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$date, episode_days = 60) # indices
#> [1] 2 1 52 9 1 5 52 4 4 33 43 5 44 22 9 19 6 17 1 56 4 15 44 46 49
#> [26] 2 49 37 26 31 48 11 25 33 5 7 9 42 56 10 35 2 51 47 40 21 50 59 50 39
#> [51] 40 24 29 57 2 14 12 31 54 18 8 53 39 58 21 54 1 38 2 12 20 52 42 41 5
#> [76] 8 15 26 32 44 21 43 5 25 50 23 22 10 43 48 11 45 12 52 26 28 23 34 8 27
#> [101] 53 46 53 23 61 16 29 11 25 46 39 3 46 57 21 45 38 43 54 33 16 4 56 38 7
#> [126] 42 7 14 31 60 3 59 59 9 56 7 38 25 4 42 33 7 49 19 14 46 24 49 47 29
#> [151] 49 6 56 59 51 45 12 60 4 53 22 17 55 59 57 30 24 58 10 48 36 6 37 32 10
#> [176] 18 57 60 37 50 58 36 9 29 33 56 58 29 37 21 57 10 9 30 13 24 34 22 6 18
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [13] FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE
#> [37] TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE TRUE
#> [49] TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [61] FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE
#> [73] FALSE TRUE FALSE TRUE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [97] TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [109] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [121] TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [133] FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [145] TRUE TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE
#> [157] FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
#> [169] FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [193] FALSE FALSE TRUE FALSE FALSE FALSE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-09-08 B8CB09 60 F Outpatie… B_STPHY_CONS S NA S NA
#> 2 2002-11-04 304347 62 M Clinical B_STRPT_PNMN S NA NA S
#> # … 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 833842 2002-05-04 B FALSE
#> 2 858515 2002-01-17 B TRUE
#> 3 A84726 2015-08-14 C FALSE
#> 4 570937 2004-02-18 B FALSE
#> 5 614772 2002-02-27 B FALSE
#> 6 6BC362 2003-04-25 C FALSE
#> 7 D43214 2015-06-16 B FALSE
#> 8 384360 2003-01-09 C FALSE
#> 9 89B405 2003-01-04 B FALSE
#> 10 4F6830 2010-03-12 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 ICU 2002-05-04 833842 1 TRUE
#> 2 ICU 2002-01-17 858515 1 TRUE
#> 3 Clinical 2015-08-14 A84726 1 TRUE
#> 4 Clinical 2004-02-18 570937 1 TRUE
#> 5 Clinical 2002-02-27 614772 1 TRUE
#> 6 ICU 2003-04-25 6BC362 1 FALSE
#> 7 Clinical 2015-06-16 D43214 1 TRUE
#> 8 Clinical 2003-01-09 384360 1 TRUE
#> 9 Clinical 2003-01-04 89B405 1 TRUE
#> 10 Clinical 2010-03-12 4F6830 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 118 15 50 75
#> 2 ICU 51 12 32 40
#> 3 Outpatient 14 8 14 14
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 [195]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 833842 B_STPHY_AURS ICU TRUE
#> 2 858515 B_STPHY_EPDR ICU TRUE
#> 3 A84726 B_STPHY_EPDR Clinical TRUE
#> 4 570937 B_KLBSL_PNMN Clinical TRUE
#> 5 614772 B_STPHY_HMNS Clinical TRUE
#> 6 6BC362 B_CRYNB ICU TRUE
#> 7 D43214 B_STPHY_HMNS Clinical TRUE
#> 8 384360 B_STPHY_CONS Clinical TRUE
#> 9 89B405 B_STRPT_PNMN Clinical TRUE
#> 10 4F6830 B_ESCHR_COLI Clinical TRUE
#> # … with 190 more rows
# }