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$date, episode_days = 60) # indices
#> [1] 42 57 32 28 63 47 37 2 59 51 13 12 5 50 32 50 60 15 52 59 43 39 10 19 54
#> [26] 31 58 4 30 24 4 2 19 55 60 49 55 42 63 60 38 53 35 13 57 29 60 36 19 19
#> [51] 22 7 11 19 28 15 9 9 16 52 55 33 55 2 11 55 45 62 38 59 57 13 27 30 40
#> [76] 30 37 63 45 35 14 9 50 1 44 6 63 4 30 41 55 55 14 24 10 37 30 31 38 38
#> [101] 16 29 17 23 54 15 40 24 15 52 62 10 58 40 60 3 28 28 51 17 59 37 13 51 26
#> [126] 38 48 13 27 8 8 25 28 30 8 33 6 31 40 8 57 59 25 61 2 52 60 1 21 36
#> [151] 23 15 11 32 21 3 46 42 15 48 61 19 27 34 32 51 47 27 55 4 63 40 54 11 8
#> [176] 13 53 20 31 21 19 31 35 56 13 2 34 20 46 35 41 13 61 31 50 3 13 8 18 40
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [13] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
#> [25] TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [37] TRUE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE
#> [49] TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [61] FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [73] TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE
#> [85] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [97] FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [109] TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [121] FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [133] FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE
#> [145] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE
#> [157] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE
#> [169] FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [181] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE
#> [193] FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE
# 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 2002-07-16 241328 78 M Outpatie… B_STPHY_CONS R NA S R
#> 2 2002-07-15 C42193 84 M ICU B_STPHY_HMNS R NA R R
#> 3 2002-08-14 785317 51 F ICU B_ESCHR_COLI R NA NA 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 616685 2012-12-27 B FALSE
#> 2 A79917 2016-05-21 A FALSE
#> 3 3F562C 2009-05-25 B FALSE
#> 4 945BD5 2008-05-06 B FALSE
#> 5 5DB1C8 2017-12-28 A FALSE
#> 6 082622 2014-02-08 A FALSE
#> 7 77FC41 2010-11-13 A FALSE
#> 8 077552 2002-05-14 A FALSE
#> 9 D08278 2016-11-18 C FALSE
#> 10 5D1690 2014-12-27 A TRUE
#> # … 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 [182]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2012-12-27 616685 1 TRUE
#> 2 Clinical 2016-05-21 A79917 1 TRUE
#> 3 Clinical 2009-05-25 3F562C 1 TRUE
#> 4 Clinical 2008-05-06 945BD5 1 TRUE
#> 5 Clinical 2017-12-28 5DB1C8 1 TRUE
#> 6 ICU 2014-02-08 082622 1 TRUE
#> 7 Outpatient 2010-11-13 77FC41 1 TRUE
#> 8 Clinical 2002-05-14 077552 1 TRUE
#> 9 ICU 2016-11-18 D08278 1 TRUE
#> 10 Outpatient 2014-12-27 5D1690 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 107 14 48 68
#> 2 ICU 57 13 38 44
#> 3 Outpatient 18 8 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 [192]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 616685 B_STPHY_EPDR Clinical TRUE
#> 2 A79917 B_ENTRC_FACM Clinical TRUE
#> 3 3F562C B_STPHY_CONS Clinical TRUE
#> 4 945BD5 B_ENTRBC_CLOC Clinical TRUE
#> 5 5DB1C8 B_STPHY_HMNS Clinical TRUE
#> 6 082622 B_ESCHR_COLI ICU TRUE
#> 7 77FC41 B_KLBSL_PNMN Outpatient TRUE
#> 8 077552 B_STRPT_PNMN Clinical TRUE
#> 9 D08278 B_ESCHR_COLI ICU TRUE
#> 10 5D1690 B_ESCHR_COLI Outpatient TRUE
#> # … with 190 more rows
# }