Summary

‘Think Glucose’ is a national initiative to improve in-patient diabetes care, including the use of a ‘traffic light’ system to give guidance to hospital staff as to which patients should be referred to the in-patient diabetes specialist team (IPDST). Keeping track of the 31 types of case referenced in the ‘traffic light’ scheme can be difficult, however, leading the Trust to develop an electronic form using its iSoft Clinical Manager (iCM) for ‘Think Glucose assessment’. This comprises a simple and mandatory tick box process to assess diabetes patients using the traffic light scheme, with the completed form then sent to the IPDST. Admin staff then add any patients who need to be seen by the IPDST to round lists. There has been a dramatic increase in the number of in-patients seeing the IPDST, while at the same time ensuring that the patients seen are only those requiring it according to the Think Glucose criteria.

Results

There were significant improvements in the overall quality of care, patient safety, bed efficiency, staff knowledge and satisfaction, use of the specialist team and cost reduction. The length of hospital stay and insulin administration errors decreased, as did cancelled procedures, complaints, delays in discharge and inappropriate referrals.

The numbers of patients being appropriately seen by the IPDST, according to the national referral criteria, increased dramatically. For example, in September and October 2010, before this initiative was implemented, there were 83 and 113 referrals. A year later these figures had risen to 452 and 390, an increase of 544% and 345% respectively.

The initiative assumes that the recommendations of the national Think Glucose project are worthy of adoption by all NHS Hospital Trusts and has not attempted to re-visit nationally established norms. Rather, this initiative was designed to facilitate local implementation of broader national guidelines.

Challenge

In the UK the number of people with diabetes increased by 18% between 2005 and 2008, with more than 4% of the population affected. Diabetes is much more common in South Asian and Afro-Caribbean ethnic groups, who are strongly represented in our West Birmingham/Sandwell locality. Nationally, between 10-20% of acute Trust beds are occupied by patients with diabetes but in our catchment area it is nearer 28%. In addition, diabetic in-patients stay on average 2.6 days longer and insulin administration is one of the treatments most frequently involved in medication-related adverse incidents. The NHS Institute for Innovation and Improvement therefore collaborated with clinical diabetes teams to develop the National Think Glucose Project, with a view to improving the care, outcomes and experience of people with diabetes who are admitted to hospital.

This West Birmingham/Sandwell initiative aimed to implement the National Think Glucose policies using an IT solution.

Objectives

The project aimed to offer a clear focus on the patient by quickly identifying people with existing diabetes and hypo- hyperglycaemia via a comprehensive and standardised assessment of their diabetes needs. This was to be achieved using a jointly agreed care pathway involving effective use of an inpatient diabetes specialist team (IPDST). It involved staff education offering appropriate training using adult learning modules, and effective commissioning and planning through good communication between the IPDST, the hospital management and commissioners.

The project was trialled in test sites across the UK and resulted in a 37% improvement in appropriate referrals, a 25% reduction in inappropriate referrals, a 25% increase in early blood glucose testing, a reduction in insulin administration errors by 3 to 1 and a reduction in the average length of hospital stay of at least two days. This saved an average hospital Trust around £1,000,000 per annum.

Solution

The clinical software used is iSoft Clinical Manager (iCM). A unique function was developed within iCM to facilitate electronic assessment and referral in line with national traffic light criteria. This traffic light system guides staff about which patients should be referred to the IPDST. The traffic lights include 31 types of case ¬– 16 types of ‘always-refer’ (red), eight types of ‘sometimes-refer’ (amber) and seven types of ‘rarely-refer’ (green). Previously, referral involved phone, fax or internal mail with inherent built-in delays and the possibility of patients being missed. This project developed an electronic form in iCM for ‘Think Glucose assessment’ and mandated all clinical staff to assess all diabetes patients as soon as possible after admission. The quick-and-easy form incorporates the 31 traffic light cases with simple tick boxes and once submitted it appears instantly in an email account accessed by the IPDST.

Evaluation

The national Think Glucose project could still be implemented in any hospital Trust using reminder cards, leaflets and posters to advertise the 31 types of case in their traffic light categories, and use phones, faxes and internal mail for referrals. However, this initiative represents a considerable improvement over those traditional methods.

Prior to the initiative there was no fail-safe system to identify people with existing diabetes at an early stage, no timely, comprehensive and standardised assessment of diabetes needs, no jointly agreed and effectively implemented care pathway, and the IPDST was used ineffectively with many inappropriate referrals – while at the same time many appropriate referrals were not made.

The initiative believes the suggested approach from the national Think Glucose project, based on data from pilot sites, is worthy of emulating. It enhances the national project by using IT to: Correctly categorise patients according to the national Think Glucose criteria; create daily ward round lists for consultants, specialist registrars and diabetes specialist nurses; and ensure that the IPDST is appropriately involved as soon as possible.

Impact

The new iCM Think Glucose assessment tool went live in August 2010, took three months to bed in and had significant impact from November. Implementation led to a dramatic increase in the number of in-patients seeing the IPDST while ensuring that these referrals were appropriate. All hospitals using iCM could emulate this and those using other clinical systems could adapt them accordingly. The clinical software offers the opportunity to order investigations and allows electronic messages to be sent to relevant staff and departments, such as phlebotomists, laboratories and imaging services.

‘Think Glucose Champions’ were recruited on every ward to be given targeted education and to take the lead in improved diabetes care, while all nursing staff were trained to update their skills. An insulin storage audit was undertaken which revealed multiple problems and required, just from one day, £1,000 worth of insulin to be disposed of because it was poorly managed (not labelled, out of date and so on). A ‘safer use of insulin’ policy was established, along with assessment for safe self medication by diabetes patients, and protocols such as the hypoglycaemia guidelines were updated.