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$date, episode_days = 60) # indices
#> [1] 21 58 44 52 10 38 12 52 61 27 13 62 24 43 11 21 56 19 15 21 44 56 62 5 14
#> [26] 36 10 54 22 36 46 12 24 13 59 34 48 29 55 8 26 5 14 1 25 37 7 60 15 57
#> [51] 36 44 62 5 40 41 6 40 6 43 40 58 56 42 10 21 47 8 33 35 15 28 22 47 33
#> [76] 51 37 33 42 41 50 8 43 31 7 7 23 60 7 51 19 20 33 61 22 46 30 18 57 21
#> [101] 49 23 51 53 13 12 17 13 40 9 35 58 31 12 30 17 61 46 45 39 28 58 18 57 27
#> [126] 30 13 1 16 7 26 28 61 57 18 16 52 31 2 3 46 59 50 6 6 58 3 4 28 21
#> [151] 48 11 16 49 54 11 27 25 27 45 3 38 31 47 3 17 28 21 55 11 29 13 20 54 32
#> [176] 53 60 25 56 10 8 26 3 50 1 7 21 13 55 4 35 45 29 32 42 62 44 2 50 41
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE
#> [13] FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE
#> [25] FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE TRUE TRUE
#> [37] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [49] FALSE TRUE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [61] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE
#> [73] FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE TRUE
#> [85] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [97] TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE TRUE
#> [109] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [121] TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [133] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [145] FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE
#> [157] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [169] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [181] FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [193] FALSE TRUE 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: 5 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-07-28 F54261 69 M Clinical B_STPHY_CONS R NA S NA
#> 2 2002-08-19 A49852 70 M Clinical B_ESCHR_COLI R NA NA NA
#> 3 2002-08-31 149442 80 F ICU B_STPHY_AURS R NA S R
#> 4 2002-07-23 F35553 51 M ICU B_STPHY_AURS R NA S R
#> 5 2002-07-24 F35553 51 M 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 A68B33 2007-04-14 C FALSE
#> 2 422833 2017-03-21 C FALSE
#> 3 545388 2013-07-29 B FALSE
#> 4 A76045 2015-10-07 C FALSE
#> 5 F35553 2004-09-22 C TRUE
#> 6 50C8DB 2011-09-01 C FALSE
#> 7 848254 2005-03-29 B FALSE
#> 8 A84726 2015-08-14 A TRUE
#> 9 D20588 2017-08-17 C FALSE
#> 10 C7C641 2008-12-27 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 [176]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 ICU 2007-04-14 A68B33 1 TRUE
#> 2 ICU 2017-03-21 422833 1 TRUE
#> 3 Clinical 2013-07-29 545388 1 TRUE
#> 4 ICU 2015-10-07 A76045 1 TRUE
#> 5 ICU 2004-09-22 F35553 4 TRUE
#> 6 Clinical 2011-09-01 50C8DB 1 TRUE
#> 7 ICU 2005-03-29 848254 1 TRUE
#> 8 Clinical 2015-08-14 A84726 1 TRUE
#> 9 ICU 2017-08-17 D20588 1 TRUE
#> 10 Outpatient 2008-12-27 C7C641 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 101 14 51 67
#> 2 ICU 62 13 36 43
#> 3 Outpatient 13 7 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 [191]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 A68B33 B_STPHY_AURS ICU TRUE
#> 2 422833 B_STPHY_EPDR ICU TRUE
#> 3 545388 B_ENTRC Clinical TRUE
#> 4 A76045 B_ENTRC_FACM ICU TRUE
#> 5 F35553 B_STPHY_AURS ICU FALSE
#> 6 50C8DB B_STPHY_EPDR Clinical TRUE
#> 7 848254 B_STPHY_EPDR ICU TRUE
#> 8 A84726 B_STPHY_EPDR Clinical TRUE
#> 9 D20588 B_STPHY_HMNS ICU TRUE
#> 10 C7C641 B_STPHY_CONS Outpatient TRUE
#> # … with 190 more rows
# }