Create a dataset based on an algorithm and CEPHIA recency testing data.

Get data frame using CEPHIA data for a given recency algorithm specified by the user. Optional subtype and ART exclusions.

createRitaCephia(
  assays,
  algorithm,
  subtype = NULL,
  ever_art = NULL,
  filepath = NULL
)

Arguments

assays: A vector of assays to include
algorithm: A function that defines the recency indicator with arguments in the same order as the assays vector. Arguments do not need to have the same name as assays. E.g., if you have assays = c("BED", "viral_load"), you can have algorithm = function(b, v) ... where b indicates BED and v indicates viral load.
subtype: HIV subtypes to include (one of "A1", "B", "C", "D"). By default includes everyone.
ever_art: Subset data to only those who have used ARTs or have not. By default includes everyone.
filepath: Optional full filepath to downloaded CEPHIA dataset (possibly obtained via the download_cephia). If NULL, then uses CEPHIA version in this package.

Value

A data frame with the following columns:

id: ID column (there can be multiple measurements per individual)
ui: infection duration in days (note, if using with other functions in this package, may have to convert to years)
ri: identified as recent based on algorithm

Examples

f <- function(b, l, v){
  ifelse((b > 1 & l < 3 & !is.na(v)), 1, 0)
}
createRitaCephia(assays=c("BED", "LAg-Sedia", "viral_load"),
             algorithm=f)
#> There are 53 missing values for viral_load 
#>             id   ui ri
#>    1: 72748497  716  0
#>    2: 72748497  884  0
#>    3: 87007074  360  0
#>    4: 87007074  696  0
#>    5: 87007074 1190  0
#>   ---                 
#> 1870: 41616271  134  0
#> 1871: 36452124   46  0
#> 1872: 43155152  114  0
#> 1873: 95702775  181  1
#> 1874: 13028433   67  0
f <- function(l, v){
  v <- ifelse(l > 1.5, 0, v)
  return(
    ifelse((l <= 1.5) & (v > 1000), 1, 0)
  )
}
test <- createRitaCephia(assays=c("LAg-Sedia", "viral_load"), algorithm=f)
#> There are 53 missing values for viral_load 
#> Removing 10 observations with missing recency indicator after application of the algorithm.