231 Pre-treatment anxiety and depression levels are correlated with less improvement in self-reported disease activity in rheumatoid arthritis patients treated with biological therapy: results from the BRAGGSS cohort

Abstract

Background: We have previously reported that psychological factors, including anxiety and depression, correlate with the subjective components (i.e. patient global visual analogue scale (PG-VAS), tender joint count) of the DAS28 in patients about to begin biologics. The current study aimed to investigate whether pre-treatment depression and anxiety is predictive of future biologics response. Methods: The study included patients from the Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate (BRAGGSS) Cohort. Demographics and clinical data were collected at pre-treatment and following six months on-drug. PG-VAS was measured on a 100mm scale. Anxiety and depression were measured using the Hospital Anxiety and Depression Scale (HADS). The relationship between anxiety or depression with improvement in the DAS28 and its sub-components was analysed using linear regression, adjusted for pre-treatment HAQ, disease activity, gender, BMI and synthetic DMARD use. Improvement in HADS at six months was compared between EULAR good and non-responder groups using a two sample t-test. Results: 2919 patients were analysed. 76% were female. At baseline, mean age was 57.3 years (SD12.3), median disease duration was 6.9 years (IQR 2.8, 14.0) and mean DAS28 was 5.73 (SD 0.86). 83.1% of patients received synthetic DMARD concurrently. Biologics included etanercept (n=942), infliximab (n=55), adalimumab (n=591), tocilizumab (n=267), rituximab (n=438), abatacept (n=118), golimumab (n=112), certolizumab (n=355), and other biologics (n=4). Higher pre-treatment HADS depression correlated with less improvement in the PG-VAS by six months. Higher pre-treatment HADS anxiety was associated with comparatively smaller improvements in DAS28, tender joint count and PG-VAS by six months. Importantly, HADs showed no association with swollen joint count or CRP (Table 1). EULAR goodresponders had a greater improvement in mean depression (-1.78 vs -0.03, p<0.0001) and anxiety (-1.73 vs -0.28, p<0.0001) scores compared to non-responders. Conclusion: Higher depression and anxiety scores at pre-treatment were associated with less improvement in the subjective components of the DAS28. Anxiety and depression are potentially modifiable and may be amenable to treatment as an adjunct to biologic therapy in some patients. These findings now require replication. (Table Presented).