My project

Pressure on the National Health Service is one of the most important issues facing the UK but data showing how hospitals are coping with that pressure is hard for the public to find and interpret.
The NHS Tracker provides the audience with this information in a format that is easy to use and understand – and is tailored to their local area.
Users simply enter a postcode (zipcode) to check whether their nearest hospital trust is meeting three key waiting time targets: Accident and emergency, cancer care and planned operations such as hip replacements.
Because of the way the NHS is organised, we have four branches of the tracker – one each for England, Scotland, Wales and Northern Ireland. The audience sees information appropriate to the postcode they entered.
The results are headlined with a summary “scorecard” with checks and crosses, which shows at a glance whether the targets have been met. The content stacks on mobile.
This is followed by detailed interactive bar charts of all the hospital trusts in the relevant nation of the UK, showing how they are performing relative to each other and where the selected trust fits into the overall picture.
Important contextual information is also provided so that the audience can understand the wider picture. These are the national average for each target, a year-on-year comparison and the date the target was last met.
We were also able to delve deeper into the data we gathered for the tracker and show performance across the NHS had worsened across a four year period. This generated strong lead news lines for launch day.

What makes this project innovative?

The tracker is one of the largest data projects ever run by BBC News in terms of the number of sources. In order to check when each hospital trust last met its target we had to analyse up to nine years of data across 600 spreadsheets. Each time we publish the tracker - several times a month - we re-download past data to ensure we have captured all revisions. This functionality has been implemented in the data tool as R packages to enable us to automate the data extraction and processing in a reproducible manner.

What was the impact of your project? How did you measure it?

The tracker is proving popular with the audience, with 4.5 million page views to date, of which 1.5 million were on the day of publication.
The analysis work we carried out for the tracker spread far beyond the page itself - it generated lead broadcast news stories across the BBC, including the flagship national TV news programmes as well as regional news.
The British Health Secretary Jeremy Hunt admitted that the health service must improve its record on waiting times after our analysis revealed the extent of the targets being missed.
We provided regional cuts of the data for local BBC teams, meaning that broadcast audiences were served the story in a way that made sense for their location.

Source and methodology

We worked closely with the BBC’s Health Correspondent Nick Triggle to ensure we selected the appropriate data from the four national NHS websites (England, Wales, Scotland, Northern Ireland).
The four nations have different targets and publish the data in different formats meaning we needed to use a range of processes to arrive at the same result for each nation.
There are many measures we could have looked at but we chose to narrow the data down to A&E, cancer and planned operations.
For each nation we navigated to the relevant section of the website for that target and then made sure we were downloading the latest data.
This is not as easy as it sounds. There is such a wealth of NHS data that selecting the correct spreadsheet and columns requires specialist knowledge.
England data is now issued monthly but was issued weekly until 2015, so for older data we needed to devise a method of dealing with this and other inconsistencies in the data.
At every stage, we confirmed our process with data managers in the NHS.
Our data analyst wrote R packages to perform the ranking and year-on-year comparisons. These output a csv which we upload to update the tracker.
The journalist checks and updates the national figures and also confirms the data analyst’s work.

Technologies Used

Our data analyst wrote four R packages to perform the main analysis and journalists use Excel or R to check the outcomes.
The tracker is a bespoke interactive feature, built using Javascript and HTML.
Updates are published using our own data FTP system.

Project members

Ransome Mpini, John Walton, Nick Triggle, Sumi Senthinathan, William Dahlgreen, Chris Ashton, Evisa Terziu, Becky Rush, Alvin Ourrad.

Project owner administration

Contributor username

Followers

Click Follow to keep up with the evolution of this project:
you will receive a notification anytime the project leader updates the project page.