Example 1 (mean=0, sd =365)

library(dplyr)

## 
## Attaching package: 'dplyr'

## The following objects are masked from 'package:stats':
## 
##     filter, lag

## The following objects are masked from 'package:base':
## 
##     intersect, setdiff, setequal, union

set.seed(100)

age <- sample(seq(as.Date('1960-01-01'), as.Date('1985-01-01'), by="day"), 1000)
date1 <- sample(seq(as.Date('2000-01-01'), as.Date('2020-01-01'), by="day"), 1000)
status <- sample(c(0, 1), size = 1000, replace = TRUE, prob = c(0.2, 0.8))
df <- data.frame(age, date1, status)
df <- df %>% mutate(id = row_number())


head(df, n=5)

##          age      date1 status id
## 1 1970-05-13 2003-01-01      0  1
## 2 1961-05-17 2010-03-22      1  2
## 3 1969-05-22 2005-06-05      1  3
## 4 1970-02-12 2007-12-16      0  4
## 5 1971-03-13 2018-09-18      1  5

Mean average entery to VA care for 0= epi group vs. 1= non-epi group without simulation

library(ggplot2)

df %>%
ggplot(aes(x    = factor(status), 
           y    = as.numeric(difftime(date1, age, unit = 'w'))/52,
           fill = factor(status))) + 
  geom_boxplot(width = 0.6) +
  guides(fill = guide_none()) +
  labs(x = 'Status', y = 'Age (years)')

df$date2 <- df$date1 - df$status * round(rnorm(nrow(df), 0, 365))

Mean average entery to VA care for 0= epi group vs. 1= non-epi group with simulation

df %>% 
  ggplot(aes(x    = factor(status), 
             y    = as.numeric(difftime(date2, age, unit = 'w'))/52,
             fill = factor(status))) + 
  geom_boxplot(width = 0.6) +
  guides(fill = guide_none()) +
  labs(x = 'Status', y = 'Age (years)')

hist(df$date1, breaks = 10)

hist(df$date2, breaks = 10)

Mean age entry

# real data
df %>% 
  filter(status==1) %>% 
  mutate( y    = as.numeric(difftime(date1, age, unit = 'days'))) %>% 
    summarise(
        n = n(),
        mean = mean(y, na.rm=T)/365)

##     n     mean
## 1 808 37.22442

# sim data

df %>% 
  filter(status==1) %>% 
  mutate( y    = as.numeric(difftime(date2, age, unit = 'days'))) %>% 
    summarise(
        n = n(),
        mean = mean(y, na.rm=T)/365)

##     n     mean
## 1 808 37.23418

Example 2 (mean=-978, sd =365)

Mean average entery to VA care for 0= epi group vs. 1= non-epi group without simulation

library(ggplot2)

df %>%
ggplot(aes(x    = factor(status), 
           y    = as.numeric(difftime(date1, age, unit = 'w'))/52,
           fill = factor(status))) + 
  geom_boxplot(width = 0.6) +
  guides(fill = guide_none()) +
  labs(x = 'Status', y = 'Age (years)')

df$date2 <- df$date1 - df$status * round(rnorm(nrow(df), -978, 365))

Mean average entery to VA care for 0= epi group vs. 1= non-epi group with simulation

df %>% 
  ggplot(aes(x    = factor(status), 
             y    = as.numeric(difftime(date2, age, unit = 'w'))/52,
             fill = factor(status))) + 
  geom_boxplot(width = 0.6) +
  guides(fill = guide_none()) +
  labs(x = 'Status', y = 'Age (years)')

hist(df$date1, breaks = 10)

hist(df$date2, breaks = 10)

Mean age entry

# real data
df %>% 
  filter(status==1) %>% 
  mutate( y    = as.numeric(difftime(date1, age, unit = 'days'))) %>% 
    summarise(
        n = n(),
        mean = mean(y, na.rm=T)/365)

##     n     mean
## 1 808 37.22442

# sim data

df %>% 
  filter(status==1) %>% 
  mutate( y    = as.numeric(difftime(date2, age, unit = 'days'))) %>% 
    summarise(
        n = n(),
        mean = mean(y, na.rm=T)/365)

##     n     mean
## 1 808 39.91363

Example of simulation

Ali

4/25/2022

Example 1 (mean=0, sd =365)

Mean age entry

Example 2 (mean=-978, sd =365)

Mean age entry