Developing a Predictive Algorithm to Identify the Key Factors Impacting on the Transmission and Fatalities from COVID-19
Author(s)
Kamgar F1, Hughes R2, Lucherini S1
1Adelphi Values Ltd, Bollington, UK, 2Adelphi Values Ltd, Bollington, CHE, UK
OBJECTIVES: Coronavirus disease (COVID-19) is a novel infectious disease that has led to a global pandemic. The majority of people who contract COVID-19 experience mild to moderate respiratory illness and recover without special treatment. However, for many COVID-19 has proved to be fatal. Despite the global challenge, policy responses across different countries have varied considerably. This study explored the impact of policy decisions by comparing the transmission rates and COVID-19 related fatalities between 1st and 30th April 2020 within different countries, based on the government policies implemented. METHODS: A predictive analytics algorithm was developed to identify factors impacting on the transmission and fatalities of COVID-19. This utilized data related to the disease spread and policy responses in 185 countries around the world from Johns Hopkins Center for Systems Science and Engineering (CSSE) and the European Centre for Disease Prevention and Control (ECDC). Bespoke algorithms were used for pre-processing the data for extracting relevant government policies, imputing missing values, eliminating less contributory factors, and selecting the most informative variables. RESULTS: Stay at home policies, restriction on gathering and international travel controls were associated with a reduced number of confirmed cases and resulting fatalities. However, factors including the number of daily tests performed and contact tracing appeared to be less contributory compared against alternative measures implemented to contain the pandemic. CONCLUSIONS: Experimental results suggested that proactive approaches to implementing all policy decisions result in a greater impact on the outcomes achieved. This is particularly true for policies such as test and trace, which is more effective at the beginning of an outbreak when there are fewer chains of transmission. Once the COVID-19 cases rise significantly, the testing process becomes ineffective and has a minimal impact on the disease spread and therefore lockdown implementation becomes a more efficient way of controlling the pandemic.
Conference/Value in Health Info
2021-05, ISPOR 2021, Montreal, Canada
Value in Health, Volume 24, Issue 5, S1 (May 2021)
Code
PIN49
Topic
Epidemiology & Public Health, Health Service Delivery & Process of Care, Methodological & Statistical Research
Topic Subcategory
Confounding, Selection Bias Correction, Causal Inference, Disease Management, Public Health
Disease
Infectious Disease (non-vaccine), Respiratory-Related Disorders