Background/Purpose: It is generally thought that the early treatment of patients with rheumatoid arthritis (RA) leads to improved patient outcomes over time. Our study aims to better understand if early treatment alone is sufficient to achieve these goals and assess whether measures such as DAS, joint count and perceived severity are sufficiently sensitive for this purpose.

Methods: We used data collected as part of an online treatment survey conducted among a panel of 500 European rheumatologists between April 2008 and May 2014 across EU5 (Fr, Ge, It, Sp, UK). In order to assess whether earlier treatment has an effect on patients current disease status we ran a linear regression using patients’ current DAS, current disease severity (as perceived by their physicians) and the number of joints affected by their disease as dependent variables (DV) and included the following as independent variables (IV):

• Time from cDMARDs initiation until first bDMARDs initiation (β₁)
• Time from diagnosis until cDMARDs initiation(β₂)
• Time from diagnosis until bDMARDs initiation(β₃)
• Severity at diagnosis(β₄)
• Number of cDMARDs before bDMARDs initiation(β₅)
• Current bDMARDs (Brand) (β₆)

We focused our analyses on patients diagnosed post 1998 to ensure patients had access to bDMARDs and only considered those currently prescribed their first line of bDMARD therapy to avoid any confounding effects caused by patients’ treatment history. We also accounted for possible differences in patients’ disease severity, both at diagnosis and at the time of their first bDMARD treatment, as well as the length of time they had been on their current bDMARD treatment.

Results: We considered data from a sample of 43,769 patient record forms and on average, our patients had a DAS of 2.9 with 85.8% classified as having moderate to severe RA. In addition, the mean time from diagnosis to cDMARD initiation and bDMARD initiation was 12.1 and 45.5 months, respectively.

Preliminary results from the regression analysis show that we can explain very little of the variability in our DV with an adjusted R square of just 5% for DAS, 2% for current severity and 5% for the number of affected joints. Therefore, we considered the coefficients of each model to measure the effect of our IV.

β₁ and β₂ have a low positive effect on DAS and current disease severity however, the choice of Enbrel, Humira and RoActemra as a first bDMARD has a strong negative effect on these two DV. In addition, β₂ and β₅ both have a strong positive effect on patients’ number of affected joints, while β₁has a low positive effect.

Finally, we see that patients who have been on their first bDMARD for a shorter period of time (up to 2 years) better explained the variation seen in our DV.

Conclusion: Our analyses demonstrate that simply initiating treatment with cDMARDs and bDMARDs early on in patients’ disease is not enough to optimise patient outcomes. Instead, early treatment must be combined with close monitoring and aggressive step-up treatment strategies such a treat-to-target to maximise patients’ response to treatment and control their disease. In addition, our data also suggest that the impact that early treatment may have could be limited in time with disease statuses becoming more similar as the duration of bDMARD therapy increases.

---

« Back to 2014 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/early-treatment-in-rheumatoid-arthritis-and-its-effect-on-patient-outcomes/