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$date, episode_days = 60) # indices
#>   [1] 10  6 35 25 25 48 41 23 28 27 51 26 47 12  1 52  2 16 34 47 57 66 62 34 41
#>  [26] 27 12 12 59 49 15 13 51 14 14  5 13 10 66 34 65 26 44 51  5 27 50 42 58 45
#>  [51]  9 53 51 65 24  8 13 50  1 40 31  7 62 32 40 28 19 55 34 36 24  7 28 60 31
#>  [76] 10 12 41 43 58 64 57 60  8  9 47 39 30 52  4 67 61 55 13 37  4 57 32 44 67
#> [101] 27 10  8 56 15  6 66 67 54 38 34 17  3 64 49 18 50 65 40 60  3 49 10 17 21
#> [126] 12 12 51  4 41 60 12 63 18  9 45 11 47 21 37 65 61 65 57  5 58 57 12 28 17
#> [151] 58 25 66 60 24  8 29 58 39 30 60 45 23 50 12 38 55 63 20 43 63  2 36 31 11
#> [176]  7 10 16 61 33 20 18 44 35 26 40 43 14 22 33 42 28 46 40 39  2 42 32  7 12
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#>   [1]  TRUE  TRUE FALSE FALSE  TRUE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE
#>  [13] FALSE FALSE  TRUE  TRUE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE
#>  [25] FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE  TRUE
#>  [37]  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE  TRUE FALSE
#>  [49] FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE
#>  [61]  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE
#>  [73]  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
#>  [85]  TRUE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE
#>  [97] FALSE  TRUE  TRUE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE
#> [109]  TRUE FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE FALSE FALSE FALSE
#> [121]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [133] FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE  TRUE FALSE FALSE
#> [145] FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE
#> [157]  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE
#> [169] FALSE  TRUE  TRUE  TRUE  TRUE FALSE  TRUE  TRUE FALSE FALSE FALSE  TRUE
#> [181]  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
#> [193]  TRUE FALSE FALSE 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: 2 × 46
#>   date       patient   age gender ward  mo           PEN   OXA   FLC   AMX  
#>   <date>     <chr>   <dbl> <chr>  <chr> <mo>         <rsi> <rsi> <rsi> <rsi>
#> 1 2002-07-30 218912     76 F      ICU   B_ESCHR_COLI R     NA    NA    NA   
#> 2 2002-07-15 C42193     84 M      ICU   B_STPHY_HMNS R     NA    R     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 189363  2004-06-22 C         FALSE      
#>  2 F35553  2003-09-05 C         FALSE      
#>  3 2E6911  2010-06-05 A         FALSE      
#>  4 451990  2008-03-22 A         FALSE      
#>  5 192790  2008-02-15 C         FALSE      
#>  6 B43936  2013-05-16 C         FALSE      
#>  7 5B78D5  2011-09-19 C         FALSE      
#>  8 179451  2007-09-15 A         TRUE       
#>  9 329273  2008-10-27 A         FALSE      
#> 10 B12441  2008-09-21 A         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 [183]
#>    ward       date       patient new_index new_logical
#>    <chr>      <date>     <chr>       <dbl> <lgl>      
#>  1 Clinical   2004-06-22 189363          2 TRUE       
#>  2 ICU        2003-09-05 F35553          2 TRUE       
#>  3 Clinical   2010-06-05 2E6911          1 TRUE       
#>  4 Clinical   2008-03-22 451990          1 TRUE       
#>  5 Outpatient 2008-02-15 192790          1 TRUE       
#>  6 ICU        2013-05-16 B43936          1 TRUE       
#>  7 Clinical   2011-09-19 5B78D5          1 TRUE       
#>  8 ICU        2007-09-15 179451          1 TRUE       
#>  9 Clinical   2008-10-27 329273          1 TRUE       
#> 10 Clinical   2008-09-21 B12441          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          117             15            54            78
#> 2 ICU                56             12            36            43
#> 3 Outpatient         10              6             9             9
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 [192]
#>    patient mo           ward       flag_episode
#>    <chr>   <mo>         <chr>      <lgl>       
#>  1 189363  B_STPHY_AURS Clinical   FALSE       
#>  2 F35553  B_ENTRC      ICU        TRUE        
#>  3 2E6911  B_ESCHR_COLI Clinical   TRUE        
#>  4 451990  B_ESCHR_COLI Clinical   TRUE        
#>  5 192790  B_STPHY_CONS Outpatient TRUE        
#>  6 B43936  B_BCTRD_FRGL ICU        TRUE        
#>  7 5B78D5  B_STPHY_AURS Clinical   TRUE        
#>  8 179451  B_ESCHR_COLI ICU        TRUE        
#>  9 329273  B_STRPT_PNMN Clinical   TRUE        
#> 10 B12441  B_STPHY_CONS Clinical   TRUE        
#> # … with 190 more rows
# }