Analysis of data of COVID lockdown period: Comorbidity and fatality rates in a few districts of Assam, India

Abstract

In many regions of the world, significant data collection, analysis, and availability on comorbidity and fatality incidents caused by COVID-19 during the lockdown period (2020–2022) is rare. This is especially true for hospitals and COVID treatment facilities in India. This lack of understanding impedes the development of appropriate treatment options, potentially resulting in inferior planning, patient recovery results, and a load on healthcare resources. This project intends to bridge the gap and enhance patient care in Assam, India, in light of the COVID pandemic. Furthermore, this study aims to determine baseline patient characteristics associated with an elevated risk of death among hospitalized COVID-19 patients in Assam. We employed machine learning (ML) and deep learning (DL) approaches to discover hidden patterns in patient data that could predict which individuals are more sensitive to severe consequences. This knowledge has the potential to transform patient care by allowing doctors to personalize treatment plans and prioritize resources for individuals who are most at risk. A retrospective observational analysis was performed using data from 5329 individuals hospitalized with SARS-CoV-2 illness between April and December 2021. ML and DL algorithms could be used to examine patient characteristics and identify risk factors for death (in this case, 554). We expect this to help us better understand the risk factors for in-hospital death among COVID19 patients in Assam. The findings could be useful in building risk assessment tools to guide patient care.

Keywords

• Comorbidity
• COVID-19
• Fatality rate
• Pandemic
• P- value