Delay to diagnosis in axial spondyloarthritis: are we improving in the UK?: Fig. 1

Sir, Diagnostic delay is a major challenge in axial spondyloarthritis (axial SpA)—including AS—with an extended interval between symptom onset and diagnosis, generally reported as 8–10 years [1, 2]. Recent studies have suggested that this delay may be improving [3]; however, concerns have been raised about the methodology used to support these conclusions [4]. The 2009 Assessment of SpondyloArthritis international Society classification criteria for axial SpA formally recognized the utility of MRI in diagnosis [5]. Given that MRI changes emerge earlier than radiographic changes, the expectation is that this would reduce the diagnostic delay. In order to determine whether there had been any recent improvement in the interval between symptom onset and diagnosis, we reviewed routinely recorded clinical data held on axial SpA patients attending two large UK secondary care centres as part of an assessment of service provision. We included 1193 patients with a physician-verified diagnosis of axial SpA (including 800 patients with AS) who had both date of symptom onset and date of diagnosis recorded in the notes: 640 patients from Bath and 553 from Norwich. There was a significant increase in the number of new diagnoses during the past 5 years (P < 0.001); overall, 289 patients were diagnosed between 2009 and 2013, representing a 51% increase over the number diagnosed in the preceding 5 years (n = 191). The mean (s.d.) delay was 8.53 (9.04) years (95% CI 8.02, 9.05) and median delay 5.0 years (interquartile range 2–12). For patients diagnosed in the past 5 years, the mean delay to diagnosis was 9.39 years (P = 0.097). Overall, 30.3% (n = 361) were diagnosed within 2 years of onset of symptoms, 21.0% in 3–5 years, 19.9% in 6–10 years, 19.2% after 11–20 years and 9.6% after 20 years. The range for diagnostic delay was 0–54 years. Figure 1 depicts the cumulative distribution in age at symptom onset and age at diagnosis, comparing patients diagnosed before and after 2009. The cumulative distribution in age at symptom onset and age at diagnosis Although previous studies have always quoted the mean delay, the non-normality of the data raises questions about the appropriateness of using the mean, as almost one-third of patients were diagnosed within 2 years of symptom onset and >50% by 5 years. We therefore believe the median delay to be more informative. The median delay has remained stable at 5.0 years for patients diagnosed between 1999–2003, 2004–08 and 2009–13. Analysis of 3-year groupings produced similar results, with a median delay of 5.0 years for patients diagnosed between 1999 and 2001, 2005 and 2007, 2008 and 2010 and 2011 and 13, with a discrepancy in the years 2002–04 when the median delay was higher, at 7.0 years. Detailed clinical characteristics were available for the Norwich cohort, and subgroup analysis showed that the only predictors of a shorter delay to diagnosis were the presence of peripheral arthritis (median delay 4.0 vs 6.0 years, log rank test P = 0.025; mean 7.63 vs 9.40 years, P = 0.045) and IBD (median delay 4.0 vs 6.0 years, P = 0.024; mean 6.45 vs 9.17 years, P = 0.012). The presence of uveitis was associated with a longer median (10.0 vs 5.0 years; median test P = 0.005) and mean delay to diagnosis (10.33 vs 8.41; P = 0.033), although this reflected divergence between the two survival distributions at the mid-point only; overall, they were not significantly different (log rank test P = 0.073). Male gender (median delay 5.0 vs 6.0 years, P = 0.077; mean 8.27 vs 9.43 years; P = 0.097) was also non-significantly associated with a shorter delay to diagnosis. We have identified a significant increase in new cases of axial SpA within the past 5 years. Potential explanations for this include the following: increased awareness amongst health-care professionals; adoption and implementation of the Assessment of SpondyloArthritis international Society classification criteria; and more widespread use of MRI with appropriate sequencing. However, we cannot specifically say whether more of the recent diagnoses were made on MRI scans because these data were not available. Why has the delay not improved? We considered a number of possible explanations. Firstly, axial SpA remains a relatively uncommon cause of a very common symptom: 60–80% of the general population report back pain at some point in their lives [6]. General practitioners have difficulty distinguishing inflammatory-type back pain from other types of back pain and are often unaware of the extra-articular features, so patients may not be investigated appropriately or referred to rheumatologists [7]. The observation that patients with a peripheral arthritis had a shorter delay to diagnosis may be because general practitioners have been constantly reminded about the importance of early referral for patients with swollen joints via the early arthritis initiative. Secondly, although the National Institute for Health and Care Excellence recommends the use of MRI for persistent non-specific low back pain [8], routine MRI scans do not generally include the SI joints or sequences specific for detecting inflammation. In conclusion, delay to diagnosis has not yet improved and remains a huge challenge for patients with axial SpA. There is still a need for further targeted education of health-care professionals in order to address the issue. Rheumatology key message The delay to diagnosis of axial SpA has not yet improved in the UK. Funding: No specific funding was received from any funding bodies in the public, commercial or not-for-profit sectors to carry out the work described in this manuscript. Disclosure statement: The authors have declared no conflicts of interest.