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$date, episode_days = 60) # indices
#> [1] 47 61 41 30 22 55 11 62 22 52 58 11 11 17 15 28 11 13 37 59 31 18 8 48 41
#> [26] 45 3 64 1 53 8 64 55 62 41 23 58 64 51 18 16 8 64 40 47 50 20 9 15 31
#> [51] 24 58 8 37 7 21 67 30 17 14 1 43 41 9 17 45 44 13 21 31 29 39 2 66 36
#> [76] 12 61 52 30 67 58 65 63 30 56 26 60 66 40 60 43 51 35 65 64 7 64 25 54 47
#> [101] 19 9 7 9 63 52 41 67 5 33 21 20 42 4 17 35 66 19 34 37 18 23 33 8 7
#> [126] 51 41 60 38 2 67 58 10 42 38 1 29 49 10 22 46 67 1 3 39 61 27 32 5 63
#> [151] 12 16 3 53 7 49 31 23 11 22 29 37 30 45 32 35 45 1 56 46 23 18 49 64 30
#> [176] 27 8 1 6 32 9 10 18 12 20 37 46 49 9 24 10 67 57 8 55 48 62 65 31 61
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE
#> [13] FALSE FALSE TRUE TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE
#> [25] FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE FALSE
#> [37] FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE TRUE TRUE TRUE
#> [49] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [61] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE FALSE
#> [73] TRUE FALSE TRUE TRUE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [85] TRUE TRUE TRUE TRUE TRUE FALSE TRUE FALSE TRUE TRUE FALSE TRUE
#> [97] FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE TRUE FALSE FALSE
#> [109] TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE
#> [121] FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE FALSE FALSE TRUE
#> [133] FALSE TRUE FALSE TRUE FALSE FALSE TRUE TRUE FALSE FALSE FALSE TRUE
#> [145] TRUE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE
#> [157] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE
#> [169] FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [181] FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE FALSE
#> [193] TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE
# filter on results from the third 60-day episode only, using base R
df[which(get_episode(df$date, 60) == 3), ]
#> # A tibble: 3 × 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 2002-06-06 24D393 20 F Clinical B_ESCHR_COLI R NA NA NA
#> 3 2002-06-19 402950 53 F Clinical B_STPHY_HMNS 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 023456 2012-11-24 A FALSE
#> 2 A79917 2016-05-21 C FALSE
#> 3 257844 2011-05-22 B TRUE
#> 4 F50462 2008-05-11 A FALSE
#> 5 3CF3C4 2006-06-26 C FALSE
#> 6 E29972 2014-11-22 C FALSE
#> 7 406502 2004-02-29 A FALSE
#> 8 D10538 2016-08-13 A FALSE
#> 9 336698 2006-07-02 C FALSE
#> 10 DF7736 2014-02-19 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 2012-11-24 023456 1 TRUE
#> 2 Clinical 2016-05-21 A79917 1 TRUE
#> 3 Clinical 2011-05-22 257844 1 TRUE
#> 4 ICU 2008-05-11 F50462 1 TRUE
#> 5 Clinical 2006-06-26 3CF3C4 1 TRUE
#> 6 Outpatient 2014-11-22 E29972 1 TRUE
#> 7 Clinical 2004-02-29 406502 1 TRUE
#> 8 ICU 2016-08-13 D10538 1 TRUE
#> 9 Clinical 2006-07-02 336698 1 TRUE
#> 10 ICU 2014-02-19 DF7736 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 54 79
#> 2 ICU 48 12 32 34
#> 3 Outpatient 11 6 9 10
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 [185]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 023456 B_PROTS_MRBL Clinical TRUE
#> 2 A79917 B_ESCHR_COLI Clinical TRUE
#> 3 257844 B_STPHY_CONS Clinical TRUE
#> 4 F50462 B_STPHY_CONS ICU TRUE
#> 5 3CF3C4 B_STPHY_CONS Clinical TRUE
#> 6 E29972 B_STPHY_HMNS Outpatient TRUE
#> 7 406502 B_STRPT_PNMN Clinical TRUE
#> 8 D10538 B_ESCHR_COLI ICU TRUE
#> 9 336698 B_NESSR_SICC Clinical TRUE
#> 10 DF7736 B_STRPT_PYGN ICU TRUE
#> # … with 190 more rows
# }