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$date, episode_days = 60) # indices
#> [1] 44 49 56 3 47 61 4 6 14 60 50 34 8 28 25 60 61 8 37 29 11 9 22 27 7
#> [26] 35 56 52 12 49 11 43 6 34 39 8 56 23 39 10 45 28 5 48 2 25 45 57 14 33
#> [51] 32 46 8 59 27 56 39 35 9 6 1 49 40 26 44 12 18 34 54 7 8 2 27 7 36
#> [76] 40 52 2 10 52 63 12 35 2 41 26 40 60 31 6 53 23 58 30 46 19 7 22 54 22
#> [101] 33 63 27 45 12 60 59 61 36 24 17 8 15 17 25 22 56 6 23 36 62 55 4 22 43
#> [126] 52 38 13 20 26 59 8 9 37 27 8 35 14 60 37 43 1 52 22 17 61 50 20 47 49
#> [151] 50 60 61 5 61 16 46 24 55 17 56 42 60 33 4 47 34 18 6 51 29 60 47 53 12
#> [176] 16 56 56 1 4 34 27 23 50 10 30 58 3 10 13 49 60 56 57 21 22 19 15 25 6
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [13] FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE
#> [25] FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [37] FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [49] FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [61] FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE
#> [73] FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE TRUE
#> [85] TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE
#> [97] TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [121] TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE
#> [133] FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE
#> [145] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [157] TRUE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [169] TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE
#> [181] FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [193] FALSE TRUE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-08-19 A49852 70 M Clinical B_ESCHR_COLI R NA NA NA
#> 2 2002-07-15 C42193 84 M ICU B_STPHY_HMNS R NA R R
#> # … 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 3B8892 2012-06-11 A FALSE
#> 2 835073 2013-09-27 B TRUE
#> 3 375975 2015-11-19 C FALSE
#> 4 A49852 2002-08-19 C FALSE
#> 5 660761 2013-03-05 A FALSE
#> 6 BF4515 2017-06-12 C FALSE
#> 7 956065 2002-11-18 C FALSE
#> 8 F35553 2003-09-05 A FALSE
#> 9 904485 2005-04-18 B FALSE
#> 10 5F7C80 2017-03-20 B 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 [181]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2012-06-11 3B8892 1 TRUE
#> 2 Clinical 2013-09-27 835073 1 TRUE
#> 3 Clinical 2015-11-19 375975 1 TRUE
#> 4 Clinical 2002-08-19 A49852 1 TRUE
#> 5 Clinical 2013-03-05 660761 1 TRUE
#> 6 ICU 2017-06-12 BF4515 1 TRUE
#> 7 Clinical 2002-11-18 956065 1 TRUE
#> 8 ICU 2003-09-05 F35553 2 FALSE
#> 9 ICU 2005-04-18 904485 1 TRUE
#> 10 Clinical 2017-03-20 5F7C80 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 117 15 58 72
#> 2 ICU 51 12 34 40
#> 3 Outpatient 13 7 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 [190]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 3B8892 B_ESCHR_COLI Clinical TRUE
#> 2 835073 B_STPHY_HMNS Clinical TRUE
#> 3 375975 B_STPHY_AURS Clinical TRUE
#> 4 A49852 B_ESCHR_COLI Clinical TRUE
#> 5 660761 B_STRPT_DYSG Clinical TRUE
#> 6 BF4515 B_ENTRC_FACM ICU TRUE
#> 7 956065 B_ESCHR_COLI Clinical TRUE
#> 8 F35553 B_ENTRBC_CLOC ICU FALSE
#> 9 904485 B_STRPT_ANGN ICU TRUE
#> 10 5F7C80 B_STPHY_AURS Clinical TRUE
#> # … with 190 more rows
# }