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$date, episode_days = 60) # indices
#> [1] 33 41 43 1 11 64 38 26 48 21 47 13 9 28 7 19 13 28 25 8 12 62 38 41 22
#> [26] 12 15 51 39 38 49 19 28 43 42 61 57 1 14 3 61 6 15 44 6 48 11 59 18 24
#> [51] 54 55 8 37 37 60 42 59 18 3 61 28 45 5 42 48 2 38 54 53 40 26 52 63 29
#> [76] 3 45 61 7 16 53 32 44 9 49 10 6 4 10 17 26 14 48 3 60 59 22 35 39 60
#> [101] 9 11 21 35 1 10 60 36 58 49 55 44 17 34 55 29 45 53 57 1 60 14 9 10 8
#> [126] 38 34 9 55 36 36 9 58 52 31 56 60 30 20 43 48 23 7 24 60 33 61 37 19 1
#> [151] 19 52 38 40 35 5 20 60 27 27 6 33 7 24 39 59 29 25 38 7 18 42 36 7 34
#> [176] 45 50 58 22 35 55 29 50 62 62 18 52 46 44 63 32 40 57 23 35 61 26 15 64 25
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [13] TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [25] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [37] TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [49] FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE
#> [61] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [73] FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE
#> [85] FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [97] TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE
#> [109] FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [121] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [133] TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE
#> [145] TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE
#> [157] TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [169] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [181] FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE TRUE TRUE FALSE
#> [193] FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE
# 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 2002-10-20 F35553 51 M ICU B_STPHY_AURS S NA S NA
#> 2 2002-11-11 D80753 74 F Outpatie… B_STPHY_CONS R NA S NA
#> 3 2002-11-16 762305 87 F Clinical B_BCTRD_FRGL R NA NA R
#> 4 2002-09-23 F35553 51 M ICU B_STPHY_AURS 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 725779 2010-09-04 B TRUE
#> 2 E4F322 2012-10-17 A TRUE
#> 3 672020 2013-04-01 B FALSE
#> 4 614772 2002-02-27 C TRUE
#> 5 B8F499 2004-08-22 B FALSE
#> 6 CFCF65 2017-12-04 C FALSE
#> 7 A81782 2011-08-18 B FALSE
#> 8 501361 2008-11-01 C FALSE
#> 9 280834 2014-09-03 A FALSE
#> 10 122506 2007-08-10 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 [183]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Outpatient 2010-09-04 725779 1 TRUE
#> 2 ICU 2012-10-17 E4F322 1 TRUE
#> 3 Clinical 2013-04-01 672020 1 TRUE
#> 4 Clinical 2002-02-27 614772 1 TRUE
#> 5 Clinical 2004-08-22 B8F499 1 TRUE
#> 6 ICU 2017-12-04 CFCF65 1 TRUE
#> 7 Clinical 2011-08-18 A81782 1 TRUE
#> 8 Clinical 2008-11-01 501361 1 TRUE
#> 9 Clinical 2014-09-03 280834 1 TRUE
#> 10 Clinical 2007-08-10 122506 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 112 13 50 68
#> 2 ICU 62 13 36 48
#> 3 Outpatient 9 5 7 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 725779 B_STRPT_PNMN Outpatient TRUE
#> 2 E4F322 B_STPHY_CONS ICU TRUE
#> 3 672020 B_STPHY_EPDR Clinical TRUE
#> 4 614772 B_STPHY_HMNS Clinical TRUE
#> 5 B8F499 B_STPHY_CONS Clinical TRUE
#> 6 CFCF65 B_ACNTB_BMNN ICU TRUE
#> 7 A81782 B_STPHY_CONS Clinical TRUE
#> 8 501361 B_ESCHR_COLI Clinical TRUE
#> 9 280834 B_STPHY_CONS Clinical TRUE
#> 10 122506 B_STPHY_AURS Clinical TRUE
#> # … with 190 more rows
# }