ISSN 1470-3947 (print) | ISSN 1479-6848 (online)

Endocrine Abstracts (2019) 66 P28 | DOI: 10.1530/endoabs.66.P28

Using quality improvement (QI) to improve the care pathway and outcomes for children newly diagnosed with type I diabetes mellitus

Edward Coxson1, Clare Edmonds1, Lynn Diskin1, Gillian Purcell1, Hannah Kyprios1, Hayley Carter1, Rey Fong1, Vineeta Gupta1, Karen Bradshaw1,2, Clare Vass1, Laura Bird1, Anna Zatchij1 & Helen Edwards1


1Royal United Hospital Bath NHS Foundation Trust, Bath, UK; 2Helen, Edwards, UK


Background: Early glycaemic control improves long-term outcomes in children with Type I diabetes. The NICE target for children with T1DM is HbA1c ≤ 48 mmol/mol. 2018 data from our newly diagnosed patients (pre-QI) demonstrated mean HbA1c 50 mmol/mol at 3 months and 62 mmol/mol at 12 months.

Aims and methods: Our aim is to improve average blood glucose levels at day 28 post diagnosis and achieve a median HbA1c of <48 mmol/mol at 3 and 12 months post diagnosis in 75% of our newly diagnosed patients by November 2020.

Drivers for improvement included: 1) Implementing Carbohydrate Counting and Expert meters at diagnosis of Type I diabetes.

2) Intensive inpatient management with multiple insulin correction doses and overnight corrections.

3) Achieving blood glucose levels in target prior to discharge to emphasise the importance of this to our families.

4) Additional MDT clinic 6 weeks post diagnosis.

5) Setting up home downloading and Diasend accounts prior to discharge for new patients.

We used Fishbone analysis to undertake a needs assessment for implementing carbohydrate counting and Expert meters at diagnosis. The main barriers were ward staff training and communicating key changes to the wider paediatric team. ‘Tea Trolley’ training and ‘Newly Diagnosed Packs’ helped facilitate change. Outcome data on average blood levels, HbA1c, days to carbohydrate counting and balancing measures including length of stay were collected.

Results: 10 new patients have been managed along the new care pathway. Mean average blood glucose at D28 post diagnosis has improved from 8.0 mmol to 5.9 mmol. Mean HbA1c at 3 months post QI is now 45.9 mmol/mol (n=7). Average length of initial hospital stay increased from 3.2 to 4.3 days. All patients had the facility to home download at discharge.

Lessons learnt: Weekly QI meetings and maintaining run charts of average blood glucose and HbA1c for new patients has been a powerful team motivator. Working with other teams in the National QI Collaborative has made us braver to implement change and initial project results are encouraging. We are now using QI methodology to improve clinic experience and education around download interpretation.