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$date, episode_days = 60) # indices
#> [1] 14 23 55 63 62 57 7 50 31 59 21 49 23 61 2 31 5 12 54 27 9 47 54 21 11
#> [26] 58 39 1 2 62 47 64 33 22 38 11 46 57 13 17 38 44 46 46 14 11 33 58 41 43
#> [51] 1 63 17 20 1 39 21 32 22 21 8 56 60 61 27 29 35 53 59 2 39 53 59 29 8
#> [76] 11 6 46 11 30 15 31 62 59 62 56 7 15 41 9 28 11 30 15 26 36 31 17 34 11
#> [101] 43 44 18 43 60 27 26 37 7 45 58 8 17 16 24 8 34 19 12 30 58 61 54 51 47
#> [126] 49 9 63 45 51 6 12 10 45 23 27 48 40 12 42 45 52 33 42 46 51 5 48 51 26
#> [151] 37 22 4 25 6 6 30 38 29 9 52 3 27 2 55 10 46 30 8 48 18 11 4 23 18
#> [176] 61 32 36 26 4 42 40 61 8 53 12 2 10 38 3 16 63 62 59 35 62 61 4 29 20
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE FALSE
#> [13] FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE
#> [25] FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE
#> [37] FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [49] FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE
#> [61] FALSE TRUE TRUE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [73] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [85] TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE
#> [97] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [109] TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE
#> [121] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [133] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [145] TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [157] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [169] TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [181] TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [193] FALSE FALSE FALSE FALSE FALSE FALSE TRUE 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 2002-07-23 F35553 51 M ICU B_STPHY_AURS R NA S R
#> 2 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 105248 2005-06-16 B FALSE
#> 2 3C8163 2007-06-26 B FALSE
#> 3 062027 2015-09-21 B FALSE
#> 4 8DD375 2017-09-13 A FALSE
#> 5 944337 2017-07-08 A FALSE
#> 6 D36589 2016-02-14 B TRUE
#> 7 C27336 2003-09-22 A FALSE
#> 8 658640 2014-05-30 A FALSE
#> 9 EC9741 2009-06-18 B FALSE
#> 10 644292 2016-10-26 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 [180]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2005-06-16 105248 1 TRUE
#> 2 Clinical 2007-06-26 3C8163 1 TRUE
#> 3 ICU 2015-09-21 062027 1 TRUE
#> 4 ICU 2017-09-13 8DD375 1 TRUE
#> 5 Clinical 2017-07-08 944337 1 TRUE
#> 6 ICU 2016-02-14 D36589 1 TRUE
#> 7 ICU 2003-09-22 C27336 1 TRUE
#> 8 Clinical 2014-05-30 658640 1 TRUE
#> 9 Outpatient 2009-06-18 EC9741 1 TRUE
#> 10 ICU 2016-10-26 644292 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 113 14 52 69
#> 2 ICU 52 11 29 42
#> 3 Outpatient 15 7 14 14
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] TRUE
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 105248 B_ESCHR_COLI Clinical TRUE
#> 2 3C8163 B_PSDMN_AERG Clinical TRUE
#> 3 062027 B_STPHY_CPTS ICU TRUE
#> 4 8DD375 B_ESCHR_COLI ICU TRUE
#> 5 944337 B_GLBCT_SNGN Clinical TRUE
#> 6 D36589 B_KLBSL_OXYT ICU TRUE
#> 7 C27336 B_BCTRD_FRGL ICU TRUE
#> 8 658640 B_STPHY_AURS Clinical TRUE
#> 9 EC9741 B_ESCHR_COLI Outpatient TRUE
#> 10 644292 B_STPHY_AURS ICU TRUE
#> # … with 190 more rows
# }