LAUSR

Background In patients with rheumatic diseases, suicidal ideation (Si) and suicide attempt (Sa) have been observed more frequently than in the general population. Objectives 1. To determine the prevalence of Si and Sa in patients with biological therapies. 2. To evaluate possible factors associated with the Si. Methods Observational cross-sectional study in patients with intravenous BT from Day Hospital of a tertiary hospital in Madrid. Sociodemographic, disease features [weight, height, work disability (WD), diagnosis, treatment] and PRO (FiRST, FACIT-F, HAD, and GHQ-28) were collected. We defined the Si according to the answer at the 6th and 7th questions of section D of the GHQ-28. The Sa was registered according the patient`s clinical history. The categorical variables were analyzed with the Chi square test. A logistic regression analysis was performed to evaluate the possible factors associated with suicidal behavior (SB). To evaluate the variance of discrete quantitative variables between groups the Kruskal-Wallis test was used, with a post-hoc analysis with U-Mann Whitney, to determinate the difference between pairs. Results We included 321 patients, 65% women, with a mean age (range, SD) of 56 years (89-15, 14.1), and a mean disease duration (range, SD) of 16 years, DS (53-1, 9.7). The sociodemographic and clinical characteristics are described in Table 1 and 2. Of all patients, 4% had had 1 or more Sa and 11% had Si. 23% of patients had associated fibromyalgia (FiRST), 47% fatigue (FACIT-F), 27.4% anxiety (A-HAD), 16.2% depression (D-HAD) and a 48.6% a probable psychic disorder (GHQ-28). The final logistic regression model includes FACIT-F, IL, D-HAD, and first biological with Cox R2 and Snell 20.1% and Nagelkerke 35.5%. The area under the curve was 0.849 with significance p <0.0001 [95% CI (0.786 - 0.912)]. Regarding the GHQ-28 score, a significant difference was observed among the diagnostic groups (p <0.001); in post-Hoc, a lower score was observed among patients with ankylosing spondylitis p <0.001. We also observed a significant difference among the treatment groups (p <0.001) for the GHQ-28 score. In the post-hoc analysis, the Abatacept group had a significantly higher mean p = 0.042 of GHQ-28 score, however, in the multivariate analysis no significant difference was observed p = 0.416. Conclusion There is a high prevalence of Si and Sa in patients with BT. Depression, fatigue, sleep disorders, fibromyalgia and work disability were associated with a higher prevalence of SB.Table 1 Demographic characteristics No ideation/No attempt (273) Ideation (36) Attempt (12)Table 2 Clinical characteristics No ideation/No attempt (273) Ideation (36) Attempt (12) Diagnostic (%) Rheumatoid arthritis 125 (45,78) 20 (55,55) 5 (41,66) Ankylosing spondylitis 94 (34,43) 6 (16.6) 3 (25) Psoriatic arthritis 20 (7,3) 4 (11,11) 0 (0) Others 34 (12,4) 6 (16,6) 4 (33,33) Treatment (%) Infliximab 159 (58,24) 16 (44,44) 3 (25) Tocilizumab 46 (16,84) 8 (22,2) 2 (16,67) Abatacept 12 (4,3) 6 (16,67) 2 (16,67) Rituximab 49 (17,94) 5 (13,88) 5 (41,67) Others 7 (2,5) 1 (2,7) 0 (0) Previous biologic (yes) 107 (39,19) 25 (69,44) 8 (66,67) FiRST >5 (%) 49 (17,94) 15 (41,66) 9 (75) FACIT-F < 34 (%) 109 (39,92) 31 (86,11) 11 (91,67) GHQ-28 >5 (%) 114 (41,75) 34 (94,44) 8 (66,67) CGHQ-28 >12 (%) 216 (79,12) 36 (100) 11 (91,67) A-HAD >11 (%) 61 (22,34) 21 (58,33) 6 (50) D-HAD >11 (%) 26 (9,5) 20 (55,55) 6 (50) Disclosure of Interests Amparo Lopez Esteban : None declared, Roberto Daniel Gonzalez Benitez: None declared, Aurora Alonso Amigo: None declared, Iustina Janta: None declared, Juan Carlos Sanchez Zanotti: None declared, Florentina Garca Calle: None declared, Juan Carlos Nieto: None declared, Beln Serrano Benavente: None declared, Juan Ovalles: None declared, Tamara Del Ro Blasco: None declared, Alfonso Ariza: None declared, Carlos Gonzalez Consultant for: Celgene, Gilead, Janssen, Merk, Novartis, Pfizer, Speakers bureau: Celgene, Roche, UCB, Indalecio Monteagudo Sez: None declared