key: cord-0686190-b4kqji3a
authors: Murugesan, M.; Venkatesan, P.; Kumar, S.; Thangavelu, P.; Rose, W.; John, J.; Mohan, V. R.; Rupali, P.
title: Epidemiological investigation of the COVID-19 outbreak in Vellore district in South India using Geographic Information Surveillance (GIS)
date: 2022-04-22
journal: nan
DOI: 10.1101/2022.04.21.22274138
sha: 89b17e30ee45845bd0e3443eba0b7a641d257a45
doc_id: 686190
cord_uid: b4kqji3a

Objectives: Geographical Information Surveillance (GIS) is an advanced digital technology tool that maps location-based data and helps in epidemiological modeling. We applied GIS to analyze patterns of spread and hotspots of COVID-19 cases in Vellore district in South India. Methods: Laboratory-confirmed COVID-19 cases from the Vellore district and neighboring taluks from March 2020 to June 2021 were geo-coded and spatial maps were generated. Time trends exploring urban-rural burden with an age-sex distribution of cases and other variables were correlated with outcomes. Results: A total of 45,401 cases of COVID-19 were detected with 20730 cases during the first wave and 24671 cases during the second wave. The overall incidence rates of COVID-19 were 462.8 and 588.6 per 100,000 populations during the first and second waves respectively. The pattern of spread revealed epicenters in densely populated urban areas with radial spread sparing rural areas in the first wave. The case fatality rate was 1.89% and 1.6% during the first and second waves that increased with advancing age. Conclusions: Modern surveillance systems like GIS can accurately predict the trends and pattern of spread during future pandemics. A real-time mapping can help design risk mitigation strategies thereby preventing the spread to rural areas.

. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 22, 2022. ; https://doi.org/10.1101/2022.04.21.22274138 doi: medRxiv preprint waves of which 19,021 (44.3%) and 23,900 (55.7%) had occurred during the first and 159 second waves respectively. A higher number of cases resided in Vellore, Katpadi and 160 Gudiyattam taluks (Supplementary Figure 2) . We mapped weekly and monthly trends in 161 the epidemiological spread of infection and were able to identify geographical clusters, 162 hotspots and spatial-temporal trends using GIS. 163

Of the 938 smaller geographical areas within the study region, 188 were classified as 165 reserved forest areas and the rest comprised villages, urban localities, town panchayat and a 166 city corporation located in Vellore taluk. Throughout both waves, COVID-19 cases were 167 documented from 636 of the inhabited villages and urban areas in the Vellore district. The 168 mean (SD) number of cases in these smaller units was 67 (278) with the minimum being 169 one case documented in one of the villages and the maximum documented caseload was 170 3323 in Konavattam urban ward in Vellore city. Ten urban areas had recorded over 1000 171 cases each; of which one, four, and five of them were Gudiyattam, Katpadi and Vellore 172 taluks in the region. Village and urban ward-wise caseloads during both the waves and 173 overall study period are presented in Figure 1 and a similar trend was observed with the 174 same urban and peri-urban neighborhoods experiencing higher burden during both the 175 waves and with relative sparing of rural areas. 176 177 Heatmaps generated estimating the point density around geolocations of cases in the initial 178 stages of the pandemic from April to June 2020 revealed that the epicenters were in the 179 middle of Vellore city, a densely populated city corporation and the town panchayat area of 180 Gudiyattam taluk. Between July and September 2020 when the first wave was at its peak, it 181 was observed that more cases occurred in the urban corporation and town panchayat areas 182 and spread to neighboring smaller urban areas sparing the rural areas in Vellore, 183 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) cases as compared to the first wave. Although the pattern of spread was similar with the 190 epicenters in urban geographies spreading radially and sparing rural areas, heat maps 191 revealed much higher densities at these epicenters when compared to the first wave and 192 with more peri-urban and rural area involvement ( Figure 2 ). Most of them belonged to the age group of 21-60 years. Among the study population, 3.1% 206 in the first wave and 1.99% in the second wave were children less than 10 years. Elderly 207 (>60 years) were 16% in the first wave and 17.3% in the second wave. Overall, COVID-19 208 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 22, 2022. ; https://doi.org/10.1101/2022.04.21.22274138 doi: medRxiv preprint incidence was higher in males (first wave 58.71%; second wave 56.86%) when compared 209 to females (first wave 41.29%; second wave 42.90%). Figure 4 shows the overall time 210 trends of the outbreak in this location. The first case documented to be COVID positive 211 was on 28 March 2020. The outcome details were documented for 20730 cases in the first 212 wave and 21672 cases in the second wave. Case fatality rate was 1.89% during the first 213 wave and 1.6% during the second wave and steadily increased with advancing age, i.e., 214 7.38% were aged more than 60 years in the first wave and 5.02% in the second wave (Table  215 2). Case fatality rates were higher in men; first wave 2.40%, and 1.76% in the second wave 216 as compared to women first wave 1.16%, second wave 1.38%. When correlating 217 comorbidities with mortality, data from the first wave revealed that the fatality rates were 218 highest among those who had >2 comorbidities (9.52%). 219 220

The data on risk factors, comorbidities and clinical symptoms, and outcomes were available 221 only for 17748 out of 19092 cases during the first wave of the pandemic but were not 222 available for the second wave as recording this was not mandatory during the second wave. 223

On analyzing risk factors for acquiring COVID-19 infection, 4% of them had a recent 224 history of travel outside India and 2.1% reported a history of travel within India but 62.9% 225 had a history of contact with a COVID-19 infected individual. The comorbidity profile data 226 available for 17748 cases showed that 14.4% of the patients had at least one co-morbidity 227 but 7% of the patients had more than one co-morbidity. Diabetes, hypertension, and asthma 228 were the most seen co-morbidities in these patients (Supplementary Table 1 ). Among the 229 61.93% of patients who were symptomatic, fever (39.7%), cough (24.8%) and sore throat 230 (15.9%) were the most common symptoms. Breathlessness was reported in 9% of all cases 231 (Supplementary Table 2 ). 99 out of 17748 (0.6%) were pregnant among the study 232 population during the first wave of the pandemic. 233 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In our spatial-temporal analysis, we noted that the first wave of the pandemic started with a 237 returned traveler from the UK in March after which a peak was noted in travelers who had 238 attended a religious meet in another part of the country (Tablighi Jamaat, 2020). The rapid transmission left many countries and states unprepared. Hence, the usage of 257 digital methods to map the cluster areas, predict hotspots would have enabled contact 258 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 22, 2022.

tracing in real-time helping to cordon off affected households to prevent ongoing 259 transmission. 260

The geospatial mapping in our study population showed that the pattern of spread of the 262 cases is mainly in highly populated urban areas followed by less populated semi-urban 263 areas and then rural areas. In both the first and the second wave, the initial cluster of cases 264 is seen in the urban population. But in the second wave, the peri-urban and rural areas were 265 also affected. Since 60% of population in India is rural it is imperative to curb spread from 266 urban to rural areas. Similar trend has been shown worldwide where the metropolitan cities 267 and urban areas are affected earlier due to a higher population density, indoor crowding, 268 that although, the transmission was higher during the second wave, the CFR is probably 282 lower due to a predominantly younger population being affected, pre-existing immunity 283 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. leading to 500 deaths (Scary measles history has been forgotten by many -The Washington 308 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The application of geographic information system (GIS) in the field of public health. n.d. 490

https://ieeexplore.ieee.org/document/5603111/ (accessed March 29, 2022). 491 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 22, 2022. ; https://doi.org/10.1101/2022.04.21.22274138 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Table 1 : Age and gender of COVID-19 patients during the first and second wave 540 . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 22, 2022. ; https://doi.org/10.1101/2022.04.21.22274138 doi: medRxiv preprint . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) . CC-BY-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 22, 2022. ; https://doi.org/10.1101/2022.04.21.22274138 doi: medRxiv preprint

Estimate of the Basic Reproduction Number for 392

COVID-19: A Systematic Review and Meta-analysis

ArcGIS Desktop quick start guide-ArcMap | Documentation

and Case Fatality Rate in Pune India: An Analysis of First and Second Wave of the 401 Pandemic

Estimation of the reproductive 404 number of the Spanish flu epidemic in

The reproductive number of the Delta variant of SARS-CoV-2 is far 441 higher compared to the ancestral SARS-CoV-2 virus

Estimating the basic reproduction number for COVID-444 19 in Western Europe

1.617.2 Delta variant replication and immune evasion

Use of GIS Mapping 452 as a Public Health Tool-From Cholera to Cancer

Government of India

Health Surveillance: A Tool for Targeting and Monitoring Interventions

Control Priorities in Developing Countries

The epidemic diseases act, 2020, Government of India

The Impact of COVID-19 on children

Vellore City Population Census 2011-2022 | Tamil Nadu

Estimating the reproductive number in the presence of 501 spatial heterogeneity of transmission patterns

WHO Coronavirus (COVID-19) Dashboard. n.d

World Health Organization. Modes of transmission of virus causing COVID-19: 508 implications for IPC precaution recommendations: scientific brief

SARS-CoV-2 Viral

Upper Respiratory Specimens of Infected Patients

Supplementary figure 1: Map of undivided Vellore district showing rural and urban 656

Supplementary figure 2: Spatial mapping of COVID-19 cases in the study region for 671