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$date, episode_days = 60) # indices
#> [1] 51 65 58 14 10 39 37 52 60 27 70 33 39 63 23 30 51 56 66 11 34 57 38 65 65
#> [26] 44 19 67 18 27 53 14 22 49 14 54 3 31 21 8 14 50 3 67 4 42 8 59 28 33
#> [51] 48 24 62 31 68 6 30 31 36 1 46 48 60 2 46 35 28 4 67 71 40 42 21 70 24
#> [76] 51 42 42 47 52 13 57 29 4 56 29 64 8 52 9 42 69 50 2 66 12 30 23 56 57
#> [101] 56 10 51 41 68 7 56 21 17 65 32 34 14 43 12 45 54 64 71 68 15 56 1 21 15
#> [126] 71 44 65 25 6 68 43 10 22 43 14 36 25 65 9 10 2 26 37 2 1 47 70 58 1
#> [151] 26 37 61 6 18 70 20 61 59 16 56 8 5 22 50 36 45 46 16 17 63 21 40 46 1
#> [176] 7 55 40 23 12 30 9 1 27 4 18 63 53 19 42 3 22 54 28 61 25 61 11 16 33
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE FALSE FALSE
#> [13] FALSE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE
#> [25] FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [37] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [49] FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE
#> [61] FALSE TRUE FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
#> [73] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE
#> [85] FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [97] TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE TRUE
#> [109] FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [121] FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE
#> [133] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [145] FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE TRUE TRUE
#> [157] TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE TRUE TRUE
#> [169] TRUE TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE
#> [193] TRUE FALSE FALSE TRUE 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: 3 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-08-14 785317 51 F ICU B_ESCHR_COLI R NA NA NA
#> 2 2002-07-30 218912 76 F ICU B_ESCHR_COLI R NA NA NA
#> 3 2002-06-19 402950 53 F Clinical B_STPHY_HMNS 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 964129 2013-09-11 C TRUE
#> 2 B46416 2016-08-24 C FALSE
#> 3 523893 2015-04-13 A FALSE
#> 4 F35553 2004-12-29 A FALSE
#> 5 419655 2004-02-21 C FALSE
#> 6 F54287 2010-07-05 B FALSE
#> 7 662978 2010-02-20 A FALSE
#> 8 966513 2013-11-12 A FALSE
#> 9 618515 2015-08-09 C FALSE
#> 10 778D04 2007-09-30 A 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 [184]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2013-09-11 964129 1 TRUE
#> 2 ICU 2016-08-24 B46416 1 TRUE
#> 3 Clinical 2015-04-13 523893 1 TRUE
#> 4 ICU 2004-12-29 F35553 4 TRUE
#> 5 ICU 2004-02-21 419655 1 TRUE
#> 6 Clinical 2010-07-05 F54287 1 TRUE
#> 7 Clinical 2010-02-20 662978 1 TRUE
#> 8 Clinical 2013-11-12 966513 1 TRUE
#> 9 Clinical 2015-08-09 618515 1 TRUE
#> 10 ICU 2007-09-30 778D04 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 14 55 80
#> 2 ICU 61 13 42 48
#> 3 Outpatient 5 4 5 5
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 964129 B_ESCHR_COLI Clinical TRUE
#> 2 B46416 B_STPHY_EPDR ICU TRUE
#> 3 523893 B_STPHY_AURS Clinical TRUE
#> 4 F35553 B_SERRT_MRCS ICU TRUE
#> 5 419655 B_STRPT_PNMN ICU TRUE
#> 6 F54287 B_KLBSL_OXYT Clinical TRUE
#> 7 662978 B_STPHY_CONS Clinical TRUE
#> 8 966513 B_STPHY_HMNS Clinical TRUE
#> 9 618515 B_STPHY_EPDR Clinical TRUE
#> 10 778D04 B_STRPT_PNMN ICU TRUE
#> # … with 190 more rows
# }