Which is the best service model to ensure that people with atrial fibrillation receive the recommended anticoagulation?
28 July 2021
DPhil Health Services Research Students
Departmental DPhil student, Nick Jones, reports on findings from a recent systematic review that formed the first part of his thesis, providing data inputs for future economic evaluations, including decision-analytical modeling studies, to reduce stroke risk through anticoagulation prescribing for people with heart failure and atrial fibrillation.
The results of this systematic review will tie in with a CPRD analysis to inform future decision modeling work.
Why was a review needed?
Atrial fibrillation is associated with a five-fold increased risk of ischaemic stroke but anticoagulation can reduce this risk by over 65% among eligible patients. People with co-morbid diseases, such as heart failure, are at a particularly high risk of stroke. However, up to a third of people with atrial fibrillation who have a high stroke risk are currently not prescribed anticoagulation. In the UK, improving stroke prevention treatment is one of the key aims of the Long Term Plan, which aims to prevent nearly 50,000 strokes over a ten-year period. As part of the plan, the NHS has invested £9 million into an anticoagulation review scheme, run by pharmacists and specialist nurses in the community. We wanted to know whether this approach was likely to be the most cost-effective approach to improving anticoagulation treatment. We, therefore, conducted a systematic review in which we aimed to summarise the economic evidence and compare the cost-effectiveness of anticoagulation service interventions to provide results that could inform future resource allocation.
What did the review find?
We undertook a comprehensive review incorporating medical and economic databases as well as searching ‘grey’ literature sources, guideline committees, and health technology appraisals. Despite this, we found there are only very limited data as to the cost or the cost-effectiveness of anticoagulation service interventions. This lack of existing research is an important finding as it identifies that substantial investments are being made to change clinical practice without clear evidence of the most cost-effective approach to service improvement, meaning limited funds are not necessarily being allocated as effectively as possible.
We did identify 13 original studies that were eligible for inclusion in the review. We categorised these into four distinct groups of interventions; anticoagulation clinics, decision support tools, patient-focused interventions, and complex interventions. The studies were all drawn from studies in high-income settings but there were important differences in terms of the study date, setting, comparison group, and population that limit the generalisability of results. Overall, we did find some evidence to support the use of nurse or pharmacist-led anticoagulation clinics, as these could be cost-saving compared to usual care and resulted in comparable or improved health outcomes. These clinics also offered the opportunity for a more holistic overview of the atrial fibrillation treatment with some wider benefits reported around cardiovascular risk management and patient education. We also found some evidence that reported patient-focused interventions could also be cost-effective, but not complex interventions, which largely reflected the higher staff costs involved. Service interventions that were targeted at groups of patients at particularly high stroke risk were most likely to result in cost savings.
What does this mean for healthcare services?
The high costs of treatment, long-term care, and loss of earnings related to stroke for patients and the healthcare system means that there are large potential cost savings from service interventions that improve anticoagulation management. Our review suggests that a two-tiered approach to anticoagulation delivery might be most cost-effective. More intensive and costly interventions may be best targeted at a minority of high-risk patients who are not adequately treated. For most patients a low cost, light-touch interventions may be suitable, such as new pieces of software integrated into a GP’s computer or regular audit, possibly supported by a nurse or pharmacist-led anticoagulation clinics.
These conclusions are only tentative given the limited evidence available. This is an important gap in the current research and developing cost-effective approaches to anticoagulation delivery should be seen as a research priority to inform healthcare decision-makers who plan care pathways. Importantly, we found no direct comparison of any two service interventions. Our review may provide data inputs for future economic evaluations, including decision-analytical modeling studies.
What to read next
DPhil student Georgia Richards argues why its time for doctors to rethink the prescribing of high-dose opioids for people with chronic pain.