Potential of age distribution profiles for the prediction of COVID-19 infection origin in a patient group

The COVID-19 pandemic is a serious and global public health concern. It is now well known that COVID-19 cases may result in mild symptoms leading to patient recovery. However, severity of infection, fatality rates, and treatment responses across different countries, age groups, and demographic groups suggest that the nature of infection is diverse, and a timely investigation of the same is needed for evolving sound treatment and preventive strategies. This paper reports an the analysis of age distribution patterns in six groups of Indian COVID-19 patient populations based on their likely geographical origin of infection viz. the United Kingdom, North America, the European Union, the Middle East, and Asian countries. It was observed that patient groups stratified in this way had a distinct age profile and that some of these groups e.g. patient groups from Asia, the European Union, and the United Kingdom formed a different cluster than those from North America, the Middle East, and other regions. Patient age profiles of a population were found to be highly predictive of the group they belong to, and there are indications of their distinct recovery and fatality rates across gender. Altogether this study provides a scalable framework to estimate the source of infection in a new population of COVID-19 patients with unknown origin. It is also concluded that greater public availability of age and other demographic profile details of patients may be helpful in gaining robust insights into COVID-19 infection origins. Datasets and scripts used in this work are shared at http://covid.sciwhylab.org.