PMS62 - COST-EFFECTIVENESS OF REDUCING ARTHRITIS FATIGUE BY CLINICAL TEAMS (RAFT) USING COGNITIVE–BEHAVIOURAL APPROACHES: A RANDOMISED CONTROLLED TRIAL

Abstract

Patients with rheumatoid arthritis (RA) report fatigue as a key symptom. RAFT is a group cognitive-behavioural approach to help patients manage fatigue impact. This study assessed the cost-effectiveness of RAFT alongside usual care (information booklet) compared with usual care alone. Patients in seven UK hospitals were randomly assigned to RAFT or usual care. The primary clinical outcome was fatigue impact (Bristol RA Fatigue Numerical Rating Scale, BRAF-NRS). EQ-5D-5L was used to derive quality-adjusted life years (QALYs). Costs associated with the intervention, medications, health care, social care, lost productivity and patient expenses were measured. A cost–utility analysis from the societal perspective was conducted after 26 weeks. A secondary analysis considered the cost–utility of RAFT from the healthcare payer perspective. 175 patients were randomised to receive RAFT and 158 to usual care. After 26 weeks, fatigue impact was improved by -0.59 BRAF-NRS units (95%CI -1.11 to -0.06) for RAFT, sustained for 2 years. RAFT was associated with a small increase in QALYs (0.008; 95%CI -0.008 to 0.023), and costs (£434; 95%CI -389 to 1258). The incremental cost-effectiveness ratio (ICER) was £55202 per QALY for the primary analysis, and the probability that RAFT was cost-effective was 0.35 at a societal willingness-to-pay threshold of £30000 per QALY. From the perspective of the healthcare payer, the ICER was £34878 per QALY, with the probability of cost-effectiveness 0.46. There was a weak to moderate negative correlation between the change in utility score and the change in fatigue impact at 26 weeks (Pearson correlation coefficient, -0.4, p<0.001). RAFT is unlikely to be cost-effective compared with usual care using conventional thresholds in the UK. However, the sustained improvement in fatigue impact should be considered given the paucity of treatment options. These findings add to the ongoing debate about whether QALYs capture fatigue in long-term conditions.