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$date, episode_days = 60) # indices
#> [1] 49 45 39 34 32 25 14 10 65 5 28 63 26 42 9 56 25 35 9 5 24 67 19 40 1
#> [26] 53 31 61 38 67 8 42 28 40 45 33 4 1 50 5 44 49 26 40 22 64 29 51 58 4
#> [51] 57 31 7 44 53 13 5 64 10 7 11 23 30 59 20 11 5 37 67 60 32 60 59 10 7
#> [76] 44 61 49 55 20 3 12 62 20 27 41 46 24 56 13 19 34 47 52 12 11 59 5 54 12
#> [101] 17 30 26 43 25 25 56 17 47 18 67 40 27 25 45 23 64 43 8 46 60 8 15 25 6
#> [126] 67 11 2 9 50 4 46 60 59 46 49 14 7 28 9 36 36 50 60 2 12 11 66 48 21
#> [151] 31 65 66 12 54 42 62 3 64 42 61 2 7 9 25 1 16 41 63 37 38 39 51 60 24
#> [176] 59 17 62 41 55 62 59 45 25 33 21 50 62 35 57 57 15 52 40 13 48 33 67 63 33
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [13] FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [25] FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE TRUE FALSE
#> [37] TRUE FALSE TRUE TRUE FALSE TRUE TRUE FALSE TRUE FALSE TRUE TRUE
#> [49] TRUE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [61] FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [73] FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [85] TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE TRUE TRUE FALSE
#> [97] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [109] FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE
#> [121] FALSE TRUE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [133] FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE
#> [145] FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE
#> [157] TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE TRUE TRUE FALSE
#> [169] FALSE TRUE TRUE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE
#> [181] FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE
#> [193] FALSE FALSE TRUE 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 2002-06-23 798871 82 M Clinical B_ENTRC_FCLS NA NA NA NA
#> 2 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 545388 2013-07-29 A FALSE
#> 2 107DD1 2012-09-03 B FALSE
#> 3 829641 2011-01-03 A FALSE
#> 4 B54813 2009-10-15 B FALSE
#> 5 942999 2009-02-17 C FALSE
#> 6 0E2483 2007-08-10 A FALSE
#> 7 871360 2004-12-13 B FALSE
#> 8 C58921 2004-01-01 A FALSE
#> 9 468282 2017-05-31 C TRUE
#> 10 762305 2002-11-16 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 [178]
#> ward date patient new_index new_logical
#> <chr> <date> <chr> <dbl> <lgl>
#> 1 Clinical 2013-07-29 545388 1 TRUE
#> 2 Clinical 2012-09-03 107DD1 1 TRUE
#> 3 Clinical 2011-01-03 829641 1 TRUE
#> 4 ICU 2009-10-15 B54813 1 TRUE
#> 5 Clinical 2009-02-17 942999 1 TRUE
#> 6 ICU 2007-08-10 0E2483 1 TRUE
#> 7 Clinical 2004-12-13 871360 1 TRUE
#> 8 Clinical 2004-01-01 C58921 1 TRUE
#> 9 Clinical 2017-05-31 468282 1 TRUE
#> 10 Clinical 2002-11-16 762305 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 15 55 72
#> 2 ICU 55 13 37 44
#> 3 Outpatient 12 8 12 12
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 [191]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 545388 B_KLBSL_PNMN Clinical TRUE
#> 2 107DD1 B_STPHY_CONS Clinical TRUE
#> 3 829641 B_ESCHR_COLI Clinical TRUE
#> 4 B54813 B_ENTRC ICU TRUE
#> 5 942999 B_STRPT_PNMN Clinical TRUE
#> 6 0E2483 B_ENTRC_FACM ICU TRUE
#> 7 871360 B_KLBSL_PNMN Clinical TRUE
#> 8 C58921 B_ESCHR_COLI Clinical TRUE
#> 9 468282 B_STPHY_AURS Clinical TRUE
#> 10 762305 B_PROTS_MRBL Clinical TRUE
#> # … with 190 more rows
# }