O-02-01

Risk of suicide after attempted suicide

S. Ziherl, B. Zalar. University Psychiatric Hosp.. Ljubljana, Slovenia

Objective: All suicide attempts cannot predict suicide commitment. The aim of our study was to examine those characteristics of suicide attempt which could most accurately predict completed suicide.

Methods: Subjects were all individuals registered as committed suicides (N=16.522) or attempted suicides (N=15.057) in the Register of Suicides of the Republic of Slovenia between 1970 and 1996. Log linear analysis of a frequency table was used to uncover relationship between categorical variables.

Results: We found the model which fit between variables: mode, number of repetitions and type, then between number of repetitions, type and gender, and between mode, type and gender.

Conclusion: Our data suggest that clinicians should heighten their awareness that any suicide attempt can in some 20% predict suicide. At a highest risk is someone who has attempted suicide by hanging.

O-02-02

Suicide attempts in Basel (Switzerland) 2003-2004

P. Berger, M. Eichhom, A. Riecher-Rossler. University Hospital Psychiatric Outpatient Dept., Basel, Switzerland

Objective: To analyse the psychiatric and cultural specificity of suicide attempts of Turkish immigrants.

Methods: In the context of the WHO/EURO multicentre study on parasuicide we recorded all suicide attempts of inhabitants of Kanton Basel-Stadt in the years 2003 and 2004. Most of the suicide attempters were treated at first in the emergency department of the University Hospital. From there, if medically indicated, they were transfered to the intensive care unit, the crisis intervention unit (KIS) or the University Psychiatric Hospital of Basel. To assure a complete recruitment we regularly contacted all hospitals of Basel and surroundings, general practitioners and psychiatrists. We recorded sociodemographic data, information about context and methods of the suicide attempt as well as psychiatric and medical diagnoses. Suicide attempts were defined according to the WHO criteria.

Results: In the 2003 we recorded 228 suicide attempts, 156 women and 72 men (quotient 2,16), which correspond to a rate of 144,1/100'000 inhabitants for women and 87,4/100’000 for men. The rate of suicide attempts for immigrants from Germany, Ex- Yugoslavia and Southern Europe was comparable to data from the respective countries of origin. But Turkish immigrants showed a rate of 362/100'000 Turkish inhabitants, which is not only significantly higher than the rate of Swiss people, but also higher than the rate known from Turkey. Most of these Turkish suicide attempters were young women.

Conclusion: Identifying risk factors for deliberate self-harm, with reference to specific cultural influences, could permit selective preventive interventions.

O-02-03

Homicide of women by intimitate partners - femicide

R. Kovacevic, B. Kecman. KPD-bolnica (Prison hospital) Forensic psychiatry, Belgrade, Yugoslavia

Objective: Homicide of women (femicide) by intimate partners is only one type of family homicide. Marriage and family are legal and habitual arranged institutions. Homicide is the act of malign aggression and present negation of marriage, family and life at all.

Methods: We studies 90 male offenders who committed homicide of women (femicide). All of them are examined on the Department of forensic psychiatry in Prison hospital in Belgrade, during the period between January 1, 1992, and December 31, 2002. For this research, we construed special questionnaire with the different groups of questions for detail analyzing lives of offenders. The results are work out by descriptive and differential statistical methods. Aim: Identification of psychopathological and criminological characteristics of offenders of femicide, identification crim- inogen factors in psychics status of offenders during the homicide, and analyze eventual role of victim in homicides.

Results: Male who committed homicide of women, are persons with normal intelligentsia, but persons with emotional instable, egotism and paranoid characteristics. The most frequent factors that start homicidal behavior are psychotic symptoms, reactive affective state generated by situation and alcohol intoxication. Mostly, victims were not act provocative.

O-02-04

Newer antidepressants - a principal factor in preventing suicide

L. Sondergard, L. Kessing, K. Kvist, P. K. Andersen. Rigshospitalet Psychiatric Department 623100, Copenhagen 0, Denmark

Objective: As in many developed countries, the use of antidepressants in Denmark has been substantially increasing during recent years coinciding with a decreasing suicide rate. This suggests a putative protective effect of antidepressants. In this study we aim to investigate this relationship on individualized data.

Methods: We obtained data from three Danish registers in an historic prospective national pharmacoepidemiological register linkage study. We included 438,625 patients who had purchased antidepressants over a five-year period, according to prescriptions and number of dispatches, and compared them with 1,073,862 population based control persons who had not purchased antidepressants. The risk of suicide was estimated using Poisson regression analyses.

Results: For patients treated with SSRIs and newer non-SSRI antidepressants, respectively, we found a significantly decreased suicide rate among those who purchased two or more dispatches compared to the rate for those who purchased only a single dispatch. Purchasing SSRIs at least twice was associated with a 0.59 [95% Cl 0.50-0.68] and 0.72 [95% Cl 0.58-0.88] reduced rate of suicide for men and women, respectively, compared to the rate when purchasing SSRIs once only. Further, the suicide rate decreased with increasing number of purchased dispatches. In contrast, for patients treated with older antidepressants, those who purchased two or more dispatches had a suicide rate similar to the rate for those who purchased a single dispatch.

Conclusion: For the individual patient, antidepressant treatment with SSRIs and newer noradrenergic acting antidepressants seems to be a principal factor in preventing suicide.

O-02-05

Mortality in relation to anxiety and depression

A. Mykletun, Research Centre for Health Pro Dep. of Psychology, Bergen, Norway

Objective: From the literature we know that depression is a risk-factor for mortality in the general population as well as in various clinical populations. (1) Whether anxiety has a similar effect on mortality is still unclear, as is also (2) the effect of milder levels of depression. (3) Beyond coronary heart diseases (CHD) and suicide, there is also little knowledge about the mechanisms that are involved in the association between depression and mortality. These three issues constitute the aims of the present study.

Methods: Information on anxiety, depression and somatic health status was obtained by self-report and clinical examinations in the context of the Health Study of Nord-Tr0ndelag Conunty (HUNT-II) in 1995-97. Mortality as the end-point was obtained from the National Mortality Registry by record linkage. Anxiety and depression was measured by the Hospital Anxiety and Depression Scale (HADS).

Results: (1) Adjusted for physical health at HUNT-II, high levels of both anxiety and depression were associated with mortality, though stronger for depression than for anxiety. However, also low level of anxiety (that is within the first quartile) was associated with mortality. (2) The impact of depression on mortality was found also for very low levels of depression, far lower than the conventional clinically used cut-offs for

Conclusion: There are three new findings from our study: (1) The U-shaped effect of anxiety on mortality (which we suggest might be an effect of risk-taking behaviour in individuals with very low levels of anxiety), (2) the effect of very low levels of depression on mortality (levels commonly regarded below any clinical relevance), and (3) that a broad range mortality diagnoses are associated with anxiety/depression.

O-02-06

Severe personality disorders in the offspring of antenatally depressed mothers - a 31 year follow-up of the Northern Finland 1966 Birth Cohort

P. Maki, P. Maki, J. Veijola, M. Joukamaa, P. Rantakallio, J. Jokelainen, L. Kantojarvi, M. Isohanni. University of Oulu Psychiatry, University of Oulu, Finland

Objective: Maternal depression is common both during the ante and postnatal period. Prenatal depression has been connected to externalizing problems in children. Postpartum depression has been considered a risk for mental distress of the children and young adults. We studied the association between maternal antenatal depression and severe personality disorders in the young adult offspring.

Methods: At midgestation mothers of 12 058 babies in the Northern Finland 1966 Birth Cohort were asked at the antenatal clinic if they felt depressed. The general population birth cohort was followed up 31 years. Those offspring of the cohort, who appeared on the Finnish Hospital Discharge Register between the years 1983-97, were identified. All psychiatric diagnoses were checked against DSM-III-R criteria.

Results: Of the mothers of the offspring, 14 % felt depressed during pregnancy. The cumulative incidence of the hospital-treated personality disorders was 2.4 % in the male offspring of depressed mothers and 0.6 % in the sons of non-depressed mothers (p<0.001). The corresponding numbers for female offspring were 0.8 % and 0.3 % respectively (p<0.05). When adjusted for mother's marital status and place of residence at the child's birth, maternal smoking, age and parity and parental socioeconomic status during pregnancy and mother's wantedness of pregnancy and perinatal complications, the risk was elevated for borderline personality disorders 10-foldly and for antisocial personality disorders 3-foldly in the male offspring of antenatally depressed mothers.

Conclusion: Mothers’ self-reported depression during pregnancy predicted severe personality disorders in their offspring, especially hospital-treated borderline and antisocial personality disorders in men.

O-02-07

Outcome and follow up of treatment of personality disorders E. Fabian. Klinik Menlerschwaige, München, Germany

Objective: Evaluation of outcome of an inpatient dynamic psychiatric treatment of patients with personality disorders.

Methods: The outcome and effectiveness of treatment of 144 patients with personality disorders is to be evaluated in three times of measurement: admission, discharge and follow up of three to four years. Included to the study are all patients of the Dynamic Psychiatric Hospital, Munich, in the year 2001 who meet the criteria of diagnosis, regularly discharge and a mimimum treatment time of 14 days.

Results: As shown with psychometric instruments and symptom scales like the SCL 90, the BDI and the IIP, social data (occupational state, social relationships) and data of further treatment, patients do best, who continue therapy in an outpatient setting and who are able to activate their constructive, healthy ressources. Duration of inpatient treatment has a strong impact on outcome.

Conclusion: Treatment of patients with personality disorders takes time and has to integrate different inpatient and outpatient therapeutic concepts.

O-02-08

Anticipatory fear in violent schizophrenia and personality disorder subjects: A functional mri study

D. Mrigendra, V. Kumari, I. Barkakati, P. Taylor, T. Sharma. Broadmoor Hospital, Crowthome, United Kingdom

Objective: Incidents of violence in severe mental disorders like schizophrenia and, personality disorders are higher than in the general population. Previous research has reported impaired emotion processing in these disorders, and neural circuits involved in emotional processing may underlie violent behaviour. This study therefore explored the neural correlates of anticipatory fear, in four groups of men: (i) Personality disorder with history of extreme violence (n=13), (ii) schizophrenia with history of extreme violence (n=13), (iii) schizophrenia with no history of violence (n= 13), and (iv) matched control subjects (n=14).

Methods: The experimental paradigm involved inducing a state of anticipatory fear in the subjects inside the scanner by repeated presentation of a control condition (‘safe’) followed by presentation of an activating condition (‘shock’). Before going into the scanner, subjects were delivered a mild electric shock by a stimulator in the ‘shock’ condition off line. The subjects were then instructed that they would receive a similar shock (stronger than the one example they received before the scan) during the ‘shock’ condition inside the scanner.

Results: Results in the healthy controls in the shock versus safe conditions indicated activation in the right medial/inferior frontal gyrus, left superior temporal gyrus, and right Insula. In the nonviolent schizophrenia patients activation was noted in the bilateral inferior frontal gyrus and left insula. Patients with history of violence did not show any significant areas of activation during this condition.

O-02-09

Amygdala and emotions - investigation of gender and cultural differences in facial emotion recognition

B. Hoheisel, U. Habel, C. Windischberger, S. Robinson, I. Kryspin- Exner, E. Moser. Kompetenzzentrum Hochfeld MR Medizinische Universitat Wien, Wien, Austria

Objective: Recognizing emotions in facial expressions is essential for successful social interaction. Patients suffering from psychiatric disorders characterized by social and emotional impairments such as schizophrenia, show a reduced capability in recognizing facial emotional expressions. This social-cognitive competency relies on a widespread subcortical-limbic and cortical network. The amygdala seems to play a crucial role in this, especially for processing fearful stimuli. Its impact on the recognition of other „basic emotions” is not fully clear. Understanding the neurobiological basis of discriminating different emotional qualities in a facial expression seems to be highly relevant for characterizing psychiatric disorders with emotion recognition impairment aspects. Also, methodological influences such as the accuracy of measurement, data quality and culture or gender differences must be taken into account in order to adequately interpret clinical findings.

Methods: To analyze and consider these factors we investigated 40 healthy subjects (20 female, 20 male) of various cultural backgrounds during a facial emotion recognition task with a 3 Tesla whole body scanner (Medspec Bruker, Biospin). We used a specially optimized protocol to reliably detect amygdala activation. Sex hormone assessment was used for the correlation with functional data to control for such influences on cerebral correlates of gender differences. Cultural differences in brain activation during emotion discrimination have also been analyzed.

Results: Our data show robust amygdala activation and emphasize the critical role of the amygdala in processing emotions in facial expressions independent of the expressed valence as well as culture and gender differences.

Conclusion: The amygdala emerged as a crucial node of emotional evaluation in the early stages of processing. Nevertheless, gender and culture seem to exert a modulating influence on its activation, indicating relevant factors that have to be considered in clinical studies.

O-02-10

The serotonin transporter polymorphism may modulate severity of adult ADHD symptoms in conjunction with adverse life events

D. J. Muller. Charite Berlin Abt. flir Psychiatrie, Berlin, Germany

Childhood ADHD may persist into adulthood where it reaches prevalence rates of 1-4%. Family, twin and adoption studies strongly suggest that ADHD is a complex genetic disorder, thus occurrence of specific environmental circumstances is likely to influence etiology/severity of ADHD. We tested whether severity of adult ADHD (measured by the Brown Attention Deficit Disorder Scale) was associated with adverse life events (ALE) moderated by a functional promoter polymorphism of the serotonin transporter gene (5-HTTLPR). The study included 110 patients and occurrence of ALE were assessed for a large variety of circumstances (e.g., physical/sexual/verbal abuse; separation/death of parent/close relatives, medical conditions). Overall, findings showed a significant association between higher scores of ALE and severity of ADHD (F=4,16, 2df, p=.01). The 5-HTTLPR was not associated with adult ADHD (F =2.11, 2df, p=. 12). However, a trend was noted between severity of ADHD and ALE moderated by 5- HTTLPR: Patients with the 1/1 genotypes were less severely affected with ADHD symptoms when they experienced fewer or moderate ALE than patients with s/s genotypes, although this relationship was not seen in patients with highest ALE scores (F=2.47, 4df, p=.05). Our preliminary Findings suggest that the 5- HTTLPR may moderate the association of ALE with ADHD symptoms severity.