---
title: "COVID-19: Mortality, Vaccination, and Population"
output: 
  flexdashboard::flex_dashboard:
    storyboard: true
    social: menu
    source: embed
    css: ~/Desktop/Git/Personal/datamustflow/public/css/coder.min.a4f332213a21ce8eb521670c614470c58923aaaf385e2a73982c31dd7642decb.css
    theme: spacelab
---


```{r setup, include=FALSE}
# Load required packages
library(flexdashboard) # dashboard wrapping
library(readxl) # data import
library(stringr) # string manipulation
library(tidyverse) # data wrangling
library(tidygeocoder) # geocoding
library(crosstalk) # interactivity
library(highcharter) # charts
library(broom) #output of built-in functions cleanup
library(jsonlite) # json wrangling
library(geojsonR) # geojson wrangling
library(DT) # Table formatting
library(leaflet) # maps
library(leaflet.extras) # map tweaks
library(htmltools) # widgets
```

```{r, warning=FALSE, message=FALSE}
# Data import

## Vaccination
## Source: https://ourworldindata.org/grapher/cumulative-covid-vaccinations?tab=chart&stackMode=absolute&time=earliest..latest&region=World

`%notin%` <- Negate(`%in%`)

vaccination <- read.csv('cumulative-covid-vaccinations.csv') %>%
                mutate(Date = as.Date(Date),
                       country = case_when(country %in% c('England', 'Scotland', 'Wales', 'Northern Ireland', 'Gibraltar') ~ 'United Kingdom',
                                           TRUE ~ country)) %>%
                select(-Code) %>%
                filter(country %notin% c('European Union', 'World')) %>%
                group_by(country, Date) %>%
                summarise(total_vaccination = sum(total_vaccinations)) 

vaccination_latest <- vaccination %>%
                        group_by(country) %>%
                        filter(Date == max(Date))

## Mortality
## Source: https://coronavirus.jhu.edu/data/mortality

mortality <- read.csv('mortality.csv')

## Population
## Source: https://www.worldometers.info/world-population/population-by-country/

population <- read.csv('population.csv') %>%
                mutate(country = case_when(country == 'Czech Republic (Czechia)' ~ 'Czechia',
                                           TRUE ~ country))

## Combination
vaccination_mortality_population <- vaccination_latest %>%
                          left_join(mortality, by = 'country') %>%
                          left_join(population, by = 'country') %>%
                          mutate(vaccinated_per_100k_inhabitants = round((total_vaccination / population) * 100000, 2),
                                 deaths_per_100_inhabitants = round((deaths / population) * 100000, 2),
                                 confirmed_per_100_inhabitants = round((confirmed / population) * 100000, 2),
                                 vaccinated_per_deaths = round((total_vaccination / deaths) * 100000, 2),
                                 vaccinated_per_100K_confirmed = round((total_vaccination / confirmed) * 100000, 2),
                                 urban_pop = as.numeric(str_remove(urban_pop, ' %')))
```

### **The number of vaccinated and deaths caused by COVID-19.**

```{r}
vaccination_mortality_population %>%  
  hchart('scatter',
         hcaes(x = deaths_per_100_inhabitants,
               y = vaccinated_per_100k_inhabitants, 
               size = deaths_per_100k, 
               color = deaths_per_100k), 
         maxSize = "10%",  name = "Country")  %>%   
         hc_chart(zoomType = "xy") %>%
      hc_yAxis(type = "logarithmic", title = list(text = "Number of vaccinated per 100k inhabitants")) %>%
      hc_xAxis(type = "logarithmic", title = list(text = "Number of deaths caused by COVID-19 per 100k inhabitants")) %>%
      hc_tooltip(pointFormat = '{point.country}: {point.total_vaccination:,.0f} vaccinated, 
{point.population:,.0f} inhabitants
') %>%
      hc_chart(style = list(
                                      fontWeight = "bold",
                                      fontSize = 16)) %>%
      hc_add_theme(hc_theme_ffx())
```

***

**Data** *(2021-02-04)*

* [*Vaccination*](https://ourworldindata.org/grapher/cumulative-covid-vaccinations?tab=chart&stackMode=absolute&time=earliest..latest&region=World). `r nrow(vaccination)` observations (multiple records per 1 country). `r ncol(vaccination)` variables (*Country*, *Date*, *Number of vaccinated*).
* [*Mortality*](https://coronavirus.jhu.edu/data/mortality) - `r nrow(mortality)` countries. `r ncol(mortality)` variables (*Country* or *Number of confirmed cases*).
* [*Population*](https://www.worldometers.info/world-population/population-by-country/) - `r nrow(population)` countries. `r ncol(population)` variables (*Country* or *Population*).

**Chart**

*X-axis*. The number of deaths caused by COVID-19 per 100k inhabitants in the respective country (i.e. "mortality").

*Y-axis*. The number of vaccinated per 100k inhabitants in the respective country (i.e. "vaccination rate").

*Circle size* and *color*. The number of deaths caused by COVID-19 per 100k confirmed COVID-19 cases in the respective country. Higher numbers result in circles with larger diameters and brighter colors (i.e. "lethality").

*Zoom* by dragging a cursor, click on *Reset zoom* to reset the view.


### **The number of reported vaccines.**

```{r}
vaccination_mortality_population %>%  
  hchart(
    "treemap",
    hcaes(x = country, value = total_vaccination, color = urban_pop),
    ) %>%
  hc_chart(zoomType = "xy", style = list(
                                      fontWeight = "bold",
                                      fontSize = 16)) %>%
  hc_colorAxis(stops = color_stops(colors = viridis::viridis(10))) %>%
  hc_tooltip(useHTML = TRUE,pointFormat = "{point.name}

                             Number of reported vaccinations: {point.value:,.0f}

                             % of urban population: {point.urban_pop:,.0f}") %>%
                hc_add_theme(hc_theme_ffx()) %>%
  hc_legend(align = "center")
```

***

*How many vaccinations have been reported per country?*

*Can we assume that countries with higher share of population living in urban areas are more likely administer more vaccines?*

Note the rounding of reported vaccination numbers in countries like China *(i.e. `r format(vaccination_mortality_population$total_vaccination[vaccination_mortality_population$country == 'China'], big.mark = ' ')`)* or Russia *(i.e. `r format(vaccination_mortality_population$total_vaccination[vaccination_mortality_population$country == 'Russia'], big.mark = ' ')`)* and compare it with the rest.

**Chart**

*Box size*. The absolute number of reported vaccinations.

*Box color*. Share of population living in urban areas in the respective country. Higher numbers result in brighter colors of the boxes.

*Zoom* by dragging a cursor, click on *Reset zoom* to reset the view.


### **The number of days with recorded vaccinations.**

```{r}
# Load the world Map data
data(worldgeojson, package = "highcharter")

reporting_interval_map <- vaccination %>%
  mutate(interval=Date-lag(Date)) %>%
  group_by(country) %>%
  summarize(no_of_entries = n()) %>%
  mutate(country = case_when(country == 'Czechia' ~ 'Czech Republic',
                             country == 'United States' ~ 'United States of America',
                             TRUE ~ country))

hc <- highchart() %>%
  hc_add_series_map(
    worldgeojson, reporting_interval_map, value = "no_of_entries", joinBy = c('name','country'),
    name = "Number of days with recorded vaccinations"
    )  %>% 
  hc_chart(style = list(
                                      fontWeight = "bold",
                                      fontSize = 16)) %>%
  hc_colorAxis(stops = color_stops(colors = viridis::viridis(10))) %>% 
                hc_add_theme(hc_theme_ffx()) %>%
  hc_mapNavigation(enabled = TRUE) %>%
  hc_legend(align = "center") 

hc

```

***

```{r}
vaccination_first <- vaccination %>%
  group_by(country) %>%
  filter(Date == min(Date))
```


*For how many times have countries recorded the number of vaccinated so far?*

There are **`r length(unique(vaccination$country))` countries** with recorded vaccination numbers. The **first** vaccination was recorded for **`r vaccination$country[vaccination$Date == min(vaccination$Date)]`** on **`r min(vaccination$Date)`**. The **last initial** record is on **`r max(vaccination_first$Date)`** for **`r vaccination_first$country[vaccination_first$Date == max(vaccination_first$Date)]`**.

**Map**

*Area color*. The absolute value of days with reported number of vaccinated. Higher numbers result in brighter colors of the columns.

*Zoom in* by clicking on the "+" button in the upper left corner of the map, *Zoom out* by clicking on the "-" button there. Alternatively, use *scrolling*.


### **Frequency of reporting on vaccination.**

```{r}
reporting_interval <- vaccination %>%
  mutate(interval=Date-lag(Date)) %>%
  group_by(country) %>%
  summarize(mean_interval = mean(interval, na.rm = TRUE))

hchart(as.numeric(reporting_interval$mean_interval)) %>%
  hc_colorAxis(stops = color_stops(colors = viridis::viridis(10))) %>%
  hc_tooltip(borderWidth = 1, sort = TRUE, crosshairs = TRUE,
             headerFormat = "",
             pointFormatter = JS("function() {
  return 'Between values '  + (this.x - 0.5) + ' and ' + (this.x + 0.5) + '
 Count: ' +  this.y;    
               }")) %>%
  hc_yAxis(title = list(text = "Count")) %>%
  hc_xAxis(title = list(text = "Average difference between two reporting points in time (in days)")) %>%
  hc_chart(style = list(
                        fontWeight = "bold",
                        fontSize = 16)) %>%
  hc_add_theme(hc_theme_smpl()) %>%
  hc_legend(align = "center") %>%
  hc_add_theme(hc_theme_ffx())
```

***

```{r}
# Create the function.
getmode <- function(v) {
   uniqv <- unique(v)
   uniqv[which.max(tabulate(match(v, uniqv)))]
}
```


*How often countries report the numbers of vaccinated?*

50% of countries with at least two vaccination records inform each `r round(median(as.numeric(reporting_interval$mean_interval), na.rm = TRUE), 2)` day(s) and more frequently.

**Chart**

*Column size*. Interval of 5.

*Column color*. The absolute value of days with reported number of vaccinated. Higher numbers result in brighter colors of the columns.

*Zoom* by dragging a cursor, click on *Reset zoom* to reset your view.

**Created by**
Vít Gabrhel
https://www.data-must-flow.com/