Merge pull request #4 from opensafely/milanwiedemann/measures-suggestions

Suggestions for improving plotting measures

Author: viv3ckj

analysis/dataset_definition.py

@@ -0,0 +1,31 @@
+plot_measures <- function(
+    data,
+    title = NULL,
+    x_label = NULL,
+    y_label = NULL) {
+
+  # Create plot
+  plot <- ggplot(
+    data,
+    aes(
+      x = interval_end,
+      y = numerator,
+      color = measure,
+      group = measure
+    )
+  ) +
+    geom_line() +
+    labs(
+      title = title,
+      x = x_label,
+      y = y_label,
+      color = "Condition"
+    ) +
+    # Setting the minimum y-value
+    ylim(y_min, NA) +
+    # Applying the minimal theme
+    theme_minimal()
+
+  # Return plot
+  plot
+}

analysis/stata.do

@@ -1,20 +1,23 @@
+# In this file we're trying out some code
 library(tidyverse)
 library(ggplot2)
-library(gridExtra)  # For arranging multiple plots
+library(gridExtra) # For arranging multiple plots
 
 # Load data
 df_measures <- readr::read_csv(
   here::here("output", "report", "conditions_measures.csv")
 )
+
+# Develop plotting function 
 p <- ggplot(df_measures, aes(x = interval_end, y = numerator, color = measure, group = measure)) +
-    geom_line() +
-    labs(title = paste("Number of Consultations for each Pharmacy First Clinical Condition per month"),
-         x = "Date",
-         y = "Number of Consultations",
-         color = "Condition") +
-         ylim(0, NA)
-    theme_minimal()
-  
+  geom_line() +
+  labs(
+    title = paste("Number of Consultations for each Pharmacy First Clinical Condition per month"),
+    x = "Date",
+    y = "Number of Consultations",
+    color = "Condition"
+  ) +
+  ylim(0, NA)
+
 source(here::here("lib", "functions", "graph_func.R"))
 plot1_conditions <- plot_pharmacy_first_conditions(df_measures)
-  

lib/functions/function_plot_measures.R

@@ -5,12 +5,6 @@ expectations:
   population_size: 1000
 
 actions:
-  generate_dataset:
-    run: ehrql:v1 generate-dataset analysis/dataset_definition.py --output output/dataset.csv.gz
-    outputs:
-      highly_sensitive:
-        dataset: output/dataset.csv.gz
-
   generate_pf_measures:
     run: >
       ehrql:v1 generate-measures analysis/report_measures.py