Rename objects (#35)

Introduce consistent object names for snomed, ICD-10 and OPSC-4 codes
---------

Co-authored-by: Milan Wiedemann <[email protected]>

R/data.R

@@ -9,7 +9,7 @@ NULL
 #' \describe{
 #'   \item{start_date}{Start date of code usage count}
 #'   \item{end_date}{End date of code usage count}
-#'   \item{snomed_concept_id}{SNOMED Concept ID}
+#'   \item{snomed_code}{SNOMED Concept ID}
 #'   \item{usage}{Yearly summary of code usage.
 #'   Note that counts are rounded to the nearest 10.
 #'   Counts of 5 or below are displayed as 5.}
@@ -35,6 +35,9 @@ NULL
 #' Yearly summary of 4-character ICD-10 code usage from 1st April 2013 to 31st March 2024.
 #' The code usage represents the annual count of all episodes which record the given ICD-10 code in any primary or secondary position.
 #' Restricted codes for which annual usage is not published have been removed.
+#' Yearly summary of 4-character ICD-10 code usage from 1st April 2013 to 31st March 2024.
+#' The code usage represents the annual count of all episodes which record the given ICD-10 code in any primary or secondary position.
+#' Restricted codes for which annual usage is not published have been removed.
 #' @format A data frame with 135,951 rows and 5 columns:
 #' \describe{
 #'   \item{start_date}{Start date of code usage count}
@@ -43,6 +46,8 @@ NULL
 #'   Note that the punctuation from the code has been removed for compatibility with OpenCodelists.}
 #'   \item{usage}{Annual count of code usage.}
 #'   \item{description}{Description of the ICD-10 Code}
+#'   \item{usage}{Annual count of code usage.}
+#'   \item{description}{Description of the ICD-10 Code}
 #' }
 #' @source <https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity>
 #' @examples
@@ -59,25 +64,39 @@ NULL
 #' "https://www.opencodelists.org/codelist/opensafely/pregnancy-icd10-aurum/5a7d8d12/download.csv")
 #' icd10_usage |>
 #' dplyr:: filter(icd10_code %in% codelist$code) 
+#' @examples
+#' # Filter to codes in the ICD-10 Chapter XIX: "Injury, poisoning..." 
+#' # (codes begin with letters "S" or "T"), with usage > 10,000.
+#' # For each of these, select the year with the highest count. 
+#' icd10_usage |>
+#' dplyr:: filter(grepl("^[ST]", icd10_code) & usage > 10000)|>
+#' dplyr:: group_by(description) |>
+#' dplyr:: slice_max(usage)
+#' # Filter to codes present in the CPRD Aurum ICD-10 pregnancy codelist.
+#' # This codelist is available in OpenCodelists.org
+#' codelist<- read.csv(
+#' "https://www.opencodelists.org/codelist/opensafely/pregnancy-icd10-aurum/5a7d8d12/download.csv")
+#' icd10_usage |>
+#' dplyr:: filter(icd10_code %in% codelist$code) 
 "icd10_usage"
 
-#' Yearly OPCS Code Usage from Hospital Admitted Patient Care Activity in England
+#' Yearly OPCS-4 Code Usage from Hospital Admitted Patient Care Activity in England
 #'
-#' Yearly summary of 4-character OPCS code usage from 1st April 2013 to 31st March 2024.
+#' Yearly summary of 4-character OPCS-4 code usage from 1st April 2013 to 31st March 2024.
 #' The code usage represents the total annual count of each procedure, recorded across the primary and the secondary procedure positions.
 #' Restricted codes for which annual usage is not published have been removed.
 #' @format A data frame with 107,376 rows and 5 columns:
 #' \describe{
 #'   \item{start_date}{Start date of code usage count}
 #'   \item{end_date}{End date of code usage count}
-#'   \item{opcs_code}{The 4-character OPCS code.
+#'   \item{opcs4_code}{The 4-character OPCS-4 code.
 #'   Note that the punctuation from the code has been removed for compatibility with OpenCodelists.}
 #'   \item{usage}{Annual count of code usage.}
-#'   \item{description}{Description of the OPCS Code}
+#'   \item{description}{Description of the OPCS-4 Code}
 #' }
 #' @source <https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity>
 #' @examples
 #' # Filter to procedures involving "biopsy" after March 2020 (note each year runs April - March).
-#' opcs_usage |>
+#' opcs4_usage |>
 #' dplyr:: filter(grepl("biopsy", description, ignore.case = TRUE) & lubridate:: year(end_date) > 2020)
-"opcs_usage"
+"opcs4_usage"

R/server.R

@@ -30,13 +30,13 @@ app_server <- function(input, output, session) {
   selected_data <- reactive({
     if (input$dataset == "snomedct") {
       codeusage::snomed_usage |>
-        select(start_date, end_date, code = snomed_concept_id, description, usage)
+        select(start_date, end_date, code = snomed_code, description, usage)
     } else if (input$dataset == "icd10") {
       codeusage::icd10_usage |>
         select(start_date, end_date, code = icd10_code, description, usage)
     } else if (input$dataset == "opcs4") {
-      codeusage::opcs_usage |>
-        select(start_date, end_date, code = opcs_code, description, usage)
+      codeusage::opcs4_usage |>
+        select(start_date, end_date, code = opcs4_code, description, usage)
     }
   })
 

R/ui.R

@@ -22,7 +22,7 @@ app_ui <- function(request) {
               "Select dataset",
               bs_icon("info-circle")
             ),
-            "SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms); ICD (International Classification of Diseases); OPCS Classification of Interventions and Procedures",
+            "SNOMED CT (Systematized Nomenclature of Medicine Clinical Terms); ICD-10 (International Classification of Diseases); OPCS-4 Classification of Interventions and Procedures",
             options = list(
               customClass = "left-align-tooltip"
             )
@@ -32,10 +32,10 @@ app_ui <- function(request) {
           choices = c(
             # Systematized Nomenclature of Medicine Clinical Terms (SNOMED CT)
             "SNOMED-CT" = "snomedct",
-            # International Classification of Diseases (ICD)
+            # International Classification of Diseases (ICD-10)
             "ICD-10" = "icd10",
-            # OPCS Classification of Interventions and Procedures (OPCS)
-            "OPCS" = "opcs4"
+            # OPCS-4 Classification of Interventions and Procedures (OPCS-4)
+            "OPCS-4" = "opcs4"
           )
         )
       ),

R/zzz.R

@@ -6,11 +6,11 @@ utils::globalVariables(c(
   "annual_proportion",
   "start_date",
   "end_date",
-  "snomed_concept_id",
+  "snomed_code",
   "description",
   "usage",
   "icd10_code",
-  "opcs_code",
+  "opcs4_code",
   "total_usage",
   "full_slug"
 ))

README.Rmd

@@ -21,12 +21,12 @@ knitr::opts_chunk$set(
 
 <!-- badges: end -->
 
-The goal of `codeusage` is to make yearly summaries of **SNOMED Code Usage in Primary Care** and **ICD-10 and OPCS Code Usage in Secondary Care** in England, published by NHS Digital, available in R for research.
+The goal of `codeusage` is to make yearly summaries of **SNOMED Code Usage in Primary Care** and **ICD-10 and OPCS-4 Code Usage in Secondary Care** in England, published by NHS Digital, available in R for research.
 The interactive [Code Usage Explorer](https://milanwiedemann.shinyapps.io/codeusage/) dashboard provides different options to explore these datasets.  
 The original data is available from NHS Digital at:
 
 -   [SNOMED Code Usage in Primary Care](https://digital.nhs.uk/data-and-information/publications/statistical/mi-snomed-code-usage-in-primary-care)
--   [ICD-10 and OPCS Code Usage in Secondary Care](https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity)
+-   [ICD-10 and OPCS-4 Code Usage in Secondary Care](https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity)
 
 ## Installation
 
@@ -63,13 +63,13 @@ This is only a selection of the full dataset published by NHS Digital, for the d
 icd10_usage
 ```
 
-### Dataset: OPCS Code Usage in Secondary Care in England
+### Dataset: OPCS-4 Code Usage in Secondary Care in England
 
-This is the total annual count of each instance that each 4-character OPCS code is listed across all primary and secondary procedure positions in the Finished Consultant Episodes (FCE) of the Hospital Episode Statistics in England.
+This is the total annual count of each instance that each 4-character OPCS-4 code is listed across all primary and secondary procedure positions in the Finished Consultant Episodes (FCE) of the Hospital Episode Statistics in England.
 
-This is only a selection of the full dataset published by NHS Digital, for the data pre-processing see `/data-raw/opcs_usage.R`.
+This is only a selection of the full dataset published by NHS Digital, for the data pre-processing see `/data-raw/opcs4_usage.R`.
 
-```{r opcs-usage}
-# Return OPCS code usage data
-opcs_usage
+```{r opcs-4-usage}
+# Return OPCS-4 code usage data
+opcs4_usage
 ```

README.md

@@ -10,7 +10,7 @@
 <!-- badges: end -->
 
 The goal of `codeusage` is to make yearly summaries of **SNOMED Code
-Usage in Primary Care** and **ICD-10 and OPCS Code Usage in Secondary
+Usage in Primary Care** and **ICD-10 and OPCS-4 Code Usage in Secondary
 Care** in England, published by NHS Digital, available in R for
 research. The interactive [Code Usage
 Explorer](https://milanwiedemann.shinyapps.io/codeusage/) dashboard
@@ -19,7 +19,7 @@ The original data is available from NHS Digital at:
 
 - [SNOMED Code Usage in Primary
   Care](https://digital.nhs.uk/data-and-information/publications/statistical/mi-snomed-code-usage-in-primary-care)
-- [ICD-10 and OPCS Code Usage in Secondary
+- [ICD-10 and OPCS-4 Code Usage in Secondary
   Care](https://digital.nhs.uk/data-and-information/publications/statistical/hospital-admitted-patient-care-activity)
 
 ## Installation
@@ -46,7 +46,7 @@ for the data pre-processing see `/data-raw/snomed_code_usage.R`.
 # Return SNOMED code usage data
 snomed_usage
 #> # A tibble: 1,523,967 × 7
-#>    start_date end_date   snomed_concept_id description     usage active_at_start
+#>    start_date end_date   snomed_code description     usage active_at_start
 #>    <date>     <date>     <chr>             <chr>           <int> <lgl>          
 #>  1 2023-08-01 2024-07-31 279991000000102   Short message… 4.41e8 TRUE           
 #>  2 2023-08-01 2024-07-31 184103008         Patient telep… 1.91e8 TRUE           
@@ -90,21 +90,21 @@ icd10_usage
 #> # ℹ 135,941 more rows
 ```
 
-### Dataset: OPCS Code Usage in Secondary Care in England
+### Dataset: OPCS-4 Code Usage in Secondary Care in England
 
 This is the total annual count of each instance that each 4-character
-OPCS code is listed across all primary and secondary procedure positions
+OPCS-4 code is listed across all primary and secondary procedure positions
 in the Finished Consultant Episodes (FCE) of the Hospital Episode
 Statistics in England.
 
 This is only a selection of the full dataset published by NHS Digital,
-for the data pre-processing see `/data-raw/opcs_usage.R`.
+for the data pre-processing see `/data-raw/opcs4_usage.R`.
 
 ``` r
-# Return OPCS code usage data
-opcs_usage
+# Return OPCS-4 code usage data
+opcs4_usage
 #> # A tibble: 107,376 × 5
-#>    start_date end_date   opcs_code description                             usage
+#>    start_date end_date   opcs4_code description                             usage
 #>    <date>     <date>     <chr>     <chr>                                   <int>
 #>  1 2023-04-01 2024-03-31 A011      Hemispherectomy                             7
 #>  2 2023-04-01 2024-03-31 A012      Total lobectomy of brain                   36

_pkgdown.yml

@@ -7,7 +7,7 @@ reference:
   contents:
   - snomed_usage
   - icd10_usage
-  - opcs_usage
+  - opcs4_usage
 - title: Codelists
   contents:
   - get_codelist

data-raw/opcs_code_usage.R → data-raw/opcs4_code_usage.R

@@ -5,7 +5,7 @@ library(httr)
 
 url_start <- "https://files.digital.nhs.uk/"
 
-opcs_code_usage_urls <- list(
+opcs4_code_usage_urls <- list(
   "fy23to24" = list(
     url = paste0(url_start, "92/DB66C9/hosp-epis-stat-admi-proc-2023-24-tab-v2.xlsx"),
     sheet = 6,
@@ -81,7 +81,7 @@ opcs_code_usage_urls <- list(
 )
 
 # Function to download and read the xlsx files
-read_opcs_usage_xlsx_from_url <- function(url_list, ...) {
+read_opcs4_usage_xlsx_from_url <- function(url_list, ...) {
   temp_file <- tempfile(fileext = ".xlsx")
   GET(
     url_list$url,
@@ -101,7 +101,7 @@ read_opcs_usage_xlsx_from_url <- function(url_list, ...) {
 select_all_diag_counts <- function(data, url_list) {
   dplyr::select(
     data,
-    opcs_code = 1,
+    opcs4_code = 1,
     description = 2,
     usage = url_list$usage_col
   ) |>
@@ -111,13 +111,13 @@ select_all_diag_counts <- function(data, url_list) {
 }
 
 # Combine both functions
-get_opcs_data <- function(url_list, ...) {
-  df_temp <- read_opcs_usage_xlsx_from_url(url_list, ...)
+get_opcs4_data <- function(url_list, ...) {
+  df_temp <- read_opcs4_usage_xlsx_from_url(url_list, ...)
   select_all_diag_counts(df_temp, url_list)
 }
 
-opcs_usage <- opcs_code_usage_urls |>
-  map(get_opcs_data) |>
+opcs4_usage <- opcs4_code_usage_urls |>
+  map(get_opcs4_data) |>
   bind_rows(.id = "nhs_fy") |>
   separate(nhs_fy, c("start_date", "end_date"), "to") |>
   mutate(
@@ -127,12 +127,12 @@ opcs_usage <- opcs_code_usage_urls |>
     end_date = as.Date(
       paste0("20", str_extract_all(end_date, "\\d+"), "-03-31")
     ),
-    opcs_code = str_replace_all(opcs_code, "\\.", "")
+    opcs4_code = str_replace_all(opcs4_code, "\\.", "")
   ) |>
   filter(!is.na(usage))
 
 usethis::use_data(
-  opcs_usage,
+  opcs4_usage,
   compress = "bzip2",
   overwrite = TRUE
 )

data-raw/snomed_code_usage.R

@@ -28,19 +28,19 @@ snomed_code_usage_urls <- list(
 # Data dictionary from SNOMED_code_usage_metadata.xlsx
 # https://files.digital.nhs.uk/31/097702/SNOMED_code_usage_metadata.xlsx
 
-# * SNOMED_Concept_ID (Text string of digits up to 18 characters long)
+# * snomed_code (Text string of digits up to 18 characters long)
 #   SNOMED concepts which have been added to a patient record in a general practice system during the reporting period.
 # * Description (Text string)
-#   The fully specified name associated with the SNOMED_Concept_ID on the final day of the reporting period (31 July).
+#   The fully specified name associated with the snomed_code on the final day of the reporting period (31 July).
 # * Usage (Numeric (integer) or *)
-#   The number of times that the SNOMED_Concept_ID was added into any patient record within the reporting period, rounded to the nearerst 10.
+#   The number of times that the snomed_code was added into any patient record within the reporting period, rounded to the nearerst 10.
 #   Usage of 1 to 4 is displayed as *.
 # * Active_at_Start
-#   Active status of the SNOMED_Concept_ID on the first day of the reporting period.
+#   Active status of the snomed_code on the first day of the reporting period.
 #   This is taken from the most recent UK clinical extension, or associated International extention, which was published up to the start of the reporting year (1 August).
 #   1 = SNOMED concept was published and was active (active = 1).
 #   0 = SNOMED concept was either not yet available or was inactive (active = 0).
-# * Active_at_End	"Active status of the SNOMED_Concept_ID on the first day of the reporting period.
+# * Active_at_End	"Active status of the snomed_code on the first day of the reporting period.
 #   This is taken from the most recent UK clinical extension, or associated International extention, which was published up to the end of the reporting year (31 July).
 #   1 = SNOMED concept was published and was active (active = 1).
 #   0 = SNOMED concept was either not yet available or was inactive (active = 0).
@@ -49,7 +49,7 @@ snomed_code_usage_urls <- list(
 snomed_usage <- snomed_code_usage_urls %>%
   map(read_tsv,
     col_types = list(
-      SNOMED_Concept_ID = "c",
+      snomed_code = "c",
       Description = "c",
       Usage = "i",
       Active_at_Start = "l",
@@ -65,6 +65,11 @@ snomed_usage <- snomed_code_usage_urls %>%
     usage = replace_na(usage, 5)
   )
 
+# Manipulation required due to variable name change
+snomed_usage <- snomed_usage |>
+  rename(snomed_code = snomed_concept_id)|>
+  mutate(snomed_code = as.character(snomed_code))
+
 usethis::use_data(
   snomed_usage,
   compress = "bzip2",