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$date, episode_days = 60) # indices
#> [1] 18 57 46 24 3 28 41 20 44 57 51 63 66 9 10 6 40 61 57 42 60 2 8 28 39
#> [26] 39 3 4 21 57 8 54 12 15 20 66 36 35 21 55 22 48 46 58 55 26 13 62 6 15
#> [51] 39 34 17 15 36 54 8 10 51 38 35 4 13 30 17 4 37 35 53 43 62 25 64 7 46
#> [76] 5 26 55 44 30 4 48 14 42 13 50 52 13 7 58 50 17 2 62 56 62 61 32 7 33
#> [101] 64 64 28 31 46 8 24 33 9 31 22 27 39 36 49 12 6 48 24 25 55 43 34 33 2
#> [126] 26 45 33 19 16 47 4 52 4 37 30 5 27 51 1 51 46 19 22 64 6 62 18 11 58
#> [151] 63 59 20 37 32 58 6 37 10 8 38 40 16 26 6 11 43 50 1 30 65 45 4 53 62
#> [176] 13 42 37 37 61 52 36 44 58 39 53 52 4 65 60 58 61 60 9 23 18 29 48 59 33
is_new_episode(df$date, episode_days = 60) # TRUE/FALSE
#> [1] FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE FALSE FALSE TRUE TRUE
#> [13] TRUE FALSE TRUE FALSE TRUE FALSE FALSE FALSE FALSE FALSE TRUE TRUE
#> [25] FALSE FALSE FALSE FALSE TRUE FALSE FALSE TRUE TRUE TRUE FALSE FALSE
#> [37] TRUE TRUE FALSE TRUE TRUE FALSE FALSE TRUE FALSE FALSE FALSE TRUE
#> [49] FALSE FALSE FALSE TRUE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [61] FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [73] TRUE FALSE FALSE TRUE FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE
#> [85] FALSE FALSE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE FALSE
#> [97] FALSE FALSE TRUE FALSE FALSE FALSE FALSE TRUE FALSE FALSE FALSE FALSE
#> [109] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE TRUE TRUE
#> [121] FALSE TRUE FALSE TRUE TRUE TRUE FALSE FALSE FALSE FALSE TRUE FALSE
#> [133] FALSE TRUE FALSE FALSE FALSE TRUE FALSE TRUE FALSE TRUE TRUE FALSE
#> [145] FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE FALSE TRUE FALSE
#> [157] FALSE FALSE FALSE FALSE TRUE FALSE TRUE FALSE FALSE TRUE FALSE FALSE
#> [169] FALSE FALSE FALSE TRUE FALSE FALSE FALSE TRUE TRUE TRUE FALSE FALSE
#> [181] FALSE FALSE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE TRUE
#> [193] TRUE TRUE TRUE FALSE TRUE TRUE 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 2002-07-21 955940 82 F Clinical B_PSDMN_AERG R NA NA R
#> 2 2002-07-30 218912 76 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 A90606 2006-01-19 B FALSE
#> 2 F76601 2015-09-20 B FALSE
#> 3 824233 2012-07-10 A FALSE
#> 4 3C8163 2007-06-26 B FALSE
#> 5 955940 2002-07-21 B FALSE
#> 6 D81577 2008-05-12 B FALSE
#> 7 675872 2011-03-03 B TRUE
#> 8 54890C 2006-05-26 B FALSE
#> 9 967247 2012-02-06 A FALSE
#> 10 0DBF93 2015-10-12 C 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 2006-01-19 A90606 1 TRUE
#> 2 Clinical 2015-09-20 F76601 1 TRUE
#> 3 Clinical 2012-07-10 824233 1 TRUE
#> 4 Clinical 2007-06-26 3C8163 1 TRUE
#> 5 Clinical 2002-07-21 955940 1 TRUE
#> 6 Clinical 2008-05-12 D81577 1 TRUE
#> 7 Clinical 2011-03-03 675872 1 TRUE
#> 8 Clinical 2006-05-26 54890C 1 TRUE
#> 9 ICU 2012-02-06 967247 1 TRUE
#> 10 Clinical 2015-10-12 0DBF93 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 117 14 58 79
#> 2 ICU 56 11 30 41
#> 3 Outpatient 10 5 10 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 [191]
#> patient mo ward flag_episode
#> <chr> <mo> <chr> <lgl>
#> 1 A90606 B_STRPT_PNMN Clinical TRUE
#> 2 F76601 B_ESCHR_COLI Clinical TRUE
#> 3 824233 B_STPHY_AURS Clinical TRUE
#> 4 3C8163 B_PSDMN_AERG Clinical TRUE
#> 5 955940 B_PSDMN_AERG Clinical TRUE
#> 6 D81577 B_HMPHL_INFL Clinical TRUE
#> 7 675872 B_BCTRD_FRGL Clinical TRUE
#> 8 54890C B_ESCHR_COLI Clinical TRUE
#> 9 967247 B_STPHY_CONS ICU TRUE
#> 10 0DBF93 B_STPHY_CPTS Clinical TRUE
#> # … with 190 more rows
# }