An increasing number of people require support to manage their diabetes due to other conditions that affect their ability to self-care, such as dementia and arthritis (Diabetes UK, 2010). Community staff have a growing caseload of people who require this support. It is therefore important that they have the knowledge and skills to give the right care to people with diabetes, which could even involve delegation of insulin administration. This article briefly describes a project being conducted within a clinical commissioning group (CCG) in East Kent with the aim of improving the community nursing team's insulin administration service for vulnerable homebound adults with diabetes.
Community nursing teams in East Kent are commissioned (on a ‘block’ contract) by the CCGs to provide services to patients who are housebound, including the administration of insulin. Most patients are prescribed a once- or twice-daily insulin regimen but can often be discharged from hospital on three or four insulin injections per day. Blood glucose levels are checked at the time of insulin administration. However, these patients often do not have a comprehensive annual review, which in turn, makes it difficult to set individualised targets for them.
The setting of this project was a large (three hub) CCG. From the outset, it was clear that there was poor overall service provision for those with diabetes living in a residential care setting, as well as those who lived alone and required support from the community nursing team. These findings were similar to those of a report of the Institute of Diabetes in Older People (IDOP) (2014). Effective management of diabetes in such complex patients requires appreciation of the changing circumstances of their health status and other comorbidities (Strain et al, 2018). Within this area, housebound patients were not receiving the same parity of care as those who were not housebound, despite having daily contact with a member of the community nurse team.
To improve the practice, the clinical multidisciplinary team lead was successful in gaining funding to implement a project—led by a nurse consultant in diabetes—who would assess, implement and evaluate the changes made throughout the project. Training is essential for any staff member to effectively undertake an annual review and to fulfil the nine key care processes listed below that make up the Quality Outcomes Framework data in primary care and are measured in the National Diabetes Audit (NHS Digital, 2018):
Part of the annual review is also to formulate a care plan and provide a high standard of diabetes care. During the project, unregistered practitioners were trained in the annual review process, as well as the other areas of diabetes care (listed below).
Training was delivered, both in the workplace and classroom settings, by the nurse consultant to enable six healthcare assistants (HCAs)/associate practitioners to become competent in a basic standard in these nine key competency areas. The key steps and activities undertaken during development of the project are outlined in Table 1.
| Engagement of GP practices by clinical MDT lead | Engagement of community nurse teams by clinical MDT lead | Timeline of project agreed between nurse consultant and clinical MDT lead |
|---|---|---|
|
|
|
|
MDT: multidisciplinary team; HCA: healthcare assistant
The nurse consultant role (Trend UK, 2019) was introduced to identify service shortfalls, develop strategies with local commissioning bodies and promote evidence-based practice and cost-effectiveness. There were also associated responsibilities such as auditing and reporting on the project, producing information on the relevant outcomes of interventions and identifying the need for change.
The engagement of GP practices and community nursing teams was time consuming, partly because there are so many surgeries within the CCGs and hubs, but mostly because attempts were made to ensure that all practices and teams involved were aware of the process and the timeline. While the clinical MDT lead was negotiating the terms and conditions of the project, the nurse consultant developed the training programme for the HCAs, adequate assessment tools for the project and a database.
Data collection would be in relation to the following:
These areas were chosen as, in the health service, it is important to prove financial gains or savings. However, as the project was developed, it was clear that while financial savings are important, they should not be the only focus—it became apparent that the quality of visits and education of staff needed improvement, in addition to cross-disciplinary working.
Community nursing teams are under enormous pressure to deliver their service. Costing of this service was difficult, as the community service works on a ‘block contract’. Therefore, to drill down to individual costs and potential savings was extremely difficult to accomplish. Additionally, it potentially posed a threat to the community nursing teams, if they perceived that they may lose funding for the service.
A project like this needs to consider sustainability. Most of the people on the caseload could be regarded as frail or vulnerable, and there are a number of socio-economic factors that affect their ongoing management and care, as well as other comorbidities. To presume that the project would resolve ongoing needs would be unrealistic, and therefore, the sustainability of the project and work need to be considered, in addition to defining responsibility for these groups of patients.
During reviews in the first hub, it was clear that daily blood glucose monitoring was not picking up overnight hypoglycaemia. Because of the intense pressures on the community nursing caseload over a wide area, many visits for insulin administration took place after the patients had had breakfast (particularly if they relied on carers to prepare meals), and therefore some hypoglycaemic episodes were missed (see the case study described below).
Case study
Julia, aged 58 years, has had type 2 diabetes for 5 years. When reviewed as part of the project, she was prescribed metformin, gliclazide and long-acting insulin once daily. Julia is housebound due to severe mental health issues, learning disabilities and agoraphobia.
Community nurses were visiting her in pairs (because of Julia's anxiety and previous allegations made by her) every morning to administer the insulin. NHS 111 had also contacted the surgery, as Julia was making ‘excessive’ calls to the service overnight, with vague symptoms of not feeling well, but also reporting that she was extremely lonely. A best-interest meeting had been scheduled with the GP, the practice manager, frailty team and 111 to discuss management of her case.
Julia had not had a full annual review for over a year and had only had one key care process met. No care plan or individual targets had been set for her.
When her case was reviewed as part of the project, it was found that community nurses were performing a blood glucose test once a day at the time when the insulin was administered. The time varied between 9:00 and 11:00 am, so the blood glucose tests were mainly after breakfast, which was prepared by carers.
The findings of the blood glucose tests were generally between 4.9 and 8 mmol. HbA1c testing and a full annual review were carried out during the first visit. It was thought that hypoglycaemia was being missed because of limited blood glucose monitoring.
The use of a Libre sensor, which scans blood glucose levels using interstitial fluid, was agreed with Julia, and her carers were advised on how to use the device (no recording of blood glucose levels needed to be carried out as the sensor reader recorded this).
Julia was reviewed 2 weeks later by the unregistered practitioner who was part of the community nursing team, and it was found that her overnight blood glucose levels were between 2.7 and 3.4 mmol. She did not have any specific symptoms of hypoglycaemia, but just felt unwell and not right. Her HbA1c level was 25 mmol. This was in line with the readings that the Libre sensor gave after 2 weeks. After the hypoglycaemic episodes were identified, her insulin was stopped immediately, and she continued on metformin and gliclazide. The Libre sensor still showed overnight hypoglycaemia with blood glucose readings between 3 and 3.9 mmol, so the gliclazide was also stopped. The Libre sensor was taken off after 2 weeks, and her glucose levels settled between 6 and 12 mmol on metformin alone.
Community nurses visited sporadically initially, but they no longer visit her for insulin administration. Her calls to NHS 111 have also reduced.
Two months later, Julia was restarted on her anti-psychotic medication, which is known to cause hyperglycaemia. Potentially, Julia will need to restart the insulin. Because of the increased knowledge within the community team and after discussing with the frailty team and GP, her case will now be re-reviewed.
Unregistered practitioners and competencies
It is widely recognised—and accepted—that for the community nursing team to be able to administer insulin to a large number of people on a daily basis, much of this work must be delegated to unregistered practitioners, who are either HCAs or associate practitioners who have been taught insulin administration by registered nurses within the team. Because of the high volume of the caseload, insulin administration is task driven and often viewed as requiring a minor or routine visit.
In the authors' experience of working with vulnerable groups who rely on community nursing teams to administer insulin, it is not unusual for the timing of visits and mealtimes to be mismatched, or for insulin to be omitted if the patient is hypoglycaemic or not on a covering caseload. These issues have a significant impact on blood glucose control and increase the risk of hypoglycaemia in the case of mismatched meals and insulin timings.
Project status
At the time of writing the summary, the project is halfway through its timeline. It has been completed at one hub (Table 2) and begun at the next, with a total of 32 new reviews and 12 follow-ups having been carried out by the nurse consultant and unregistered practitioners.
| Number of patients | Number of reviews during the project | Prescribing costs | Community nurse reviews | ||
|---|---|---|---|---|---|
| Before | After | Before | After | ||
| 18 | 46 | £600.93 | £441.27 | 21 | 12.25* |
Before and after indicate before and after project implementation.
One patient died
Although prescribing costs are included in the data collection, it is clear that insulin is not one of the most expensive treatments for diabetes; what counts more towards the cost is the time of the community nursing team. The outcomes of the project have been similar to what was expected, but the use of the Libre sensor for some of the vulnerable patients involved in this project has highlighted the real risk of undetected hypoglycaemia and the potential for reducing some visits from the community nurse. Simultaneously, the need to monitor the quality of these visits has also been highlighted.
Reviewing the unregistered practitioners' reflections and competencies throughout the project made it clear that they now view the management of these vulnerable patients differently, and one of the most successful components so far has been the sharing of their knowledge with other members of the team, including the development of ‘top tips’ as part of community nurse documentation.
Conclusion
GP practices, commissioners, community nurse teams, secondary care teams and pharmacy teams are all now aware of the project, and there has been a real interest in the work that is being carried out. It is clear that the project is already of benefit, but the challenging aspect is now to ensure that it is sustained with the right health care professional carrying out these reviews in the future.