Simulating a kidney transplant waiting list is easy to do, with functions from {simK}.
Candidates
The minimum information needed for kidney transplant candidates encompasses: HLA typing, ABO blood group, cPRA, age, time on dialysis, and clinical urgency.
We can start by defining our wait list size (n) and patients’ HLA according to its origin:
candidates <- hla_sample(n = 100, replace = TRUE, origin = 'API')
#> New names:
#> • `api` -> `api...4`
#> • `afa` -> `afa...5`
#> • `cau` -> `cau...6`
#> • `his` -> `his...7`
#> • `api` -> `api...11`
#> • `afa` -> `afa...12`
#> • `cau` -> `cau...13`
#> • `his` -> `his...14`
# lets create also an unique identifier for each patient
candidates$ID <- paste0('R', rownames(candidates))
candidates
#> # A tibble: 100 × 7
#> A1 A2 B1 B2 DR1 DR2 ID
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr>
#> 1 1 33 61 44 7 7 R1
#> 2 24 2 35 13 12 3 R2
#> 3 2 24 60 62 15 4 R3
#> 4 29 29 7 7 3 10 R4
#> 5 11 11 54 52 4 15 R5
#> 6 24 24 38 7 15 1 R6
#> 7 2 11 62 60 15 15 R7
#> 8 33 33 44 58 13 13 R8
#> 9 30 24 13 75 7 12 R9
#> 10 2 3 51 44 16 13 R10
#> # ℹ 90 more rowsPatients’ blood group can be added with abo() and their ages with function ages():
# we just have to define sample size and probabilities for A, AB, B and O groups (in this order)
candidates$bg <- abo(n = nrow(candidates), probs = c(0.4, 0.04, 0.06, 0.5))
# for patients' ages we define a trimmed normal distributed value with mean and sd
candidates$age <- ages(n = nrow(candidates), lower = 18, upper = 80, mean = 65, sd = 30)
# let's take a look at our transplant candidates
candidates
#> # A tibble: 100 × 9
#> A1 A2 B1 B2 DR1 DR2 ID bg age
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <dbl>
#> 1 1 33 61 44 7 7 R1 O 48
#> 2 24 2 35 13 12 3 R2 A 58
#> 3 2 24 60 62 15 4 R3 O 67
#> 4 29 29 7 7 3 10 R4 A 69
#> 5 11 11 54 52 4 15 R5 B 79
#> 6 24 24 38 7 15 1 R6 O 27
#> 7 2 11 62 60 15 15 R7 A 44
#> 8 33 33 44 58 13 13 R8 A 52
#> 9 30 24 13 75 7 12 R9 A 76
#> 10 2 3 51 44 16 13 R10 O 77
#> # ℹ 90 more rowsNow we must define groups of sensitization to transplant candidates according to a 4 elements vector of probabilities corresponding to cPRA = 0%; 1% < cPRA < 50%; 50% < cPRA < 85%; and cPRA > 85%, respectively.
candidates$cPRA <- cpra(n = nrow(candidates), probs = c(0.7, 0.1, 0.1, 0.1))
# and we can also classify those hipersensitised patients
candidates$hiper <- candidates$cPRA > 85Time on dialysis (in months) is computed according to blood group and hipersensitation status:
candidates <- candidates |>
dplyr::rowwise() |>
dplyr::mutate(dialysis = dial(hiper = hiper, bg = bg, seed.number = NA)) |>
dplyr::ungroup()
candidates
#> # A tibble: 100 × 12
#> A1 A2 B1 B2 DR1 DR2 ID bg age cPRA hiper dialysis
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <dbl> <dbl> <lgl> <dbl>
#> 1 1 33 61 44 7 7 R1 O 48 0 FALSE 66
#> 2 24 2 35 13 12 3 R2 A 58 73 FALSE 57
#> 3 2 24 60 62 15 4 R3 O 67 0 FALSE 90
#> 4 29 29 7 7 3 10 R4 A 69 37 FALSE 40
#> 5 11 11 54 52 4 15 R5 B 79 82 FALSE 23
#> 6 24 24 38 7 15 1 R6 O 27 0 FALSE 49
#> 7 2 11 62 60 15 15 R7 A 44 21 FALSE 40
#> 8 33 33 44 58 13 13 R8 A 52 0 FALSE 16
#> 9 30 24 13 75 7 12 R9 A 76 0 FALSE 2
#> 10 2 3 51 44 16 13 R10 O 77 0 FALSE 82
#> # ℹ 90 more rowsAnd we can finish it defining patients clinical urgent (if any):
# let´s asume we have 5% of candidates that are clinical urgent for transplantation
candidates$urgent <- sample(c(0,1), size = 100, replace = TRUE, prob = c(0.95,0.05))
# reording the columns
candidates |>
dplyr::select(ID, bg, A1, A2, B1, B2, DR1, DR2, age, cPRA, hiper, dialysis, urgent)
#> # A tibble: 100 × 13
#> ID bg A1 A2 B1 B2 DR1 DR2 age cPRA hiper dialysis
#> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <chr> <dbl> <dbl> <lgl> <dbl>
#> 1 R1 O 1 33 61 44 7 7 48 0 FALSE 66
#> 2 R2 A 24 2 35 13 12 3 58 73 FALSE 57
#> 3 R3 O 2 24 60 62 15 4 67 0 FALSE 90
#> 4 R4 A 29 29 7 7 3 10 69 37 FALSE 40
#> 5 R5 B 11 11 54 52 4 15 79 82 FALSE 23
#> 6 R6 O 24 24 38 7 15 1 27 0 FALSE 49
#> 7 R7 A 2 11 62 60 15 15 44 21 FALSE 40
#> 8 R8 A 33 33 44 58 13 13 52 0 FALSE 16
#> 9 R9 A 30 24 13 75 7 12 76 0 FALSE 2
#> 10 R10 O 2 3 51 44 16 13 77 0 FALSE 82
#> # ℹ 90 more rows
#> # ℹ 1 more variable: urgent <dbl>