OP0282 COST-EFFECTIVENESS ANALYSIS OF A CAFASPA REFERRAL MODEL FOR AXIAL SPONDYLOARTHRITIS

Background: Chronic low back pain (CLBP) poses a significant individual and socio-economic burden. A substantial amount of patients with CLBP have axial spondyloarthritis (axSpA), but early recognition of these patients is difficult for general practitioners (GPs). Guidelines form primary care and secondary care differ in criteria for referral recommendation. The Dutch primary care guideline is restrictive in referring CLBP patients to secondary care whereas ASAS recommend to refer CLBP patients having at least 1 axSPA feature1. Therefore several referral models have been developed to assist GPs. Although the validated CaFaSpA referral model2 is able to identify CLBP patients at risk for axSpA, its cost-effectiveness is yet unknown and essential before implementation in daily clinical practice. Objectives: Primary objective to assess the cost-effectiveness of the CaFaSpA referral model for axSpA in primary care. Secondary objective to evaluate the costs made for screening by following the CaFaSpA vs ASAS referral model. Methods: A clustered randomized controlled trial was performed with GPs as clusters. Clusters were randomized into the intervention (CaFaSpA referral, CS) or usual care (UC). Cost-effectiveness analysis from a societal perspective was performed to compare the CS and UC. Clinical outcomes were disability (Roland-Morris Disability Questionnaire (RMDQ)) and health-related quality of life (EuroQol (EQ-5D)) after 12 months. Direct (Medical Consumption Questionnaire IMCQ) and indirect healthcare (Productivity Cost Questionnaire IPCQ) costs were evaluated. Complete case analysis was performed. Incremental cost-effectiveness ratios (ICERs) were calculated for both clinical effects. Fictive costs according to the Dutch standard prices were assessed if the ASAS guideline would be followed (screening costs)3. Results: Of all 679 patients sixty-four percent were female and mean age was 36 (SD) years. In the CS 333 patients were included and in the UC. Non-significant differences in clinical outcomes were for RMDQ: 0.78 (95% CI: -0.38-2.07) and for EQ5D 0.03 (95% CI: -0.04-0.11). Costs were significantly higher in the UC group €19,748 (95% CI: € 15,327-25,022) vs CS € 14,169 (95% CI: € 10,723-18,066). Productivity loss was the largest contributor to the total costs (CS group: 62%, UC group: 96%). The majority of the bootstrapped ICERs presented were located in the south-eastern quadrant of the cost-effectiveness planes (Figure 1a and 1b), indicating that the CS is cost-effective. The ICER for RMDQ was €-5,579, indicating that per point improvement on the RMDQ the intervention saved €5,579. The difference in QALY’s between the CS and UC was very small resulting in a large ICER of €16,9583. The fictive screening costs by using the ASAS referral advice, i.e. referring 85% of 679 patients, results in €876 per patient. The total screening costs per patient by using the CaFaSpA model, i.e. referring 60% of 679 patients is €618. Conclusion: Although the clinical effects between the CaFaSpA referral strategy and usual care were comparable, the CaFaSpA referral strategy resulted in a better cost-effectiveness. Lower costs were mainly driven by the increased productivity. References: [1]Poddubnyy D et al. Ann Rheum Dis 2015;74:1483–7. [2]van Hoeven L et al. PLoS One 2015; 22;10(7):e0131963. [3]van Hoeven L et al. Ann Rheum Dis 2015;74(12):e68. Disclosure of Interests: None declared