eurostat.Rmd

---
title: "Eurostat R package"
output: html_notebook
editor_options: 
  chunk_output_type: inline
---
# Introduction
Eurostat data and `trelliscopej` package to explore the Eurostat dataset



```{r "setup"}
knitr::opts_chunk$set(cache = TRUE, echo = TRUE)

library(eurostat)
library(dplyr)
library(ggplot2)
library(tidyr)
library(rvest)
library(httr)
library(trelliscopejs)

options(scipen = 3)
```



```{r "view available health data"}

toc <- get_eurostat_toc()

toc %>% filter(grepl("hlth", code) & type == "dataset")

```

```{r "hospitalisation rate"}

## Extract data


data <- get_eurostat("hlth_co_disch2t", type = "label", time_format = "num") ## this downloads > 6.6 million rows of data

Hmisc::describe(data)

## Data is inpatients admission rate, by age, by region, for 150 reasons for admission, from 2000 - 2014
```

To plot all the time series potentially means `r 150 * 184 * 23` charts, which we can create with `trelliscopejs` (although this will take several hours). To illustrate we'll look at admissions for acloholic liver disease (takes several minutes).

```{r "trellis charts"}
data %>%
  filter(icd10 == "Alcoholic liver disease") %>%
  ggplot(aes(time, values)) +
  geom_line() 
  facet_trelliscope(~geo + age)
  


```