Kamal Kishore1, Vidushi Jaswal2, Madhur Verma3, Vipin Kaushal1. 1. Post Graduate Institute of Medical Education and Research, Chandigarh, Chandigarh, IN. 2. Mehr Chand Mahajan DAV College, Chandigarh, IN. 3. All India Institute of Medical Sciences Bathinda, Jodhpur Romana Road,, Bathinda, IN.
Abstract
BACKGROUND: Association between human mobility and disease transmission for COVID-19 is established, but quantifying the levels of mobility over large geographical areas is difficult. Google released Community Mobility Report (CMR) data about the movement of people collated from mobile devices. OBJECTIVE: To explore the use of CMR to assess the role of mobility in spreading COVID-19 infection in India. METHODS: An Ecological study analyzed CMR for human mobility between March - October 2020. The data compared before (Between 14th -25th March 2020), during (25th March - 7th June 2020), and after lockdown (8th June 2020 - 15th October 2020) phases with the reference periods, i.e. (3rd January 2020 - 6th February 2020). Another dataset depicting the burden of COVID-19 as per various disease severity indicators was derived from a crowd-sourced Application Programming software. The relationship between the two datasets was investigated using Kendall's tau correlation to depict the correlation between mobility and disease severity. RESULTS: At the national level, mobility decreased from -38% to -77% for all but residential areas (an increase of 24.6%) during the lockdown compared to the reference period. At the beginning of unlocking, the state of Sikkim (minimum cases- seven) with a -60% reduction in mobility depicted more mobility compared to -82% in Maharashtra (maximum cases-1.59 million). Residential mobility negatively correlated (-0.05 to -0.91) with all other measures of mobility. The magnitude of correlations for intra-mobility indicators was comparatively low for the lockdown phase (corr ≥ 0.5 for 12 indicators) compared to other phases (corr ≥ 0.5 for 45 and 18 indicators in pre-lockdown and unlock phase, respectively). A high correlation coefficient between epidemiological and mobility indicators is observed for the lockdown and unlock phases compared to the pre-lockdown. CONCLUSIONS: We can use mobile-based open-source mobility data to assess the effectiveness of social distancing in mitigating the disease spread. CMR data depicted an association between mobility and disease severity, and we suggest that this technique supplement future COVID-19 surveillance.
BACKGROUND: Association between human mobility and disease transmission for COVID-19 is established, but quantifying the levels of mobility over large geographical areas is difficult. Google released Community Mobility Report (CMR) data about the movement of people collated from mobile devices. OBJECTIVE: To explore the use of CMR to assess the role of mobility in spreading COVID-19infection in India. METHODS: An Ecological study analyzed CMR for human mobility between March - October 2020. The data compared before (Between 14th -25th March 2020), during (25th March - 7th June 2020), and after lockdown (8th June 2020 - 15th October 2020) phases with the reference periods, i.e. (3rd January 2020 - 6th February 2020). Another dataset depicting the burden of COVID-19 as per various disease severity indicators was derived from a crowd-sourced Application Programming software. The relationship between the two datasets was investigated using Kendall's tau correlation to depict the correlation between mobility and disease severity. RESULTS: At the national level, mobility decreased from -38% to -77% for all but residential areas (an increase of 24.6%) during the lockdown compared to the reference period. At the beginning of unlocking, the state of Sikkim (minimum cases- seven) with a -60% reduction in mobility depicted more mobility compared to -82% in Maharashtra (maximum cases-1.59 million). Residential mobility negatively correlated (-0.05 to -0.91) with all other measures of mobility. The magnitude of correlations for intra-mobility indicators was comparatively low for the lockdown phase (corr ≥ 0.5 for 12 indicators) compared to other phases (corr ≥ 0.5 for 45 and 18 indicators in pre-lockdown and unlock phase, respectively). A high correlation coefficient between epidemiological and mobility indicators is observed for the lockdown and unlock phases compared to the pre-lockdown. CONCLUSIONS: We can use mobile-based open-source mobility data to assess the effectiveness of social distancing in mitigating the disease spread. CMR data depicted an association between mobility and disease severity, and we suggest that this technique supplement future COVID-19 surveillance.
Authors: Gavin Leech; Charlie Rogers-Smith; Joshua Teperowski Monrad; Jonas B Sandbrink; Benedict Snodin; Robert Zinkov; Benjamin Rader; John S Brownstein; Yarin Gal; Samir Bhatt; Mrinank Sharma; Sören Mindermann; Jan M Brauner; Laurence Aitchison Journal: Proc Natl Acad Sci U S A Date: 2022-05-31 Impact factor: 12.779