Project description

A&Es and ambulance services have been under growing pressure in recent years, and statistics play a vital role in monitoring how performance has deteriorated.

This story looked at one element of the interaction between ambulance services and hospitals. An increasing number of handovers from ambulance to A&E taking longer than the recommended 15 minutes is often a sign that A&Es are struggling, with patients already in departments facing waits for treatment or a lack of beds, which in itself may be the impact of lack of resources and high demand in other parts of the hospital.

In order to analyse how long individual handovers had taken I used Freedom of Information requests to build a database of all of the individual handovers, the priority of the call being answered and how long the handover took.

Analysis of this database revealed the huge increases in the number of ambulances facing long waits to handover patients to A&Es – a warning sign that highlights the increasing levels of pressure A&Es have been under in recent years. Working from call level data, meant stories could be produced for a number of papers, showing how local hospitals were faring and revealing the shocking and distressing long waits being faced by some patients.

What makes this project innovative?

This project is innovative in the scale of the investigation and the way in which I used Freedom of Information requests to gather the huge quantity of data needed to properly analyse the situation in terms of handovers at different hospitals. I used Freedom of Information requests - making dozens of requests over several years and battling public bodies that refused to hand over the data - to build a comprehensive database of nearly 30 million patient handovers by ambulances at Britain’s A&Es, covering the past five years. Making requests systematically, over a number of years and asking for raw data, allowed me to build a dataset that only exists because of the way in which I use the powers available.

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

The story was one of the most read items on several of the sites that published it, including the Liverpool Echo and WalesOnline. On such sites, the story was read thousands of times, with high engagement levels, shown by readers spending more than a minute on the stories. The stories were also shared nearly 1,000 times on Facebook, an excellent level of engagement for the type of content.

Source and methodology

Freedom of Information requests to public bodies across Britain, collated in spreadsheets. I kept careful records to keep track of the responses received, and ensure bodies that had yet to reply could be chased, as well as refusals that needed to be challenged. In order to get all of the data I wanted I had to fight several refusals, where public bodies did not want to supply all of the call level data but as I felt it Once received, information was then checked, with particular information that looked like outliers or that was reported in unusual formats queried with the public body. The information was then analysed in order to draw out national and local stories. Building the database meant sense-checking the raw data provided to exclude data where some or all of the information had not been entered by crews or where the data strongly appeared to have been wrongly entered (where dates had been entered wrong meaning waits of several days, or where the patient was handed over before the ambulance arrived at the A&E). Comparing the proportions of over 30 minute and over 1 hour waits to the limited data published by NHS England on winter ambulance handovers also helped as an accuracy check. This was to ensure that the statistics better reflected actual waits experienced by patients, rather than poorly recorded data.

Technologies Used

The data is collated and analysed in spreadsheets (either OpenOffice Calc or Excel). Stories were then sent out using the Data Unit's bulletin system. The bulletin system is based on stories written in Google Docs, which are then sent out via Gmail using a script that collects the data from the Google Doc and creates emails that are sent out based on a spreadsheet of contact details.

Project members

Paul Whitelam, Mark Smith, Damon Wilkinson, Kelly Williams, Harry, Howard, Hannah Robinson, Katy Hallam


Additional links


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