Stringency measurements and COVID19 spreading in Sri Lanka

Resource persons: Prof. Achala Upendra Jayatilleke, Sanjeewa Dayarathne, Dr.Pandula Siribaddana, Dr.Rushan A B Abeygunawardana, Ms. Prof. Nilanthi de Silva, Prof. Janaka de Silva

The aim of using the SIR model is to predict the number of individuals who are susceptible to infection, actively infected, and recovered from infection at any given time. It is the one of simplest forms out of many complex epidemic models. Due to this simplicity, it can be used and understand by many people who are not in the medical domain, such as administrative and policymakers.

SIR model contained two key parameters namely beta and gamma. Where b governs the disease velocity in the population in other words the speed of moving people from the susceptible compartment to the infected compartment. Many scholars have given different identification to this parameter eg. effective contact rate, effective transmission rate, etc.?On the other hand, the transition from the infected compartment to the recovered compartment is governed by the parameter gamma.?

Many different approaches have been followed to estimate these two parameters in the empirical studies as well as conceptual modeling.?The parameters are labeled as transmission rate and recovery rate, which can be estimated with different operational variables. Since the beta parameter accounts for the transmissibility of the disease, the value is closely associated with the impact of community mitigation strategies. Similarly, the gamma parameter accounts for activities that influence the ability to recover a person from the disease which is the development related to the medical domain, essentially the development of vaccine and its success.

In our previous study (https://www.medrxiv.org/content/10.1101/2020.05.20.20103887v1.full.pdf) we proposed to use the Oxford COVID-19 Government stringency index as the proxy to estimate the beta as it appears to be associated with community mitigation strategies and we used fixed value for estimating gamma parameter. The new version of our study proposed to use the percentage of population vaccinated as the proxy to gamma parameter.

We have considered the COVID19 cases from 1st April 2021 as we identified it as the best and optimal starting point for analyzing the spreading of the 3rd wave of the COVID19 in Sri Lanka and corresponding values for the Oxford COVID-19 Government stringency index. further, we assume at least 50% of the target population of Sri Lanka will gradually vaccinate by end of this year.?

Three scenarios of the stringency measurements and respective three outcomes of the active infected population?were calculated based on the SIR model. The error between the fitted curves and the actual data is considered as the objective function?while the least square method was used on the objective function.

Estimated stringency index values were based on the stringency index calculation methodology which is based on key two sub-indexes namely Containment and closure policies and Health system policies. All the eight indicators that measure the Containment and closure policies and one indicator from Health system policies were considered to construct the Government stringency Index.

Priority was given to reduce the burden on the economic dimension when reducing the stringency index by five basis points in the interval of every two weeks. If the country adopted tight stringency measurments (the current level is 85) till the end of June (green line) will reduce the peak active caseload by half compared with if the country adopte a policy to relax stringecy immediately and shapely ( Blue line).

Finally, we construed the caseload chart assuming if the country adopted the higher stringency measurement in early April, just after we found the UK variant B.1.1.7 on 8th April from the samples obtained from Boralesgamuwa, what would have been the peak caseload. And also, if the country adopts a strict stringency measurement after New year.

The following tables indicats the proposed actions in order to reduce the stringency as sugested above.