MAOA and MAOB polymorphisms and anger-related traits in suicidal participants and controls

MAOA and, to a lesser extent, MAOB polymorphisms have been related to aggression traits and suicidality. We aimed to investigate the role of MAOA and MAOB in suicidal versus non-suicidal participants and interactions between genetic variation and suicidal status on aggression and anger-related traits. The sample was composed of three groups: one group of suicide attempters (n = 171, males 35.1 %), one group of suicide completers (n = 90, males 57.8 %) and a healthy control group (n = 317, males 43.8 %). We examined the following markers: MAOA rs909525, rs6323, and rs2064070, and MAOB rs1799836. Anger traits were measured with the state-trait anger expression inventory (STAXI) and aggression traits with the questionnaire for measuring factors of aggression (FAF). Associations were separately examined for males and females. Variation in the three MAOA variants was associated with higher levels of anger expressed outwards (STAXI “anger-out” subscale) in male suicidal patients compared to controls (p < 0.001). In females, the C allele of rs6323 showed higher scores on the same subscale (“anger out”) (p = 0.002). Allele frequencies of the MAOA rs909525 were associated with suicidality (p < 0.007). Our findings show an association between genetic variation in three polymorphisms of the MAOA and anger traits in suicidal males and one replication for the functional variant rs6323 in females. This relationship was stronger than a direct genetic association with suicide status. Future studies incorporating endophenotypic measures of anger and aggression in suicidal participants are warranted.     

Genetic polymorphisms in the SCN8A gene are associated with suicidal behavior in psychiatric disorders in the Chinese population

Abstract

Objectives. Suicidal behavior is a serious public health problem which is partly heritable. Identifying the genes and the neurobiologic pathways relevant to suicidal behavior is important for preventative strategies. One family-based study reported an association between sodium channel voltage gated type VIII alpha (SCN8A) and suicidal behavior. In the present study, we aimed to search for SCN8A polymorphisms conferring genetic susceptibility to suicide in the Chinese population. Methods. A total of 626 subjects was recruited for the study, including 297 suicide attempters and 329 non-attempters from Shanghai, China. We conducted a case-control association analysis of five SNPs (rs10506302, rs1601012, rs4762004, rs12581041, rs17126078) within the region of SCN8A gene. Results. we found that two genetic polymorphisms showed statistically significant differences between cases and controls (rs1601012, P=0.004; rs12581041, P=0.01). Moreover, no haplotypes were significantly associated with suicidal behavior in psychiatric disorders after the false discovery rate (FDR) correction. In the analysis of schizophrenia subgroup, three genetic polymorphisms showed statistically significant differences between cases and controls (rs10506302, P=0.024; rs1601012, P=0.004; rs12581041, P=0.004). Conclusions. Our findings suggest that the SCN8A gene may be involved in the susceptibility to suicidal behavior among psychiatric disorder patients in the Han Chinese population.     

Factors associated with deliberate self-harm behaviour among depressed adolescent outpatients

This study examined whether depressed adolescent outpatients with deliberate self-harm behaviour (DSH) differed from non-suicidal depressed adolescent outpatients in depressive and anxiety symptoms, alcohol use, perceived social support and number of negative life-events. Depressed adolescent outpatients (n = 155) aged 13–19 years were interviewed using K-SADS-PL for DSM-IV Axis I diagnoses and completed self-report questionnaires. Suicidal behaviour was assessed by K-SADS-PL suicidality items. Depressed adolescents with DSH were younger, perceived less support from the family, had more severe depressive symptoms and used more alcohol than non-suicidal depressed adolescents. Adolescents with DSH and suicidal ideation or suicide attempts had more depressive and anxiety symptoms than adolescents with DSH only. Adolescents with severe internalizing distress symptoms are at risk not only for DSH, but also additional suicidal behaviour. Family interventions may be needed in the treatment of depressed adolescents with DSH.     

5-ht2c receptor and mao-a interaction analysis: no association with suicidal behaviour in bipolar patients

The serotonin 2C (HTR2C) receptor has been implicated in suicide-related behaviours, however there are not many studies to date about HTR2C and suicidality. We studied HTR2C haplotypes in suicide attempters, where our sample composed of 306 families with at least one member affected by bipolar disorder. HTR2C (Cys23Ser and a common STR in the promoter) variants were analyzed with respect to attempter status and the severity of suicidal behaviour. The X-linked haplotype analysis in relation to suicide attempt did not reveal any significant association. Furthermore, we performed a particular gene-gene interaction for the X-linked serotonergic genes (HTR2C and MAOA), and found no association among this intergenic haplotype combination and suicidal behaviour in bipolar disorder. Drs. De Luca and Tharmaligam have contributed equally to this work.     

Heavy daily smoking among under 18-year-old psychiatric inpatients is associated with increased risk for suicide attempts.

PURPOSE: To investigate the relationship between heavy daily smoking and suicidality among adolescent psychiatric inpatients in Finland. METHODS: Data were collected from 411 patients (age 12-17 years) admitted to inpatient psychiatric hospitalization between April 2001 and July 2005. The number of daily cigarettes (>15) and the time of first cigarette after waking up (within 30 min) were used as indicators of heavy daily smoking. RESULTS: After adjusting for psychiatric diagnoses an over twofold risk for suicide attempts was found among adolescents who smoked over 15 cigarettes a day. Additionally, if an adolescent also smoked the first cigarette immediately after waking up the risk was over threefold. Suicidal ideation was not associated with smoking behaviour. DISCUSSION AND CONCLUSIONS: Among adolescents with severe psychiatric illnesses, heavy daily smoking may increase the risk of suicidal behaviour independently of current psychiatric diagnosis. The degree of nicotine dependence of an adolescent should therefore be carefully assessed as part of psychiatric evaluation.     

Suicidal and self-harm behaviour associated with adolescent attention deficit hyperactivity disorder—A study in the Northern Finland Birth Cohort 1986

Background: Suicidal behaviour, i.e. suicidal ideation and suicidal acts, as well as self-harm behaviour, are relatively common among adolescents. Depression and/or female gender seem to be risk factors for suicidal behaviour. However, the role of attention deficit hyperactivity disorder (ADHD) in these behaviours is still unclear. Aim: To study the effect of ADHD on suicidal or self-harm behaviour in adolescents from a general population sample. Methods: The sample was derived from a population-based Northern Finland Birth Cohort 1986 (n = 9432). Based on the Schedule for Affective Disorders and Schizophrenia for School-Age Children, Present and Lifetime Version (Kiddie-SADS-PL) interview performed in a subpopulation (n = 457), the associations between suicidal behaviour and deliberate self-harm (DSH) and the diagnosis of ADHD were studied. Results: Compared with adolescents without ADHD (n = 169), those with ADHD (n = 104) had more suicidal ideation (57% vs. 28%, P < 0.001) and DSH (69% vs. 32%, P < 0.001). In binary logistic models, the effect of ADHD on suicidal ideation remained strong (OR = 6.1) after controlling for several other predictors. Other contributing factors in suicidal behaviour included female gender, childhood emotional and behavioural problems, concurrent depression and anxiety, and, specifically in DSH, behavioural disorder, substance abuse and strains in family relations. Discussion and clinical implications: ADHD is a risk factor for suicidal ideation and DSH. These findings in a general population sample speak for a need to target mental health interventions at children and adolescents with relevant symptoms of ADHD.     

Multimethod assessment of suicidality in adolescent psychiatric inpatients: preliminary results

OBJECTIVE: To examine agreement among multiple assessments of adolescent suicidal ideation and suicidal behavior for adolescent psychiatric inpatients, including pencil/paper checklists; structured and unstructured interviews; and adolescent, clinician, and parent reports, and to provide suggestions for the accurate and reliable assessment of suicidality in adolescence. METHOD: Participants included 153 adolescent psychiatric inpatients (54 boys, 99 girls) between the ages of 12 and 17 years. Measurement of suicidal ideation and behavior included common assessment instruments and standard clinical practices, including the Suicidal Ideation Questionnaire, NIMH Diagnostic Interview Schedule for Children, clinician interview, and parent report (Behavior Assessment Scale for Children). RESULTS: Results revealed significantly different rates of suicidality across each instrument and poor to moderate agreement between similar measures of adolescent suicidal ideation and suicidal behavior. Agreement between measures was generally best for boys, for older adolescents, and for assessments relying on a single informant. Reporters were most likely to agree on the presence of suicidality for more severely suicidal adolescents; this finding suggests that agreement in itself may be a useful marker for adolescent suicide. CONCLUSIONS: Results of this preliminary study supported the use of multiple measurement approaches when examining adolescent suicidality, particularly those that rely on clinician judgment and adolescents' own reports. Implications for future research and for clinical practice are also discussed.     

Mania,Hypomania, and Suicidality: Findings from a Prospective Community Study

ABSTRACT

We examined prospectively whether mania and hypomania are associated with an elevated risk for suicidality in a community sample of adolescents and young adults. Baseline and four-year follow-up data were used from the Early-Developmental-Stages-of-Psychopathology (EDSP) Study, a prospective longitudinal study of adolescents and young adults in Munich. Suicidal tendencies (ideation/attempts), mania, and hypomania were assessed using the standardized Munich-Composite-International-Diagnostic-Interview. At baseline, mania/hypomania was associated to a different degree with suicidality (Odds ratios [OR] range from 1.9 to 13.7). In the prospective analyses, the risk for subsequent incident suicidal ideation was increased in the presence of prior mania (38.0% vs. 14.1%; OR = 4.4; 95% CI = 1.4–13.5). No associations could be found between prior mania/hypo-mania and incident suicide attempts. The prospective analyses revealed a remarkable relationship between preexisting mania and increased risk for subsequent suicidal ideation.     

The Double-Edged Sword - Outward and Inward Directed Aggression among War Combatants

Inspired by the two-stage model of countervailing forces, we explored the mechanism underlining inward and outward aggression among ex-combatants. Israeli veterans (N = 230) reported their partner’s violence (outward aggression), suicidal ideation (inward aggression), aggressive impulses, posttraumatic stress disorder (PTSD), paranoid ideation and guilt. Suicidal ideation was related to aggressive impulses only in the presence of PTSD, or under high guilt; whereas paranoid ideation buffered these effects. Violence towards the partner, on the other hand, was related to aggressive impulses under low guilt and in the absence of PTSD. Aggressive impulses underline both suicidal ideation and partner violence. The inter-relations between PTSD, guilt, and paranoid ideation influence the manifestation of aggression and the direction it takes.     

Suicidal ideation in Huntington disease: the role of comorbidity

Huntington disease (HD) is a neurodegenerative condition characterized by cognitive impairments, motor abnormalities, and psychiatric disturbance. An increased risk for suicide has been documented. The majority of HD research has focused on cognitive and motor features of HD; the implications of psychiatric manifestations have received less consideration. Recent studies have sought to identify the stages of HD in which patients are at increased risk to experience suicidal ideation, though no study has examined possible risk factors for suicidality. The current study examines the presence of psychiatric comorbidity and its involvement in suicidal ideation. Suicidal ideation was examined in 1941 HD patients enrolled in the Huntington Study Group. Of those, 19% (N=369) endorsed current suicidal ideation. Logistic regression analyses indicated that depression/anxiety and aggression/irritability are significant predictors of suicidal ideation. In a subsample with the greatest suicidal ideation, alcohol and drug abuse were also predictive. It is recommended that all individuals with HD (specifically those with features of depression, aggression, substance abuse) have routine suicide assessment; further research is needed to understand the high rate of suicide in HD.     

Association of VNTR polymorphisms in the MAOA promoter and DRD4 exon 3 with heroin dependence in male Chinese addicts

Objective. To explore the involvement of variable number tandem repeat (VNTR) polymorphisms in the monoamine oxidase A (MAOA) promoter and exon 3 of the dopamine D4 receptor (DRD4) gene in heroin addiction modulate the vulnerability of individuals to heroin addiction. Methods. Eight hundred and ninety-four male heroin addicts without other psychiatric disorders, were recruited as subjects. Another community 180 males were selected randomly as controls. Results. The geno-distribution of the DRD4 exon 3 VNTR polymorphism in controls was in Hardy–Weinberg equilibrium (HWEχ²=0.925), but the distribution in heroin addicts was not (HWEχ²=28.35). The long-repeat alleles of the DRD4 exon 3 VNTR polymorphism were found more frequently in the heroin addicts (P=0.019). However, the long-repeat alleles of the MAOA promoter VNTR polymorphism were not (P=0.828). No interaction between these two VNTR polymorphisms was found by using multiple logistic regression analysis (P=0.261). Conclusion. The long-repeat allelic variants (>4-repeats) and 2-repeat allele of the DRD4 exon 3 VNTR polymorphism might be risk alleles for individual vulnerability to heroin addiction in Chinese men, but the MAOA promoter VNTR polymorphism does not mean that the partial dominant inherited mode might involved in the genetics of heroin dependence.     

Dopaminergic system genes in childhood aggression: Possible role for DRD2

Abstract

Excessive or deficient levels of extracellular dopamine have been hypothesized to contribute to a broad spectrum of mood, motor, and thought abnormalities, and dopaminergic system genes have been implicated in aggressive behaviour from animal and human studies. Objective. We examined selected members of the dopaminergic system genes for association with child aggression. Method. We analyzed polymorphisms in the dopamine transporter DAT1/SLC6A3, dopamine receptor DRD2, and DRD4 genes in our sample of pervasive childhood aggression consisting of 144 cases paired with 144 healthy controls, matched for sex and ethnicity. Results. Aggressive children were significantly more likely to have the at least one copy of the G allele for the DRD2 A-241G polymorphism (genotypic P=0.02; allelic P=0.01). The DRD2 rs1079598 CC genotype was overrepresented in aggressive children compared to controls (genotype P=0.04). The DRD2 TaqIA T allele (P=0.01) and the TT genotype (P=0.01) were also significantly overrepresented in aggressive children. Conclusions. Our preliminary results suggest that three polymorphisms in DRD2 are associated with childhood aggression. Future studies are required to replicate the current results and to further explore the relationship between the dopamine system and aggressive behaviour in children.     

Serotonin 1A receptor genetic variations, suicide, and life events in the Iranian population

Aim: The association of serotonin 1A receptor (5‐HTR1A) gene polymorphisms with suicidal behavior has been reported in several previous studies, but the results have been inconsistent, which might be due to ethnic differences. The aim of the present study was therefore to investigate the association between polymorphisms ?1019C>G, 47C>T (Pro16Leu) and 815G>A (Gly272Asp) and suicidal behavior, taking into account age, gender, and the presence of stressful life and loss events in 1 year prior to suicide. Methods: A total of 191 suicide victims and 218 healthy control subjects were included in the present study. 5‐HT1RA gene polymorphisms were determined on polymerase chain reaction–restriction fragment length polymorphism. Results: The distribution of ?1019C>G genotypes was significantly different in suicide victims and healthy controls (P = 0.002), and the GG genotype was associated with a significantly higher number of more stressful life and loss events in the suicide victims (P = 0.017, P = 0.037, respectively). The distribution of 47C>T (Pro16Leu) and 815G>A (Gly272Asp) genotypes was not significantly different in the suicide victims and control subjects (P > 0.05). Moreover, these genotypes were not associated with stressful life and loss events (P > 0.05). Conclusion: The frequency of the ?1019G allele in the 5‐HTR1A gene was higher in suicide victims (with stressful life events) as compared with the control group. In contrast, neither 47C>T (Pro16Leu) nor 815G>A (Gly272Asp) polymorphisms were related with suicide and stressful life events.     

Remember Me to Life

Objective: Despite growing awareness of adverse mental health consequences, the scarce existing evidence on the link of disaster exposure and suicidality has remained inconclusive, and the differential suicidality risk associated with distinct levels of natural and man-made disaster exposure is unknown. We therefore investigated the lifetime prevalence and risk of suicidal behavior associated with natural and man-made disaster exposure in Australia. Method: We utilized data from a nationally representative mental health survey (n = 8,841). Univariate and multivariate logistic regression analyses examined the lifetime risk of suicidal thoughts, plans, and attempts associated with varied types and levels of disaster exposure. We focused explicitly on natural and man-made disasters while controlling for other types of trauma exposure, including established risk factors for suicidality. Results: Multivariate analyses indicated that those exposed to multiple natural (adjusted odds ratio [AOR] = 2.21, 95% confidence interval [CI] = [1.04, 4.71], p < .05) or man-made disasters (AOR 3.4, 95% CI = [1.20, 9.58] p < .05) were at significantly greater risk of making suicide attempts, whereas single natural or man-made disaster exposure was not associated with an increased risk of suicidal behavior. Conclusions: Our study findings establish the differential suicidality risk associated with natural and man-made disaster exposure in Australia and highlight the critical role of repeat disaster exposure across distinct disaster types. Suicidal behavior may warrant increasing attention in psychosocial recovery schemes, particularly in the context of disaster-prone areas and for population groups at elevated risk of repeat disaster exposure.     

Suicidal feelings and work environment in psychiatric nursing personnel

Suicidal feelings, attempted suicide and aspects of work environment and well-being in Swedish psychiatric nursing personnel were studied using a questionnaire. The questionnaire, containing 190 questions, was mailed to all 242 nurses and attendants working in psychiatric care at the department of psychiatry at Karolinska Hospital. Eighty-one percent (n=197) returned the questionnaire. Suicidal feelings last year were lower than in the general population, but suicidal feeling and attempted suicide earlier than last year were much more common, and 13% reported that they had attempted suicide earlier in life. In order to study the possible association between work environment and suicide, a factor analysis was performed. Four factors were extracted and labelled: suicidality, quality of work, negative work environment and burn out/depression. The correlation between the factors suggests that negative work environment is associated with burn out/depression, which in turn is related to suicidality. No direct link was demonstrated between suicidality and work environment, and completed suicide was not investigated. The study provides some indirect evidence that a negative work environment may increase suicidal feelings.     

Peripheral serotonergic markers in acutely suicidal patients. 1. Comparison of serotonergic platelet measures between suicidal individuals,nonsuicidal patients with major depression and healthy subjects

Summary Background. A robust association between “suicidality” and deficits of the serotoninergic neurotransmission has been claimed in the past. However, many studies having investigated the relationship between suicidality and peripheral indicators of serotoninergic neurotransmission suffer from considering only one or a very small number of potentially useful serotoninergic parameters, whereas a synoptic multidimensional approach appears to be more appropriate. Furthermore, the psychiatric context within which suicidal behaviour occurs should be considered when interpreting biochemical findings of patients with suicidal ideation and suicide attempts. Methods. In the present study 5 peripheral serotonergic markers, (platelet 5HT concentration, 5HT uptake activity, 5HT_2A receptor binding characteristics, MAO-B activity and tryptophan concentration in plasma) were assessed simultaneously. Of the 60 acutely suicidal inpatients (ICD-10: F43.xx, n = 52; F31/32/33, n = 8), 45 were suicide attempters. Data of 28 nonsuicidal patients with major depression (F31, n = 4; F32, n = 14; F33, n = 10) and 123 healthy volunteers represented the control groups. Results. Mean platelet 5HT concentration was significantly lower in suicidal inpatients when compared to nonsuicidal depressed patients, but did not differ from the figure in healthy subjects. Nonsuicidal depressed patients showed significantly higher mean platelet-5HT concentration than healthy controls. Mean V_max of 5HT uptake in washed platelets, but not in platelet-rich plasma, was significantly higher in suicidal patients than in healthy controls, not, however, when compared to nonsuicidal depressed patients. Mean K_D for the platelet 5HT_2A receptor and MAO-B activity were significantly lower in suicidal patients as compared to nonsuicidal depressed patients and healthy controls. The observed differences in peripheral serotonergic markers between groups are partially due to a significant gender effect. A lower MAO-B activity was observed only in suicidal females, while the higher V_max of 5HT uptake in washed platelets of suicidal patients was due to suicidal males. Conclusions. In view of conflicting observations made by other authors and the present findings on suicidal patients with adjustment disorder it remains doubtful whether and if so to which extent platelet studies can provide valid information on serotonergic mechanisms related to suicidal behaviour.     

Bullying Victimization (Being Bullied) Among Adolescents Referred for Urgent Psychiatric Consultation: Prevalence and Association With Suicidality

Objective:

To examine the prevalence of bullying victimization among adolescents referred for urgent psychiatric consultation, to study the association between bullying victimization and suicidality, and to examine the relation between different types of bullying and suicidality.

Method:

A retrospective chart review was conducted for all adolescents referred to a hospital-based urgent consultation clinic. Our study sample consisted of adolescents with a history of bullying victimization. The Research Ethics Board of Queen’s University provided approval. Data analysis was conducted using SPSS (IBM SPSS Inc, Armonk, NY). Chi-square tests were used for sex, suicidal ideation, history of physical and sexual abuse, and time and type of bullying, and an independent sample t test was used for age.

Results:

The prevalence of bullying victimization was 48.5% (182 of 375). There was a significant association between being bullied and suicidal ideation (P = 0.01), and between sex and suicidal ideation (P ≤ 0.001). Victims of cyberbullying reported more suicidal ideation than those who experienced physical or verbal bullying (P = 0.04).

Conclusions:

Bullying victimization, especially cyberbullying, is associated with increased risk of suicidal ideation among adolescents referred for psychiatric risk assessment. The detailed history of the type and duration of bullying experienced by the victims should be considered when conducting a psychiatric risk assessment.     

The association of attempted suicide with genetic variants in the SLC6A4 and TPH genes depends on the definition of suicidal behavior: a systematic review and meta-analysis

The global prevalence of suicide has increased substantially over the last four decades. Suicidal behavior manifests owing to a combination of biological, behavioral and social factors; however, the etiology of suicidality remains elusive. Even though twin studies have reported a significant heritability of 30–50%, meta-analyses have not highlighted a common genetic variant associated with the spectrum of suicidal behavior. Here, we performed a systematic review of the literature (n=112) to assess the association between serotonergic and non-serotonergic genetic polymorphisms and suicidal behavior. Using an inverse variance random-effects model, we developed pooled odds ratios for the 10 most commonly studied genetic variants related to suicidal behavior, each with at least five independent studies that met our stringent inclusion criteria. Our pooled results indicate no significant correlation between genetic polymorphisms and overall suicidal behavior. However, subgroups of suicide attempts demonstrated actual significance with the serotonin transporter (SLC6A4) 5HTTLPR (OR=1.13 (95% confidence interval=1.05–1.21), P=0.001) and reached nominal significance with the tryptophan hydroxylase rs1800532 (1.22 (1.05–1.41), P=0.007) variant. Subgroups of suicidal behavior (completions and attempts) displayed reduced heterogeneity compared with the overall suicidal behavior spectrum. Our findings suggest that the 5HTTLPR and rs1800532 polymorphisms are significantly associated with suicide attempts, but not associated with completed suicides. The high degree of heterogeneity in past studies may be attributed to the lack of a phenotypic distinction between suicidal attempts and completions. Consequently, we have identified an important source of phenotypic heterogeneity that provides a rationale for the current lack of a common genetic variant associated with suicidal behavior.     

Longitudinal analysis of the suicidal behaviour risk in short-term placebo-controlled studies of mirtazapine in major depressive disorder

Abstract

Objective. To examine suicidal behaviour risk in the short-term placebo-controlled studies of mirtazapine in Major Depressive Disorder (MDD). Method. Longitudinal Generalized Estimating Equations analyses were performed on pooled data from 15 placebo-controlled, randomized, double-blind, short-term trials of mirtazapine, using the suicide item scores from the Hamilton Depression Rating Scale (HAMD) as a proxy outcome measure for suicidality risk. Results. The overall analysis using the convention that a patient is at risk if the HAMD suicide item score is ≥3, and excluding patients at risk at baseline, demonstrated a statistically significantly lower risk for mirtazapine- compared to placebo-treated patients on the HAMD (odds ratio mirtazapine versus placebo 0.38; 95% confidence interval 0.21-0.66; P= 0.0008). Conclusion. Our results based on pooled data from 15 placebo-controlled, short-term studies of mirtazapine in MDD using the suicide item scores from the HAMD as a proxy outcome measure for suicidality risk, demonstrate that mirtazapine was associated with statistically significantly lower suicidality risk compared to placebo.     

Psychopathological distress predicts suicidal ideation and self-harm in Adolescent Eating Disorder outpatients

AbstractObjective This study evaluated the differences in suicidal behaviour between adolescent anorexia nervosa (AN) and bulimia nervosa (BN), and the association of age, menarche timing, duration of eating disorder (ED), depression and general psychopathological symptoms (GSI) with suicidal behaviour in adolescent ED.Methods The study group comprised 57 adolescent outpatients (girls) attending for assessment because of eating disorders. Suicidal ideation, deliberate self-harm and suicidal attempts were assessed in self-report questionnaires.Results In both ED groups, one adolescent had attempted suicide before assessment. Suicidal ideation and/or deliberate self-harm were reported in over half of the cases. Bulimics had significantly more suicidal ideation and deliberate self-harm than anorectics. In multivariate analysis, BN and depression predicted suicidal ideation, but only GSI persisted as predicting deliberate self-harm.Conclusions Suicidal behaviour is common in adolescent ED. Type of ED (BN), depression and higher GSI are strongly associated with suicidal ideation and deliberate self-harm. Our results point to the need to evaluate psychopathological symptoms in adolescent ED, especially in BN, in the initial assessment to prevent severe suicidal behaviour.