OBJECTIVES: Several international studies suggest that the time between symptom onset and DMARD initiation in RA patients is longer than is considered optimal. We sought to assess the health economic impact of this delay in an Australian context. METHODS: The delay in DMARD initiation was estimated from a 2005 study of 96 Australian RA patients referred to one public and four private rheumatology practices. RA-associated utilities and costs were sourced from published data. Patients not taking and taking DMARD therapy were assumed to have utilities of 0.443 and 0.543, respectively. The annual direct costs of RA, excluding DMARDs, was AUD $3780, and of DMARD therapy was $2658. It was conservatively assumed that DMARD therapy did not reduce non-DMARD RA costs. RESULTS: In the 2005 study, the mean time from symptom onset to initiation of DMARD therapy was 1.48 years. Over this time a mean of 0.65 QALYs would have been lived per patient and $5579 of direct healthcare costs incurred. Had DMARDs been commenced at symptom onset, 0.80 QALYs would have been lived per patient, and $9503 of direct healthcare costs incurred. Hence early initiation of DMARDs would have saved 0.15 QALYs at a cost of $3924 per person, equating to an incremental cost-effectiveness ratio (ICER) of $26,583 per QALY saved. An additional $3400 could be spent per patient to reduce the time to DMARD initiation before the ICER breached the arbitrary cut-off of $50,000 per QALY saved. Our analysis was conservative in it did not consider the long-term health and cost savings associated with avoidance of permanent joint damage. CONCLUSIONS: The considerable delay in the initiation of DMARD therapy among patients with RA leads to significant health loss. Reducing the time to initiation of DMARDs represents a cost-effective means of reducing the burden of RA.