Background/Purpose:   Axial spondyloarthritis (axSpA) is diagnosed an average of 9 years after symptom onset (1), partly because inflammatory back pain (IBP) can be difficult for primary care providers to differentiate from mechanical back pain (MBP). Untreated, axSpA can result in irreversible structural damage, functional loss and reduced quality of life. We evaluated a stratified screening process for early axSpA identification. Objectives: 1) measure time to diagnosis by a rheumatologist; 2) measure referral wait times from primary care physician (PCP) to rheumatology screening; 3) determine the incremental precision and accuracy of a stratified screening process from primary to rheumatology care.

Methods:   Adults (18+ years) with low back pain visited their PCP or a dedicated interprofessional back pain model of care (www.isaec.org) and underwent primary screening, consisting of a standardized clinical assessment that incorporated ASAS criteria for IBP. At the primary care level, patients with back pain >3 months duration, onset age <50 years, with at least one other IBP feature were referred for a secondary screen by a physiotherapist with advanced rheumatology training. This screen included standardized history, physical examination and baseline investigations. The likelihood of axSpA risk (vs MBP) was determined at each screening level and defined as low, medium, or high. Precision and accuracy of primary and secondary screens were measured against the clinical judgement of a rheumatologist with axSpA expertise. The utility of HLA-B27 was assessed as an independent screen. Sensitivity, specificity and predictive values were calculated.

Results:  410 patients underwent primary and secondary screening over a 3-year period. Mean age: 36.9 years (±9.8); 55% female; average back pain duration 7 years (±7.2).  HLA-B27 was present in 14.4% of patients. Mean time from onset of back pain to diagnosis for patients with medium or high risk (as per rheumatologist) was 6 years (±6.3), with a median delay of 3 years. Median wait time from primary to secondary screen was 22 days.  AxSpA risk assignment by rheumatologist was: 63.6% MBP or low risk and 36.4% medium or high risk, with 18.0% of all patients receiving a final diagnosis of axSpA. HLA-B27 performed poorly as an independent screen (28% sensitivity). The best combination of sensitivity, specificity, and predictive values was found with the secondary screen (see table 1).  

Conclusion:  The inclusion of a secondary screening process utilizing a stratified interprofessional model can shorten time to diagnosis, with high precision and accuracy in patients with axSpA.  Findings provide a platform to bridge the gap between onset of back pain and diagnosis and thereby improve long-term outcomes in this patient population.

References: (1) Feldtkeller E, Bruckel J, Khan M.  Scientific contributions of ankylosing spondylitis patient advocacy groups.  Curr Opin Rheumatol. 2000;12:239-47