Background: Juvenile idiopathic arthritis (JIA) is a heterogeneous group of chronic inflammatory arthritides with onset before 16 years of age. Its wide clinical spectrum makes early recognition difficult and may delay referral. We characterised the presentation and diagnostic pathway of non-systemic JIA and examined factors associated with time to diagnosis.
Methods: Retrospective observational study of all patients diagnosed with JIA before 16 years of age (ILAR criteria) who began follow-up at a tertiary Paediatric Rheumatology referral centre between January 2010 and June 2025. Systemic JIA (n=14), inflammatory bowel disease-associated arthritis (n=6), other arthropathies (n=2) or insufficient clinical data (n=47) were excluded. RF-positive and RF-negative polyarticular JIA were combined for reporting. Statistical analysis was performed using IBM SPSS Statistics version 28.
Results: A total of 148 patients were included; 94 (63.5%) female. Oligoarticular JIA was the commonest subtype (43.2%), followed by enthesitis-related arthritis (ERA; 27.1%) and polyarticular JIA (14.9%). Median age at symptom onset was 9.0 years (IQR 3.0-13.0). Presenting features were pain (81.1%), joint swelling (77.7%), stiffness (50.7%) and limping (39.2%). ANA was positive in 43.9% and rheumatoid factor in 3.4%; 47.5% of ERA patients were HLA-B27 positive.
Before specialist assessment, 74 patients (50.0%) had been seen by Orthopaedics, of whom 22 (29.7%) were referred to Paediatric Rheumatology and 9 (12.2%) underwent arthrocentesis; 92 (62.2%) attended the Emergency Department, of whom 42 (45.7%) were referred. NSAIDs had been prescribed in 77.7% and antibiotics in 7.4%. Imaging preceded specialist assessment in 92 patients (62.2%): radiography in 52/92 (56.5%) and joint ultrasound in 45/92 (48.9%). Median waiting time from referral to assessment was 22.5 days (IQR 10-33). The commonest immediate referral sources were another hospital (32.4%) and the Emergency Department (28.6%); 24.5% came from primary care.
Median time to diagnosis was 4 months (IQR 2-12). No association was found with age at onset, sex, family history of autoimmune disease, degree of urbanisation, distance to the referral centre, number of affected joints, previous NSAID use or prior orthopaedic assessment. Time to diagnosis was shorter in patients with uveitis at presentation (1.75 vs. 5.0 months, p=0.014), prior joint ultrasound (3.0 vs. 6.0 months, p=0.028) and positive ANA (3.0 vs. 9.0 months, p<0.001). Uveitis at any time during follow-up was more frequent in ANA-positive patients (32.3% vs. 8.4%, p<0.001).
Conclusions: Time to diagnosis was not explained by demographic, geographic or care-pathway variables, including distance and urbanisation, and the interval from referral to specialist assessment was short. The factors associated with earlier diagnosis - uveitis, ANA positivity and prior ultrasound - largely identify patients in whom rheumatic disease was already suspected and are confounded by subtype, rather than modifiable steps. The bottleneck appears to lie upstream, in initial clinical recognition: half of the patients were assessed by Orthopaedics but fewer than a third of these were referred, and only a quarter reached the clinic from primary care. Awareness and referral criteria outside Paediatric Rheumatology are the more plausible targets. This single-centre study was not powered to exclude moderate associations.

