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$date, episode_days = 60) # indices
#> [1] 39 28 17 35 24 12 22 42 33 12 15 14 62 30 28 37 35 21 57 31 56 7 64 24 30
#> [26] 57 32 31 62 38 64 22 19 16 55 19 1 53 21 64 34 41 15 23 4 8 47 4 32 10
#> [51] 56 17 5 23 1 59 56 16 63 35 11 22 58 48 10 59 57 14 18 44 36 5 20 50 37
#> [76] 7 30 59 55 5 47 59 64 41 1 6 24 24 25 3 49 51 32 49 21 52 50 27 5 13
#> [101] 45 9 17 1 12 28 31 26 3 64 44 34 32 25 62 65 42 59 2 34 21 59 20 6 15
#> [126] 14 29 19 39 30 7 45 16 34 31 18 35 32 20 17 7 33 31 4 35 9 25 27 35 60
#> [151] 14 41 40 61 58 25 64 53 9 46 26 51 65 61 35 61 18 47 52 48 6 1 13 58 44
#> [176] 53 60 54 65 16 11 14 52 40 53 64 67 17 22 43 39 1 55 32 32 23 17 65 66 64
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE
#> [13] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [25] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [37] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [49] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE
#> [61] FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE
#> [73] TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE TRUE
#> [85] FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE
#> [97] TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [109] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE TRUE
#> [121] FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE FALSE TRUE
#> [133] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [145] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE
#> [157] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [169] FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
#> [181] TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [193] FALSE FALSE FALSE TRUE TRUE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-06-18 012595 30 M ICU B_CRYNB I NA NA NA
#> 2 2002-06-22 FD8039 75 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 F15984 2011-04-01 B FALSE
#> 2 800264 2008-04-10 B TRUE
#> 3 277241 2005-08-31 B FALSE
#> 4 662978 2010-02-20 C FALSE
#> 5 0E2483 2007-04-06 B TRUE
#> 6 189363 2004-06-22 B FALSE
#> 7 E19253 2006-08-22 B FALSE
#> 8 582258 2012-03-12 C TRUE
#> 9 006827 2009-07-24 A FALSE
#> 10 189363 2004-06-22 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 [185]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Outpatient 2011-04-01 F15984 1 TRUE
#> 2 Outpatient 2008-04-10 800264 1 TRUE
#> 3 ICU 2005-08-31 277241 1 TRUE
#> 4 Clinical 2010-02-20 662978 1 TRUE
#> 5 Clinical 2007-04-06 0E2483 1 TRUE
#> 6 Clinical 2004-06-22 189363 1 TRUE
#> 7 Clinical 2006-08-22 E19253 1 TRUE
#> 8 ICU 2012-03-12 582258 1 TRUE
#> 9 Clinical 2009-07-24 006827 1 TRUE
#> 10 Clinical 2004-06-22 189363 1 FALSE
#> # … 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 114 15 52 74
#> 2 ICU 57 13 36 42
#> 3 Outpatient 14 8 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 F15984 B_STPHY_CONS Outpatient TRUE
#> 2 800264 B_ESCHR_COLI Outpatient TRUE
#> 3 277241 B_STPHY_AURS ICU TRUE
#> 4 662978 B_STPHY_CONS Clinical TRUE
#> 5 0E2483 B_ESCHR_COLI Clinical TRUE
#> 6 189363 B_STPHY_AURS Clinical TRUE
#> 7 E19253 B_STPHY_CONS Clinical TRUE
#> 8 582258 B_STPHY_CONS ICU TRUE
#> 9 006827 B_ENTRC_FCLS Clinical TRUE
#> 10 189363 B_STPHY_AURS Clinical FALSE
#> # … with 190 more rows
# }