Business intelligence: handling big data in emergency care

The Leeds Teaching Hospitals NHS Trust (LTH) treats 200,000 patients every year in their Emergency Department. It is essential that every service, including the Emergency Department, is performing to the best of its capabilities.

Over the last seven years, with the use of EMIS Health’s unscheduled care application Symphony, LTH had collected a large amount of clinical and patient data from the Emergency Department but did not have the resources to analyse it in a timely manner. In order to deal with an increasing number of requests for up to date data, a system to review the stored data was required.

The big data challenge

In order to meet these demands, EMIS Health’s analytics team approached LTH to take part in a pilot project to transform this data into meaningful reports. In conjunction with Microsoft, Intel and Two10degrees; EMIS Health used Natural Language Programming (NLP) to develop a system to search through the free text fields of patient and clinical records for keywords, and to quantify the data to a structured output.

Tailored to the trust

The NLP system can be configured to understand individual trust’s abbreviations in order to incorporate as much data as possible in the reports.

“One of the challenges is that NLP is labour intensive, it takes a lot of human hours to ‘teach’ the NLP, but it’s absolutely worth it.”

Iain MacBrairdy, Business Manager of Urgent Care at Leeds

The NLP system can be used effectively for a number of different purposes, including:

  • identifying key trends from historic data sets to improve resource allocation
  • monitoring the impact of new services by measuring the frequency the service is mentioned in patient records
  • improving the billing and income processes by recognising improperly recorded investigations.

Improving patient care

By using EMIS Health’s Business Intelligence, Leeds Teaching hospitals were able to simplify the process of data requests, saving time and money which can be repurposed to directly improve patient care.

“Big Data and NLP has delivered many benefits to the ED, but the most important thing is improving patient care.”

Iain MacBrairdy, Business Manager of Urgent Care at Leeds

Natural Language Programming at Leeds Teaching Hospitals NHS Trust