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$date, episode_days = 60) # indices
#> [1] 35 26 56 13 15 11 32 52 50 56 20 21 2 25 53 12 29 1 40 26 28 48 53 23 34
#> [26] 6 34 22 27 57 28 42 41 6 55 12 48 5 30 49 54 45 51 4 54 55 18 12 58 32
#> [51] 12 25 38 1 41 5 29 16 42 17 7 44 24 46 42 57 59 52 32 42 58 45 61 57 53
#> [76] 19 48 19 58 53 56 34 7 55 40 60 30 8 16 46 59 35 45 23 48 22 9 35 48 15
#> [101] 31 47 59 8 55 36 34 61 61 1 57 57 25 31 1 19 53 51 3 37 32 35 11 38 60
#> [126] 43 4 12 6 25 2 14 29 58 40 63 27 27 19 25 59 45 51 58 13 48 2 14 28 36
#> [151] 37 5 1 16 39 61 35 21 49 54 15 49 10 10 11 48 32 12 23 50 44 20 15 56 33
#> [176] 56 1 21 57 18 21 34 58 24 54 51 20 18 55 55 9 62 25 19 2 9 13 33 63 15
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE TRUE FALSE
#> [13] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [37] FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE TRUE FALSE FALSE
#> [49] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [61] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE
#> [73] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [85] FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE FALSE TRUE FALSE TRUE
#> [97] TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [109] FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE
#> [121] FALSE FALSE TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE FALSE TRUE
#> [133] FALSE FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [145] FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE
#> [157] FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [169] FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE
#> [181] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [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 2002-11-04 304347 62 M Clinical 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 1D4C00 2011-04-04 C FALSE
#> 2 B13757 2009-02-12 C FALSE
#> 3 960787 2016-03-27 A FALSE
#> 4 904485 2005-04-18 A FALSE
#> 5 6D6319 2005-08-22 B TRUE
#> 6 C82247 2004-09-10 A FALSE
#> 7 317826 2010-05-14 B FALSE
#> 8 384779 2015-04-07 B FALSE
#> 9 0F9638 2014-09-22 A FALSE
#> 10 329C35 2016-04-03 C 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 [182]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2011-04-04 1D4C00 1 TRUE
#> 2 Outpatient 2009-02-12 B13757 1 TRUE
#> 3 Clinical 2016-03-27 960787 1 TRUE
#> 4 ICU 2005-04-18 904485 1 TRUE
#> 5 Outpatient 2005-08-22 6D6319 1 TRUE
#> 6 ICU 2004-09-10 C82247 1 TRUE
#> 7 ICU 2010-05-14 317826 1 TRUE
#> 8 Clinical 2015-04-07 384779 1 TRUE
#> 9 Clinical 2014-09-22 0F9638 1 TRUE
#> 10 Clinical 2016-04-03 329C35 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 111 14 49 76
#> 2 ICU 59 13 34 45
#> 3 Outpatient 12 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 [192]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 1D4C00 B_KLBSL_AERG Clinical TRUE
#> 2 B13757 B_STPHY_CONS Outpatient TRUE
#> 3 960787 B_KLBSL_PNMN Clinical TRUE
#> 4 904485 B_STRPT_ANGN ICU TRUE
#> 5 6D6319 B_STPHY_EPDR Outpatient TRUE
#> 6 C82247 B_STPHY_CONS ICU TRUE
#> 7 317826 B_ESCHR_COLI ICU TRUE
#> 8 384779 B_STRPT_PNMN Clinical TRUE
#> 9 0F9638 B_ESCHR_COLI Clinical TRUE
#> 10 329C35 B_ESCHR_COLI Clinical TRUE
#> # … with 190 more rows
# }