ggplot_efficient.qmd

---
title: "HealthyR demo: efficient ggplotting"
format: html
editor: visual
execute:
  echo: true
  warning: false
editor_options: 
  chunk_output_type: console
---

# Dataset

We are using the same dataset as our last few demos - licorice gargle RCT.

```{r}
#| echo: false
library(tidyverse)

# Load dataset
licodata_raw = medicaldata::licorice_gargle

# reading in 0,1, recoding
licodata = licodata_raw %>% 
  mutate(preOp_gender.factor = preOp_gender %>% 
           factor() %>% 
           fct_recode("Male" = "0",
                      "Female" = "1"),
         preOp_asa.factor = preOp_asa %>% 
           factor() %>% 
           fct_recode("a normal healthy patient" = "1",
                      "a patient with mild systemic disease" = "2",
                      "a patient with severe systemic disease" = "3"),
         treat.factor = treat %>% 
           factor() %>% 
           fct_recode("Sugar 5g" = "0",
                      "Licorice 0.5g" = "1"),
         preOp_smoking.factor = preOp_smoking %>% 
           factor() %>% 
           fct_recode("Current" = "1",
                      "Past"    = "2",
                      "Never"   = "3")
  )
```

# Plot of smokers

```{r}
barplot_count = function(df, var, xlab = ""){
  df %>% 
    ggplot(aes(x = {{var}} %>% 
                 fct_infreq() %>% 
                 fct_rev())) + 
    geom_bar(fill = "seagreen3",
             colour = "seagreen4") + 
    coord_flip() + 
    theme_classic() + 
    geom_label(aes(label = after_stat(count / sum(count)) %>% scales::percent(1)),
               stat = "count",
               hjust = 1,
               size = 5) + 
    ylab("N Patients") + 
    xlab(xlab) + 
    scale_y_continuous(expand = c(0, 0))
}

```

```{r}

barplot_count(licodata, preOp_smoking.factor, "Smoking")

barplot_count(licodata, treat.factor, "Treatment") 

barplot_count(licodata, preOp_asa.factor, "ASA score")

barplot_count(licodata, preOp_gender.factor, "Gender")

```