The value of referral letter information in predicting inflammatory arthritis--factors important for effective...

The aim of this study is to identify factors from referral information predictive of patients with inflammatory arthritis (IA) requiring early review. Four hundred twenty-six consecutive rheumatologist-triaged referrals from February to June 2012 were retrospectively reviewed to identify patients with rheumatologist-diagnosed IA correctly triaged for review within 2 weeks from referral date. Information from referral was analyzed descriptively followed by univariate logistic regression adjusted for age and sex to identify predictors of IA. Of the 108 patients with rheumatologist-confirmed diagnoses seen within 2 weeks, 76 patients (70.4%) were correctly triaged with 44.7% having rheumatoid arthritis (RA); 9.2%, psoriatic arthritis; 9.2%, spondyloarthritis; and 18.4%, undifferentiated inflammatory arthritis. The majority were females (63.2%), with median age of 52.8 years (Q1; Q3 38.4; 61.3) with referrers indicating presence of morning stiffness in 71.4% and symmetrical distribution in 74.6%. Five or more joints were involved in 65.7% with suspected metacarpophalangeal joint (MCPJ) (44.7%) or proximal interphalangeal joint (PIPJ) (59.6%) involvement. Of the referrals with laboratory results, erythrocyte sedimentation rate (ESR) was raised with median 43.5 mm/h (Q1; Q3 24.8; 77.5) and normal median uric acid of 312.5 μmol/L (Q1; Q3 249.5; 363.5). Univariate analysis revealed that presence of ≥5 joints affected (p = 0.001), symmetrical distribution (p = 0.006), MCPJ (p = 0.003), PIPJ (p = 0.003), and elevated ESR (p = 0.001) were predictive of IA after adjustment for age and sex. Specific information including number, pattern, and location of joint involvement with relevant laboratory investigations should be included in referral letters to assist with effective triaging of patients with IA.