Retaining the skills of junior doctors through AI and machine learning
Transform’s Director of Data Engineering and Technology, Cyril Law, and Senior Data Analyst, Patrick Greenway, discuss about building machine learning predictive models for Health Education England through the latest data and artificial intelligence technologies.
Recruiting the nation’s best healthcare workers is just the beginning for Health Education England. Demand is growing, current recruitment is insufficient, training is expensive, even more so when individuals drop out. The institution trains employees for seven years, investing as much as £250,000 into the careers of junior doctors, that’s why reducing turnover is critical.
To support that Transform and Health Education England are using data and advanced analytics to deliver breakthrough insights, building confidence in AI to tackle workforce challenges cost-effectively. After understanding drivers of attrition through surveys and leaver interviews, ingesting over 10 million records from multiple NHS and HEE systems and creating a scalable platform that data scientists across departments can use collaboratively, we managed to predict with 63% accuracy the risk of attrition.
Key takeaways of the session will be:
1. Empowering HEE to tackle its workforce challenges through reduction of attrition using innovative data and artificial intelligence technologies.
2. Approach to building a machine learning attrition predictive model.
3. Overview of a Machine Learning framework for modelling.
4. Getting the data foundations right to support the model.
5. A clear explanation of Supervised Machine Learning.
6. Modelling approaches and diagnostic tools to evaluate the accuracy of the model.
This project won the Health Tech Award in 2022 for “Best Use of AI and Automation Tools”.