The UK’s National Health Service is the publicly-funded healthcare system in the UK. It is by far the largest employer in the country, with over 1.2 million employees. The NHS was founded in 1948 by health minister Aneurin Bevan in the aftermath of the Second World War. This was the first time that free healthcare was offered to all citizens of a country, paid for by taxes rather than by insurance premiums.
As well as providing healthcare, the NHS is also responsible for the training, professional development, and certification of the country’s doctors, nurses and other healthcare professionals. The NHS takes in junior doctors, who have recently completed a medical degree, and puts them on a training pathway until they qualify in a specialty, such as cardiology, urology, or general practice. It costs around £200,000 and takes about six years to train a junior doctor to reach the grade of consultant or GP. Naturally, the NHS loses money when junior doctors leave the organisation before completing a training pathway.
Higher Education England (HEE), a department of the NHS, needed a way of anticipating and understanding employee attrition. Fast Data Science designed and trained a machine learning model in Azure ML which was able to predict which employees are at risk of leaving the NHS at a given time.
We discovered a number of surprising factors behind employee attrition and communicated these back to management where they are being incorporated into the NHS’s future workforce strategy.
We also provided some upskilling and consultation services in data science for the analytics teams in HEE, introducing some analysts to the power of no-code ML with Microsoft Azure ML.
You can read in more detail about the techniques we employed in this project, from analytics to machine learning, deep learning, and survival analysis, in this article on predicting employee attrition.