OBJECTIVES: Atrial fibrillation (AF) is a major risk factor for stroke and is managed via oral anticoagulation (OAC) after a confirmed diagnosis. Continuous monitoring with an insertable cardiac monitor (ICM) in the REVEAL AF study demonstrated a high incidence of AF in high-risk patients where previous external monitoring (≥24 hours) was negative. Currently it is unknown whether ICM monitoring to detect AF is cost-effective in this population.

METHODS: A Markov model was designed to synthesise clinical evidence with costs and patient quality of life. The UK NHS perspective was used. The model input was based on a large, prospective, single-arm, multicenter study that included patients with an ICM (REVEAL AF). The diagnostic yield of alternative testing strategies was simulated from ICM data. The model included treatment with OAC after AF detection; the incidence of ischaemic strokes and major and minor bleeding events was based on published literature. One-way deterministic and probabilistic sensitivity analyses were undertaken to test key model inputs.

RESULTS: For the average patient (aged 71.3 and with a virtual CHADS score of 2.94), the total cost for the ICM device was £2,891. When considering all costs, including treatment and downstream events, the total per-patient cost for an ICM-based strategy was £13,370, compared with £12,946 for an annual 24h Holter (Standard-of-Care, “SoC”). ICMs generated 6.51 QALYs vs. 6.31 for SoC, giving an incremental cost-effectiveness ratio of £7,128/QALY. In probabilistic sensitivity analysis, using a £20,000 willingness-to-pay threshold, the probability of ICMs being cost-effective was 78.4%. The model results were sensitive to the rate patients initiated and discontinued treatment after a diagnosis, the type of OAC used, and the intensity of monitoring methods assumed for SoC.

CONCLUSIONS: The use of ICMs to identify AF in a high-risk population is cost-effective for the UK NHS.