Background/Purpose: Early vs. delayed referral/start of therapy within 3-4 months has been shown beneficial for outcomes in rheumatoid arthritis (RA) (Lard et al. AM J Med 2001, Gremese at al. ARD 2013, Ehrmann Feldman et al. Rheumatology 2013). However, in a real world setting, this applies for a minority of RA patients. In the FIN-RACo study, a delay of 4 months significantly influenced a 2-year remission rate in the monotherapy arm (remission rate was 11% vs 35%) while in the combination arm, 42% were in remission regardless of a delay indicating that early intensive therapy may overcome a delay (Möttönen et al. A&R 2002). Therefore, our aim was to study whether a delay of starting therapy in early RA affects long-term outcomes in a clinic with a T2T approach including preferring methotrexate based combinations over monotherapy, long-term low dose glucocorticoids over bridging, intra articular glucocorticoid injections to all swollen joints at every visit, patient education by specialist nurses, and routine monitoring including disease activity and self-reported outcomes.

Methods: A clinical database in a district of 275.000 population was analyzed for patients with a new diagnosis of RA in 1993-2013. Variables included demographics, clinical course, outcomes, medications and the date of first symptoms of RA, a question which was asked from the patient at the first visit and recorded in the database. Duration of symptoms (delay) at initiation of therapy was categorized as 0-3mo, 4-6mo, 7-12mo, 13-24mo, and >2yr. Outcome variables and medication data were available in 60-68% of patients at a mean of 7.3 years after diagnosis.

Results: In 1993-2013, 2374 patients (mean age 56yr, 67%F, 59%RF/CCP+) were diagnosed and treatment started with a median delay of 5.5mo (4.5mo in 2005 to 7.4 in 2003 with no trend of delay over years). Existing musculoskeletal disease and younger age were associated with longer delay, adjusted for sero+/-, gender, and year of diagnosis. Overall, 32%, 24%, 24%, 10%, and 10% of patients had a delay of 0-3mo, 4-6mo, 7-12mo, 13-24mo, and >2yr, respectively (in 98 patients delay data was missing).

After a mean follow-up of 7 years, the mean swollen and tender joint count (28JC) was <1, the mean ESR and CRP were normal, symptom level was low, functional capacity well maintained, and 66% of patients met the DAS28 remission (Table). No significant differences in outcomes and treatments were observed between the delay groups.

Conclusion: Clinical outcomes were good after 7 years in patients with early RA who were treated extensively using a T2T approach, including 66% of patients in DAS28 remission. Delay of treatment start did not influence outcomes in this clinic. Further analyses will include radiographic outcomes and work disability.