Anticipating the novel coronavirus disease (COVID-19) pandemic

Taranjot Kaur; Sukanta Sarkar; Sourangsu Chowdhury; Sudipta Kumar Sinha; Mohit Kumar Jolly; Partha Sharathi Dutta

In this work, we analyse how the timing of strict controlling strategies influence the COVID-19 growth curve of the total confirmed cases in different countries. We use statistical tools to calculate the return rate and lag-1 autocorrelation function of the time-series data of the cumulative confirmed cases each in nine different countries. We investigate the EWSs for the alarming situations observed in the growth curves in each of the countries and record the timing of implementation of containment strategies to slow down the outbreak. Our work suggests that the dynamics of growth curve in the initial 40 days since the first reported case can signal an upcoming sudden rise in the cumulative number of infected cases. Thus, preliminary actions of at least 20 days before the timing of observed EWSs is crucial for an effective and timely containment of the disease. Delay in the strict surveillance and control measures can increase the time to contain the spread, which in turn will affect a larger proportion of the population. Furthermore, the proportion of the affected cases on the commencement of public health measures plays a significant role in containing the epidemic in each country. The timeline of implementation of strict intervention strategies coincided with that of emergence of EWSs for many countries such as India, Italy and Germany. However, the relatively low proportion of the affected cases in the case of India compared to Italy or Germany can be a significant factor explaining the slow rise for India but a relatively disruptive situation in the other countries. Thus, a combination of these two factors for India may restrict the extent of COVID-19 spread in the country, as compared to many other countries across the world. We conclude that model-independent forecasting systems can be applied to clinical data sets for predictability of the disease re-occurrence and formulate control policies.