Project bed occupancy from admissions

This function projects bed occupancy using admission incidence and a distribution of length of stay (los).

project_beds(x, ...)

# S3 method for projections
project_beds(x, r_los, n_sim = 10, last_date = NULL, ...)

# S3 method for incidence
project_beds(x, r_los, n_sim = 10, last_date = NULL, ...)

Arguments

x	Either a \link[projections:build_projections]{projections} object storing one or more integer forecasts of admissions or a \link[incidence:incidence]{incidence} object of known daily admissions.
...	Additional arguments passed to other methods.
r_los	A function with a single parameter n returning n integer values of lenth of hospital stay (LoS) in days. Ideally, this should come from a discrete random distribution, such as rexp or any distcrete object.
n_sim	The number of times duration of hospitalisation is simulated for each admission. Defaults to 10. Only relevant for low (<30) numbers of initial admissions, in which case it helps accounting for the uncertainty in LoS.
last_date	the last date to simulate until (defaults to the maximum date of x).

Value

A \link[projections:build_projections]{projections} object produced from the admission trajectories.

Examples

  ## fake LoS; check `\link[distcrete:distcrete]{distcrete::distcrete}`
  ## for discretising existing distributions
  r_los <- function(n) rgeom(n, prob = .3)


  # Incidence input

  ## fake data
  dates <- Sys.Date() - 1:10
  admissions <- sample(1:100, 10, replace = TRUE)
  x <- incidence(rep(dates, admissions))
  x
#> <incidence object>
#> [500 cases from days 2020-07-11 to 2020-07-20]
#> 
#> $counts: matrix with 10 rows and 1 columns
#> $n: 500 cases in total
#> $dates: 10 dates marking the left-side of bins
#> $interval: 1 day
#> $timespan: 10 days
#> $cumulative: FALSE
#> 
  plot(x)

  ## project bed occupancy
  beds <- project_beds(x, r_los)
  beds
#> 
#> /// Incidence projections //
#> 
#>   // class: projections, matrix, array
#>   // 10 dates (rows); 10 simulations (columns)
#> 
#>  // first rows/columns:
#>            sim_1 sim_2 sim_3 sim_4 sim_5 sim_6
#> 2020-07-11     4     3     2     4     5     4
#> 2020-07-12    49    55    48    55    49    51
#> 2020-07-13    85    94    84    95    84    89
#> 2020-07-14    94   119   103   107   111   115
#>  .
#>  .
#>  .
#> 
#>  // dates:
#>  [1] "2020-07-11" "2020-07-12" "2020-07-13" "2020-07-14" "2020-07-15"
#>  [6] "2020-07-16" "2020-07-17" "2020-07-18" "2020-07-19" "2020-07-20"
#> 
  plot(beds)


  # Projections input

  ## make fake data - each column after the first is a separate forecast
  admissions <- data.frame(
      date = Sys.Date() - 1:10,
      as.data.frame(replicate(30, sample(1:100, 10, replace = TRUE))))

  x <- build_projections(x = admissions[, -1], dates = admissions$date)
  x
#> 
#> /// Incidence projections //
#> 
#>   // class: projections, matrix, array
#>   // 10 dates (rows); 30 simulations (columns)
#> 
#>  // first rows/columns:
#>            V1 V2 V3 V4  V5 V6
#> 2020-07-11  8 17 36 95 100 89
#> 2020-07-12 36 52 99  2  80 57
#> 2020-07-13 45 16 87 37  11 73
#> 2020-07-14 56 43  1 22  35 65
#>  .
#>  .
#>  .
#> 
#>  // dates:
#>  [1] "2020-07-11" "2020-07-12" "2020-07-13" "2020-07-14" "2020-07-15"
#>  [6] "2020-07-16" "2020-07-17" "2020-07-18" "2020-07-19" "2020-07-20"
#> 
  plot(x)

  ## project bed occupancy
  beds <- project_beds(x, r_los)
  beds
#> 
#> /// Incidence projections //
#> 
#>   // class: projections, matrix, array
#>   // 10 dates (rows); 300 simulations (columns)
#> 
#>  // first rows/columns:
#>            sim_1 sim_2 sim_3 sim_4 sim_5 sim_6
#> 2020-07-11     7     6     4     5     7     6
#> 2020-07-12    33    27    30    30    32    25
#> 2020-07-13    55    58    52    54    57    49
#> 2020-07-14    81    78    84    87    73    76
#>  .
#>  .
#>  .
#> 
#>  // dates:
#>  [1] "2020-07-11" "2020-07-12" "2020-07-13" "2020-07-14" "2020-07-15"
#>  [6] "2020-07-16" "2020-07-17" "2020-07-18" "2020-07-19" "2020-07-20"
#> 
  plot(beds)