Reworking code, file names, and folder structure

.gitignore

@@ -9,4 +9,4 @@ venv/
 .DS_Store
 .Rhistory
 .Rproj.user/
-reports/pf_report.html
+reports/pharmacy_first_report.html

analysis/report_measures.py → analysis/measures_definition_pf_codes_conditions.py

@@ -1,10 +1,13 @@
 from ehrql import INTERVAL, create_measures, months, codelist_from_csv
-from ehrql.tables.tpp import clinical_events, patients, practice_registrations
+from ehrql.tables.tpp import clinical_events, practice_registrations
 
 measures = create_measures()
 measures.configure_dummy_data(population_size=1000)
 
-# Dictionary of pharmacy first codes
+start_date = "2023-11-01"
+monthly_intervals = 8
+
+# Create dictionary of pharmacy first codes
 pharmacy_first_event_codes = {
     # Community Pharmacy (CP) Blood Pressure (BP) Check Service (procedure)
     "blood_pressure_service": ["1659111000000107"],
@@ -16,27 +19,21 @@
     "pharmacy_first_service": ["983341000000102"],
 }
 
-# Import the codelist from CSV
-pharmacy_first_codelist = codelist_from_csv(
+# Import pharmacy first conditions codelist
+pharmacy_first_conditions_codelist = codelist_from_csv(
     "codelists/user-chriswood-pharmacy-first-clinical-pathway-conditions.csv",
-    column="code", category_column = "term"
+    column="code",
+    category_column="term",
 )
 
-pharmacy_first_conditions_codes = {}
-# Iterate through codelist, forming a dictionary 
-for codes, term in pharmacy_first_codelist.items():
-    normalised_term = term.lower().replace(" ", "_")
-    codes = [codes]
-    pharmacy_first_conditions_codes[normalised_term] = codes
-
 registration = practice_registrations.for_patient_on(INTERVAL.end_date)
 
 # Select clinical events in interval date range
 selected_events = clinical_events.where(
     clinical_events.date.is_on_or_between(INTERVAL.start_date, INTERVAL.end_date)
 )
 
-# Loop through each CLINICAL SERVICE to create a measure
+# Create measures for pharmacy first services
 for pharmacy_first_event, codelist in pharmacy_first_event_codes.items():
     condition_events = selected_events.where(
         clinical_events.snomedct_code.is_in(codelist)
@@ -52,10 +49,16 @@
         name=f"count_{pharmacy_first_event}",
         numerator=numerator,
         denominator=denominator,
-        intervals=months(8).starting_on("2023-11-01")
+        intervals=months(monthly_intervals).starting_on(start_date),
     )
 
-# Loop through each CLINICAL CONDITION to create a measure
+# Create measures for pharmacy first conditions
+pharmacy_first_conditions_codes = {}
+for codes, term in pharmacy_first_conditions_codelist.items():
+    normalised_term = term.lower().replace(" ", "_")
+    codes = [codes]
+    pharmacy_first_conditions_codes[normalised_term] = codes
+
 for condition_name, condition_code in pharmacy_first_conditions_codes.items():
     condition_events = selected_events.where(
         clinical_events.snomedct_code.is_in(condition_code)
@@ -71,5 +74,5 @@
         name=f"count_{condition_name}",
         numerator=numerator,
         denominator=denominator,
-        intervals=months(8).starting_on("2023-11-01")
-    )
+        intervals=months(monthly_intervals).starting_on(start_date),
+    )

lib/functions/function_plot_measures.R

@@ -1,21 +1,19 @@
 #' Plot Measures Over Time
-#' 
+#'
 #' Creates a line plot of measures over time, with customisable labels and colours.
-#' 
+#'
 #' @param data A dataframe containing the data to plot.
 #' @param measure_names Strings specifiying the names of measure columns to be plotted.
-#' @param custom_labels Strings specifying the names of legend labels. 
-#' @param title A string specifying the title of the plot. Default is NULL. 
+#' @param custom_labels Strings specifying the names of legend labels.
+#' @param title A string specifying the title of the plot. Default is NULL.
 #' @param x_label A string specifying the label for the x-axis. Default is NULL.
-#' @param y_label A string specifying the label for the y-axis. Default is NULL. 
+#' @param y_label A string specifying the label for the y-axis. Default is NULL.
 #' @param color_label A string specifying the label for the color legend. Default is NULL.
 #' @param value_col The name of the dataframe column which contains the y-axis values. Default is "numerator".
 #' @param measure_col The name of the dataframe column which contains the categorical variable. Default is "measure".
-#' 
-#' 
+#'
 #' @return A ggplot object.
 
-# Define the function
 plot_measures <- function(
     data,
     measure_names,
@@ -28,7 +26,6 @@ plot_measures <- function(
     y_label = NULL,
     color_label = NULL,
     axis_x_text_size = 7) {
-
   # Check if the necessary columns exist in the data
   if (date_col %in% names(data) == FALSE) {
     stop("Data does not have a column with the name '", date_col, "'")
@@ -80,24 +77,27 @@ plot_measures <- function(
       limits = c(0, NA),
     ) +
     theme_minimal() +
-    theme(axis.text.x = element_text(size = axis_x_text_size), 
-    legend.position="bottom",
-    legend.key.size = unit(0.5, "cm"),
-    legend.text = element_text(size = 8),
-    legend.title = element_text(size = 8)) +
+    theme(
+      axis.text.x = element_text(size = axis_x_text_size),
+      legend.position = "bottom",
+      legend.key.size = unit(0.5, "cm"),
+      legend.text = element_text(size = 8),
+      legend.title = element_text(size = 8)
+    ) +
+    # Adjust number of rows in the legend
     guides(
-    color = guide_legend(nrow = 2)  # Adjust number of rows in the legend
-) +
+      color = guide_legend(nrow = 2)
+    ) +
     geom_vline(
       xintercept = lubridate::as_date("2024-02-01"),
       linetype = "dotted",
       colour = "orange",
-      linewidth = .7) +
+      linewidth = .7
+    ) +
     scale_x_date(
-      date_breaks = "1 month",  
+      date_breaks = "1 month",
       date_labels = "%b %Y"
     )
 
-
   plot1
-}
+}

project.yaml

@@ -7,8 +7,8 @@ expectations:
 actions:
   generate_pf_measures:
     run: >
-      ehrql:v1 generate-measures analysis/report_measures.py
-      --output output/report/conditions_measures.csv
+      ehrql:v1 generate-measures analysis/measures_definition_pf_codes_conditions.py
+      --output output/measures/pf_codes_conditions_measures.csv
     outputs:
       moderately_sensitive:
-        measure: output/report/conditions_measures.csv  
+        measure: output/measures/pf_codes_conditions_measures.csv