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$date, episode_days = 60) # indices
#> [1] 62 16 19 48 19 56 61 35 35 19 63 41 12 60 21 53 7 28 10 50 8 2 24 11 16
#> [26] 36 5 56 41 12 9 50 27 61 26 53 1 54 1 52 34 27 23 61 56 66 14 33 64 47
#> [51] 37 1 5 60 1 11 32 25 3 44 56 29 30 27 7 64 11 28 56 6 17 31 38 51 40
#> [76] 61 10 51 38 43 48 55 43 54 24 37 11 7 3 64 48 6 64 10 55 14 7 39 9 49
#> [101] 31 16 15 58 21 42 62 8 59 48 56 28 1 47 22 64 57 13 49 57 58 9 35 19 64
#> [126] 28 29 5 16 37 7 65 14 38 8 29 63 26 13 8 45 24 19 52 5 22 3 57 55 33
#> [151] 29 44 23 31 64 12 53 59 38 29 41 59 26 12 18 15 9 1 29 20 47 5 43 39 33
#> [176] 6 36 4 52 46 33 39 61 65 56 57 13 3 25 22 6 42 22 49 5 6 44 35 7 63
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE TRUE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [13] FALSE TRUE TRUE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE
#> [25] FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE
#> [37] FALSE TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [49] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [61] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE
#> [73] FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [85] TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [97] FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [121] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [133] FALSE FALSE FALSE FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE
#> [145] FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [157] FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE
#> [169] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE TRUE
#> [181] FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE
#> [193] FALSE TRUE FALSE FALSE FALSE 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: 4 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-05-16 D25302 65 F ICU B_STRPT_ANGN S NA NA S
#> 2 2002-05-22 F35553 50 M ICU B_STPHY_AURS R NA S R
#> 3 2002-06-06 24D393 20 F Clinical B_ESCHR_COLI R NA NA NA
#> 4 2002-06-19 402950 53 F Clinical B_STPHY_HMNS R NA S 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 927431 2017-01-12 C FALSE
#> 2 161740 2005-06-21 A TRUE
#> 3 968584 2006-01-11 C FALSE
#> 4 479516 2013-09-30 B FALSE
#> 5 F32657 2006-03-09 A FALSE
#> 6 193126 2015-12-18 B FALSE
#> 7 E84349 2016-11-25 A FALSE
#> 8 699321 2010-07-23 B FALSE
#> 9 727637 2010-09-09 C FALSE
#> 10 F32657 2006-03-09 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 2017-01-12 927431 1 TRUE
#> 2 Clinical 2005-06-21 161740 1 TRUE
#> 3 ICU 2006-01-11 968584 1 TRUE
#> 4 Clinical 2013-09-30 479516 1 TRUE
#> 5 ICU 2006-03-09 F32657 1 TRUE
#> 6 Clinical 2015-12-18 193126 1 TRUE
#> 7 ICU 2016-11-25 E84349 1 TRUE
#> 8 Clinical 2010-07-23 699321 1 TRUE
#> 9 Clinical 2010-09-09 727637 1 TRUE
#> 10 ICU 2006-03-09 F32657 1 FALSE
#> # … 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 116 14 55 75
#> 2 ICU 57 12 33 43
#> 3 Outpatient 15 7 12 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 [192]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 927431 B_STPHY_CONS Clinical TRUE
#> 2 161740 B_STPHY_CONS Clinical TRUE
#> 3 968584 F_CANDD_ALBC ICU TRUE
#> 4 479516 B_STPHY_HMNS Clinical TRUE
#> 5 F32657 B_CRYNB ICU TRUE
#> 6 193126 B_STPHY_EPDR Clinical TRUE
#> 7 E84349 B_ESCHR_COLI ICU TRUE
#> 8 699321 B_STPHY_CONS Clinical TRUE
#> 9 727637 B_ESCHR_COLI Clinical TRUE
#> 10 F32657 B_STPHY_EPDR ICU TRUE
#> # … with 190 more rows
# }