Objectives To study the incidence of suicidal ideation in patients with SLE and its relationship with anxiety/depression score, disease activity and organ damage

Methods Consecutive patients who fulfilled the ACR criteria for SLE were recruited for a questionnaire study on suicidal ideation, which was assessed by:(1) Three questions on suicidal thoughts/plans in the past month; and(2) The validated Chinese version of the Beck Scale for Suicidal Ideation(BSSI), which consists of 19 questions that assess suicidal intention (each scores 0-2 and the final score is the summation of the 19 questions[0-38 points]). The higher the score is, the greater is the suicidal intention. Anxiety/depressive symptoms were assessed by the Hospital Anxiety and Depression(HAD) scale (0-21 points for each of depression and anxiety). Disease activity of SLE was assessed by SLEDAI and physicians’ global assessment(PGA), whereas organ damage since SLE diagnosis was assessed by the ACR SLICC damage index(SDI). The BSSI score was correlated with the demographic and clinical features of the participants, HAD, SLEDAI and SDI score. Linear regression models were established to study the independent factors associated with suicidal intention.

Results 319 SLE patients were studied (94% women). The mean age at SLE onset was 30.2±12.2 years and the disease duration was 9.0±7.0 years. 43(13%) patients had clinically active SLE (SLEDAI score of ≥5 points). 106(33%) patients had organ damage (SDI score≥1 point). 29(9%) patients had suicidal thoughts and 4(1%) patients had suicidal thoughts together with solid plans within 1 month of this study. 39(12%) patients had previous suicidal thoughts/attempts whereas 18(6%) had documented suicidal attempts. The mean BSSI score was 1.37±3.62 points (range0-24). Compared to those without suicidal thoughts within 1 month of study, patients with suicidal thoughts on direct questioning had significantly higher BSSI scores (6.0±6.1vs0.89±2.9points;p<0.001), HAD-anxiety (13.1±5.2vs5.70±4.2;p<0.001) and HAD-depression (12.4±4.3vs4.50±3.9;p<0.001) scores. SDI damage score was higher in those with suicidal thoughts but the difference was not statistical significant (1.2±1.6vs0.64±1.2;p=0.10). Linear regression analysis revealed that the total BSSI scores (suicidal intention) correlated significantly with HAD-depression (Beta0.52;p<0.001), HAD-anxiety (Beta0.47;p<0.001), increasing age (Beta0.15;p=0.009), male sex (Beta-0.13;p=0.02), previous suicidal attempts (Beta0.17;p=0.003) and SDI score (Beta0.21;p<0.001) but not with SLEDAI score (Beta0.05;p=0.44). In a regression model, only HAD-depression score (Beta0.30;p<0.001), HAD-anxiety score (Beta0.28;p<0.001), increasing age (Beta0.12;p=0.02) and SDI damage score (Beta0.11;p=0.03) correlated positively and significantly with the total BSSI score.

Conclusions In this study, suicidal ideation in the preceding month occurs in 9% of patients with SLE, whereas previous suicidal thoughts/attempts are present in 12% of participants. The intention of suicidal attempt is stronger in older patients, men, and those with more end organ damage, anxiety or depressive symptoms. Suicidal intention does not correlate with disease activity. Further study on the contribution of socioeconomic factors to suicidal ideation in SLE patients is in progress.

Disclosure of Interest None Declared