BACKGROUND: The National Health Service (NHS) in England spends over £9 billion on prescription medicines dispensed in primary care, of which over two-thirds is accounted for by repeat prescriptions. Recently, GPs in England have been urged to limit the duration of repeat prescriptions, where clinically appropriate, to 28 days to reduce wastage and hence contain costs. However, shorter prescriptions will increase transaction costs and thus may not be cost saving. Furthermore, there is evidence to suggest that shorter prescriptions are associated with lower adherence, which would be expected to lead to lower clinical benefit. The objective of this study is to estimate the cost-effectiveness of 3-month versus 28-day repeat prescriptions from the perspective of the NHS. METHODS: We adapted three previously developed UK policy-relevant models, incorporating transaction (dispensing fees, prescriber time) and drug wastage costs associated with 3-month and 28-day prescriptions in three case studies: antihypertensive medications for prevention of cardiovascular events; drugs to improve glycaemic control in patients with type 2 diabetes; and treatments for depression. RESULTS: In all cases, 3-month prescriptions were associated with lower costs and higher QALYs than 28-day prescriptions. This is driven by assumptions that higher adherence leads to improved disease control, lower costs and improved QALYs. CONCLUSION: Longer repeat prescriptions may be cost-effective compared with shorter ones. However, the quality of the evidence base on which this modelling is based is poor. Any policy rollout should be within the context of a trial such as a stepped-wedge cluster design.