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$date, episode_days = 60) # indices
#> [1] 61 15 27 7 41 50 27 14 30 57 61 25 26 60 34 11 53 45 4 9 6 11 25 37 59
#> [26] 14 63 44 23 23 36 48 5 14 46 2 33 33 60 39 62 57 52 52 56 34 66 22 57 5
#> [51] 62 64 47 12 7 65 1 39 12 34 12 64 13 59 22 6 17 63 48 14 23 45 36 32 4
#> [76] 34 24 50 42 8 3 14 53 45 21 55 21 53 52 6 35 19 43 48 6 55 3 5 29 21
#> [101] 22 47 38 65 30 51 8 54 62 2 40 28 11 45 8 31 4 63 53 14 57 57 62 45 57
#> [126] 9 3 52 31 53 40 51 49 8 52 36 18 54 20 11 10 7 59 18 43 63 63 51 16 49
#> [151] 12 10 48 33 43 50 1 58 33 43 8 42 66 56 44 13 60 7 40 41 34 8 45 13 12
#> [176] 55 43 45 18 33 1 32 62 9 34 57 42 30 57 36 4 42 24 28 35 42 28 4 3 43
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE
#> [13] TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE TRUE
#> [25] TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE
#> [37] FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE
#> [49] FALSE TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [61] FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE
#> [73] FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [85] FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE
#> [97] FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE
#> [109] FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE
#> [121] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [133] FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE
#> [145] FALSE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [157] TRUE TRUE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [169] TRUE TRUE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE
#> [181] FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE
#> [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 2002-07-23 F35553 51 M ICU B_STPHY_AURS R NA S R
#> 2 2002-08-19 A49852 70 M Clinical B_ESCHR_COLI R NA NA NA
#> 3 2002-07-21 955940 82 F Clinical B_PSDMN_AERG R NA NA R
#> 4 2002-08-14 785317 51 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 D08278 2016-11-18 B TRUE
#> 2 4047F6 2005-01-19 A FALSE
#> 3 B66559 2007-11-28 A FALSE
#> 4 E35356 2003-06-11 A TRUE
#> 5 807228 2011-06-10 C FALSE
#> 6 966513 2013-11-12 B FALSE
#> 7 9C1B92 2008-01-25 A FALSE
#> 8 C56827 2004-12-05 B FALSE
#> 9 E95747 2008-06-30 C FALSE
#> 10 A76045 2015-10-06 A 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 [182]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Outpatient 2016-11-18 D08278 1 TRUE
#> 2 ICU 2005-01-19 4047F6 1 TRUE
#> 3 ICU 2007-11-28 B66559 1 TRUE
#> 4 ICU 2003-06-11 E35356 1 TRUE
#> 5 Clinical 2011-06-10 807228 1 TRUE
#> 6 Clinical 2013-11-12 966513 1 TRUE
#> 7 Clinical 2008-01-25 9C1B92 1 TRUE
#> 8 Clinical 2004-12-05 C56827 1 TRUE
#> 9 Clinical 2008-06-30 E95747 1 TRUE
#> 10 ICU 2015-10-06 A76045 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 111 14 55 74
#> 2 ICU 57 13 37 42
#> 3 Outpatient 14 9 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] 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 [187]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 D08278 B_ESCHR_COLI Outpatient TRUE
#> 2 4047F6 B_ESCHR_COLI ICU TRUE
#> 3 B66559 B_ESCHR_COLI ICU TRUE
#> 4 E35356 B_STPHY_CONS ICU TRUE
#> 5 807228 B_STRPT_PNMN Clinical TRUE
#> 6 966513 B_STPHY_HMNS Clinical TRUE
#> 7 9C1B92 B_STPHY_CONS Clinical TRUE
#> 8 C56827 B_ESCHR_COLI Clinical TRUE
#> 9 E95747 B_KLBSL_PNMN Clinical TRUE
#> 10 A76045 B_ENTRC_FACM ICU TRUE
#> # … with 190 more rows
# }