldwide with specific mortality and burdens worldwide public health. Due to the non-stationarity and complicated nature of the novel coronavirus, it is challenging to model such a phenomenon, delivering reliable long-term forecasts of the extreme death rate. The present study describes a novel bio-system reliability approach, particularly suitable for multi-regional environmental health systems, observed over a sufficient period, resulting in a reliable long-term forecast of the novel coronavirus registration rate. This study analysed COVID-19 patient numbers from different US states, constituting an example of a multi-state model observed during the years 2020-2022. Traditional statistical methods dealing with temporal observations of multi-regional processes do not have the advantage of dealing efficiently with extensive regional dimensionality and cross-correlation between different regional observations. The present study presents a novel statistical method to analyse raw clinical data using a multicenter, population-based and medical survey data-based bio-statistical approach. Non-dimensional parameter λ is introduced to measure multi-state risk level. The suggested method predicts a 100-year return period risk level along with its 95% confidence interval band. This methodology can be used in various public health applications based on clinical survey data.