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$date, episode_days = 60) # indices
#> [1] 35 62 35 6 5 61 64 22 27 58 10 64 57 60 29 58 21 9 15 3 46 19 32 23 55
#> [26] 5 58 29 21 36 62 41 18 12 20 41 62 6 38 26 37 41 33 35 22 15 20 61 4 19
#> [51] 25 41 51 48 50 44 11 42 10 57 45 2 27 16 53 35 41 62 28 54 36 12 7 39 5
#> [76] 12 58 19 14 63 30 28 10 8 54 17 46 30 53 32 14 31 30 17 46 16 39 59 48 52
#> [101] 49 1 12 57 33 37 51 12 22 13 27 16 60 36 9 64 10 33 2 42 40 4 24 15 60
#> [126] 31 63 49 2 43 45 53 53 38 48 38 39 48 60 61 12 4 27 37 37 27 28 56 34 28
#> [151] 47 11 14 58 24 54 64 16 59 15 7 31 45 23 10 32 21 61 43 21 14 38 58 13 14
#> [176] 12 1 59 24 23 60 6 19 7 5 3 7 32 37 24 54 24 52 24 39 15 58 32 9 14
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [13] TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE
#> [25] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [37] TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE TRUE FALSE
#> [49] FALSE TRUE TRUE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [61] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
#> [73] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
#> [97] FALSE TRUE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [109] FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [121] TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE TRUE
#> [133] FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [145] FALSE FALSE FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [157] TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE
#> [169] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE
#> [181] FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [193] FALSE FALSE FALSE FALSE TRUE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-10-11 871360 78 M Clinical B_STPHY_EPDR R NA S NA
#> 2 2002-11-11 D80753 74 F Outpatie… B_STPHY_CONS R 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 387C40 2010-12-06 A TRUE
#> 2 6FC260 2017-06-07 C FALSE
#> 3 829641 2011-01-03 C FALSE
#> 4 914520 2003-07-31 B FALSE
#> 5 114570 2003-04-08 B FALSE
#> 6 612042 2017-03-15 A FALSE
#> 7 C80762 2017-12-21 C FALSE
#> 8 650870 2007-09-10 B FALSE
#> 9 CD4354 2009-02-19 B FALSE
#> 10 C6F894 2016-07-11 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 [179]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2010-12-06 387C40 1 TRUE
#> 2 ICU 2017-06-07 6FC260 1 TRUE
#> 3 Clinical 2011-01-03 829641 1 TRUE
#> 4 Clinical 2003-07-31 914520 1 TRUE
#> 5 ICU 2003-04-08 114570 1 TRUE
#> 6 ICU 2017-03-15 612042 1 TRUE
#> 7 ICU 2017-12-21 C80762 1 TRUE
#> 8 ICU 2007-09-10 650870 1 TRUE
#> 9 ICU 2009-02-19 CD4354 1 TRUE
#> 10 ICU 2016-07-11 C6F894 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 106 14 52 73
#> 2 ICU 60 12 36 48
#> 3 Outpatient 13 9 13 13
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 [192]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 387C40 B_KLBSL_PNMN Clinical TRUE
#> 2 6FC260 B_STPHY_HMNS ICU TRUE
#> 3 829641 B_ESCHR_COLI Clinical TRUE
#> 4 914520 B_STPHY_AURS Clinical TRUE
#> 5 114570 B_STRPT_GRPA ICU TRUE
#> 6 612042 B_ESCHR_COLI ICU TRUE
#> 7 C80762 B_ESCHR_COLI ICU TRUE
#> 8 650870 B_ENTRBC_CLOC ICU TRUE
#> 9 CD4354 B_ESCHR_COLI ICU TRUE
#> 10 C6F894 B_STPHY_AURS ICU TRUE
#> # … with 190 more rows
# }