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$date, episode_days = 60) # indices
#> [1] 46 38 48 21 23 4 29 56 29 37 60 63 64 60 4 40 58 22 2 26 17 22 7 58 8
#> [26] 21 21 60 50 1 23 36 52 43 6 68 7 43 47 28 57 6 51 60 62 65 59 28 29 31
#> [51] 13 62 55 4 9 54 58 15 49 31 10 63 36 54 8 17 58 40 49 25 10 7 14 23 8
#> [76] 68 60 19 65 12 16 8 47 14 58 9 65 52 53 59 6 48 62 47 20 42 67 9 13 22
#> [101] 55 16 13 26 60 27 65 1 29 25 41 53 42 32 10 62 54 10 68 42 29 13 34 65 64
#> [126] 44 39 31 13 18 61 14 62 1 36 54 33 9 67 61 43 15 37 48 16 11 61 38 66 58
#> [151] 3 24 4 68 35 54 41 7 7 36 5 30 1 61 22 39 34 45 19 13 5 53 13 7 61
#> [176] 4 20 11 45 7 21 49 52 64 60 9 3 35 30 45 11 65 10 58 5 13 13 18 13 41
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE FALSE TRUE FALSE TRUE
#> [13] FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE
#> [37] FALSE TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [49] FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE TRUE TRUE TRUE FALSE
#> [61] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE
#> [73] FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE TRUE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
#> [97] TRUE TRUE FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [121] FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [133] FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [145] FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [157] TRUE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE TRUE FALSE
#> [169] TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE
#> [181] FALSE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE TRUE 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: 2 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-08-31 149442 80 F ICU B_STPHY_AURS R NA S R
#> 2 2002-07-23 F35553 51 M ICU B_STPHY_AURS R NA S 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 299616 2012-11-13 C TRUE
#> 2 296170 2010-07-23 C FALSE
#> 3 545388 2013-07-29 C FALSE
#> 4 3F5B3F 2006-09-06 C FALSE
#> 5 89F578 2007-03-17 A FALSE
#> 6 974319 2002-10-11 B FALSE
#> 7 A42180 2008-04-21 B FALSE
#> 8 900485 2015-05-27 A FALSE
#> 9 181687 2008-05-04 A TRUE
#> 10 E58549 2010-03-06 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 2012-11-13 299616 1 TRUE
#> 2 ICU 2010-07-23 296170 1 TRUE
#> 3 Clinical 2013-07-29 545388 1 TRUE
#> 4 ICU 2006-09-06 3F5B3F 1 TRUE
#> 5 Clinical 2007-03-17 89F578 1 TRUE
#> 6 Outpatient 2002-10-11 974319 1 TRUE
#> 7 Clinical 2008-04-21 A42180 1 TRUE
#> 8 ICU 2015-05-27 900485 1 TRUE
#> 9 Outpatient 2008-05-04 181687 1 TRUE
#> 10 Clinical 2010-03-06 E58549 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 105 15 50 72
#> 2 ICU 64 13 39 47
#> 3 Outpatient 14 7 13 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 [190]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 299616 B_STPHY_AURS Clinical TRUE
#> 2 296170 B_STRPT_GRPB ICU TRUE
#> 3 545388 B_PROTS_MRBL Clinical TRUE
#> 4 3F5B3F B_STPHY_CONS ICU TRUE
#> 5 89F578 B_STPHY_CONS Clinical TRUE
#> 6 974319 B_STPHY_EPDR Outpatient TRUE
#> 7 A42180 B_KLBSL_PNMN Clinical TRUE
#> 8 900485 B_STRPT_MITS ICU TRUE
#> 9 181687 B_ESCHR_COLI Outpatient TRUE
#> 10 E58549 B_KLBSL_PNMN Clinical TRUE
#> # … with 190 more rows
# }