Skip to contents

These functions determine which items in a vector can be considered (the start of) a new episode, based on the argument episode_days. This can be used to determine clinical episodes for any epidemiological analysis. The get_episode() function returns the index number of the episode per group, while the is_new_episode() function returns values TRUE/FALSE to indicate whether an item in a vector is the start of a new episode.

Usage

get_episode(x, episode_days, ...)

is_new_episode(x, episode_days, ...)

Arguments

x

vector of dates (class Date or POSIXt), will be sorted internally to determine episodes

episode_days

required episode length in days, can also be less than a day or Inf, see Details

...

ignored, only in place to allow future extensions

Value

  • get_episode(): a double vector

  • is_new_episode(): a logical vector

Details

Dates are first sorted from old to new. The oldest date will mark the start of the first episode. After this date, the next date will be marked that is at least episode_days days later than the start of the first episode. From that second marked date on, the next date will be marked that is at least episode_days days later than the start of the second episode which will be the start of the third episode, and so on. Before the vector is being returned, the original order will be restored.

The first_isolate() function is a wrapper around the is_new_episode() function, but is more efficient for data sets containing microorganism codes or names and allows for different isolate selection methods.

The dplyr package is not required for these functions to work, but these functions do support variable grouping and work conveniently inside dplyr verbs such as filter(), mutate() and summarise().

See also

Examples

# `example_isolates` is a data set available in the AMR package.
# See ?example_isolates
df <- example_isolates[sample(seq_len(2000), size = 200), ]

get_episode(df$date, episode_days = 60) # indices
#>   [1] 56 58 21 46 12 43 36 38 26 65 24 10 57 39  9 61 62 19 16  6 63 35 46 61 27
#>  [26] 65 54 57 60  5 52 11 35 47  4 53 61 23 63 10 28 15  1  4 18 59  9 41 51 38
#>  [51] 52 60 10 19 38  2 45 34 41 30 38 27 29  8  8 24 39  2 26 52 34 41 63 16  3
#>  [76] 58 24 17 38 42 30 52 13 18  5 10 63 13 55 48  7 47 40 52 34 38 15 62 28  9
#> [101] 16 56 40  6 53 32  5  2 50 23  6 16  1 64 39 27 47 12 19 62 25 52 56 48 28
#> [126] 58 21 62 65 44 10 15 10 15 62 55 65 50 28 12 42 61 62 63  6 49 59 33 23 53
#> [151] 38 15 20 56 26 46 60 36 40 32 18 29  8 16 37  4 31 15  4 38 62 36 22 43 33
#> [176] 40 38 37 64 17 19  9 14  2 22 59 37 11 22 28 53 31 12 56 42 52  7 13 30 22
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#>   [1]  TRUE  TRUE FALSE  TRUE  TRUE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE
#>  [13] FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE
#>  [25]  TRUE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE FALSE FALSE  TRUE FALSE FALSE
#>  [37]  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE  TRUE
#>  [49]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE
#>  [61] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE
#>  [73] FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#>  [85] FALSE  TRUE  TRUE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#>  [97] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [109]  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE
#> [121]  TRUE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE
#> [133] FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [145] FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
#> [157]  TRUE FALSE  TRUE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE
#> [169] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE
#> [181] FALSE  TRUE  TRUE  TRUE  TRUE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE
#> [193] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE

# filter on results from the third 60-day episode only, using base R
df[which(get_episode(df$date, 60) == 3), ]
#> # A tibble: 1 × 46
#>   date       patient   age gender ward     mo           PEN   OXA   FLC   AMX  
#>   <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <rsi> <rsi> <rsi> <rsi>
#> 1 2002-12-14 144280     76 F      Clinical B_STPHY_AURS R     NA    S     R    
#> # … with 36 more variables: AMC <rsi>, AMP <rsi>, TZP <rsi>, CZO <rsi>,
#> #   FEP <rsi>, CXM <rsi>, FOX <rsi>, CTX <rsi>, CAZ <rsi>, CRO <rsi>,
#> #   GEN <rsi>, TOB <rsi>, AMK <rsi>, KAN <rsi>, TMP <rsi>, SXT <rsi>,
#> #   NIT <rsi>, FOS <rsi>, LNZ <rsi>, CIP <rsi>, MFX <rsi>, VAN <rsi>,
#> #   TEC <rsi>, TCY <rsi>, TGC <rsi>, DOX <rsi>, ERY <rsi>, CLI <rsi>,
#> #   AZM <rsi>, IPM <rsi>, MEM <rsi>, MTR <rsi>, CHL <rsi>, COL <rsi>,
#> #   MUP <rsi>, RIF <rsi>

# the functions also work for less than a day, e.g. to include one per hour:
get_episode(c(
  Sys.time(),
  Sys.time() + 60 * 60
),
episode_days = 1 / 24
)
#> [1] 1 2

# \donttest{
if (require("dplyr")) {
  # is_new_episode() can also be used in dplyr verbs to determine patient
  # episodes based on any (combination of) grouping variables:
  df %>%
    mutate(condition = sample(
      x = c("A", "B", "C"),
      size = 200,
      replace = TRUE
    )) %>%
    group_by(condition) %>%
    mutate(new_episode = is_new_episode(date, 365)) %>%
    select(patient, date, condition, new_episode)
}
#> # A tibble: 200 × 4
#> # Groups:   condition [3]
#>    patient date       condition new_episode
#>    <chr>   <date>     <chr>     <lgl>      
#>  1 0DBF93  2015-10-12 B         TRUE       
#>  2 F61180  2016-02-15 C         FALSE      
#>  3 351193  2007-01-15 B         TRUE       
#>  4 685398  2013-04-16 A         TRUE       
#>  5 978187  2004-09-28 A         FALSE      
#>  6 743093  2012-09-14 B         FALSE      
#>  7 725779  2010-09-04 C         FALSE      
#>  8 E16523  2011-06-25 A         FALSE      
#>  9 F3BD65  2008-01-30 A         FALSE      
#> 10 3D2C93  2017-12-13 C         TRUE       
#> # … with 190 more rows
if (require("dplyr")) {
  df %>%
    group_by(ward, patient) %>%
    transmute(date,
      patient,
      new_index = get_episode(date, 60),
      new_logical = is_new_episode(date, 60)
    )
}
#> # A tibble: 200 × 5
#> # Groups:   ward, patient [181]
#>    ward       date       patient new_index new_logical
#>    <chr>      <date>     <chr>       <dbl> <lgl>      
#>  1 Clinical   2015-10-12 0DBF93          1 TRUE       
#>  2 ICU        2016-02-15 F61180          1 TRUE       
#>  3 ICU        2007-01-15 351193          1 TRUE       
#>  4 Clinical   2013-04-16 685398          1 TRUE       
#>  5 ICU        2004-09-28 978187          1 TRUE       
#>  6 Outpatient 2012-09-14 743093          1 TRUE       
#>  7 Outpatient 2010-09-04 725779          1 TRUE       
#>  8 Clinical   2011-06-25 E16523          1 TRUE       
#>  9 Outpatient 2008-01-30 F3BD65          1 TRUE       
#> 10 ICU        2017-12-13 3D2C93          1 TRUE       
#> # … with 190 more rows
if (require("dplyr")) {
  df %>%
    group_by(ward) %>%
    summarise(
      n_patients = n_distinct(patient),
      n_episodes_365 = sum(is_new_episode(date, episode_days = 365)),
      n_episodes_60 = sum(is_new_episode(date, episode_days = 60)),
      n_episodes_30 = sum(is_new_episode(date, episode_days = 30))
    )
}
#> # A tibble: 3 × 5
#>   ward       n_patients n_episodes_365 n_episodes_60 n_episodes_30
#>   <chr>           <int>          <int>         <int>         <int>
#> 1 Clinical          113             15            51            67
#> 2 ICU                53             12            32            38
#> 3 Outpatient         15              8            13            14
if (require("dplyr")) {

  # grouping on patients and microorganisms leads to the same
  # results as first_isolate() when using 'episode-based':
  x <- df %>%
    filter_first_isolate(
      include_unknown = TRUE,
      method = "episode-based"
    )

  y <- df %>%
    group_by(patient, mo) %>%
    filter(is_new_episode(date, 365)) %>%
    ungroup()

  identical(x, y)
}
#> Including isolates from ICU.
#> [1] FALSE
if (require("dplyr")) {

  # but is_new_episode() has a lot more flexibility than first_isolate(),
  # since you can now group on anything that seems relevant:
  df %>%
    group_by(patient, mo, ward) %>%
    mutate(flag_episode = is_new_episode(date, 365)) %>%
    select(group_vars(.), flag_episode)
}
#> # A tibble: 200 × 4
#> # Groups:   patient, mo, ward [190]
#>    patient mo            ward       flag_episode
#>    <chr>   <mo>          <chr>      <lgl>       
#>  1 0DBF93  B_STPHY_EPDR  Clinical   TRUE        
#>  2 F61180  B_ESCHR_COLI  ICU        TRUE        
#>  3 351193  B_ESCHR_COLI  ICU        TRUE        
#>  4 685398  B_STRPT_GRPB  Clinical   TRUE        
#>  5 978187  B_STRPT_SLVR  ICU        TRUE        
#>  6 743093  B_ENTRBC_CLOC Outpatient TRUE        
#>  7 725779  B_STRPT_PNMN  Outpatient TRUE        
#>  8 E16523  B_STRPT_EQNS  Clinical   TRUE        
#>  9 F3BD65  B_STRPT_ANGN  Outpatient TRUE        
#> 10 3D2C93  B_STPHY_EPDR  ICU        TRUE        
#> # … with 190 more rows
# }