search

epiforecasts / covid19.nhs.data

Trust-level COVID-19 hospitalisations in England
https://epiforecasts.io/covid19.nhs.data
Other
8 stars 3 forks source link
covid19 hospitalisations local-authority nhs openscience r-package uk

readme

NHS trust level Covid-19 data aggregated to a range of spatial scales

Lifecycle: experimental R build status Codecov test coverage DOI

This package contains a many-to-many mapping between local authority districts and NHS Acute Trusts in England; details of this mapping (including a summary of the methods and a quick-start guide) can be found in vignettes/mapping-summary.

This package also has functionality to download trust-level hospital admissions data, published weekly on the NHS COVID-19 Hospital Activity webpage. Data published on date YYYY-MM-DD can be downloaded using the function get_admissions(release_date = "YYYY-MM-DD"). This function can also be used to return estimated admissions by upper-tier and lower-tier local authorities. See the quick start below, the vignettes, and the package documentation for more.

Installation

Install the stable development version of the package from our r-universe:

install.packages(
  "covid19.nhs.data",
  repos = c(ropensci = 'https://epiforecasts.r-universe.dev',
            CRAN = 'https://cloud.r-project.org')
)

Or from GitHub:

remotes::install_github("epiforecasts/covid19.nhs.data")

Quick start

Load the package.

library(covid19.nhs.data)

Download the latest admissions mapped to lower-tier local authority (LTLA) using the default mapping. Note: This data is updated weekly each Thursday and the mapping is a probabilistic estimate.

adm <- get_admissions("ltla")

Map the latest available estimates by LTLA using one of the built in package shapefiles.

map_admissions(adm, england_ltla_shape)

Plot the time series of estimated admissions in an example LTLA (here Derby).

library(ggplot2)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union

adm %>% 
  filter(geo_name %in% "Derby") %>% 
  ggplot(aes(x = date, y = admissions)) + 
  geom_col(width = 0.9, col = "grey50", fill = "grey85") +
  theme_minimal() +
  labs(x = "Date", y = "Daily Hospital Admissions",
       title = "Covid-19 Admissions in Derby", 
       subtitle = "Estimated using a probabilistic mapping from NHS Trusts to lower-tier local authority level")