Human contact network analytics and COVID-19 hospital incidence in France

This data set contains COVID-19 hospital incidence, temperature and human mobility and contact data recorded between 2020-03-24 and 2021-03-30 used in the paper:

Selinger et al. 2021: Predicting COVID-19 incidence in French hospitals using human contact network analytics. 10.1016/j.ijid.2021.08.029

See methods in the article for detailed descriptions and the data curation process.

1) cov_mob_tst_national.csv contains national-level data
 

The columns comprise:

incid_hosp: hospital admission incidence                         

incid_rea: ICU admission incidence

incid_dc: hospital death incidence 

incid_rad: incidence of those returned home

within_departement_colocation_X%: X%-quantile of colocation probabilities with départements

between_departement_colocation_X%: X%-quantile of colocation probabilities between départements

fb_population_coverage_X%: X%-quantile of ratio of fb_population over census population in département

null_links_X%: X%-quantile of null links across départements

clustering_X%: X%-quantile of clustering coefficients across départements

ricci_X%: X%-quantile of curvature across départements

ricci_min_X%: X%-quantile of minimum curvature across départements

ricci_mean_X%: X%-quantile of average curvature across départements

ricci_max_X%: X%-quantile of maximum curvature across départements

strength_X%: X%-quantile of network strengths across départements

betweenness_centrality_X%: X%-quantile of betweenness_centrality scores across départements

positive_test_ratio_weekly: ratio of weekly cumulated positive tested over weekly cumulated tests

retail_and_recreation_percent_change_from_baseline: Google Mobility Reports

grocery_and_pharmacy_percent_change_from_baseline: Google Mobility Reports

parks_percent_change_from_baseline:  Google Mobility Reports        

transit_stations_percent_change_from_baseline: Google Mobility Reports

workplaces_percent_change_from_baseline: Google Mobility Reports

residential_percent_change_from_baseline: Google Mobility Reports

mean_temperature_X%: X% quantile of mean daily temperatures averaged over the week across départements

min_temperature_X%: X% quantile of minimum daily temperatures averaged over the week across départements

max_temperature_X%: X% quantile of maximum daily temperatures averaged over the week across départements

2) cov_mob_dep.csv contains département-level data

The columns comprise:

dep: département code

incid_hosp: hospital admission incidence                         

incid_rea: ICU admission incidence

incid_dc: hospital death incidence 

incid_rad: incidence of those returned home

week: week (matched to colocation data recording usually on Tuesdays)

dep_name: name of the département

null_links: number of null links

betweenness_centrality: betweenness centrality

clustering: clustering coefficient

strength: network strength

ricci_mean: minimum curvature among all edges incident to a département

ricci_min: mean curvature across all edges incident to a département             

ricci_X%: X%-quantile curvature among all edges incident to a département

fb_population: number of facebook users                       

facebook_colocation_within_dep: colocation probability within département

fb_population_coverage: ratio of fb_population over census population in département

facebook_colocation_between_dep_X%: X%-quantile of facebook colocation among all edges incident to the département

min_temperature: minimum daily temperature averaged over the week

max_temperature: maximum daily temperature averaged over the week

mean_temperature: mean daily temperature averaged over the week

incid_hosp_Y: incidence of hospital admission from Ynd most colocated département

incid_rea_Y: incidence of ICU admission from Ynd most colocated département

incid_dc_Y: incidence of hospital deaths from Ynd most colocated département

incid_rad_Y: incidence of returned home from Ynd most colocated département