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

# filter on results from the third 60-day episode only, using base R
df[which(get_episode(df$date, 60) == 3), ]
#> # A tibble: 4 × 46
#>   date       patient   age gender ward     mo           PEN   OXA   FLC   AMX  
#>   <date>     <chr>   <dbl> <chr>  <chr>    <mo>         <rsi> <rsi> <rsi> <rsi>
#> 1 2002-07-21 955940     82 F      Clinical B_PSDMN_AERG R     NA    NA    R    
#> 2 2002-06-04 082413     78 M      ICU      B_STRPT_PNMN S     NA    NA    S    
#> 3 2002-07-28 F54261     69 M      Clinical B_STPHY_CONS R     NA    S     NA   
#> 4 2002-06-05 24D393     20 F      Clinical B_ESCHR_COLI R     NA    NA    NA   
#> # … 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 784436  2016-05-28 B         TRUE       
#>  2 533225  2008-10-01 C         FALSE      
#>  3 B95551  2004-12-17 B         FALSE      
#>  4 D65308  2004-11-03 A         FALSE      
#>  5 BF4515  2010-12-29 A         FALSE      
#>  6 F35553  2002-10-20 A         FALSE      
#>  7 D97054  2003-04-13 C         FALSE      
#>  8 452212  2005-04-22 C         TRUE       
#>  9 083080  2012-04-16 A         FALSE      
#> 10 914520  2003-07-31 B         FALSE      
#> # … 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 [173]
#>    ward       date       patient new_index new_logical
#>    <chr>      <date>     <chr>       <dbl> <lgl>      
#>  1 ICU        2016-05-28 784436          1 TRUE       
#>  2 Clinical   2008-10-01 533225          1 TRUE       
#>  3 Outpatient 2004-12-17 B95551          1 TRUE       
#>  4 ICU        2004-11-03 D65308          1 TRUE       
#>  5 Clinical   2010-12-29 BF4515          1 TRUE       
#>  6 ICU        2002-10-20 F35553          1 TRUE       
#>  7 Clinical   2003-04-13 D97054          1 TRUE       
#>  8 ICU        2005-04-22 452212          1 TRUE       
#>  9 Clinical   2012-04-16 083080          1 TRUE       
#> 10 Clinical   2003-07-31 914520          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          107             14            49            69
#> 2 ICU                55             13            34            44
#> 3 Outpatient         11              7            11            11
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 [186]
#>    patient mo           ward       flag_episode
#>    <chr>   <mo>         <chr>      <lgl>       
#>  1 784436  B_STPHY_HMNS ICU        TRUE        
#>  2 533225  B_ESCHR_COLI Clinical   TRUE        
#>  3 B95551  B_STPHY_CONS Outpatient TRUE        
#>  4 D65308  B_STPHY_EPDR ICU        TRUE        
#>  5 BF4515  B_ESCHR_COLI Clinical   TRUE        
#>  6 F35553  B_STPHY_AURS ICU        TRUE        
#>  7 D97054  B_STPHY_AURS Clinical   TRUE        
#>  8 452212  B_ENTRC      ICU        TRUE        
#>  9 083080  B_ESCHR_COLI Clinical   TRUE        
#> 10 914520  B_STPHY_AURS Clinical   TRUE        
#> # … with 190 more rows
# }