Skip to content

This is a data processing pipeline for the NHS Mental Health Services Data Set

License

Notifications You must be signed in to change notification settings

HFAnalyticsLab/MHSDS-pipeline

 
 

Repository files navigation

Mental Health Services Dataset Pipeline

Project Status: [In progress]

Project Description

  • Each month, NHS Digital releases key statistics on the performance of mental health services in the NHS. Those are released as multiple spreadsheets each month, with a separate file for each type of measure (e.g. main performance stats, restrictive interventions, eating disorders, etc). See example below. Each sub-page contains multiple more links.

  • Finding all the right files going back months can be time-consuming. If you're interested in time series data for more than one type of measure, this isn't made publicly available (to my knowledge) and there isn't a publicly available API either.

  • To make things easier, I have put together a data pipeline that:

    • Web-scrapes the links to the CSVs you want, starting from the MHSDS landing page and working its way through the underlying pages for each monthly release
    • Downloads the required files into a local folder of your choice
    • Appends the CSVs into a single time-series file for each type of measure
  • By running this pipeline each month, the code provided here identifies new files hosted on the NHS Digital website and downloads those (and only those!) before appending them to the time series

  • This pipeline downloads two metrics for each monthly series:

    • The main performance indicators
    • Eating disorder indicators
  • It could be adpated to include more types of metrics (e.g. Mental Health Act metrics or restrictive interventions).

  • I have also provided some code that produces a simple dashboard powered by flexdashboard. You can adapt this at the end of your pipeline to build a dashboard with the mental health metrics of your choice.

How does it work?

  1. Edit 0. File locations.R with the location of the directory you would like to save your MHSDS files to. Start with an empty folder. For example, I have chosen 'rawdatadir' as "/Users/sgpeytrignet/Documents/MHSDS data/".
  2. Run 1. MHSDS-data-cleaning.R to download the desired MHSDS data locally, append them into a practical time-series files and do some basic data cleaning. This relies on a new function that fetches monthly series from the NHS Digital plage (e.g. MHSDS_monthly_series_download("december 2020")). If you are doing this for the first time (or if new data has been released and it's time to refresh your data), set the 'refresh_data' parameter to YES. If you want to leave your data as is, set it to NO.
  3. Now, you should be ready to start running some simple analysis or doing some visualizations, for example using a dashboard.
  4. If you would like to create your own dashboard, feel free to adapt the 2. MHSDS-dashboard.Rmd script!

Requirements

The following R packages (available on CRAN) are needed:

  • tidyverse
  • lubridate
  • here
  • data.table
  • DescTools
  • pbapply
  • rvest
  • downloader
  • curl

If you're interested in building your own dashboard, you'll also need the following packages:

  • plotly
  • hrbrthemes
  • flexdashboard

Useful references

If you would like a starter on building dashboards using flexdashboard, you can get started with this resource.

Authors

License

This project is licensed under the MIT License.

About

This is a data processing pipeline for the NHS Mental Health Services Data Set

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 100.0%