Background/Purpose: Differences in access to care that influence the timing and quality of treatment interventions may create outcome inequities for Aboriginal patients with inflammatory arthritis. Our study compares Aboriginal and Caucasian patients in disease presentation, treatment strategy, and outcomes over five years.

Methods: Participants were enrolled in a prospective multi-center early arthritis cohort, and treated with routine care. Inclusion criteria for the present study were < 1 year symptom duration, self-identified as Aboriginal or Caucasian, and completion of >1 follow-up visit. Baseline demographics, clinical characteristics, and therapy escalation for moderate (DAS28 >3.2) or high (DAS28 >5.1) disease activity states (defined as any of increased dose of methotrexate, addition of a DMARD, and/or addition or switching biologic) were compared using standard descriptive statistics. The frequency of remission and use of DMARD, biologic and steroid therapy were compared between groups. Mixed-model repeated measures and Poisson regression analysis were used to determine rates of change for disease activity measures over five years, with adjustment for baseline demographics and disease activity measures.

Results: The study sample included a total of 2173 patients (Aboriginal n=100; Caucasian n=2073), 70% female with mean(sd) age of 54(15) years, symptom duration of 179(91) days, and baseline DAS28 4.87(1.48). Differences in current smoking status, body mass index, education, and household income disfavoured Aboriginal patients (Table 1, all p<0.01). Aboriginal patients were more frequently seropositive and less likely to have erosions at baseline, but did not differ in symptom duration, number of comorbid conditions, nor baseline HAQ and DAS28 scores. Therapy was escalated at ~50% and 60% of visits where patients were in moderate and high disease activity states respectively, with no differences between groups in the frequency or type of strategy used (i.e. use of oral steroids, combination DMARD therapies or biologics). DAS28 remission was less frequent in Aboriginal patients at all visits up to 36 months (3 months 16% vs 30%; 12-months 16% vs 50%; 36-months 40% vs 59%, p values <0.01). This was driven by higher values for all DAS28 components. In particular, swollen joint counts in Aboriginal patients improved at a significantly slower rate (slope difference between groups p=0.029), and patient global scores did not improve significantly (p=0.115) in Aboriginal patients.  

Conclusion: We observed differences in disease phenotype in Aboriginal patients, and worse disease outcomes despite having a treatment escalation strategy similar to the Caucasian population. This may reflect disparities in socioeconomic status and differences in environmental exposures associated with worse disease outcomes.