Note: Data for covid-19 last updated 5/23/2020.

Data Set-up

Key for the dfs:

df_unfiltered: USA as states, no as country

df_world: USA as country, no as states

df_filtered: Social variables (population, age, urban, density, HDI) and USA as country, no as states.

VARIABLES

dependent_variables <- names(df_unfiltered)[c(23:28,30:33,35:50)]
dependent_variables
##  [1] "total_deaths_per_million"                   
##  [2] "days_to_reached_0.1_deaths_per_million"     
##  [3] "days_to_reached_1_deaths_per_million"       
##  [4] "mean_daily_deaths_per_million"              
##  [5] "median_daily_deaths_per_million"            
##  [6] "max_daily_deaths_per_million"               
##  [7] "total_deaths_week_3_per_million"            
##  [8] "median_daily_deaths_per_million_week_3"     
##  [9] "mean_daily_deaths_per_million_week_3"       
## [10] "max_daily_deaths_per_million_week_3"        
## [11] "total_deaths_month_1_per_million"           
## [12] "median_daily_deaths_per_million_month_1"    
## [13] "mean_daily_deaths_per_million_month_1"      
## [14] "max_daily_deaths_per_million_month_1"       
## [15] "log_total_deaths_per_million"               
## [16] "log_mean_daily_deaths_per_million"          
## [17] "log_median_daily_deaths_per_million"        
## [18] "log_max_daily_deaths_per_million"           
## [19] "log_total_deaths_week_3_per_million"        
## [20] "log_median_daily_deaths_per_million_week_3" 
## [21] "log_mean_daily_deaths_per_million_week_3"   
## [22] "log_max_daily_deaths_per_million_week_3"    
## [23] "log_total_deaths_month_1_per_million"       
## [24] "log_median_daily_deaths_per_million_month_1"
## [25] "log_mean_daily_deaths_per_million_month_1"  
## [26] "log_max_daily_deaths_per_million_month_1"
independent_variables_continuous <- names(df_unfiltered)[17:18]
independent_variables_continuous
## [1] "BCG_mean_coverage"   "BCG_median_coverage"
independent_variables_categorical <- names(df_unfiltered)[12]
independent_variables_categorical
## [1] "BCG_policy"
confounding_varibales <- names(df_unfiltered)[c(7,51,6,10)]
confounding_varibales
## [1] "population_density_2018" "urban_percentage_2018"  
## [3] "HDI_2018"                "ages_65_up"
dependent_variables_label <- df_labels$New.label[df_labels$Variable %in% dependent_variables]

confounding_varibales_label <- df_labels$New.label[df_labels$Variable %in% confounding_varibales]

Models with UNFILTERED data

All data, including states of USA

Linear Models Unfiltered

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_continuous), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_continuous, "_LMU")


for(i in 1:length(independent_variables_continuous)){
  for(j in 1:length(dependent_variables)){
    linear_model <- lm(df_unfiltered[,dependent_variables[j]]~df_unfiltered[,independent_variables_continuous[i]])
    model_results[j,i] <- glance(linear_model)$p.value
    }
}
BCG_mean_coverage_LMU BCG_median_coverage_LMU
total_deaths_per_million 0.2525661 0.3111653
days_to_reached_0.1_deaths_per_million 0.0045333 0.0027851
days_to_reached_1_deaths_per_million 0.0009710 0.0010236
mean_daily_deaths_per_million 0.3281958 0.3814714
median_daily_deaths_per_million 0.5989774 0.6878797
max_daily_deaths_per_million 0.0827120 0.0923322
total_deaths_week_3_per_million 0.1197871 0.1487372
median_daily_deaths_per_million_week_3 0.0936020 0.1180391
mean_daily_deaths_per_million_week_3 0.1197871 0.1487372
max_daily_deaths_per_million_week_3 0.2077102 0.2517715
total_deaths_month_1_per_million 0.5544449 0.6131759
median_daily_deaths_per_million_month_1 0.0907533 0.1167336
mean_daily_deaths_per_million_month_1 0.5544449 0.6131759
max_daily_deaths_per_million_month_1 0.7810019 0.8059700
log_total_deaths_per_million 0.0015710 0.0022768
log_mean_daily_deaths_per_million 0.0092376 0.0114251
log_median_daily_deaths_per_million 0.0294174 0.0339341
log_max_daily_deaths_per_million 0.0008927 0.0007768
log_total_deaths_week_3_per_million 0.0023889 0.0034696
log_median_daily_deaths_per_million_week_3 0.0121006 0.0126740
log_mean_daily_deaths_per_million_week_3 0.0023889 0.0034696
log_max_daily_deaths_per_million_week_3 0.0044500 0.0057825
log_total_deaths_month_1_per_million 0.0031469 0.0036257
log_median_daily_deaths_per_million_month_1 0.0069309 0.0089112
log_mean_daily_deaths_per_million_month_1 0.0031469 0.0036257
log_max_daily_deaths_per_million_month_1 0.0036552 0.0033046

Anova Models Unfiltered

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_categorical), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_categorical, "_AMU")


for(i in 1:length(independent_variables_categorical)){
  for(j in 1:length(dependent_variables)){
    aov_model <- aov(df_unfiltered[,dependent_variables[j]]~df_unfiltered[,independent_variables_categorical[i]])
    model_results[j,i] <- glance(aov_model)$p.value

  }
}
BCG_policy_AMU
total_deaths_per_million 0.0000000
days_to_reached_0.1_deaths_per_million 0.0001084
days_to_reached_1_deaths_per_million 0.0000000
mean_daily_deaths_per_million 0.0000000
median_daily_deaths_per_million 0.0000000
max_daily_deaths_per_million 0.0000001
total_deaths_week_3_per_million 0.0000830
median_daily_deaths_per_million_week_3 0.0000000
mean_daily_deaths_per_million_week_3 0.0000826
max_daily_deaths_per_million_week_3 0.0065881
total_deaths_month_1_per_million 0.0000001
median_daily_deaths_per_million_month_1 0.0000002
mean_daily_deaths_per_million_month_1 0.0000001
max_daily_deaths_per_million_month_1 0.0000267
log_total_deaths_per_million 0.0000000
log_mean_daily_deaths_per_million 0.0000000
log_median_daily_deaths_per_million 0.0000000
log_max_daily_deaths_per_million 0.0000000
log_total_deaths_week_3_per_million 0.0000000
log_median_daily_deaths_per_million_week_3 0.0000000
log_mean_daily_deaths_per_million_week_3 0.0000000
log_max_daily_deaths_per_million_week_3 0.0000000
log_total_deaths_month_1_per_million 0.0000000
log_median_daily_deaths_per_million_month_1 0.0000000
log_mean_daily_deaths_per_million_month_1 0.0000000
log_max_daily_deaths_per_million_month_1 0.0000000

T-test Models Unfiltered

df_unfiltered$BCG_TF <- 1
df_unfiltered[which(df_unfiltered$BCG_policy == "never"),]$BCG_TF <- 0
df_unfiltered[which(df_unfiltered$BCG_policy == "interrupted"),]$BCG_TF <- 0

df_unfiltered$BCG_TF_2 <- 1
df_unfiltered[which(df_unfiltered$BCG_policy == "never"),]$BCG_TF_2 <- 0
df_unfiltered[which(df_unfiltered$BCG_policy == "interrupted"),]$BCG_TF_2 <- NA
independent_variables_t_test <- names(df_unfiltered)[c(ncol(df_unfiltered)-1, ncol(df_unfiltered))] #BCG_TF
independent_variables_t_test
## [1] "BCG_TF"   "BCG_TF_2"
model_results <- data.frame(matrix(NA, ncol = length(independent_variables_t_test), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_t_test, "_TMU")

for(i in 1:length(independent_variables_t_test)){
  for(j in 1:length(dependent_variables)){
    t_test_model <- t.test(df_unfiltered[,dependent_variables[j]]~df_unfiltered[,independent_variables_t_test[i]], paired = FALSE, na.action = na.pass, alternative="greater", var.equal = TRUE)
    model_results[j, i] <- glance(t_test_model)$p.value
  }
}
BCG_TF_TMU BCG_TF_2_TMU
total_deaths_per_million 0.0000000 0.0000000
days_to_reached_0.1_deaths_per_million 0.9999692 0.9999679
days_to_reached_1_deaths_per_million 1.0000000 1.0000000
mean_daily_deaths_per_million 0.0000000 0.0000000
median_daily_deaths_per_million 0.0000000 0.0000000
max_daily_deaths_per_million 0.0000000 0.0000000
total_deaths_week_3_per_million 0.0000071 0.0000110
median_daily_deaths_per_million_week_3 0.0000000 0.0000000
mean_daily_deaths_per_million_week_3 0.0000071 0.0000110
max_daily_deaths_per_million_week_3 0.0009609 0.0009861
total_deaths_month_1_per_million 0.0000000 0.0000000
median_daily_deaths_per_million_month_1 0.0000000 0.0000000
mean_daily_deaths_per_million_month_1 0.0000000 0.0000000
max_daily_deaths_per_million_month_1 0.0000075 0.0000060
log_total_deaths_per_million 0.0000000 0.0000000
log_mean_daily_deaths_per_million 0.0000000 0.0000000
log_median_daily_deaths_per_million 0.0000000 0.0000000
log_max_daily_deaths_per_million 0.0000000 0.0000000
log_total_deaths_week_3_per_million 0.0000000 0.0000000
log_median_daily_deaths_per_million_week_3 0.0000000 0.0000000
log_mean_daily_deaths_per_million_week_3 0.0000000 0.0000000
log_max_daily_deaths_per_million_week_3 0.0000000 0.0000000
log_total_deaths_month_1_per_million 0.0000000 0.0000000
log_median_daily_deaths_per_million_month_1 0.0000000 0.0000000
log_mean_daily_deaths_per_million_month_1 0.0000000 0.0000000
log_max_daily_deaths_per_million_month_1 0.0000000 0.0000000

Models without USA states (UNFILTEREDish data)

All data, excluding states of USA

Linear Models without USA states

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_continuous), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_continuous, "_LMU2")


for(i in 1:length(independent_variables_continuous)){
  for(j in 1:length(dependent_variables)){
    linear_model <- lm(df_world[,dependent_variables[j]]~df_world[,independent_variables_continuous[i]])
    model_results[j,i] <- glance(linear_model)$p.value

        }
}
BCG_mean_coverage_LMU2 BCG_median_coverage_LMU2
total_deaths_per_million 0.2525661 0.3111653
days_to_reached_0.1_deaths_per_million 0.0045333 0.0027851
days_to_reached_1_deaths_per_million 0.0009710 0.0010236
mean_daily_deaths_per_million 0.3281958 0.3814714
median_daily_deaths_per_million 0.5989774 0.6878797
max_daily_deaths_per_million 0.0827120 0.0923322
total_deaths_week_3_per_million 0.1197871 0.1487372
median_daily_deaths_per_million_week_3 0.0936020 0.1180391
mean_daily_deaths_per_million_week_3 0.1197871 0.1487372
max_daily_deaths_per_million_week_3 0.2077102 0.2517715
total_deaths_month_1_per_million 0.5544449 0.6131759
median_daily_deaths_per_million_month_1 0.0907533 0.1167336
mean_daily_deaths_per_million_month_1 0.5544449 0.6131759
max_daily_deaths_per_million_month_1 0.7810019 0.8059700
log_total_deaths_per_million 0.0015710 0.0022768
log_mean_daily_deaths_per_million 0.0092376 0.0114251
log_median_daily_deaths_per_million 0.0294174 0.0339341
log_max_daily_deaths_per_million 0.0008927 0.0007768
log_total_deaths_week_3_per_million 0.0023889 0.0034696
log_median_daily_deaths_per_million_week_3 0.0121006 0.0126740
log_mean_daily_deaths_per_million_week_3 0.0023889 0.0034696
log_max_daily_deaths_per_million_week_3 0.0044500 0.0057825
log_total_deaths_month_1_per_million 0.0031469 0.0036257
log_median_daily_deaths_per_million_month_1 0.0069309 0.0089112
log_mean_daily_deaths_per_million_month_1 0.0031469 0.0036257
log_max_daily_deaths_per_million_month_1 0.0036552 0.0033046

Anova Models without USA states

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_categorical), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_categorical, "_AMU2")

for(i in 1:length(independent_variables_categorical)){
  for(j in 1:length(dependent_variables)){
    aov_model <- aov(df_world[,dependent_variables[j]]~df_world[,independent_variables_categorical[i]])
    
    model_results[j,i] <- glance(aov_model)$p.value

  }
}
BCG_policy_AMU2
total_deaths_per_million 0.0000000
days_to_reached_0.1_deaths_per_million 0.3232947
days_to_reached_1_deaths_per_million 0.0535218
mean_daily_deaths_per_million 0.0000000
median_daily_deaths_per_million 0.0000000
max_daily_deaths_per_million 0.0000000
total_deaths_week_3_per_million 0.0000000
median_daily_deaths_per_million_week_3 0.0000000
mean_daily_deaths_per_million_week_3 0.0000000
max_daily_deaths_per_million_week_3 0.0000005
total_deaths_month_1_per_million 0.0000000
median_daily_deaths_per_million_month_1 0.0000015
mean_daily_deaths_per_million_month_1 0.0000000
max_daily_deaths_per_million_month_1 0.0000000
log_total_deaths_per_million 0.0000000
log_mean_daily_deaths_per_million 0.0000000
log_median_daily_deaths_per_million 0.0000000
log_max_daily_deaths_per_million 0.0000000
log_total_deaths_week_3_per_million 0.0000000
log_median_daily_deaths_per_million_week_3 0.0000020
log_mean_daily_deaths_per_million_week_3 0.0000000
log_max_daily_deaths_per_million_week_3 0.0000002
log_total_deaths_month_1_per_million 0.0000000
log_median_daily_deaths_per_million_month_1 0.0000002
log_mean_daily_deaths_per_million_month_1 0.0000000
log_max_daily_deaths_per_million_month_1 0.0000000

T-test Models without USA states

df_world$BCG_TF <- 1
df_world[which(df_world$BCG_policy == "never"),]$BCG_TF <- 0
df_world[which(df_world$BCG_policy == "interrupted"),]$BCG_TF <- 0

df_world$BCG_TF_2 <- 1
df_world[which(df_world$BCG_policy == "never"),]$BCG_TF_2 <- 0
df_world[which(df_world$BCG_policy == "interrupted"),]$BCG_TF_2 <- NA
independent_variables_t_test <- names(df_world)[c(ncol(df_world)-1, ncol(df_world))] #BCG_TF
independent_variables_t_test
## [1] "BCG_TF"   "BCG_TF_2"
model_results <- data.frame(matrix(NA, ncol = length(independent_variables_t_test), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_t_test, "_TMU2")


for(i in 1:length(independent_variables_t_test)){
  for(j in 1:length(dependent_variables)){
    t_test_model <- t.test(df_world[,dependent_variables[j]]~df_world[,independent_variables_t_test[i]], paired = FALSE, na.action = na.pass, alternative="greater", var.equal = TRUE)
    
    model_results[j,i] <- glance(t_test_model)$p.value
  }
}
BCG_TF_TMU2 BCG_TF_2_TMU2
total_deaths_per_million 0.0000000 0.0000000
days_to_reached_0.1_deaths_per_million 0.9225754 0.7530464
days_to_reached_1_deaths_per_million 0.9879902 0.8595931
mean_daily_deaths_per_million 0.0000000 0.0000000
median_daily_deaths_per_million 0.0000000 0.0000000
max_daily_deaths_per_million 0.0000053 0.0000000
total_deaths_week_3_per_million 0.0000156 0.0000000
median_daily_deaths_per_million_week_3 0.0000000 0.0000000
mean_daily_deaths_per_million_week_3 0.0000156 0.0000000
max_daily_deaths_per_million_week_3 0.0005770 0.0000001
total_deaths_month_1_per_million 0.0000000 0.0000000
median_daily_deaths_per_million_month_1 0.0000001 0.0000000
mean_daily_deaths_per_million_month_1 0.0000000 0.0000000
max_daily_deaths_per_million_month_1 0.0000491 0.0000000
log_total_deaths_per_million 0.0000000 0.0000008
log_mean_daily_deaths_per_million 0.0000000 0.0000017
log_median_daily_deaths_per_million 0.0000000 0.0000084
log_max_daily_deaths_per_million 0.0000000 0.0000078
log_total_deaths_week_3_per_million 0.0000000 0.0000174
log_median_daily_deaths_per_million_week_3 0.0000001 0.0040196
log_mean_daily_deaths_per_million_week_3 0.0000000 0.0000174
log_max_daily_deaths_per_million_week_3 0.0000001 0.0000901
log_total_deaths_month_1_per_million 0.0000000 0.0000067
log_median_daily_deaths_per_million_month_1 0.0000000 0.0051662
log_mean_daily_deaths_per_million_month_1 0.0000000 0.0000067
log_max_daily_deaths_per_million_month_1 0.0000000 0.0000067

Models with FILTERED data

All data, filtered as follows

df_filtered <- subsoet(df_unfiltered, 
                covid19_cumulative_deaths > 0
                  & population_million > 1
                  & ages_65_up > 15 
                  & urban_percentage_2018 > 60 
                  & population_density_2018 < 300 
                  & HDI_2018 > 0.7
                  & ISO3 != "USA_state")

Linear Models Filtered

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_continuous), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_continuous, "_LMF")


for(i in 1:length(independent_variables_continuous)){
  for(j in 1:length(dependent_variables)){
    linear_model <- lm(df_filtered[,dependent_variables[j]]~df_filtered[,independent_variables_continuous[i]])
    model_results[j,i] <- glance(linear_model)$p.value
    }
}
BCG_mean_coverage_LMF BCG_median_coverage_LMF
total_deaths_per_million 0.0928875 0.0812290
days_to_reached_0.1_deaths_per_million 0.9575894 0.8872806
days_to_reached_1_deaths_per_million 0.9977671 0.9678422
mean_daily_deaths_per_million 0.0618269 0.0545268
median_daily_deaths_per_million 0.0734482 0.0671536
max_daily_deaths_per_million 0.1506741 0.1303914
total_deaths_week_3_per_million 0.3558381 0.3541563
median_daily_deaths_per_million_week_3 0.3013023 0.2973710
mean_daily_deaths_per_million_week_3 0.3558381 0.3541563
max_daily_deaths_per_million_week_3 0.2263708 0.2278724
total_deaths_month_1_per_million 0.0882472 0.0883046
median_daily_deaths_per_million_month_1 0.9914933 0.9827768
mean_daily_deaths_per_million_month_1 0.0882472 0.0883046
max_daily_deaths_per_million_month_1 0.0254745 0.0260383
log_total_deaths_per_million 0.5910722 0.5501185
log_mean_daily_deaths_per_million 0.5436489 0.5062876
log_median_daily_deaths_per_million 0.0195320 0.0161222
log_max_daily_deaths_per_million 0.3796605 0.3471881
log_total_deaths_week_3_per_million 0.4259286 0.4513514
log_median_daily_deaths_per_million_week_3 0.2350580 0.2611631
log_mean_daily_deaths_per_million_week_3 0.4259286 0.4513514
log_max_daily_deaths_per_million_week_3 0.3204635 0.3472311
log_total_deaths_month_1_per_million 0.3735312 0.3881031
log_median_daily_deaths_per_million_month_1 0.4846813 0.4728291
log_mean_daily_deaths_per_million_month_1 0.3735312 0.3881031
log_max_daily_deaths_per_million_month_1 0.1446985 0.1500235

Anova Models Filtered

model_results <- data.frame(matrix(NA, ncol = length(independent_variables_categorical), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_categorical, "_AMF")

for(i in 1:length(independent_variables_categorical)){
  for(j in 1:length(dependent_variables)){
    aov_model <- aov(df_filtered[,dependent_variables[j]]~df_filtered[,independent_variables_categorical[i]])
    
    model_results[j,i] <- glance(aov_model)$p.value
  }
}
BCG_policy_AMF
total_deaths_per_million 0.0110433
days_to_reached_0.1_deaths_per_million 0.5138357
days_to_reached_1_deaths_per_million 0.8611822
mean_daily_deaths_per_million 0.0151171
median_daily_deaths_per_million 0.0062392
max_daily_deaths_per_million 0.0274574
total_deaths_week_3_per_million 0.2047799
median_daily_deaths_per_million_week_3 0.1090813
mean_daily_deaths_per_million_week_3 0.2047799
max_daily_deaths_per_million_week_3 0.3865571
total_deaths_month_1_per_million 0.1918745
median_daily_deaths_per_million_month_1 0.1688351
mean_daily_deaths_per_million_month_1 0.1918745
max_daily_deaths_per_million_month_1 0.1710350
log_total_deaths_per_million 0.0140011
log_mean_daily_deaths_per_million 0.0201393
log_median_daily_deaths_per_million 0.0104689
log_max_daily_deaths_per_million 0.0170379
log_total_deaths_week_3_per_million 0.5182464
log_median_daily_deaths_per_million_week_3 0.2786673
log_mean_daily_deaths_per_million_week_3 0.5182464
log_max_daily_deaths_per_million_week_3 0.6610229
log_total_deaths_month_1_per_million 0.2631596
log_median_daily_deaths_per_million_month_1 0.0470756
log_mean_daily_deaths_per_million_month_1 0.2631596
log_max_daily_deaths_per_million_month_1 0.1778639

T-test Filtered

df_filtered$BCG_TF <- 1
df_filtered[which(df_filtered$BCG_policy == "never"),]$BCG_TF <- 0
df_filtered[which(df_filtered$BCG_policy == "interrupted"),]$BCG_TF <- 0

df_filtered$BCG_TF_2 <- 1
df_filtered[which(df_filtered$BCG_policy == "never"),]$BCG_TF_2 <- 0
df_filtered[which(df_filtered$BCG_policy == "interrupted"),]$BCG_TF_2 <- NA
independent_variables_t_test <- names(df_filtered)[c(ncol(df_filtered)-1, ncol(df_filtered))] #BCG_TF
independent_variables_t_test
## [1] "BCG_TF"   "BCG_TF_2"
model_results <- data.frame(matrix(NA, ncol = length(independent_variables_t_test), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(independent_variables_t_test, "_TMF")

for(i in 1:length(independent_variables_t_test)){
  for(j in 1:length(dependent_variables)){
    t_test_model <- t.test(df_filtered[,dependent_variables[j]]~df_filtered[,independent_variables_t_test[i]], paired = FALSE, na.action = na.pass, alternative="greater", var.equal = TRUE)
    
    model_results[j, i] <- glance(t_test_model)$p.value
  }
}
BCG_TF_TMF BCG_TF_2_TMF
total_deaths_per_million 0.0046129 0.0000251
days_to_reached_0.1_deaths_per_million 0.2016341 0.1108277
days_to_reached_1_deaths_per_million 0.4230279 0.2906656
mean_daily_deaths_per_million 0.0040625 0.0000133
median_daily_deaths_per_million 0.0040814 0.0000010
max_daily_deaths_per_million 0.0038820 0.0002303
total_deaths_week_3_per_million 0.0392208 0.0827602
median_daily_deaths_per_million_week_3 0.0172725 0.0513321
mean_daily_deaths_per_million_week_3 0.0392208 0.0827602
max_daily_deaths_per_million_week_3 0.0937636 0.2309676
total_deaths_month_1_per_million 0.0329438 0.0288370
median_daily_deaths_per_million_month_1 0.0417358 0.2190385
mean_daily_deaths_per_million_month_1 0.0329438 0.0288370
max_daily_deaths_per_million_month_1 0.0361193 0.0186749
log_total_deaths_per_million 0.0043163 0.0000686
log_mean_daily_deaths_per_million 0.0054457 0.0001359
log_median_daily_deaths_per_million 0.0028382 0.0004002
log_max_daily_deaths_per_million 0.0028718 0.0030190
log_total_deaths_week_3_per_million 0.1242737 0.2443389
log_median_daily_deaths_per_million_week_3 0.0623407 0.2828274
log_mean_daily_deaths_per_million_week_3 0.1242737 0.2443389
log_max_daily_deaths_per_million_week_3 0.1860787 0.3641082
log_total_deaths_month_1_per_million 0.0525123 0.0599427
log_median_daily_deaths_per_million_month_1 0.0249550 0.4720666
log_mean_daily_deaths_per_million_month_1 0.0525123 0.0599427
log_max_daily_deaths_per_million_month_1 0.0408538 0.0396862

Models with Confounding varibales

confounding_varibales <- names(df_unfiltered)[c(7,51,6,10)]
confounding_varibales
## [1] "population_density_2018" "urban_percentage_2018"  
## [3] "HDI_2018"                "ages_65_up"

Linear Models (with UNFILTERED data, including USA states)

df_confounding_models <- data.frame()

model_results <- data.frame(matrix(NA, ncol = length(confounding_varibales), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(confounding_varibales, "_LMU")


for(i in 1:length(confounding_varibales)){
  for(j in 1:length(dependent_variables)){
    lm_model <- lm(df_unfiltered[,dependent_variables[j]]~df_unfiltered[,confounding_varibales[i]], df_unfiltered)
    model_results[j,i] <- glance(lm_model)$p.value
  }
}
population_density_2018_LMU urban_percentage_2018_LMU HDI_2018_LMU ages_65_up_LMU
total_deaths_per_million 0.5499921 0.0000000 0.00e+00 0.0000000
days_to_reached_0.1_deaths_per_million 0.5835394 0.0000343 4.50e-06 0.0020023
days_to_reached_1_deaths_per_million 0.3047965 0.0001387 0.00e+00 0.0000001
mean_daily_deaths_per_million 0.4912991 0.0000000 0.00e+00 0.0000004
median_daily_deaths_per_million 0.8803698 0.0000002 0.00e+00 0.0000010
max_daily_deaths_per_million 0.0041731 0.0000047 0.00e+00 0.0000002
total_deaths_week_3_per_million 0.1144446 0.0002162 7.30e-06 0.0000146
median_daily_deaths_per_million_week_3 0.9014046 0.0000973 7.30e-06 0.0000233
mean_daily_deaths_per_million_week_3 0.1144665 0.0002159 7.30e-06 0.0000146
max_daily_deaths_per_million_week_3 0.0003571 0.0010419 6.50e-06 0.0000043
total_deaths_month_1_per_million 0.2466448 0.0000142 2.50e-06 0.0000068
median_daily_deaths_per_million_month_1 0.9849843 0.0003136 7.42e-05 0.0000630
mean_daily_deaths_per_million_month_1 0.2467086 0.0000142 2.50e-06 0.0000068
max_daily_deaths_per_million_month_1 0.0013183 0.0001554 1.00e-07 0.0000085
log_total_deaths_per_million 0.4395827 0.0000000 0.00e+00 0.0000000
log_mean_daily_deaths_per_million 0.4017288 0.0000000 0.00e+00 0.0000000
log_median_daily_deaths_per_million 0.2195715 0.0000000 0.00e+00 0.0000000
log_max_daily_deaths_per_million 0.0991806 0.0000000 0.00e+00 0.0000000
log_total_deaths_week_3_per_million 0.1375008 0.0000000 0.00e+00 0.0000000
log_median_daily_deaths_per_million_week_3 0.2735047 0.0000002 0.00e+00 0.0000000
log_mean_daily_deaths_per_million_week_3 0.1380584 0.0000000 0.00e+00 0.0000000
log_max_daily_deaths_per_million_week_3 0.0268278 0.0000000 0.00e+00 0.0000000
log_total_deaths_month_1_per_million 0.1877320 0.0000000 0.00e+00 0.0000000
log_median_daily_deaths_per_million_month_1 0.2668111 0.0000023 0.00e+00 0.0000000
log_mean_daily_deaths_per_million_month_1 0.1887620 0.0000000 0.00e+00 0.0000000
log_max_daily_deaths_per_million_month_1 0.0523995 0.0000000 0.00e+00 0.0000000

Linear Models (with UNFILTEREDish data, excluding USA states)

model_results <- data.frame(matrix(NA, ncol = length(confounding_varibales), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(confounding_varibales, "_LMU2")


for(i in 1:length(confounding_varibales)){
  for(j in 1:length(dependent_variables)){
    lm_model <- lm(df_world[,dependent_variables[j]]~df_world[,confounding_varibales[i]], df_world)
    model_results[j,i] <- glance(lm_model)$p.value
  }
}
population_density_2018_LMU2 urban_percentage_2018_LMU2 HDI_2018_LMU2 ages_65_up_LMU2
total_deaths_per_million 0.7544052 0.0000057 0.0000002 0.0000000
days_to_reached_0.1_deaths_per_million 0.4983133 0.0008668 0.0090812 0.1233536
days_to_reached_1_deaths_per_million 0.2442380 0.0014408 0.0000006 0.0004797
mean_daily_deaths_per_million 0.6443577 0.0000140 0.0000008 0.0000000
median_daily_deaths_per_million 0.6306911 0.0002138 0.0000070 0.0000000
max_daily_deaths_per_million 0.0064831 0.0018184 0.0000020 0.0000010
total_deaths_week_3_per_million 0.1546170 0.0056824 0.0004026 0.0000000
median_daily_deaths_per_million_week_3 0.6754278 0.0029423 0.0002024 0.0000000
mean_daily_deaths_per_million_week_3 0.1546170 0.0056824 0.0004026 0.0000000
max_daily_deaths_per_million_week_3 0.0014454 0.0079651 0.0015257 0.0000001
total_deaths_month_1_per_million 0.2790044 0.0012475 0.0002928 0.0000000
median_daily_deaths_per_million_month_1 0.7592899 0.0116798 0.0061438 0.0000000
mean_daily_deaths_per_million_month_1 0.2790044 0.0012475 0.0002928 0.0000000
max_daily_deaths_per_million_month_1 0.0034264 0.0075344 0.0000538 0.0000006
log_total_deaths_per_million 0.3295656 0.0000000 0.0000000 0.0000000
log_mean_daily_deaths_per_million 0.2892904 0.0000000 0.0000000 0.0000000
log_median_daily_deaths_per_million 0.3183875 0.0000182 0.0000000 0.0000019
log_max_daily_deaths_per_million 0.0518483 0.0000000 0.0000000 0.0000000
log_total_deaths_week_3_per_million 0.0843342 0.0000000 0.0000000 0.0000000
log_median_daily_deaths_per_million_week_3 0.2986153 0.0000199 0.0000000 0.0000000
log_mean_daily_deaths_per_million_week_3 0.0843342 0.0000000 0.0000000 0.0000000
log_max_daily_deaths_per_million_week_3 0.0131376 0.0000002 0.0000000 0.0000000
log_total_deaths_month_1_per_million 0.1204841 0.0000000 0.0000000 0.0000000
log_median_daily_deaths_per_million_month_1 0.3420044 0.0000307 0.0000000 0.0000000
log_mean_daily_deaths_per_million_month_1 0.1204841 0.0000000 0.0000000 0.0000000
log_max_daily_deaths_per_million_month_1 0.0275145 0.0000001 0.0000000 0.0000000

Linear Models (with FILTERED data)

model_results <- data.frame(matrix(NA, ncol = length(confounding_varibales), nrow = length(dependent_variables)))
rownames(model_results) <- dependent_variables
colnames(model_results) <- paste0(confounding_varibales, "_LMF")


for(i in 1:length(confounding_varibales)){
  for(j in 1:length(dependent_variables)){
    lm_model <- lm(df_filtered[,dependent_variables[j]]~df_filtered[,confounding_varibales[i]], df_filtered)
    model_results[j,i] <- glance(lm_model)$p.value
  }
}
population_density_2018_LMF urban_percentage_2018_LMF HDI_2018_LMF ages_65_up_LMF
total_deaths_per_million 0.0259357 0.2853816 0.2639330 0.5671965
days_to_reached_0.1_deaths_per_million 0.8324767 0.5762872 0.5796591 0.1549722
days_to_reached_1_deaths_per_million 0.6992257 0.5553264 0.4696498 0.0135677
mean_daily_deaths_per_million 0.0297402 0.2409108 0.2182460 0.6012273
median_daily_deaths_per_million 0.0269162 0.2746109 0.2343750 0.6875859
max_daily_deaths_per_million 0.0823445 0.1470788 0.1481730 0.3796946
total_deaths_week_3_per_million 0.2066132 0.6835253 0.4011812 0.2415917
median_daily_deaths_per_million_week_3 0.1557906 0.5544300 0.2227274 0.1412244
mean_daily_deaths_per_million_week_3 0.2066132 0.6835253 0.4011812 0.2415917
max_daily_deaths_per_million_week_3 0.3503533 0.8095073 0.5237430 0.3205938
total_deaths_month_1_per_million 0.1089217 0.6012750 0.3677846 0.3569663
median_daily_deaths_per_million_month_1 0.2028617 0.9351630 0.4418470 0.2295414
mean_daily_deaths_per_million_month_1 0.1089217 0.6012750 0.3677846 0.3569663
max_daily_deaths_per_million_month_1 0.0573573 0.5930605 0.2882405 0.3567455
log_total_deaths_per_million 0.0155019 0.5043627 0.1522237 0.2313471
log_mean_daily_deaths_per_million 0.0192657 0.5401530 0.1520931 0.2023825
log_median_daily_deaths_per_million 0.0523729 0.4443593 0.1343915 0.4251434
log_max_daily_deaths_per_million 0.0658775 0.2917277 0.0381601 0.1396601
log_total_deaths_week_3_per_million 0.1321516 0.9814656 0.1843091 0.0354790
log_median_daily_deaths_per_million_week_3 0.4095748 0.3998172 0.1063668 0.0634728
log_mean_daily_deaths_per_million_week_3 0.1321516 0.9814656 0.1843091 0.0354790
log_max_daily_deaths_per_million_week_3 0.2496209 0.9511354 0.2169627 0.0416779
log_total_deaths_month_1_per_million 0.0504551 0.8859752 0.1189967 0.0564385
log_median_daily_deaths_per_million_month_1 0.2744135 0.4103420 0.0737936 0.0216169
log_mean_daily_deaths_per_million_month_1 0.0504551 0.8859752 0.1189967 0.0564385
log_max_daily_deaths_per_million_month_1 0.0494041 0.8220960 0.0770338 0.1176949

Combination Tables (p-value matrix)

Linear Models

Variable BCG_mean_coverage_LMU BCG_median_coverage_LMU BCG_mean_coverage_LMU2 BCG_median_coverage_LMU2 BCG_mean_coverage_LMF BCG_median_coverage_LMF
total_deaths_per_million Deaths/1 M (Total) 0.2525661 0.3111653 0.2525661 0.3111653 0.0928875 0.0812290
days_to_reached_0.1_deaths_per_million Days to 0.1 Death/1 M 0.0045333 0.0027851 0.0045333 0.0027851 0.9575894 0.8872806
days_to_reached_1_deaths_per_million Days to 1 Death/1 M 0.0009710 0.0010236 0.0009710 0.0010236 0.9977671 0.9678422
mean_daily_deaths_per_million Deaths/day/1 M (mean) 0.3281958 0.3814714 0.3281958 0.3814714 0.0618269 0.0545268
median_daily_deaths_per_million Deaths/day/1 M (median) 0.5989774 0.6878797 0.5989774 0.6878797 0.0734482 0.0671536
max_daily_deaths_per_million Deaths/day/1 M (max) 0.0827120 0.0923322 0.0827120 0.0923322 0.1506741 0.1303914
total_deaths_week_3_per_million Deaths/1 M/week 3 (Total) 0.1197871 0.1487372 0.1197871 0.1487372 0.3558381 0.3541563
median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (median) 0.0936020 0.1180391 0.0936020 0.1180391 0.3013023 0.2973710
mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (mean) 0.1197871 0.1487372 0.1197871 0.1487372 0.3558381 0.3541563
max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (max) 0.2077102 0.2517715 0.2077102 0.2517715 0.2263708 0.2278724
total_deaths_month_1_per_million Deaths/1 M/month 1 (Total) 0.5544449 0.6131759 0.5544449 0.6131759 0.0882472 0.0883046
median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (median) 0.0907533 0.1167336 0.0907533 0.1167336 0.9914933 0.9827768
mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (mean) 0.5544449 0.6131759 0.5544449 0.6131759 0.0882472 0.0883046
max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (max) 0.7810019 0.8059700 0.7810019 0.8059700 0.0254745 0.0260383
log_total_deaths_per_million Deaths/1 M (Total) 0.0015710 0.0022768 0.0015710 0.0022768 0.5910722 0.5501185
log_mean_daily_deaths_per_million Deaths/day/1 M log(mean) 0.0092376 0.0114251 0.0092376 0.0114251 0.5436489 0.5062876
log_median_daily_deaths_per_million Deaths/day/1 M log(median) 0.0294174 0.0339341 0.0294174 0.0339341 0.0195320 0.0161222
log_max_daily_deaths_per_million Deaths/day/1 M loglog(max) 0.0008927 0.0007768 0.0008927 0.0007768 0.3796605 0.3471881
log_total_deaths_week_3_per_million Deaths/1 M/week 3 log(Total) 0.0023889 0.0034696 0.0023889 0.0034696 0.4259286 0.4513514
log_median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(median) 0.0121006 0.0126740 0.0121006 0.0126740 0.2350580 0.2611631
log_mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(mean) 0.0023889 0.0034696 0.0023889 0.0034696 0.4259286 0.4513514
log_max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(max) 0.0044500 0.0057825 0.0044500 0.0057825 0.3204635 0.3472311
log_total_deaths_month_1_per_million Deaths/1 M/month 1 log(Total) 0.0031469 0.0036257 0.0031469 0.0036257 0.3735312 0.3881031
log_median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(median) 0.0069309 0.0089112 0.0069309 0.0089112 0.4846813 0.4728291
log_mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(mean) 0.0031469 0.0036257 0.0031469 0.0036257 0.3735312 0.3881031
log_max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(max) 0.0036552 0.0033046 0.0036552 0.0033046 0.1446985 0.1500235

Anova Models

Variable BCG_policy_AMU BCG_policy_AMU2 BCG_policy_AMF
total_deaths_per_million Deaths/1 M (Total) 0.0000000 0.0000000 0.0110433
days_to_reached_0.1_deaths_per_million Days to 0.1 Death/1 M 0.0001084 0.3232947 0.5138357
days_to_reached_1_deaths_per_million Days to 1 Death/1 M 0.0000000 0.0535218 0.8611822
mean_daily_deaths_per_million Deaths/day/1 M (mean) 0.0000000 0.0000000 0.0151171
median_daily_deaths_per_million Deaths/day/1 M (median) 0.0000000 0.0000000 0.0062392
max_daily_deaths_per_million Deaths/day/1 M (max) 0.0000001 0.0000000 0.0274574
total_deaths_week_3_per_million Deaths/1 M/week 3 (Total) 0.0000830 0.0000000 0.2047799
median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (median) 0.0000000 0.0000000 0.1090813
mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (mean) 0.0000826 0.0000000 0.2047799
max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (max) 0.0065881 0.0000005 0.3865571
total_deaths_month_1_per_million Deaths/1 M/month 1 (Total) 0.0000001 0.0000000 0.1918745
median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (median) 0.0000002 0.0000015 0.1688351
mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (mean) 0.0000001 0.0000000 0.1918745
max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (max) 0.0000267 0.0000000 0.1710350
log_total_deaths_per_million Deaths/1 M (Total) 0.0000000 0.0000000 0.0140011
log_mean_daily_deaths_per_million Deaths/day/1 M log(mean) 0.0000000 0.0000000 0.0201393
log_median_daily_deaths_per_million Deaths/day/1 M log(median) 0.0000000 0.0000000 0.0104689
log_max_daily_deaths_per_million Deaths/day/1 M loglog(max) 0.0000000 0.0000000 0.0170379
log_total_deaths_week_3_per_million Deaths/1 M/week 3 log(Total) 0.0000000 0.0000000 0.5182464
log_median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(median) 0.0000000 0.0000020 0.2786673
log_mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(mean) 0.0000000 0.0000000 0.5182464
log_max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(max) 0.0000000 0.0000002 0.6610229
log_total_deaths_month_1_per_million Deaths/1 M/month 1 log(Total) 0.0000000 0.0000000 0.2631596
log_median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(median) 0.0000000 0.0000002 0.0470756
log_mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(mean) 0.0000000 0.0000000 0.2631596
log_max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(max) 0.0000000 0.0000000 0.1778639

T-test Models

Variable BCG_TF_TMU BCG_TF_2_TMU BCG_TF_TMU2 BCG_TF_2_TMU2 BCG_TF_TMF BCG_TF_2_TMF
total_deaths_per_million Deaths/1 M (Total) 0.0000000 0.0000000 0.0000000 0.0000000 0.0046129 0.0000251
days_to_reached_0.1_deaths_per_million Days to 0.1 Death/1 M 0.9999692 0.9999679 0.9225754 0.7530464 0.2016341 0.1108277
days_to_reached_1_deaths_per_million Days to 1 Death/1 M 1.0000000 1.0000000 0.9879902 0.8595931 0.4230279 0.2906656
mean_daily_deaths_per_million Deaths/day/1 M (mean) 0.0000000 0.0000000 0.0000000 0.0000000 0.0040625 0.0000133
median_daily_deaths_per_million Deaths/day/1 M (median) 0.0000000 0.0000000 0.0000000 0.0000000 0.0040814 0.0000010
max_daily_deaths_per_million Deaths/day/1 M (max) 0.0000000 0.0000000 0.0000053 0.0000000 0.0038820 0.0002303
total_deaths_week_3_per_million Deaths/1 M/week 3 (Total) 0.0000071 0.0000110 0.0000156 0.0000000 0.0392208 0.0827602
median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (median) 0.0000000 0.0000000 0.0000000 0.0000000 0.0172725 0.0513321
mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (mean) 0.0000071 0.0000110 0.0000156 0.0000000 0.0392208 0.0827602
max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 (max) 0.0009609 0.0009861 0.0005770 0.0000001 0.0937636 0.2309676
total_deaths_month_1_per_million Deaths/1 M/month 1 (Total) 0.0000000 0.0000000 0.0000000 0.0000000 0.0329438 0.0288370
median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (median) 0.0000000 0.0000000 0.0000001 0.0000000 0.0417358 0.2190385
mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (mean) 0.0000000 0.0000000 0.0000000 0.0000000 0.0329438 0.0288370
max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 (max) 0.0000075 0.0000060 0.0000491 0.0000000 0.0361193 0.0186749
log_total_deaths_per_million Deaths/1 M (Total) 0.0000000 0.0000000 0.0000000 0.0000008 0.0043163 0.0000686
log_mean_daily_deaths_per_million Deaths/day/1 M log(mean) 0.0000000 0.0000000 0.0000000 0.0000017 0.0054457 0.0001359
log_median_daily_deaths_per_million Deaths/day/1 M log(median) 0.0000000 0.0000000 0.0000000 0.0000084 0.0028382 0.0004002
log_max_daily_deaths_per_million Deaths/day/1 M loglog(max) 0.0000000 0.0000000 0.0000000 0.0000078 0.0028718 0.0030190
log_total_deaths_week_3_per_million Deaths/1 M/week 3 log(Total) 0.0000000 0.0000000 0.0000000 0.0000174 0.1242737 0.2443389
log_median_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(median) 0.0000000 0.0000000 0.0000001 0.0040196 0.0623407 0.2828274
log_mean_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(mean) 0.0000000 0.0000000 0.0000000 0.0000174 0.1242737 0.2443389
log_max_daily_deaths_per_million_week_3 Deaths/1 M/week 3 log(max) 0.0000000 0.0000000 0.0000001 0.0000901 0.1860787 0.3641082
log_total_deaths_month_1_per_million Deaths/1 M/month 1 log(Total) 0.0000000 0.0000000 0.0000000 0.0000067 0.0525123 0.0599427
log_median_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(median) 0.0000000 0.0000000 0.0000000 0.0051662 0.0249550 0.4720666
log_mean_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(mean) 0.0000000 0.0000000 0.0000000 0.0000067 0.0525123 0.0599427
log_max_daily_deaths_per_million_month_1 Deaths/1 M/month 1 log(max) 0.0000000 0.0000000 0.0000000 0.0000067 0.0408538 0.0396862