Abstract / Summary
Objective: To assess whether the sex of the treating rheumatologist is associated with longitudinal disease activity trajectories in patients with psoriatic arthritis (PsA). Patients and Methods: We conducted a longitudinal registry-based study including 712 patients with PsA followed from 26 September 2014 to 23 December 2019 and classified by rheumatologist sex at baseline. The primary outcome was change from baseline in 28-joint disease activity score based on C-reactive protein (DAS28-CRP) (n=588) and ankylosing spondylitis disease activity score based on C-reactive protein was secondary (n=452). Missing baseline covariates were multiply imputed. Baseline imbalances were addressed using inverse probability of treatment weighting based on outcome-specific propensity scores. Longitudinal associations were estimated using inverse probability of treatment weighting-weighted generalized estimating equations with spline modeling of time and pooled using Rubin’s rules. Results: Patients (410/712 [57.6%] managed by male and 302/712 [42.4%] by female rheumatologists) had a median follow-up of 3.4 years. At 12 months, the pooled difference in ΔDAS28-CRP (women-men) was -0.08 (95% CI, −0.37 to 0.20; P =.577). Differences were larger at 24 months (−0.382; 95% CI, −0.62 to −0.14; P =.002) than at 12 months, whereas the 36-month estimate remained directionally similar (−0.26; 95% CI, −0.54 to 0.01; P =.061). The 24-month difference coincided with larger reductions in 28-joint tender joint count (−0.85; 95% CI, −1.53 to −0.17) and 28-joint swollen joint count (−0.62; 95% CI, −1.18 to −0.05). No between-group differences were observed for ankylosing spondylitis disease activity score based on C-reactive protein. Estimates were generally similar across sensitivity analyses. Conclusion: In this real-world PsA cohort, rheumatologist sex was associated with a modestly greater mean DAS28-CRP improvement at 24 months, mainly driven by joint counts.