Reducing suboptimal prescribing of glucagon-like peptide mimetics in type 2 diabetes

In 2016, spend on diabetes prescribing — and glucagon-like peptide 1 (GLP-1) mimetics, in particular — at NHS Hastings and Rother Clinical Commissioning Group (CCG) and NHS Eastbourne, Hailsham and Seaford CCG was significantly higher than expected, based on national prescribing patterns. There was also significant variation between GP practices. A project was undertaken by the medicines management (MM) team to support primary care diabetes teams to work with patients to optimise prescribing through implementation of the National Institute for Health and Care Excellence’s (NICE) guidelines for type 2 diabetes mellitus (T2DM).

The project aimed to review all patients prescribed GLP-1 mimetics at our GP surgeries to assess prescribing in accordance with NICE guidelines. The main objectives were to identify suboptimal prescribing, rationalise therapy, and improve quality of life for patients, while ensuring good value from the prescribing budget.

The project was a key part of our GP prescribing support scheme, which provides financial incentive for engagement. All key stakeholders were consulted during the project’s development. Opportunities to educate clinicians on diabetes medicines optimisation and share details of our project were utilised during GP engagement events. Inspirational speakers delivered key note sessions at our annual GP education event. Details of the project were shared with all key stakeholders through communications such as newsletters.

Before the project was implemented, training was provided for pharmacists undertaking reviews. Key resources used to standardise data collection included searches to identify appropriate patients and a template produced by EMIS Health.

Practices were required to meet with their MM pharmacist to discuss patient-level diabetes medication reviews and agree an action plan to improve prescribing in line with NICE guidelines. These reviews were intended to optimise diabetes medication, with a focus on blood pressure and cholesterol management; however, the reviews were holistic — other medication issues identified were highlighted and discussed. Action plans were agreed to ensure key messages became embedded into normal practice. An audit was then carried out to ensure the agreed actions had been implemented.

A formulary update and development of key resources in collaboration with primary and secondary care colleagues ran parallel to the patient-centred reviews.

All but two practices from across both CCGs (n=44) undertook the project and agreed an action plan with their MM pharmacist.

In total, 850 patient records were reviewed by the MM team. Half were considered suitable for a face-to-face clinician review, with the aim of stopping the GLP-1 mimetic, where appropriate, and optimising medicines for diabetes.

In 93% (n=396) of patients, recommendations to optimise therapy were implemented, which generated significant financial savings across both CCGs. The CCGs demonstrated reduced growth of GLP-1 mimetic prescribing compared with national growth.

Owing to the variation in baseline prescribing of these agents, rates of change varied between practices. Some practices achieved rates of change of 20%, while others achieved significantly higher rates of change (over 75%). It was proposed that GP practices with small patient numbers at baseline may have the least potential for change; however, in practice, there were no data to support any such correlation.

Cases of diabetic polypharmacy were identified where patients were receiving more than three blood glucose-lowering therapies, suggesting clinical inertia towards insulin. This also suggests that current therapy had been added to, rather than reviewed and stopped if the therapies gave no sufficient reduction in HbA1c.

Cases were also identified in which medication should have been stopped owing to the patient’s poor kidney function. Numerous cases of patients receiving both GLP-1 mimetics and DPP-4 inhibitors were found, which suggested a lack of awareness of mechanisms of action — both these agents work on the same pathway. These messages were highlighted to clinicians during the project.

There was a clear focus on improving quality of prescribing, which clinicians valued. Feedback on the project was very positive; clinicians enjoyed the holistic discussions and the presentation of patient cases using the EMIS Health template, which made it clear where NICE-recommended HbA1c and weight reductions had not been achieved.

A lack of awareness of NICE guidance was apparent, in relation to both initiation and monitoring of these agents. Many clinicians were unaware of the initiation criteria and the six-month targets to continue: both a reduction in HbA1c and weight are required. Additionally, it was noted that GLP-1 mimetics were frequently added to patients’ repeat prescription record in quantities not aligned to the dose, resulting in inadvertent over-ordering. This opportunity was used to develop local guidance and resources to support our clinicians.

We continue to share the legacy messages of this project and encourage clinicians to refer to our formulary GLP-1 mimetic initiation flowchart and GLP-1 mimetic patient contract when initiating these agents; document the targets for continuation at six months; and prescribe an appropriate quantity of devices. These resources are available on our formulary website.

We are working collaboratively with clinicians and managers to integrate effective medicines optimisation, including insulin initiation, into local service redesign for diabetes. In 2018, the GP support scheme continues to focus on medicines optimisation for diabetes, and MM pharmacists are undertaking polypharmacy reviews in frail older patients with T2DM.

 

Hannah Syed, medicines management adviser, NHS Hastings and Rother CCG; NHS Eastbourne, Hailsham and Seaford CCG, on behalf of the medicines management team

Last updated
Citation
The Pharmaceutical Journal, Reducing suboptimal prescribing of glucagon-like peptide mimetics in type 2 diabetes;Online:DOI:10.1211/PJ.2018.20205541

You may also be interested in