Developing the foundations of applied analytics: using a collaborative approach to model ‘black days’ Hywel Dda University Local Health Board
- Run by Hywel Dda University Local Health Board.
- Aimed to strengthen the analytical skills and awareness at Hywel Dda University Local Health Board, and improve forecasting of patient demand to reduce ‘black days’.
- Significantly increased analytical capabilities and implemented processes to allow for data-mining and rapid exploration of datasets to inform future service reconfiguration.
- Ran from August 2019 to November 2020.
‘Black alerts’ in hospitals are when the predicted demand for services exceeds the available capacity. Improving forecasting of patient demand can help to reduce the number of ‘black days’ that occur. However, carrying out this forecasting requires advanced analytical skills.
This project aimed to strengthen analytical capabilities at Hywel Dda University Local Health Board by introducing applied techniques (including machine learning and data-mining), with a focus on using analytical skills and staff experience to describe and understand ‘black days’.
The Health Analytics Team presented examples (and real output) of applied analytics to increase staff awareness and help senior decision makers appreciate the importance of using these data to solve both complex and day-to-day problems at the health board.
The cross-industry standard process for data-mining methodology (CRISP-DM) was used to develop a Systematic Healthcare Exploratory Resource for Locating and Observing Clusters (SHERLOC), along with a framework for future cluster analysis and an end-to-end tool for rapid exploration of datasets.
These products form an integral part of the organisation’s analytical toolbox by providing an effective way to visualise clinical problems, develop hypotheses and systematically explore data to identify differences and similarities between groups.
Despite setbacks relating to COVID-19, and the time consuming and unpredictable nature of learning clustering methods, the analytical capability, skills and awareness of health board staff have dramatically increased.
Importantly, links between alcohol use, a specific age group and increased emergency department use at certain times were identified. Data such as these will feed into future demand forecasting work to develop early warning systems and inform service reconfiguration to ensure adequate capacity for improved patient care.
Contact information
For more information about this project, please contact Sebastian Neale, Lead Specialist Senior Analyst, Hywel Dda University Health Board.
Work with us
We look for talented and passionate individuals as everyone at the Health Foundation has an important role to play.
View current vacanciesThe Q community
Q is an initiative connecting people with improvement expertise across the UK.
Find out more