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$date, episode_days = 60) # indices
#> [1] 19 20 54 49 35 60 26 3 18 4 43 56 55 48 60 58 47 55 59 3 6 32 27 25 56
#> [26] 10 24 9 5 60 57 62 15 1 52 7 23 5 2 46 9 14 46 56 18 38 34 20 44 41
#> [51] 21 45 1 16 28 52 2 12 31 48 7 30 3 62 54 44 36 47 14 54 4 29 13 54 11
#> [76] 53 8 3 54 14 48 37 57 16 21 36 6 23 61 45 9 46 38 26 42 27 1 44 9 21
#> [101] 10 2 31 54 7 20 61 40 57 35 62 4 1 46 5 8 49 14 37 38 44 52 18 39 54
#> [126] 22 25 20 59 57 6 54 44 3 7 50 26 17 38 26 33 33 52 53 7 54 54 48 59 17
#> [151] 45 30 43 42 19 41 5 17 15 32 7 20 62 45 14 6 13 10 1 59 28 18 14 34 60
#> [176] 54 6 13 23 55 11 19 34 26 51 15 11 22 54 10 22 28 8 26 38 39 28 37 63 56
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [13] TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [25] TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [37] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [49] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE
#> [61] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [73] TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE
#> [85] TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [97] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [109] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE
#> [121] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [133] FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE
#> [145] FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE TRUE
#> [157] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [169] TRUE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE
#> [181] FALSE TRUE TRUE FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
#> [193] FALSE FALSE TRUE FALSE TRUE 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: 5 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-11-18 956065 89 F Clinical B_ESCHR_COLI R NA NA NA
#> 2 2002-11-28 705451 56 M Clinical B_STPHY_CONS R NA S NA
#> 3 2002-11-27 496896 47 F ICU B_STPHY_CONS R NA R R
#> 4 2002-10-20 F35553 51 M ICU B_STPHY_AURS S NA S NA
#> 5 2002-10-11 871360 78 M Clinical B_STPHY_EPDR 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 E59875 2006-03-25 C FALSE
#> 2 C34072 2006-07-05 C TRUE
#> 3 0DBF93 2015-10-12 A FALSE
#> 4 2F9056 2014-05-06 A TRUE
#> 5 953526 2010-04-23 B FALSE
#> 6 976997 2017-03-02 C FALSE
#> 7 080086 2007-10-26 B FALSE
#> 8 956065 2002-11-18 B FALSE
#> 9 612575 2006-01-31 B FALSE
#> 10 285137 2002-12-13 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 [178]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 ICU 2006-03-25 E59875 1 TRUE
#> 2 Clinical 2006-07-05 C34072 1 TRUE
#> 3 Clinical 2015-10-12 0DBF93 1 TRUE
#> 4 ICU 2014-05-06 2F9056 1 TRUE
#> 5 ICU 2010-04-23 953526 1 TRUE
#> 6 Clinical 2017-03-02 976997 1 TRUE
#> 7 Clinical 2007-10-26 080086 1 TRUE
#> 8 Clinical 2002-11-18 956065 1 TRUE
#> 9 Clinical 2006-01-31 612575 1 TRUE
#> 10 ICU 2002-12-13 285137 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 115 13 49 69
#> 2 ICU 55 12 34 39
#> 3 Outpatient 8 7 8 8
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 [190]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 E59875 B_STPHY_EPDR ICU TRUE
#> 2 C34072 B_STPHY_CONS Clinical TRUE
#> 3 0DBF93 B_STPHY_AURS Clinical TRUE
#> 4 2F9056 B_HAFNI_ALVE ICU TRUE
#> 5 953526 B_STPHY_CONS ICU TRUE
#> 6 976997 B_STRPT_PYGN Clinical TRUE
#> 7 080086 B_STRPT_GRPB Clinical TRUE
#> 8 956065 B_ESCHR_COLI Clinical TRUE
#> 9 612575 B_ENTRBC_CLOC Clinical TRUE
#> 10 285137 B_ESCHR_COLI ICU TRUE
#> # … with 190 more rows
# }