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.
Arguments
- x
vector of dates (class
DateorPOSIXt), 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
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().
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$date, episode_days = 60) # indices
#> [1] 1 32 7 22 29 55 15 62 34 39 13 4 28 30 43 26 34 55 46 3 16 62 49 38 31
#> [26] 13 1 39 52 20 48 55 10 43 60 31 14 9 29 32 8 18 52 57 48 54 62 43 49 6
#> [51] 13 57 21 43 61 52 17 4 56 10 31 18 59 3 41 20 37 6 32 5 48 43 49 12 58
#> [76] 63 63 49 38 44 35 10 60 9 55 63 48 58 11 43 7 61 45 32 60 62 2 58 53 24
#> [101] 40 35 16 29 29 46 13 25 25 15 45 58 30 59 44 64 57 45 14 34 3 18 4 28 1
#> [126] 2 29 61 27 46 23 40 50 61 60 41 15 3 45 44 18 55 60 37 37 28 15 47 42 41
#> [151] 8 27 26 15 7 50 33 2 63 64 23 12 41 62 8 16 62 49 9 12 22 32 10 9 3
#> [176] 51 30 36 11 57 19 37 9 34 7 2 27 17 54 19 55 35 6 53 28 17 56 4 32 45
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [13] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [25] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [37] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [49] FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE
#> [61] FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE
#> [73] TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE
#> [85] FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [97] FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE
#> [109] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE
#> [121] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [133] TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [145] FALSE TRUE FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [157] TRUE FALSE TRUE TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [169] FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE FALSE
#> [181] TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE
#> [193] FALSE FALSE FALSE TRUE FALSE TRUE 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: 5 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-06-07 710157 76 M Outpatie… B_STPHY_CONS S NA S NA
#> 2 2002-06-22 FD8039 75 F ICU B_ESCHR_COLI R NA NA NA
#> 3 2002-07-15 C42193 84 M ICU B_STPHY_HMNS R NA R R
#> 4 2002-07-15 426426 67 F ICU B_ESCHR_COLI R NA NA NA
#> 5 2002-06-04 082413 78 M ICU B_STRPT_PNMN S NA NA S
#> # … 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 614772 2002-02-27 A TRUE
#> 2 650870 2009-11-12 C FALSE
#> 3 82C90B 2003-09-19 A FALSE
#> 4 B52127 2007-06-03 A FALSE
#> 5 B13757 2009-02-12 C TRUE
#> 6 097186 2015-10-28 C FALSE
#> 7 C82046 2005-08-08 A TRUE
#> 8 D80438 2017-07-03 C FALSE
#> 9 09B453 2010-03-21 C FALSE
#> 10 5B78D5 2011-09-19 A 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 [188]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2002-02-27 614772 1 TRUE
#> 2 Outpatient 2009-11-12 650870 1 TRUE
#> 3 Clinical 2003-09-19 82C90B 1 TRUE
#> 4 ICU 2007-06-03 B52127 1 TRUE
#> 5 Outpatient 2009-02-12 B13757 1 TRUE
#> 6 Clinical 2015-10-28 097186 1 TRUE
#> 7 ICU 2005-08-08 C82046 1 TRUE
#> 8 Clinical 2017-07-03 D80438 1 TRUE
#> 9 Clinical 2010-03-21 09B453 1 TRUE
#> 10 Clinical 2011-09-19 5B78D5 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 122 15 53 75
#> 2 ICU 49 12 33 39
#> 3 Outpatient 17 8 13 15
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 [193]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 614772 B_STPHY_HMNS Clinical TRUE
#> 2 650870 B_ESCHR_COLI Outpatient TRUE
#> 3 82C90B B_STPHY_EPDR Clinical TRUE
#> 4 B52127 B_ESCHR_COLI ICU TRUE
#> 5 B13757 B_STPHY_EPDR Outpatient TRUE
#> 6 097186 B_STPHY_CPTS Clinical TRUE
#> 7 C82046 B_ESCHR_COLI ICU TRUE
#> 8 D80438 B_CRYNB_STRT Clinical TRUE
#> 9 09B453 B_STPHY_AURS Clinical TRUE
#> 10 5B78D5 B_STPHY_AURS Clinical TRUE
#> # … with 190 more rows
# }