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$date, episode_days = 60) # indices
#> [1] 14 30 51 63 66 3 25 25 10 50 17 32 20 64 19 41 45 62 25 18 24 26 28 6 63
#> [26] 47 32 10 16 28 23 2 4 66 62 8 41 13 15 56 9 22 34 51 15 65 45 1 53 57
#> [51] 29 39 12 52 12 33 58 57 48 51 40 27 30 18 36 43 13 7 65 60 13 25 16 21 62
#> [76] 66 39 22 39 7 8 24 60 19 27 27 58 5 49 6 62 23 15 39 10 66 63 7 50 5
#> [101] 60 29 67 51 52 57 34 48 60 18 65 55 48 51 66 12 20 49 41 65 41 67 53 39 11
#> [126] 55 65 66 68 13 27 43 34 46 51 3 13 11 23 31 65 34 11 37 10 54 16 66 24 50
#> [151] 38 10 5 8 59 30 49 63 59 9 36 44 38 28 30 65 5 6 66 10 30 5 26 1 42
#> [176] 18 63 66 14 60 64 35 55 47 13 43 21 66 61 8 61 54 41 64 12 65 16 40 10 52
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [13] FALSE TRUE TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [25] FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE
#> [37] FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE
#> [49] TRUE FALSE TRUE TRUE FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE
#> [61] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [73] FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [97] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [109] FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [121] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE
#> [133] FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE
#> [145] TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [157] TRUE TRUE TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE
#> [169] FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE TRUE FALSE FALSE
#> [181] FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [193] FALSE FALSE FALSE FALSE TRUE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-10-20 F35553 51 M ICU B_STPHY_AURS S NA S NA
#> 2 2002-11-18 956065 89 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 219024 2005-02-17 C FALSE
#> 2 A31059 2008-12-04 C FALSE
#> 3 D53062 2014-07-28 A FALSE
#> 4 305134 2016-10-28 C FALSE
#> 5 D80438 2017-07-03 C FALSE
#> 6 F35553 2002-10-20 C FALSE
#> 7 ED4982 2007-07-13 B FALSE
#> 8 122506 2007-08-11 A FALSE
#> 9 05B00F 2004-05-11 A TRUE
#> 10 8F77B2 2014-01-14 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 [180]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2005-02-17 219024 1 TRUE
#> 2 Clinical 2008-12-04 A31059 1 TRUE
#> 3 Clinical 2014-07-28 D53062 1 TRUE
#> 4 ICU 2016-10-28 305134 1 TRUE
#> 5 Clinical 2017-07-03 D80438 1 TRUE
#> 6 ICU 2002-10-20 F35553 2 TRUE
#> 7 ICU 2007-07-13 ED4982 1 TRUE
#> 8 Clinical 2007-08-11 122506 1 TRUE
#> 9 ICU 2004-05-11 05B00F 1 TRUE
#> 10 Clinical 2014-01-14 8F77B2 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 112 14 55 73
#> 2 ICU 56 12 36 46
#> 3 Outpatient 12 8 11 12
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 [188]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 219024 B_ESCHR_COLI Clinical TRUE
#> 2 A31059 B_STRPT_MTNS Clinical TRUE
#> 3 D53062 B_HMPHL_INFL Clinical TRUE
#> 4 305134 B_PROTS_VLGR ICU TRUE
#> 5 D80438 B_STPHY_AURS Clinical TRUE
#> 6 F35553 B_STPHY_AURS ICU TRUE
#> 7 ED4982 B_ESCHR_COLI ICU TRUE
#> 8 122506 B_STPHY_AURS Clinical TRUE
#> 9 05B00F F_CANDD_ALBC ICU TRUE
#> 10 8F77B2 B_ENTRC_FACM Clinical TRUE
#> # … with 190 more rows
# }