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
Date
orPOSIXt
), 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$date, episode_days = 60) # indices
#> [1] 10 21 49 7 48 16 36 11 37 21 63 39 16 44 22 61 14 54 44 23 12 40 8 1 49
#> [26] 42 8 52 64 26 13 52 2 14 1 22 58 63 33 53 14 31 31 3 4 25 22 29 31 64
#> [51] 66 17 17 29 46 41 14 6 1 10 60 18 38 20 6 33 37 9 64 14 47 59 64 39 43
#> [76] 29 50 50 6 59 21 40 4 15 62 46 59 54 12 8 55 16 15 31 59 58 33 66 36 30
#> [101] 27 14 42 21 49 8 3 49 42 25 19 12 63 45 51 13 45 32 56 52 24 5 35 20 43
#> [126] 65 34 14 11 66 47 36 33 62 38 57 1 47 12 66 10 25 11 33 57 66 54 17 23 12
#> [151] 14 4 43 48 55 44 58 65 63 4 19 11 36 38 13 66 36 18 19 8 24 39 64 14 49
#> [176] 62 19 41 8 51 48 19 35 4 28 3 61 62 44 11 52 5 5 34 40 35 29 35 10 20
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
#> [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [25] FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE TRUE
#> [37] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [49] TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE
#> [61] TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [73] TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE
#> [85] TRUE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [97] TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE
#> [109] TRUE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE FALSE
#> [121] TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [133] FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [145] TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [157] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [169] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [181] FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [193] TRUE FALSE FALSE FALSE FALSE TRUE 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: 3 × 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 2002-07-30 218912 76 F ICU B_ESCHR_COLI R NA NA NA
#> 3 2002-08-31 149442 80 F ICU B_STPHY_AURS R NA S 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 419655 2004-02-21 C FALSE
#> 2 F3F09F 2006-06-03 A FALSE
#> 3 545388 2013-12-05 B FALSE
#> 4 E45CF0 2003-07-06 B FALSE
#> 5 992282 2013-10-01 A TRUE
#> 6 A92398 2005-05-14 B FALSE
#> 7 E58549 2010-03-06 C TRUE
#> 8 529296 2004-06-09 A FALSE
#> 9 F54287 2010-07-05 C FALSE
#> 10 578848 2006-05-08 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 [177]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 ICU 2004-02-21 419655 1 TRUE
#> 2 Clinical 2006-06-03 F3F09F 1 TRUE
#> 3 Clinical 2013-12-05 545388 1 TRUE
#> 4 Clinical 2003-07-06 E45CF0 1 TRUE
#> 5 Clinical 2013-10-01 992282 1 TRUE
#> 6 Clinical 2005-05-14 A92398 1 TRUE
#> 7 Clinical 2010-03-06 E58549 1 TRUE
#> 8 ICU 2004-06-09 529296 1 TRUE
#> 9 Clinical 2010-07-05 F54287 1 TRUE
#> 10 Clinical 2006-05-08 578848 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 113 15 54 71
#> 2 ICU 48 13 30 38
#> 3 Outpatient 16 7 13 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 [188]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 419655 B_STRPT_PNMN ICU TRUE
#> 2 F3F09F B_STRPT_ORLS Clinical TRUE
#> 3 545388 B_STPHY_HMNS Clinical TRUE
#> 4 E45CF0 B_KLBSL_PNMN Clinical TRUE
#> 5 992282 B_STRPT_GRPA Clinical TRUE
#> 6 A92398 B_ESCHR_COLI Clinical TRUE
#> 7 E58549 B_ESCHR_COLI Clinical TRUE
#> 8 529296 B_ENTRC_FACM ICU TRUE
#> 9 F54287 B_STRPT_ANGN Clinical TRUE
#> 10 578848 B_ESCHR_COLI Clinical TRUE
#> # … with 190 more rows
# }