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$date, episode_days = 60) # indices
#> [1] 36 10 11 17 57 60 18 7 5 18 8 60 31 28 16 41 3 43 48 61 2 46 35 51 2
#> [26] 28 44 45 3 49 25 56 44 31 16 1 4 32 57 6 57 62 19 50 14 55 38 55 24 35
#> [51] 5 15 7 36 38 28 20 5 24 12 63 39 17 13 35 20 30 20 13 44 13 23 21 36 45
#> [76] 43 52 53 50 35 24 32 5 23 40 16 63 44 17 63 13 16 51 36 60 12 1 8 10 26
#> [101] 25 56 55 60 1 37 23 30 20 58 11 25 8 21 29 27 10 54 59 59 17 12 11 44 33
#> [126] 48 54 18 39 8 52 26 40 58 33 7 16 8 16 49 12 15 61 1 28 48 18 1 4 19
#> [151] 2 38 37 36 9 58 35 37 57 42 16 14 29 30 34 35 5 62 48 12 11 55 26 57 35
#> [176] 47 31 24 60 37 31 56 52 41 61 33 26 47 11 8 55 13 59 62 35 34 11 22 38 8
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [13] FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [25] TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [37] FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [49] FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE
#> [61] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [73] TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE
#> [85] TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [97] FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE TRUE
#> [121] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [133] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [145] FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [157] FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [169] TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE TRUE FALSE FALSE
#> [193] FALSE TRUE TRUE TRUE FALSE TRUE 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 2002-07-23 F35553 51 M ICU B_STPHY_AURS R NA S R
#> 2 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 207325 2011-05-08 C FALSE
#> 2 304508 2004-05-09 B TRUE
#> 3 619705 2004-07-29 A TRUE
#> 4 3CF3C4 2006-06-26 B FALSE
#> 5 A79917 2016-05-21 C FALSE
#> 6 F5F794 2017-02-18 B FALSE
#> 7 698578 2006-11-13 A FALSE
#> 8 438785 2003-10-03 B FALSE
#> 9 A26548 2003-04-13 C FALSE
#> 10 418311 2006-11-10 C 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 [186]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 ICU 2011-05-08 207325 1 TRUE
#> 2 Clinical 2004-05-09 304508 1 TRUE
#> 3 ICU 2004-07-29 619705 1 TRUE
#> 4 Clinical 2006-06-26 3CF3C4 1 TRUE
#> 5 Clinical 2016-05-21 A79917 1 TRUE
#> 6 ICU 2017-02-18 F5F794 1 TRUE
#> 7 Clinical 2006-11-13 698578 1 TRUE
#> 8 ICU 2003-10-03 438785 1 TRUE
#> 9 ICU 2003-04-13 A26548 1 TRUE
#> 10 ICU 2006-11-10 418311 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 126 14 52 77
#> 2 ICU 47 12 30 35
#> 3 Outpatient 13 7 12 12
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 207325 UNKNOWN ICU TRUE
#> 2 304508 B_STRPT_PNMN Clinical TRUE
#> 3 619705 B_STRPT_ANGN ICU TRUE
#> 4 3CF3C4 B_STPHY_CONS Clinical TRUE
#> 5 A79917 B_ESCHR_COLI Clinical TRUE
#> 6 F5F794 B_STPHY_AURS ICU TRUE
#> 7 698578 B_ESCHR_COLI Clinical TRUE
#> 8 438785 B_STPHY_EPDR ICU TRUE
#> 9 A26548 B_STPHY_CONS ICU TRUE
#> 10 418311 B_STPHY_CONS ICU TRUE
#> # … with 190 more rows
# }