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$date, episode_days = 60) # indices
#> [1] 21 6 27 50 51 28 39 19 14 34 23 43 60 21 1 44 39 11 2 30 34 24 1 46 16
#> [26] 21 27 50 8 49 11 54 5 24 26 21 38 11 8 56 1 51 26 13 36 18 59 15 37 8
#> [51] 63 39 37 52 1 60 17 4 53 42 62 42 47 6 30 15 25 61 44 24 11 43 60 65 59
#> [76] 4 14 3 3 21 53 18 62 10 38 7 53 23 65 30 34 7 31 28 28 54 43 40 49 49
#> [101] 21 64 64 63 39 44 46 56 25 63 21 37 2 49 44 13 4 1 1 6 56 9 35 11 14
#> [126] 4 32 13 23 63 55 7 17 54 20 4 9 17 12 1 65 2 60 31 25 7 35 20 57 48
#> [151] 45 13 15 42 17 60 56 17 3 34 22 56 55 23 58 48 28 39 29 44 27 61 15 25 56
#> [176] 2 7 40 1 32 62 9 62 33 12 56 9 61 41 43 13 11 12 2 39 11 3 17 53 24
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE
#> [13] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [25] TRUE TRUE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE
#> [37] TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE
#> [49] TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE TRUE FALSE
#> [61] FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [73] FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE TRUE
#> [85] FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [97] FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [121] FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [133] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [145] TRUE FALSE FALSE TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE FALSE
#> [157] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [169] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [181] FALSE FALSE FALSE TRUE TRUE TRUE TRUE FALSE TRUE FALSE FALSE FALSE
#> [193] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
# filter on results from the third 60-day episode only, using base R
df[which(get_episode(df$date, 60) == 3), ]
#> # A tibble: 4 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-07-28 F54261 69 M Clinical B_STPHY_CONS R NA S NA
#> 2 2002-07-16 241328 78 M Outpatie… B_STPHY_CONS R NA S R
#> 3 2002-08-19 A49852 70 M Clinical B_ESCHR_COLI R NA NA NA
#> 4 2002-07-21 955940 82 F Clinical B_PSDMN_AERG R NA NA 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 0E2483 2007-05-29 C FALSE
#> 2 114570 2003-04-08 A TRUE
#> 3 189795 2008-10-19 A FALSE
#> 4 C48417 2014-02-07 A FALSE
#> 5 658640 2014-05-30 A FALSE
#> 6 B65162 2008-12-29 B TRUE
#> 7 E53320 2011-07-04 A FALSE
#> 8 418311 2006-11-10 B TRUE
#> 9 2B1C03 2005-10-23 C FALSE
#> 10 192353 2010-04-05 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 [182]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2007-05-29 0E2483 1 TRUE
#> 2 ICU 2003-04-08 114570 1 TRUE
#> 3 Clinical 2008-10-19 189795 1 TRUE
#> 4 Clinical 2014-02-07 C48417 1 TRUE
#> 5 Clinical 2014-05-30 658640 1 TRUE
#> 6 Clinical 2008-12-29 B65162 1 TRUE
#> 7 Clinical 2011-07-04 E53320 1 TRUE
#> 8 ICU 2006-11-10 418311 1 TRUE
#> 9 Clinical 2005-10-23 2B1C03 1 TRUE
#> 10 Clinical 2010-04-05 192353 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 113 14 49 70
#> 2 ICU 58 13 34 42
#> 3 Outpatient 11 5 11 11
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 [193]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 0E2483 B_ESCHR_COLI Clinical TRUE
#> 2 114570 B_STRPT_PYGN ICU TRUE
#> 3 189795 B_AERCC_URNQ Clinical TRUE
#> 4 C48417 B_LCTBC_DLBR_LCTS Clinical TRUE
#> 5 658640 B_STPHY_AURS Clinical TRUE
#> 6 B65162 B_STRPT_PNMN Clinical TRUE
#> 7 E53320 B_ESCHR_COLI Clinical TRUE
#> 8 418311 B_STPHY_CONS ICU TRUE
#> 9 2B1C03 B_STPHY_CONS Clinical TRUE
#> 10 192353 B_STRPT_PNMN Clinical TRUE
#> # … with 190 more rows
# }