District level correlates of COVID-19 pandemic in India

Abstract

AbstractBackgroundThe number of patients with coronavirus infection (COVID-19) has amplified in India. Understanding the district level correlates of the COVID-19 infection ratio (IR) is therefore essential for formulating policies and intervention.ObjectivesThe present study examines the association between socio-economic and demographic characteristics of India’s population and the COVID-19 infection ratio at district level…Data and MethodsUsing crowdsourced data on the COVID-19 prevalence rate, we analyzed state and district level variation in India from March 14 to July 31 2020. We identified hotspot and cold spot districts for COVID-19 cases and infection ratio. We have also carried out a regression analysis to highlight the district level demographic, socio-economic, infrastructure, and health-related correlates of the COVID-19 infection ratio.ResultsThe results showed that the IR is 42.38 per one hundred thousand population in India. The highest IR was observed in Andhra Pradesh (145.0), followed by Maharashtra (123.6), and was the lowest in Chhattisgarh (10.1). About 80 per cent of infected cases, and 90 per cent of deaths were observed in nine Indian states (Tamil Nadu, Andhra Pradesh, Telangana, Karnataka, Maharashtra, Delhi, Uttar Pradesh, West Bengal, and Gujarat). Moreover, we observed COVID-19 cold-spots in central, northern, western, and north-eastern regions of India. Out of 736 districts, six metropolitan cities (Mumbai, Chennai, Thane, Pune, Bengaluru, and Hyderabad) emerged as the major hotspots in India, containing around 30 per cent of confirmed total COVID-19 cases in the country. Simultaneously, parts of the Konkan coast in Maharashtra, parts of Delhi, the southern part of Tamil Nadu, the northern part of Jammu & Kashmir were identified as hotspots of COVID-19 infection. Moran’s-I value of 0.333showed a positive spatial clusteringlevel in the COVID-19 IR case over neighboring districts. Our regression analysis found that district-level population density (β: 0.05, CI:004-0.06), the percent of urban population (β:3.08, CI: 1.05-5.11), percent of Scheduled Caste Population (β: 3.92, CI: 0.12-7.72),and district-level testing ratio (β: 0.03, CI: 0.01-0.04) are positively associated with the prevalence of COVID-19.ConclusionCOVID-19 cases were heavily concentrated in 9 states of India. Several demographic, socio-economic, and health-related variables are correlated with COVID-19 prevalence rate. However, after adjusting the role of socio-economic and health-related factors, the COVID-19 infection rate was found to be more rampant in districts with a higher population density, a higher percentage of the urban population, and a higher percentage of deprived castes and with a higher level of testing ratio. The identified hotspots and correlates in this study give crucial information for policy discourse.