First numbers

  • Number of RCTs relevant to burden, in total, per region
  • Number of patients enrolled in RCTs relevant to the burden, in total, per region
  • Number of RCTs and number of patients per region per disease

1. First, we evaluate if total number of RCTs relevant to the burden is similar across simulations


In [1]:
data <- read.table("/media/igna/Elements/HotelDieu/Cochrane/MappingRCTs_vs_Burden/database_RCTs_regions_27diseases.txt")
nrow(data)


117180

In [2]:
data$Sample[data$Sample<10 | data$Sample>200000] <- NA
table(is.na(data$Sample))
sum(data$Sample,na.rm = TRUE)


 FALSE   TRUE 
107558   9622 
43952638

In [2]:
library(data.table)

In [38]:
options(repr.plot.width=9, repr.plot.height=9)
par(mfrow=c(5,5),mar=c(1,1,3,1))
DF <- fread(paste(c("/media/igna/Elements/HotelDieu/Cochrane/MappingRCTs_vs_Burden/Replicates/Metrics_over_repl/Metrics_over_replicates_",
                    as.character(d),".txt"),collapse=""))
hist(DF$RCTs[DF$Reg=="All"],xlim=c(70000,90000),xlab=NULL,main="all")
for(i in 1:length(sms)){ 
    DF <- fread(paste(c("/media/igna/Elements/HotelDieu/Cochrane/MappingRCTs_vs_Burden/Replicates/Metrics_over_repl/",
                        sms[i]),collapse=""))
    hist(DF$RCTs[DF$Reg=="All" & DF$Dis=="all"],
         xlim=c(70000,90000),xlab=NULL,
         main=substr(sms[i],25,nchar(sms[i])-4))               
                        }



In [39]:
Mgbd <- read.table("../Data/27_gbd_groups.txt")

In [42]:
Mgbd$x[12]


Neoplasms

In [43]:
PERF <- read.csv('../Tables/Performances_per_27disease_data.csv')

In [45]:
PERF[12,]


XdisGBDTP_DisFP_DisTN_DisFN_DisTP_OthFP_OthTN_OthFN_Oth
1212 12 Neoplasms933 42 1763 25 1213 214 1198 138

In [ ]:


In [ ]:


In [ ]:


In [ ]:


In [ ]:
dprp <- DF[DF$Dis=="dis",c("RCTs","Patients")]/DF[DF$Dis=="all",c("RCTs","Patients")]
dprp$Reg <- DF$Region[DF$Dis=="dis"]
tapply(dprp$RCTs,dprp$Reg,function(x){100*quantile(x,probs=c(0.025,0.5,0.975))})
tapply(dprp$Patients,dprp$Reg,function(x){100*quantile(x,probs=c(0.025,0.5,0.975))})

DF[,quantile(.SD,probs=c(0.025,0.5,0.975)),by=.(Dis,Region),.SDcols=c("RCTs","Patients")]


tapply(DF$RCTs[DF$Dis=="dis"],DF$Reg[DF$Dis=="dis"],function(x){quantile(x,probs=c(0.025,0.5,0.975))})

In [ ]:


In [ ]:


In [ ]:
regs <- strsplit(as.character(data$Regions),"&")
DRY <- do.call('cbind',tapply(regs,data$year,function(x){table(unlist(x))}))

DRY <- DRY[order(apply(DRY,1,sum)),]

barplot(DRY[rownames(DRY)!="High-income",])