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$date, episode_days = 60) # indices
#> [1] 63 57 48 42 9 59 28 19 62 54 15 56 19 61 26 5 7 28 34 1 8 49 33 4 38
#> [26] 40 40 11 20 24 5 7 17 27 48 34 40 21 46 11 13 53 61 45 17 38 24 51 45 40
#> [51] 28 27 15 14 57 13 59 50 2 39 61 15 49 39 24 48 9 25 63 12 39 55 15 27 32
#> [76] 26 21 1 48 22 16 61 5 52 55 61 5 60 26 17 53 37 53 27 25 40 3 22 17 22
#> [101] 55 28 26 18 3 30 9 27 2 29 4 60 3 15 28 62 14 20 23 55 51 23 31 63 19
#> [126] 6 33 43 40 61 53 48 54 53 59 57 15 19 63 50 45 28 34 22 36 38 51 10 33 47
#> [151] 9 44 32 41 51 35 48 23 2 44 16 3 8 41 5 51 39 26 41 62 30 58 47 45 56
#> [176] 44 44 51 49 37 52 12 42 44 50 55 59 61 3 7 62 38 26 52 16 62 60 17 2 36
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE TRUE
#> [13] TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE
#> [25] FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [37] FALSE TRUE TRUE FALSE FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE
#> [49] FALSE FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE
#> [61] FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [73] FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE TRUE TRUE FALSE FALSE
#> [85] TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [97] FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE
#> [109] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [121] TRUE TRUE TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE
#> [133] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [145] TRUE FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE TRUE
#> [157] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE
#> [169] FALSE TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [181] TRUE TRUE TRUE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE
#> [193] FALSE FALSE FALSE FALSE TRUE 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: 5 × 46
#> date patient age gender ward mo PEN OXA FLC AMX
#> <date> <chr> <dbl> <chr> <chr> <mo> <rsi> <rsi> <rsi> <rsi>
#> 1 2002-11-11 D80753 74 F Outpatie… B_STPHY_CONS R NA R R
#> 2 2002-10-11 974319 78 M Outpatie… B_MCRCCC S NA S NA
#> 3 2002-10-11 871360 78 M Clinical B_STPHY_EPDR R NA S NA
#> 4 2002-11-07 430011 82 F Clinical B_STPHY_CONS R NA R R
#> 5 2002-11-18 956065 89 F Clinical 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 871020 2017-07-20 C FALSE
#> 2 3B8892 2016-05-02 C FALSE
#> 3 479516 2013-09-30 C TRUE
#> 4 116866 2011-11-07 B FALSE
#> 5 A97510 2004-04-29 B FALSE
#> 6 5C1947 2016-08-08 C FALSE
#> 7 533225 2008-10-01 A FALSE
#> 8 3CF3C4 2006-06-26 C FALSE
#> 9 BF4515 2017-06-12 B TRUE
#> 10 A76045 2015-10-06 B 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 2017-07-20 871020 1 TRUE
#> 2 ICU 2016-05-02 3B8892 1 TRUE
#> 3 Clinical 2013-09-30 479516 1 TRUE
#> 4 Clinical 2011-11-07 116866 1 TRUE
#> 5 Clinical 2004-04-29 A97510 1 TRUE
#> 6 Clinical 2016-08-08 5C1947 1 TRUE
#> 7 Clinical 2008-10-01 533225 1 TRUE
#> 8 Clinical 2006-06-26 3CF3C4 1 TRUE
#> 9 Clinical 2017-06-12 BF4515 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 119 14 52 73
#> 2 ICU 56 11 38 43
#> 3 Outpatient 9 3 7 8
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 [189]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 871020 B_STPHY_HMNS Clinical TRUE
#> 2 3B8892 B_ESCHR_COLI ICU TRUE
#> 3 479516 B_STPHY_HMNS Clinical TRUE
#> 4 116866 B_STPHY_CONS Clinical TRUE
#> 5 A97510 B_STPHY_AURS Clinical TRUE
#> 6 5C1947 B_ESCHR_COLI Clinical TRUE
#> 7 533225 B_ESCHR_COLI Clinical TRUE
#> 8 3CF3C4 B_STPHY_CONS Clinical TRUE
#> 9 BF4515 B_ESCHR_COLI Clinical TRUE
#> 10 A76045 B_DRMBC_HMNS ICU TRUE
#> # … with 190 more rows
# }