New guidance from a group of diabetes organisations has identified how commissioners and hospital trusts can work together to reduce admissions associated with diabetes
Reversing the rise in emergency hospital admissions is an NHS priority. These admissions impact on elective capacity and waiting times and are unsustainable. The risk of hospitalisation for people with diabetes is almost twice that for others. Commissioners need to address admissions associated with diabetes and new guidance offers best-practice solutions.
Citation: Allan B (2014) Reducing admissions for people with diabetes. Nursing Times; 110: 10, 12-13.
Author: Belinda Allan is consultant diabetologist at Hull and East Yorkshire Hospitals Trust.
- This article has been double-blind peer reviewed
- Scroll down to read the article or download a print-friendly PDF including any tables and figures
The scale and cost of diabetes hospital admissions is enormous. The risk of hospitalisation for people with the condition is almost double that of those without (Moghissi et al, 2009).
Each year there are approximately one million admissions to hospital where diabetes appears as a diagnosis, costing an estimated £2.51bn (Kerr, 2011). Of these, about 250,000 are in excess of the numbers expected for an age-adjusted population without diabetes. The estimated cost to the NHS of these excess admissions is around £686m each year, notwithstanding the personal cost to the individual (Kerr, 2011).
There are significant variations between acute trusts in diabetes admission rates, which cannot be explained by differences in prevalence. This suggests that pathways of care, or their absence, must in some way be contributing to this.
The Joint British Diabetes Societies Inpatient Group (sponsored by Diabetes UK and the Association of British Clinical Diabetologists) has published guidance detailing how commissioners and acute trusts can work together to reduce excess hospital admissions (Allan et al, 2013).
The guidance describes models of care that reduce emergency admissions but are often not commissioned or widely available. These provide the potential for rapid improvement in admission rates without the need for massive societal change or large-scale immediate investment.
People with diabetes are at a much higher risk of hospitalisation than those without the condition, and are far more likely to be admitted via the emergency route.
The National Diabetes Inpatient Audit (NaDIA) has consistently shown that around 87% of inpatients with diabetes have been admitted as an emergency (Health and Social Care Information Centre, 2013). In addition, once patients with diabetes are admitted, they are likely to experience longer stays. On average the excess length of stay is around 0.8 days longer than for patients without diabetes, but can be considerably longer (Sampson et al, 2007).
Knowing the age-adjusted prevalence of diabetes, the average excess length of stay, tariff uplift for urgent care and the higher cost of admissions that are emergencies allows the costs of caring for inpatients with diabetes to be estimated.
To address the problem, it is necessary to understand local population needs and the triggers for admission. Data on local activity patterns associated with admission can be found via the National Diabetes Information Service and National Diabetes Inpatient Audit. These contain benchmarking data on overall diabetes admission rates and diabetes-specific admissions (diabetic ketoacidosis, hypoglycaemia, hospital admission rates of care home residents with diabetes and diabetic foot disease).
Ambulance trusts should have data on call-outs for severe acute hypoglycaemia, and carry-on rates to the emergency department (ED). The national register of patients with diabetes in Scotland (SCI-DC Network) provides comprehensive information on diabetes and links primary and secondary data.
Many pharmaceutical companies have population and case-mix adjusted diabetes admission data (derived from hospital episode statistics) at practice and clinical commissioning group level. This allows benchmarking to a CCG area and comparison with the non-diabetes population.
The whole system or pathway of care for those with diabetes must be understood. This is because triggers for admission need to be linked with the key decision points in GP surgeries, ambulance trusts, out-of-hours care, EDs and pre-operative assessment.
Integrated diabetes care
Last year, Diabetes UK published a document advising commissioners on the key components of an integrated diabetes service (Diabetes UK, 2013). This advocates:
- Services being close to patients’ homes;
- Services without duplication or gaps;
- Integrating primary and secondary care services;
- Ensuring that the multidisciplinary team is competent and available;
- Supporting self-management.
Integrated care will undoubtedly reduce the burden of diabetes admissions, particularly when diabetes itself is the primary reason for admission to hospital.
The JBDS document highlights several areas of care that have been shown to reduce admissions due to ketoacidosis (a potentially life-threatening complication), hypoglycaemia and diabetic foot ulceration (Allan et al, 2013). Its key recommendations are outlined in Box 1.
Box 1. Recommendation for commissioner
People with diabetes should have access to:
- Structured education for those with type 1 diabetes; half of admissions with diabetic ketoacidosis are avoidable;
- A diabetes specialist multidisciplinary team including an open-access phone line for advice during sick days or when ketosis develops. Ideally, this should be 24-hour, seven-day a service. A national helpline run by specialist diabetes teams could deliver this but it will require political will and leverage to implement.
Commissioners should also consider:
- A diabetes inpatient team and diabetes specialist team to support emergency departments and emergency admission wards and to provide immediate frontdoor management;
- Diabetes management guidelines for diabetes inpatients undergoing surgery or planning surgery. Many junior doctors lack confidence in diabetes management (George et al, 2011), so it is essential that trusts provide mandatory training to all staff;
- A diabetes service for frail older people that supports diabetes education, foot care and management in residential and nursing homes, with staff training in identifying highest-risk residents. This could nearly halve admissions from homes;
- An emphasis on pre-discharge planning on wards to prevent readmission, including early referral to the diabetes team;
- A high influenza vaccination uptake and statin use in type 2 diabetes patients aged over 40 years, and benchmarking of Quality and Outcomes Framework data against comparator areas;
- Identification of patients admitted frequently who need intensive education and psychological support;
- Better working between clinical diabetes teams, mental health trusts and clinical psychology, particularly targeted at those at highest risk;
- Blood ketones testing in people with type 1 diabetes – this is an earlier and more accurate marker of metabolic decompensation than urine ketone testing;
- A hypoglycaemia management pathway in collaboration with ambulance trusts with a single point of contact, and a clearly defined “see and treat” policy with a low carry-on rate to the ED. It should also identify frequent callers who can be referred to their GP and specialist diabetes team for further support;
- A diabetes foot care service that includes a foot protection team for primary care, a hospital-based multidisciplinary foot team for the highest-risk feet and an inpatient podiatry service to ensure that expertise is available to ward-based nurses.
Allan et al (2013)
Many nurses will recognise the lack of integrated working between agencies that makes delivering seamless diabetes care a challenge. The absence of technology allowing real-time communication between hospital and primary care remains a frustration to many.
It is difficult to know what savings could be made if all the JBDS recommendations were implemented. However, Kerr’s data (Kerr, 2011), suggests a 5% sustained reduction in admissions and associated costs could save £125m per year.
The admissions-avoidance document highlights many examples of small-scale changes that have led to significant reductions in admissions of people with diabetes. The list of “must dos” is understandably long, but diabetes affects so many people in so many ways that all these areas need to be addressed at the same time, and not in a piecemeal fashion.
Commissioners need to work with clinical teams - with nurse representation - to come to an agreement about what needs to be done to improve their local service. Eliminating the variations in the standards of care is the goal.
- The risk of hospitalisation for people with diabetes is almost twice that of those without the condition
- There are approximately one million hospital admissions each year where diabetes is a diagnosis
- Around 87% of people in hospital with diabetes have been admitted as an emergency
- There are variations in diabetes admission rates between acute trusts, which cannot be explained by diabetes prevalence
- Small-scale changes can significantly reduce admissions for people with diabetes
Allan B et al (2013) Admissions Avoidance and Diabetes: Guidance for Clinical Commissioning Groups and Clinical Teams. Joint British Diabetes Societies for Inpatient Care.
Diabetes UK (2013) Best Practice for Commissioning Diabetes Services: an Integrated Care Framework. London: Diabetes UK.
Health and Social Care Information Centre (2013) National Diabetes Inpatient Audit (NaDIA) - 2012. Leeds: HSCIC.
Kerr M (2011) Inpatient Care for People with Diabetes: the Economic Case for Change. Insight Health Economics, 2011.
Moghissi ES et al (2009) American Association of Clinical Endocrinologists and American Diabetes Association consensus statement on inpatient glycemic control. Diabetes Care; 32: 1119-1131.
Sampson MJ et al (2007) Total and excess bed occupancy by age, specialty and insulin use for nearly one million diabetes patients discharged from all English acute hospitals. Diabetes Research and Clinical Practice; 77: 92-98.