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$date, episode_days = 60) # indices
#>   [1] 67  8 32 43 12 22 22 31  7 57 67 15 29 40 20  1 43 49  7 45 64  1 31 67  1
#>  [26] 65 20 55 41  2  9 20 31 60 61 36 24 41 34 41 62 34 20  5 11 63 49 41 61 46
#>  [51] 16 33 28 29 51 50 29 61 14  7 18 10 60 18 47 59 22 16 55 33 64 33 48  3 58
#>  [76] 57  6 31 40 21 36 24 61 14 57 26  2 24 27 12 12 22 17  9 18 66 34 14  3  5
#> [101] 53 67 38  1 18 65 33 46 46 51  4 17  5  1 55 40 44 20 51 38 50 38 18 47 62
#> [126] 12 14 18 51 25 31 35 30 40 14 61 54 47 16 67 36 14 39 23 55  7 56 39 13 21
#> [151] 12 44 25 35 50 27 51  2  1 58 49 20 60 52 66 63 17 31 19 47 15 52 64 31 17
#> [176] 25 30  8 54 36 39 21 51 23 64  3 66 19 38  3 27 26 44  6 48 11 42 61  5 37
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#>   [1] FALSE  TRUE  TRUE  TRUE FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE
#>  [13]  TRUE  TRUE  TRUE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE
#>  [25] FALSE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE  TRUE  TRUE FALSE FALSE
#>  [37] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE
#>  [49]  TRUE  TRUE FALSE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE FALSE FALSE
#>  [61] FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE
#>  [73] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#>  [85]  TRUE FALSE  TRUE  TRUE FALSE FALSE  TRUE  TRUE  TRUE  TRUE FALSE FALSE
#>  [97] FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE FALSE FALSE
#> [109] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE FALSE FALSE
#> [121] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE  TRUE
#> [133] FALSE FALSE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE  TRUE  TRUE  TRUE
#> [145] FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [157]  TRUE FALSE  TRUE  TRUE FALSE FALSE FALSE  TRUE  TRUE  TRUE FALSE FALSE
#> [169] FALSE FALSE FALSE FALSE  TRUE FALSE FALSE  TRUE  TRUE FALSE  TRUE  TRUE
#> [181] FALSE  TRUE FALSE FALSE FALSE  TRUE FALSE  TRUE FALSE FALSE  TRUE  TRUE
#> [193] FALSE  TRUE  TRUE FALSE  TRUE 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 2002-07-15 426426     67 F      ICU      B_ESCHR_COLI R     NA    NA    NA   
#> 3 2002-06-06 24D393     20 F      Clinical B_ESCHR_COLI R     NA    NA    NA   
#> 4 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 5DB1C8  2017-12-28 B         FALSE      
#>  2 4DD722  2003-06-02 A         FALSE      
#>  3 C7C641  2008-12-27 A         FALSE      
#>  4 704554  2012-01-01 B         FALSE      
#>  5 304508  2004-05-09 B         FALSE      
#>  6 E27874  2006-12-19 B         FALSE      
#>  7 418311  2006-11-10 B         FALSE      
#>  8 501361  2008-11-01 A         TRUE       
#>  9 6BC362  2003-04-02 A         TRUE       
#> 10 D40405  2015-06-03 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 [177]
#>    ward       date       patient new_index new_logical
#>    <chr>      <date>     <chr>       <dbl> <lgl>      
#>  1 Clinical   2017-12-28 5DB1C8          1 TRUE       
#>  2 ICU        2003-06-02 4DD722          1 TRUE       
#>  3 Outpatient 2008-12-27 C7C641          1 TRUE       
#>  4 ICU        2012-01-01 704554          1 TRUE       
#>  5 Clinical   2004-05-09 304508          1 TRUE       
#>  6 Clinical   2006-12-19 E27874          1 TRUE       
#>  7 ICU        2006-11-10 418311          1 TRUE       
#>  8 Clinical   2008-11-01 501361          1 TRUE       
#>  9 ICU        2003-04-02 6BC362          1 TRUE       
#> 10 ICU        2015-06-03 D40405          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          115             14            50            71
#> 2 ICU                56             13            37            43
#> 3 Outpatient          6              4             5             5
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 [191]
#>    patient mo           ward       flag_episode
#>    <chr>   <mo>         <chr>      <lgl>       
#>  1 5DB1C8  B_STPHY_CPTS Clinical   TRUE        
#>  2 4DD722  B_ESCHR_COLI ICU        TRUE        
#>  3 C7C641  B_STPHY_CONS Outpatient TRUE        
#>  4 704554  B_STPHY_CONS ICU        TRUE        
#>  5 304508  B_STRPT_PNMN Clinical   TRUE        
#>  6 E27874  B_STRPT_PNMN Clinical   TRUE        
#>  7 418311  B_STPHY_EPDR ICU        TRUE        
#>  8 501361  B_ESCHR_COLI Clinical   TRUE        
#>  9 6BC362  B_STRPT_PNMN ICU        TRUE        
#> 10 D40405  B_ENTRC_FCLS ICU        TRUE        
#> # … with 190 more rows
# }