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$date, episode_days = 60) # indices
#> [1] 50 55 22 2 7 13 47 45 36 28 21 45 31 46 9 22 42 6 25 28 48 35 10 11 47
#> [26] 23 58 59 44 57 34 40 26 33 45 51 39 7 20 38 35 6 18 32 1 1 46 10 9 48
#> [51] 37 41 5 36 8 58 40 50 58 59 57 9 35 59 51 16 48 57 33 2 39 35 58 20 7
#> [76] 20 4 5 18 2 26 25 8 20 7 33 40 19 28 44 49 28 38 15 27 57 22 43 27 32
#> [101] 25 36 32 26 38 27 16 3 32 56 10 51 8 9 40 30 6 29 51 2 53 8 53 33 9
#> [126] 28 18 6 19 50 10 51 15 53 28 38 20 38 42 12 49 33 24 58 36 26 52 5 24 10
#> [151] 55 37 11 47 22 44 12 57 6 16 21 54 45 36 15 49 12 4 53 38 6 60 13 47 38
#> [176] 36 30 14 18 17 10 23 10 56 55 18 1 19 19 24 14 34 45 50 54 2 27 12 15 45
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [13] TRUE TRUE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE
#> [37] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [49] FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [61] FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE
#> [73] FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE
#> [97] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE
#> [109] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE
#> [121] FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [133] FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE
#> [145] TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE
#> [157] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE
#> [169] FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [181] TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE TRUE FALSE
#> [193] FALSE FALSE TRUE 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: 1 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 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 324415 2015-10-03 B FALSE
#> 2 671620 2016-12-04 A TRUE
#> 3 403631 2007-07-31 A FALSE
#> 4 F30196 2002-04-14 A FALSE
#> 5 A73011 2003-08-13 C FALSE
#> 6 F63168 2005-02-25 A FALSE
#> 7 C34429 2014-12-15 A FALSE
#> 8 658640 2014-05-30 B FALSE
#> 9 687590 2011-10-29 A FALSE
#> 10 408807 2009-05-07 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 [179]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 ICU 2015-10-03 324415 1 TRUE
#> 2 Clinical 2016-12-04 671620 1 TRUE
#> 3 Clinical 2007-07-31 403631 1 TRUE
#> 4 Outpatient 2002-04-14 F30196 1 TRUE
#> 5 Clinical 2003-08-13 A73011 1 TRUE
#> 6 Clinical 2005-02-25 F63168 1 TRUE
#> 7 Clinical 2014-12-15 C34429 1 TRUE
#> 8 Clinical 2014-05-30 658640 1 TRUE
#> 9 Clinical 2011-10-29 687590 1 TRUE
#> 10 Clinical 2009-05-07 408807 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 118 14 52 75
#> 2 ICU 51 12 34 45
#> 3 Outpatient 10 6 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 [188]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 324415 B_PROTS_MRBL ICU TRUE
#> 2 671620 B_ESCHR_COLI Clinical TRUE
#> 3 403631 B_ESCHR_COLI Clinical TRUE
#> 4 F30196 B_STRPT_GRPB Outpatient TRUE
#> 5 A73011 B_STPHY_CONS Clinical TRUE
#> 6 F63168 B_STRPT_EQUI Clinical TRUE
#> 7 C34429 B_PROTS_MRBL Clinical TRUE
#> 8 658640 B_STPHY_AURS Clinical TRUE
#> 9 687590 B_STPHY_CONS Clinical TRUE
#> 10 408807 B_STPHY_CONS Clinical TRUE
#> # … with 190 more rows
# }