library(readxl)
library(tidyverse)
evaluaciones_cognitivas <- read_excel("evaluaciones_cognitivas.xlsx")

evaluaciones_cognitivas$RAVLT_dif-> evaluaciones_cognitivas$CERAD_dif
base<-evaluaciones_cognitivas %>% filter(Data_type == "PN")
base$MMSE<-as.numeric(base$MMSE)
base$CERAD_dif<-as.numeric(base$CERAD_dif)

base$criterio_edad<-base$Age <77 & base$Age >65


base$criterio_mmse<-base$MMSE < 0 & base$MMSE > -1.5 

base$criterio_CERAD<-base$CERAD_dif <0


base$criterio_actual<- base$criterio_mmse == T & base$criterio_CERAD == T



base$criterio_brasil<- base$criterio_mmse == T | base$criterio_CERAD == T
library(gtsummary)

base %>% filter( Age > 60 & Age<77) %>% select(criterio_mmse, criterio_CERAD, criterio_actual,criterio_brasil) %>% tbl_summary()
## Warning: The `.dots` argument of `group_by()` is deprecated as of dplyr 1.0.0.
Characteristic N = 1,2761
criterio_mmse 200 (16%)
Unknown 11
criterio_CERAD 979 (78%)
Unknown 24
criterio_actual 179 (14%)
Unknown 11
criterio_brasil 1,000 (80%)
Unknown 24

1 n (%) 

library(ggplot2)
base_edad<-base%>% filter( Age > 60 & Age<77)

summary(base_edad$MMSE)
##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max.     NA's 
## -20.7700  -1.5400   0.0000  -0.7932   0.7700   2.3100       11
ggplot(base_edad, aes(x=MMSE, y=CERAD_dif, color=criterio_CERAD))+
  geom_jitter()+
  geom_vline(xintercept = 0, linetype="dashed")+
  geom_vline(xintercept = -1.5, linetype="dashed")+
  xlim(-10,2)
## Warning: Removed 40 rows containing missing values (geom_point).