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$date, episode_days = 60) # indices
#> [1] 31 3 57 44 3 12 59 6 2 52 18 35 56 59 21 41 52 6 51 45 16 4 7 51 56
#> [26] 20 50 9 59 41 5 5 27 7 8 43 44 16 57 1 60 33 23 7 40 63 18 59 32 39
#> [51] 42 31 7 6 60 2 48 58 41 40 12 59 55 24 40 17 60 37 16 9 47 56 51 17 9
#> [76] 15 26 14 40 38 14 9 59 59 15 6 11 18 36 18 55 60 3 40 59 12 53 36 43 44
#> [101] 34 21 10 31 44 33 39 12 44 55 11 34 46 20 9 33 52 30 29 59 3 26 21 36 40
#> [126] 43 36 58 9 52 17 61 6 48 22 28 8 41 52 27 45 46 63 19 58 46 44 27 13 9
#> [151] 25 22 54 23 36 50 5 14 39 56 57 26 60 36 30 62 16 44 49 4 5 8 34 47 16
#> [176] 27 16 52 52 49 26 23 59 29 29 31 62 12 30 62 10 28 57 15 37 26 20 31 50 1
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [13] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [25] TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE
#> [37] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [49] TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE
#> [61] FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [73] TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [85] TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [97] TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [109] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [121] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE
#> [133] TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE
#> [145] TRUE TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE
#> [157] FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [169] TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE
#> [193] FALSE FALSE TRUE FALSE 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: 4 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-11-16 762305 87 F Clinical B_BCTRD_FRGL R NA NA R
#> 2 2002-10-14 FCC668 54 F ICU B_ACNTB R NA NA NA
#> 3 2002-10-20 F35553 51 M ICU B_STPHY_AURS S NA S NA
#> 4 2002-11-21 450741 77 F ICU B_STPHY_EPDR 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 291882 2010-05-01 C FALSE
#> 2 762305 2002-11-16 A FALSE
#> 3 644292 2016-10-26 C FALSE
#> 4 B86399 2013-08-13 B TRUE
#> 5 FCC668 2002-10-14 C FALSE
#> 6 F41D7B 2005-01-12 A FALSE
#> 7 CD8C33 2017-02-25 C TRUE
#> 8 A59636 2003-07-24 A FALSE
#> 9 E52286 2002-05-16 A TRUE
#> 10 B6F683 2015-11-15 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 [183]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2010-05-01 291882 1 TRUE
#> 2 Clinical 2002-11-16 762305 1 TRUE
#> 3 ICU 2016-10-26 644292 1 TRUE
#> 4 ICU 2013-08-13 B86399 1 TRUE
#> 5 ICU 2002-10-14 FCC668 1 TRUE
#> 6 ICU 2005-01-12 F41D7B 1 TRUE
#> 7 Clinical 2017-02-25 CD8C33 1 TRUE
#> 8 Clinical 2003-07-24 A59636 1 TRUE
#> 9 Clinical 2002-05-16 E52286 1 TRUE
#> 10 Outpatient 2015-11-15 B6F683 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 102 13 47 71
#> 2 ICU 65 13 40 48
#> 3 Outpatient 16 8 14 15
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 [193]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 291882 B_STPHY_EPDR Clinical TRUE
#> 2 762305 B_BCTRD_FRGL Clinical TRUE
#> 3 644292 B_STPHY_AURS ICU TRUE
#> 4 B86399 B_LISTR_MNCY ICU TRUE
#> 5 FCC668 B_ACNTB ICU TRUE
#> 6 F41D7B B_STRPT_PNMN ICU TRUE
#> 7 CD8C33 B_STPHY_HMNS Clinical TRUE
#> 8 A59636 B_STPHY_AURS Clinical TRUE
#> 9 E52286 B_STPHY_AURS Clinical TRUE
#> 10 B6F683 B_STPHY_HMNS Outpatient TRUE
#> # … with 190 more rows
# }