PS32 - Exploring Suicidal Behavior in Clinical Settings: Results on Suicide Attempt From the SURVIVE Study
Lifetime Predictors of Suicidal Attempts: Childhood Maltreatment, Depressive Mood and Affective DisordersIntroduction: Suicide is a complex phenomenon that can occur at different stages in life, and it is the result from the interplay of multiple variables. Suicide is ranked among the foremost causes of death in the youth demographics. Recently, childhood maltreatment and mental disorders at this age seem to be among the most influential risk factors. Whereas, in adult population at risk, depressive and anxious symptoms are among the most replicated contributors related to suicide attempts. To date, a recent attempt is the best predictor of suicide consumption. For each death by suicide, there are currently 20 attempts. Secondary prevention strategies arise in this context with the goal of once suicidal behaviour has been detected due to an attempt, we can focus on this at-risk population to prevent relapses. Objectives: The aim of the present work is to explore different risk factors among a Spanish cohort of children and adults who have previously attempted suicide. First, we explored the interplay between childhood maltreatment, mental disorders and suicide in childhood and adolescence. Second, we analysed the distribution of anxious and depressive symptoms, perceived health-related quality of life and their interconnection in a sample of adult suicidal attempters. Last, we compared possible risk factors when single versus multiple suicide attempts were performed in the same adult cohort. Methods: Patients were recruited along nine Spanish Hospitals. At least, one suicide attempt was performed to be included in the SURVIVE cohort study. For each sub-study we used different statistical analysis. For the first sub-study we used regression analysis and moderation analysis in the children and adolescent's sub-sample. In the second sub-study a network analysis was used including subnetworks comparing younger and older adults; and women and men subsamples. Third sub-study objective was tested by a machine learning technique. Here we performed two different models. First model was a raw variable model where all conditions were included. Second model was conducted by means of diagnostic classifications according to major categories, e.g. Affective disorder diagnoses. Both models included were tested with the adult's sample. Main tools used for the purpose of the whole study were the Columbia-Suicide Severity Rating Scale, the Childhood Trauma Questionnaire, the Mini International Neuropsychiatric Interview, the Generalized Anxiety Disorder-7, the Patient Health Questionnaire, and the EuroQoL-5D. Sociodemographic and clinical variables were also collected for each participant. Results: For the first sub study, results revealed that higher scores of childhood maltreatment (all types except for Emotional neglect and Sexual Abuse) were associated with a higher number of attempts in the presence of a mental disorder, particularly Panic Disorder, General Anxiety Disorder and Major Depressive Disorder and were more likely to exhibit more lethal suicidal behaviour and to report a higher number of previous attempts (B = 1.17 [95% confidence interval (CI) 0.36 - 1.99, p = 0.005]; B = 1.04 [95% CI 0.32 - 1.76, p = 0.005]); and B = 0.96 [95% CI 0.34 - 1.58, p = 0.002], respectively). Additionally, those with higher childhood maltreatment scores reported less severe characteristics of suicidal ideation, when presenting along with an anxiety disorder. For the second sub study, depressed mood was the symptom with the greatest influence in the overall network (presented consistently high centrality measures), followed by anxious symptoms such as feeling nervous, worrying, restless, and having difficulties to relax (better connected the different communities of the network as denoted by the one- and two-steps bridge expected influence). Perceived general health was associated with increased suicidal ideation in the whole sample (establishing a connection with the community related to depressive symptoms). For the third sub study, results showed that best predictors for a suicide re-attempt were a diagnosis of Bipolar Disorder type II (importance of 12.6%) or binge-eating disorder (importance of 10.9%) or Schizophrenia (importance of 5.5%); whereas for the second model: i) a diagnosis of an eating disorder (importance = 18.4%); ii) to be born in Africa (importance = 12.3%); iii) a diagnosis of an affective disorder (importance = 8.4%); iv) being employed (importance = 6.4%); v) a diagnosis of a schizophrenia spectrum disorder (importance = 6.3%); and vi) a diagnosis of substance use disorder (importance = 5%). Conclusions: Our studies concluded that childhood maltreatment increases the risk of suicidal behaviour in young population. In adults, depressed mood was the symptom with the greatest influence in the suicide-symptomatology network in comparison to anxious and perceived health related quality of life. Also, most common predictors of suicide re-attempts are affective disorders (specifically bipolar disorder type-II) and eating disorders, as well as other less explored socioeconomic variables such as birthplace and employment status. These findings offer a potential key to identify and prevent suicide along all lifetime showing most common predictors of suicide attempts in both populations at risk, adolescents and adults.