Children with type 1 diabetes at fivefold risk of hospitalisation

Welsh data shows pre-schoolers and those from disadvantaged backgrounds most at risk

Children living with type 1 diabetes are nearly five times more likely to be admitted to hospital than non-diabetic children, a new study has found.

The number of children being diagnosed with type 1 diabetes is increasing 3-4% every year and more so in school-aged children. Treating the condition is complex and poor management can often lead to medical emergencies that result in hospitalisation, placing ever greater demands on health services.

To improve the efficiency of these services and reduce potentially unnecessary admissions, researchers wanted find out how often children with type 1 diabetes are admitted to hospital compared with children of the same age, gender and socioeconomic class, living in the same area.

Specialists from Cardiff University, Swansea University, the University of Bristol, Bangor University and the University Hospital of Wales set about answering this by analysing the anonymised data of 95% of all young people in Wales with type 1 diabetes, against hospital data of the entire Welsh population. They found was that those living with type 1 diabetes are 480% more at risk of hospital admission than their non-diabetic counterparts.

The data reveals that admission rates are at their highest in children aged between 0-5 years old, with risk of hospitalisation reducing by 15% for every five-year increase in a child's age at diagnosis. Consistent with the findings of a previous study in the US, children from poorer backgrounds were deemed to be at higher risk of hospitalisation, as were children receiving outpatient treatment for their condition in a smaller sized treatment facility.

"Our research shows that children with diabetes are at an unacceptably increased risk of being admitted to hospital," said Professor John W Gregory, a specialist in Paediatric Endocrinology at Cardiff University School of Medicine. "Based on this evidence, clinical services need to look at ways of supporting the care of those most at risk: the very young and those from poorer backgrounds. It is likely that greater anxiety surrounds healthcare issues in those from poorer backgrounds and in very young children who can become ill more rapidly than older children. These are potentially modifiable factors depending on the availability of expert help to support children and their families.

"Given the increased risk of admission of patients cared for in smaller outpatient clinics, we also need to examine working practices, especially 'out-of-hours' services, to see if there are better ways to deliver care in a way that avoids the need for admission to hospital. However, the associations we have shown do not show cause and effect and an interventional study would be needed to show that additional clinical resources or modified ways of working can benefit these admission rates."

According to the researchers, the study is the most accurate assessment of the increase of hospital admission ever undertaken. To reach their findings, researchers analysed two datasets: a register of 1577 Welsh children with type 1 diabetes, aged 0-15 years old, against anonymised health records from a Swansea-based databank.

The analysis is the first of several planned to allow the team to ask fundamentally important questions about the effectiveness of treatment in children with diabetes.

Professor Gregory added: "There is an increasing pressure on clinicians to measure the quality of clinical care provided to children with diabetes to ensure that we avoid a 'postcode lottery' service and allow high quality practice to be identified and be replicated nationwide.

"This is a powerful new approach, which will allow us to measure outcomes that will influence the future design and structure of clinical services. The data provide a measure of ongoing improvements in the quality of paediatric diabetes care which may lead to reduced rates of insulin related illness and death."

Professor Reinhard Holl, Paediatric Endocrinologist and Diabetologist from the University of Ulm (Germany), said:

"This is an extremely important study both from the perspective of patients and their families, and from a healthcare perspective. The treatment of children with type 1 diabetes has undoubtedly improved during the last two decades owing to new therapeutic strategies as well as pharmaceutical and technical advances.

"This study demonstrates that we are still far from our goal of providing treatment that interferes as little as possible with the lives of children and their families. Hospitalisation keeps children out of school and away from their families and friends – this should be avoided by all means if possible. In addition, costs to the health care system are high, money which should be invested to improve continuous outpatient management and family support for those affected."

The research was funded by NovoNordisk and the Welsh Government.