TED

Developing a predictive tool to promote earlier diagnosis of Type 1 diabetes in childhood for use in primary care.

Background

Early diagnosis of Type 1 Diabetes (T1D) is critical to avoid children developing diabetic ketoacidosis (DKA). In the UK, 25% of children who develop T1D are in life-threatening DKA at diagnosis. These children have far worse outcomes, with increased NHS costs. This research aims to develop and assess the usefulness of a tool to identify children who may have undiagnosed T1D. Using the number and details of GP consultations, the tool will assess the risk a child has of being diagnosed with T1D.

We will develop the tool using a modern technique called machine learning. Machine learning allows computers to uncover patterns in data in a very flexible way. It has proved to be more successful than traditional methods of data analysis in showing that a particular event is likely to happen. We will test the tool, using unidentifiable data from hospital and GP records. Results will show how successful the tool is at distinguishing between children who went on to develop T1D and those who did not. If successful, the tool could be used by GPs during consultations in primary care. This may mean that children will be diagnosed earlier.

Study design

Delayed and incorrect diagnosis are significant risk factors for children being diagnosed with T1D in DKA. Previous studies have relied on parent’s and GP’s recall of events. These data were collected after the child was diagnosed, and may be biased. This is because people may not recall events as they happened at the time. Our approach is unique because we are using routinely collected primary and secondary care records in England and Wales. These data are recorded at the time of a GP consultation or admission to hospital.

Using large data sets of routinely collected data is an effective and efficient method for conducting research, especially in uncommon conditions like Type 1 diabetes.

Involving the public and patients

Mrs Beth Baldwin, a mother whose son died from undiagnosed T1D, is a co-applicant on this application. She has co-written the lay summary and influenced the study design. She states “I am supporting this research to help raise public awareness to save lives. My son Peter was just 13 when he died from DKA, as a result of undiagnosed T1D. This was January 2015. There are finger prick monitors in GP practices. They need to be more widely used. Prompt diagnosis saves lives. This proposal can make a positive impact whilst saving lives.”