Update of epidict package to include an EWAR/EBS template

R/gen_msf_data.R

@@ -2,7 +2,7 @@
 #'
 #' @param dictionary Specify which dictionary you would like to use.
 #'   Currently supports "Cholera", "Measles", "Meningitis", "AJS",
-#'    "Mortality", "Nutrition", "Vaccination_long" and "Vaccination_short"
+#'    "Mortality", "Nutrition", "Vaccination_long", "Vaccination_short" and "EWAR"
 #'
 #' @param varnames Specify name of column that contains variable names. Currently
 #'   default set to "data_element_shortname". If `dictionary` is a survey,
@@ -857,6 +857,122 @@ gen_msf_data <- function(dictionary, dat_dict, is_survey, varnames = "data_eleme
 
   }
 
+  if (dictionary == "EWAR") {
+
+    # set date of event starting to the earliest date from those given
+    dis_output$date_event_start <- with(
+      dis_output,
+      pmin(
+        date_event_start,
+        date_signal,
+        date_triage,
+        date_verification,
+        date_assessment,
+        date_response_started,
+        date_response_ended,
+        na.rm = TRUE
+      )
+    )
+
+    # date signal
+    dis_output <- enforce_timing(dis_output,
+                                 first  = "date_signal",
+                                 second = "date_triage",
+                                 5:30
+    )
+
+    ## date triage
+    dis_output <- enforce_timing(dis_output,
+                                 first  = "date_triage",
+                                 second = "date_verification",
+                                 5:30,
+                                 inclusive = TRUE
+    )
+
+    # died verification
+    dis_output <- enforce_timing(dis_output,
+                                 first  = "date_verification",
+                                 second = "date_assessment",
+                                 5:30
+    )
+
+    ## date assessment
+    dis_output <- enforce_timing(dis_output,
+                                 first  = "date_assessment",
+                                 second = "date_response_started",
+                                 5:30,
+                                 inclusive = TRUE
+    )
+
+    ## date response started
+    dis_output <- enforce_timing(dis_output,
+                                 first  = "date_response_started",
+                                 second = "date_response_ended",
+                                 5:30,
+                                 inclusive = TRUE
+    )
+
+    ## create an initials variable of 3 random letters
+    dis_output$initials <- stringi::stri_rand_strings(nrow(dis_output), 2,
+                                                      pattern = "[a-z]")
+
+
+    ## create a signal id
+    dis_output$event_id_sig <- paste0(dis_output$initials, "_",
+                                      dis_output$location_signal, "_",
+                                      dis_output$signal_type, "_",
+                                      dis_output$date_signal)
+
+
+    ## Copy this signal id to assessment and response event id variables
+    dis_output$event_id_assess <- dis_output$event_id_sig
+
+    dis_output$event_id_res <- dis_output$event_id_sig
+
+
+    ## Add a random  number for total people affected
+    dis_output$total_affected <- sample(1:10, nrow(dis_output), replace = TRUE)
+
+
+    ## Add one person less for under 5 affected
+    dis_output$under5_affected <- dis_output$total_affected - 1
+
+
+    ## Add a random  number for total people identified by active case finding
+    dis_output$acf_total <- sample(2:15, nrow(dis_output), replace = TRUE)
+
+    ## Add one person less for total under 5 identified by active case finding
+    dis_output$acf_under5 <- dis_output$acf_total - 1
+
+
+    ## Ensure that alert_status takes a value of 1 where risk characterisation
+    ## is y (yes)
+    dis_output$alert_status[dis_output$risk_characterisation == "y"] <- 1
+
+
+    ## Ensure that alert_status takes a value of 0 where risk characterisation
+    ## is n (n) or u (unsure)
+    dis_output$alert_status[dis_output$risk_characterisation == "n" |
+                              dis_output$risk_characterisation == "u"] <- 0
+
+    ## Convert all columns NA when no intervention required
+    dis_output[dis_output$alert_status == 0, 42:49] <- NA
+
+    ## Convert all values in columns to NA when a signal is not verified
+    dis_output[dis_output$event_status == 0, 22:49] <- NA
+
+    ## Convert all values in columns to NA when a signal doesn't require
+    ## verification
+    dis_output[dis_output$need_verif == 0, 18:49] <- NA
+
+
+    ## Convert dates of start and end of response to NA if no response carried out
+    ## or if unsure if a response was carried out
+    dis_output[dis_output$response_undertaken %in% c("n", "u"),
+               c("date_response_started","date_response_ended")] <- NA
+
+  }
+
 
   # return dataset as a tibble
   dplyr::as_tibble(dis_output)

R/msf_dict.R

@@ -7,8 +7,8 @@
 #'
 #' @param disease Specify which disease you would like to use.
 #'   - `msf_dict()` supports "AJS", "Cholera", "Measles", "Meningitis"
-#'   - `msf_dict_survey()` supports "Mortality", "Nutrition", "Vaccination_long"
-#'   and "Vaccination_short" (only used in surveys if `template = TRUE`)
+#'   - `msf_dict_survey()` supports "Mortality", "Nutrition", "Vaccination_long",
+#'    "Vaccination_short" (only used in surveys if `template = TRUE`) and "EWAR"
 #'
 #' @param name the name of the dictionary stored in the package.
 #'   - `msf_dict_survey()` supports Kobo dictionaries not stored within this package,

R/msf_dict_rename_helper.R

@@ -3,7 +3,7 @@
 #' @export
 #' @param disease Specify which disease you would like to use.
 #' Currently supports "Cholera", "Measles", "Meningitis", "AJS",
-#' "Mortality", "Nutrition", "Vaccination_short" and "Vaccination_long".
+#' "Mortality", "Nutrition", "Vaccination_short", "Vaccination_long" and "EWAR".
 #'
 #' @param name The name of the dictionary stored in the package. The default
 #' will use dictionaries from the package. However you can also use
@@ -15,7 +15,7 @@
 #' @param varnames The name of column that contains variable names. The
 #' default set to "data_element_shortname".
 #' If `dictionary` is a survey ("Mortality", "Nutrition", "Vaccination_short"
-#' or "Vaccination_long") `varnames` needs to be "name"`. Otherwise if using
+#' "Vaccination_long" or "EWAR") `varnames` needs to be "name"`. Otherwise if using
 #' your own dictionary then specify.
 #'
 #' @param varnames_type The name of column that contains the variable type.
@@ -60,10 +60,11 @@ msf_dict_rename_helper <- function(disease,
   if (
     !tolower(disease) %in% c("cholera", "measles", "meningitis", "ajs",
                           "mortality", "nutrition", "vaccination_short",
-                          "vaccination_long") &
+                          "vaccination_long", "ewar") &
       template) {
         stop("disease must be one of `cholera`, `measles`, `meningitis`, `ajs`,
-           `mortality`, `nutrition`, `vaccination_short`, `vaccination_long`.
+           `mortality`, `nutrition`, `vaccination_short`, `vaccination_long`,
+           `ewar`.
            If using your own dictionary please set template to `FALSE`",
            call. = FALSE)
       }
@@ -85,7 +86,7 @@ msf_dict_rename_helper <- function(disease,
 
     # get msf disease specific survey data dictionary
     if (disease == "mortality" | disease == "nutrition" |
-        disease == "vaccination_short" | disease == "vaccination_long") {
+        disease == "vaccination_short" | disease == "vaccination_long" | disease == "ewar") {
 
       dat_dict <- msf_dict_survey(disease, compact = TRUE)
 

R/msf_dict_survey.R

@@ -22,7 +22,7 @@ msf_dict_survey <- function(disease, name = "MSF-survey-dict.xlsx",
     disease <- get_dictionary(disease, org = "MSF")$survey
 
     if (length(disease) == 0) {
-      stop("disease must be one of 'Mortality', 'Nutrition', 'Vaccination_long', 'Vaccination_short'", call. = FALSE)
+      stop("disease must be one of 'Mortality', 'Nutrition', 'Vaccination_long', 'Vaccination_short', 'EWAR'", call. = FALSE)
     }
     # get excel file path (need to specify the file name)
     path <- system.file("extdata", name, package = "epidict")
@@ -86,7 +86,7 @@ msf_dict_survey <- function(disease, name = "MSF-survey-dict.xlsx",
 
     outtie <- if (tibble) tibble::as_tibble(outtie) else outtie
 
-    # Return second option: a list with data dictionary and value options seperate
+    # Return second option: a list with data dictionary and value options separate
   } else {
     if (tibble) {
       dat_dict <- tibble::as_tibble(dat_dict)

R/utils.R

@@ -49,7 +49,7 @@ get_dictionary <- function(dictionary, org = "MSF") {
 
   # define which ones are outbreaks and which ones are survey datasets
   if (toupper(org) == "MSF") {
-    SURVEYS <- c("Mortality", "Nutrition", "Vaccination_long", "Vaccination_short")
+    SURVEYS <- c("Mortality", "Nutrition", "Vaccination_long", "Vaccination_short", "EWAR")
     OUTBREAKS <- c("Cholera", "Measles", "Meningitis", "AJS")
     # NOTE: For future collaborators, if you have other dictionaries you wish to
     #       add to this project, then you should place the names of your valid