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$date, episode_days = 60) # indices
#> [1] 11 63 45 25 27 5 60 43 43 48 65 20 11 34 1 35 24 32 57 55 12 31 41 31 2
#> [26] 5 3 19 45 49 66 55 20 49 1 4 3 16 37 35 25 61 63 1 37 36 20 22 3 32
#> [51] 64 11 12 40 61 16 15 21 26 57 2 44 1 49 35 21 29 7 17 15 13 48 10 20 48
#> [76] 29 45 14 40 59 65 26 52 49 7 33 29 14 9 17 46 51 18 65 3 38 6 12 11 61
#> [101] 52 5 64 57 22 26 58 7 21 30 16 8 54 33 19 11 41 10 59 34 64 38 64 57 20
#> [126] 51 14 18 54 9 6 23 53 44 21 36 63 48 62 19 7 36 7 11 49 58 34 34 37 47
#> [151] 31 45 22 24 4 58 54 22 22 56 14 14 62 32 23 63 1 8 61 41 28 11 30 26 43
#> [176] 39 14 42 49 25 49 52 43 18 11 38 55 11 56 59 54 15 9 48 60 50 4 7 18 57
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [13] TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE
#> [25] FALSE TRUE TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
#> [37] FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [49] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [61] TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [73] FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE
#> [85] FALSE TRUE FALSE FALSE TRUE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
#> [97] TRUE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE
#> [109] FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [121] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [133] TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE
#> [145] FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE
#> [157] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE
#> [169] TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [193] FALSE FALSE TRUE TRUE 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: 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 E52286 47 M Clinical B_STPHY_AURS R NA S R
#> 2 2002-05-22 F35553 50 M ICU B_STPHY_EPDR R NA R R
#> 3 2002-05-16 D25302 65 F ICU B_STRPT_ANGN S NA NA S
#> 4 2002-06-07 710157 76 M Outpatie… B_STPHY_CONS S 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 399817 2004-04-15 C FALSE
#> 2 C95287 2017-03-14 B FALSE
#> 3 A95779 2012-05-18 C TRUE
#> 4 636118 2007-06-17 A FALSE
#> 5 921720 2007-11-02 C FALSE
#> 6 600057 2002-11-28 A FALSE
#> 7 D10538 2016-08-13 A FALSE
#> 8 146120 2011-09-23 C FALSE
#> 9 685406 2011-09-27 B TRUE
#> 10 C01360 2013-07-27 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 [185]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2004-04-15 399817 1 TRUE
#> 2 ICU 2017-03-14 C95287 1 TRUE
#> 3 Clinical 2012-05-18 A95779 1 TRUE
#> 4 Clinical 2007-06-17 636118 1 TRUE
#> 5 Clinical 2007-11-02 921720 1 TRUE
#> 6 Outpatient 2002-11-28 600057 1 TRUE
#> 7 ICU 2016-08-13 D10538 1 TRUE
#> 8 Clinical 2011-09-23 146120 1 TRUE
#> 9 Clinical 2011-09-27 685406 1 TRUE
#> 10 Clinical 2013-07-27 C01360 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 104 14 53 67
#> 2 ICU 65 12 40 46
#> 3 Outpatient 16 9 16 16
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 [193]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 399817 B_CTRBC_FRND Clinical TRUE
#> 2 C95287 B_STPHY_HMNS ICU TRUE
#> 3 A95779 B_STPHY_AURS Clinical TRUE
#> 4 636118 B_MRGNL_MRGN Clinical TRUE
#> 5 921720 B_STPHY_CONS Clinical TRUE
#> 6 600057 B_STPHY_AURS Outpatient TRUE
#> 7 D10538 B_ESCHR_COLI ICU TRUE
#> 8 146120 B_STPHY_AURS Clinical TRUE
#> 9 685406 B_STRPT_PNMN Clinical TRUE
#> 10 C01360 B_STPHY_CONS Clinical TRUE
#> # … with 190 more rows
# }