key: cord-0911500-ir41wf1s
authors: Sarkar, Apurba; Chouhan, Pradip
title: COVID-19: District level vulnerability assessment in India
date: 2020-09-03
journal: Clin Epidemiol Glob Health
DOI: 10.1016/j.cegh.2020.08.017
sha: 7daf1238699bf44934fafd4a721a428f560e73c4
doc_id: 911500
cord_uid: ir41wf1s

OBJECTIVES: COVID-19 Pandemic has brought a threatening challenge to the world and as well as for Indian society and economy. In India, it has become a public health disaster and its intensity increasing continuously. For the disaster risk, reduction, and capacity building against the COVID-19 pandemic understanding of the relationship between socio-environmental conditions with the pandemic is very necessary. The objective of the present work is to construct a socio-environmental vulnerability index of the potential risk of community spread of COVID-19 using socio-economic and environmental variables. METHODOLOGY: In this, a cross-sectional study principal component analyses have been used to drive SoEVI. 4 uncorrelated sub-index has been extracted from 16 sub-indicators which reflects 59% of the variance. Aggregation of 4 Sub-Index has been done to obtain the final vulnerability Index. RESULTS: Results show that there is spatial variability in vulnerability based on environmental and socio-economic conditions. Districts of north and central India found more vulnerable then south India. Statistical significance has been tested using regression analysis and positive relation found in between vulnerability index and confirmed and active cases. CONCLUSION: The vulnerability index has highlighted environmental and socioeconomic backward districts. These areas will suffer more critical problems against COVID-19 pandemic for their socio-environmental problem.

District wise confirmed, active, recovered and deaths cases represent in fig 1. The figure shows that the epidemic has spread across the states, most of the districts have a 81 footprint of contamination. Death has increased considerably within the month. The total case 82 fatality rate is 2.8% and death per lack population is 0.49 22 . Which is very controlled and 83 marginalized in comparison to other countries (fig2). But the major concern is the rate of the 84 new confirmed case is very high and continuously rising. High co-morbidity is one of the 85 important aspects in India many researchers predict that the condition will be more critical in 86 winter 23,24 due to the favorable climatic condition and high co morbidity 25 . A surge in 87 transition will be expected if the vaccine has not been introduced 26 . That will create huge 88 pressure on the public health system. The present investigation has been selected to identify 89 Socio-environmentally venerable districts that will be facing most hurdles against pandemic 90 for their socio-economic backwardness. Sometimes people deny taking medical assistance from the government (Table 1) .

Before constrict PCA data normalization has been performed using Z-Score. The Z-score 142 model is a widely used statistical technique that can able to standardize a wide range of data 143 to represent the significant changes across the data 41 . Z-score data normalization has been 144 done using the following formula 42

Normalization has been done to avoid large differences in scale or variance between 147 variables. After the normalization. Those parameters have a negative relationship with a 148 vulnerability which means high value indicates low vulnerability vice versa is reversed by 149 multiplying with (-1). These make date unidirectional and make it easy for the interpreter. 

Pearson correlation Matrices of indicators has been represented in Fig. 3 . This indicates that 180 population density household density has a strong positive correlation (>0.5) and variables 181 like PM 2.5 , NO 2 and , MP 10 has also high positive relation of (>0.5). Therefore KMO [Insert Figure 4 and 5] 213 We have tried to make a relationship between the vulnerable index and confirmed case and 214 active case as on 25 June 2020 with the help of the Pearson correlation test. We observed a 215 significant positive correlation between SoEVI and Active cases is 0.24 and between SoEVI 216 and confirmed cases is 0.381 (Table 3) . We observer a positive relation between the 217 confirmed and active cases with SoEVI with R 2 is 0.099 and 0.067 respectively (Fig. 5) . 

The Impact of Social Vulnerability on COVID-19 in the U.S.: scheme determine the outcome of composite social vulnerability indices

Urban deprivation in a global south city-a neighborhood scale study of 438

Urban deprivation in a global south city-a neighborhood scale study of

Mapping of climate vulnerability of the 441 coastal region of Bangladesh using principal component analysis

Monitoring and modelling spatio-temporal urban growth of 444

Delhi using Cellular Automata and geoinformatics

Two methods to assess 447 vulnerability to climate change in the Mexican agricultural sector

Inherent vulnerability assessment of rural households 450 based on socio-economic indicators using categorical principal component analysis: A 451 case study of Kimsar region

Urban environmental quality 454 assessment using remote sensing and census data

Assessing the social vulnerability to 457 malaria in Rwanda