AMR/R/is_new_episode.R

# ==================================================================== #
# TITLE                                                                #
# Antimicrobial Resistance (AMR) Analysis for R                        #
#                                                                      #
# SOURCE                                                               #
# https://github.com/msberends/AMR                                     #
#                                                                      #
# LICENCE                                                              #
# (c) 2018-2020 Berends MS, Luz CF et al.                              #
# Developed at the University of Groningen, the Netherlands, in        #
# collaboration with non-profit organisations Certe Medical            #
# Diagnostics & Advice, and University Medical Center Groningen.       # 
#                                                                      #
# This R package is free software; you can freely use and distribute   #
# it for both personal and commercial purposes under the terms of the  #
# GNU General Public License version 2.0 (GNU GPL-2), as published by  #
# the Free Software Foundation.                                        #
# We created this package for both routine data analysis and academic  #
# research and it was publicly released in the hope that it will be    #
# useful, but it comes WITHOUT ANY WARRANTY OR LIABILITY.              #
#                                                                      #
# Visit our website for the full manual and a complete tutorial about  #
# how to conduct AMR analysis: https://msberends.github.io/AMR/        #
# ==================================================================== #

#' Determine (new) episodes for patients
#' 
#' This function determines which items in a vector can be considered (the start of) a new episode, based on the parameter `episode_days`. This can be used to determine clinical episodes for any epidemiological analysis.
#' @inheritSection lifecycle Experimental lifecycle
#' @param x vector of dates (class `Date` or `POSIXt`)
#' @param episode_days length of the required episode in days, defaults to 365. Every element in the input will return `TRUE` after this number of days has passed since the last included date, independent of calendar years. Please see *Details*.
#' @param ... arguments passed on to [as.Date()]
#' @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 `dplyr` package is not required for this function to work, but this function works conveniently inside `dplyr` verbs such as [`filter()`][dplyr::filter()], [`mutate()`][dplyr::mutate()] and [`summarise()`][dplyr::summarise()].
#' @return a [logical] vector
#' @export
#' @inheritSection AMR Read more on our website!
#' @examples
#' # `example_isolates` is a dataset available in the AMR package.
#' # See ?example_isolates.
#' 
#' is_new_episode(example_isolates$date)
#' is_new_episode(example_isolates$date, episode_days = 60)
#' 
#' if (require("dplyr")) {
#'   # is_new_episode() can also be used in dplyr verbs to determine patient
#'   # episodes based on any (combination of) grouping variables:
#'   example_isolates %>%
#'     mutate(condition = sample(x = c("A", "B", "C"), 
#'                               size = 2000,
#'                               replace = TRUE)) %>% 
#'     group_by(condition) %>%
#'     mutate(new_episode = is_new_episode(date))
#'   
#'   example_isolates %>%
#'     group_by(hospital_id) %>% 
#'     summarise(patients = n_distinct(patient_id),
#'               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)))
#'     
#'     
#'   # grouping on patients and microorganisms leads to the same results
#'   # as first_isolate():
#'   x <- example_isolates %>%
#'     filter(first_isolate(., include_unknown = TRUE))
#'     
#'   y <- example_isolates %>%
#'     group_by(patient_id, mo) %>%
#'     filter(is_new_episode(date))
#'
#'   identical(x$patient_id, y$patient_id)
#'   
#'   # but is_new_episode() has a lot more flexibility than first_isolate(),
#'   # since you can now group on anything that seems relevant:
#'   example_isolates %>%
#'     group_by(patient_id, mo, hospital_id, ward_icu) %>%
#'     mutate(flag_episode = is_new_episode(date))
#' }
is_new_episode <- function(x, episode_days = 365, ...) {
  meet_criteria(x, allow_class = c("Date", "POSIXt"))
  meet_criteria(episode_days, allow_class = c("numeric", "double", "integer"), has_length = 1)
  
  x <- as.double(as.Date(x, ...)) # as.Date() for POSIX classes
  if (length(x) == 1) {
    return(TRUE)
  } else if (length(x) == 2) {
    if (max(x) - min(x) >= episode_days) {
      return(c(TRUE, TRUE))
    } else {
      return(c(TRUE, FALSE))
    }
  }
  
  # I asked on StackOverflow:
  # https://stackoverflow.com/questions/42122245/filter-one-row-every-year
  exec <- function(x, episode_days) {
    indices <- integer()
    start <- x[1]
    ind <- 1
    indices[1] <- 1
    for (i in 2:length(x)) {
      if (isTRUE((x[i] - start) >= episode_days)) {
        ind <- ind + 1
        indices[ind] <- i
        start <- x[i]
      }
    }
    result <- rep(FALSE, length(x))
    result[indices] <- TRUE
    result
  }
  
  df <- data.frame(x = x,
                   y = seq_len(length(x))) %pm>%
    pm_arrange(x)
  df$new <- exec(df$x, episode_days)
  df %pm>%
    pm_arrange(y) %pm>%
    pm_pull(new)
}