Summary of the example
The number of older people with type 2 diabetes and frailty is growing. These patients are at increased risk of adverse effects from blood glucose lowering therapies including hospital admissions. Severe hypoglycaemia is the second commonest cause of hospital admission for drug related adverse events and is associated with an increased risk of CVD events or death, particularly in people with pre-existing CVD. This project was undertaken to support primary care diabetes teams in working with patients to optimise prescribing through implementation of NICE guidelines to adapt an individualised approach to diabetes care considering the need for relaxed individualised HbA1c targets for older people with frailty balancing the drive for tight glycaemic targets and prevention of harm generating significant savings to the prescribing budget.
Why we think it’s important
There are growing concerns that intensive treatment with insulin and sulfonylureas in older people with low HbA1c (<53mmol/mol) increases the risk of hypoglycaemia, morbidity and mortality. Frailty and dementia are risk factors for hypoglycaemia setting a vicious circle. Older people are also less likely to benefit from the long term protective effects of good glycaemic control and are at risk of inappropriate polypharmacy due to co-morbidities. Concerns over the harms from hypoglycaemic agents in the treatment of type 2 diabetes has led to revised NICE guidance recommending an individualised approach with more relaxed HbA1c targets for older people at high risk of hypoglycaemia. The guidance advises that HbA1c targets of less than 58 mmol/mol may not be appropriate for people at older ages, those with impaired renal function, co-morbidities, limited life expectancy, interacting medication, previous hypoglycaemia or inability to self-manage treatment. Similar recommendations are supported in European and American guidance.
Learn more about the example
Aims and objectives of the work
To reduce harm by moving towards an individualised approach for managing type 2 diabetes in older people with frailty. The main objective was to agree individualised HbA1C targets using shared decision making with patients by taking into account factors such as co-morbidities, impaired renal function and life expectancy to reduce the harms associated with problematic polypharmacy, improve patient safety and quality of life for older people living with type 2 diabetes.
The project was included in the CCG prescribing support scheme which incentivises GP practices to invest time in reducing unwarranted variation in diabetes prescribing. CCG pharmacists reviewed cohorts of patients with frailty within all GP practices through running locally developed searches on the prescribing systems. Practices were then required to meet with a senior CCG pharmacist to discuss the patient level diabetes medication reviews and agree action plans to individualise diabetes care in older people. Key stakeholders were consulted during the project development and updated regularly. The diabetes formulary was refreshed and clinical guidelines and pathways for blood glucose management in Type 2 diabetes were developed with individualising targets and treatment at the heart. Opportunities to educate clinicians on diabetes medicines optimisation and share details of our project were utilised during GP engagement and community pharmacy events. Inspirational clinical speakers delivered key note sessions at GP education events. Training was provided by the Effective Diabetes Education (EDEN) group for pharmacists undertaking reviews. Key resources including searches and an EMIS template were developed to standardise data collection. Searches were built to identify frail patients over 65 years old with low HbA1c prescribed blood glucose lowering agents. Pharmacists conducted notes based reviews and arranged MDT meetings with key clinicians to discuss medication optimisation recommendations and share key messages. The project focused on diabetes medication optimisation; however, other polypharmacy issues identified during the reviews were also discussed. Action plans were agreed to ensure key messages became embedded into normal practice and practices obtained feedback quarterly via the prescribing dashboard. Implementation of agreed actions was assessed by audit at year end.
55% (n= 238) of patients had a blood glucose lowering agent stopped. 45% (n=198) of patients had a medication change whereby the blood glucose lowering agent was continued, but at a lowered dose. Inappropriate prescribing due to polypharmacy, drug interactions, co-morbidities, renal impairment, lack of evidence and/or not meeting NICE targets to continue were the main reasons for therapy changes. 541 medication optimisation recommendations not relating to blood glucose management were also made. Of the elderly patients reviewed, 60% (n=204) were found to have a recent HbA1c of ≤48mmol/mol, of which 19% (n=66) had an HbA1c ≤ 42mmol/mol. All these patients were receiving blood glucose lowering therapies. Throughout the project pharmacists delivered a number of education and training sessions at practice and CCG level. During the initial stages, some diabetes leads did not feel this was a problem locally owing to individualising targets being part of their routine practice. Case presentation of severely frail patients with low HbA1c’s treated with insulin and/or sulphonylureas found within the CCG helped justify the need for this project. At practice level, we faced challenges regarding the possible negative effect of this project on QOF targets. Here, it was evident that an individualised approach to diabetes care had not been adopted resulting in therapy remaining unchanged or being escalated to meet QOF targets for patients with frailty and comorbidities. Many of the HbA1c identified were very low which was concerning. Towards the end of the project, proposed changes to QOF taking into frailty were released which supported the project. The project was not without challenges; in some practices differences in behaviour between individual clinicians was observed, appearing to reflect the individual clinician’s attitudes and beliefs therefore the practice action plan had to be flexible enough to capture different clinicians prescribing practice.