Adjusting Age Groups

Introduction

This vignette illustrates how EpiStandard can be used to adjust age groups in the standard population and in study results.

Standard Population

EpiStandard includes two standard populations, which are: - European Standard Population 2013 - World Standard Population 2025

Both use the same age groups, and can be used by using the function standardPopulation. You can choose which of these to use by setting the argument region to ‘Europe’ or ‘World’.

library(EpiStandard)
library(dplyr)

ageGroups <- standardPopulation(region = "Europe") 

ageGroups |>
  pull(age_group)
#>  [1] "0 to 4"    "5 to 9"    "10 to 14"  "15 to 19"  "20 to 24"  "25 to 29" 
#>  [7] "30 to 34"  "35 to 39"  "40 to 44"  "45 to 49"  "50 to 54"  "55 to 59" 
#> [13] "60 to 64"  "65 to 69"  "70 to 74"  "75 to 79"  "80 to 84"  "85 to 89" 
#> [19] "90 to 150"

However, some studies might use different age groups which are not compatible with the standard populations. This can be solved by using the function mergeAgeGroups().For example, if a study only uses the age groups ‘0-19’, ‘20-64’ and ‘65 to 150’, the standard population can be adjusted to match

newAgeGroups <- mergeAgeGroups(refdata = ageGroups, newGroups = c("0-19", "20-64", "65-150"))

newAgeGroups
#> # A tibble: 3 × 2
#>   age_group   pop
#>   <chr>     <int>
#> 1 0-19      21500
#> 2 20-64     59000
#> 3 65-150    19500

This will also work if using a bespoke standard population.

df_study <- data.frame(age=c('0-14','15-24','25-44','45-64','65-150'),
                       pop=c(114350,80259,133440,142670,92168))

new_df_study <- mergeAgeGroups(refdata = df_study, newGroups = c("0-24", "25-64", "65-150"),
                               age = "age",
                               pop = "pop")

new_df_study |> dplyr::glimpse()
#> Rows: 3
#> Columns: 2
#> $ age <chr> "0-24", "25-64", "65-150"
#> $ pop <dbl> 194609, 276110, 92168

Study Results

Additionally, you can adjust your study results to merge age groups, while taking into consideration additional stratifications of interest. For example, the data set below shows study results for the UK and France. If we want merge some age groups, but still look at each country separately, we can use the argument strata.

df_study <- data.frame(country=rep(c('UK',"France"), c(5,5)),
                       age=rep(c('0-14','15-24','25-44','45-64','65-150'),2),
                       deaths=c(132,87,413,2316,3425,605,279,3254,9001,8182),
                       fu=c(114350,80259,133440,142670,92168,37164,20036,32693,14947,2077))

new_df_study <- mergeAgeGroups(refdata = df_study, newGroups = c("0-24", "25-64", "65-150"),
                               age = "age",
                               pop = "fu",
                               event = "deaths",
                               strata = "country")

new_df_study |> dplyr::glimpse()
#> Rows: 6
#> Columns: 4
#> $ country <chr> "France", "UK", "France", "UK", "France", "UK"
#> $ age     <chr> "0-24", "0-24", "25-64", "25-64", "65-150", "65-150"
#> $ deaths  <dbl> 884, 219, 12255, 2729, 8182, 3425
#> $ fu      <dbl> 57200, 194609, 47640, 276110, 2077, 92168

Note: All data used in this vignette is artificial.