View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by PubMed Central

Research in Translation Trends in Suicidology: Personality as an Endophenotype for Molecular Genetic Investigations Jonathan B. Savitz*, Cinda-Lee Cupido, Rajkumar S. Ramesar

sychopathology, especially depression, is the most important Prisk factor for suicidal behaviour [1,2] with between 25% and 40% of depressed patients attempting suicide [3] and about 3.4% completing suicide eventually [4]. Given recent fi gures suggesting that the lifetime prevalence of a major depressive episode among the US population is 32.6–35.1 million [5], it is no surprise that suicide ranks among the top ten causes of death in many other countries [6]. Understanding the aetiology of a signifi cant public health issue such as suicide is important but diffi cult because of its complex and multifactorial origins. Although most suicidal behaviour occurs within the context of a mood disorder, most depressed individuals never attempt suicide. Furthermore, no linear relationship between the severity of the DOI: 10.1371/journal.pmed.0030107.g001 depressive episode and the likelihood Colours of Depression of suicide has been forthcoming [7], (Illustration: Raj Ramesar) highlighting the importance of other Funding: The authors received no specifi c funding factors in addition to psychiatric for this article. At the family level, the risk of illness. These factors include substance Competing Interests: The authors declare that they abuse or alcoholism, a head injury, a suicide is higher in individuals with have no competing interests. dysfunctional family or childhood abuse a family history of suicide [16,17], and the suicide rate of adolescents Citation: Savitz JB, Cupido C-L, Ramesar RS [8], high rates of gun ownership [9], (2006) Trends in suicidology: Personality as an smoking [10], socioeconomic adversity is highly correlated with the suicide endophenotype for molecular genetic investigations. [11], and personality factors [12]. rate among their relatives [18]. Even PLoS Med 3(5): e107. Genetic factors may also be very studies that have controlled for levels DOI: 10.1371/journal.pmed.0030107 important [13,14]. Marusic and Farmer of psychopathology have shown that relatives of suicide completers and Copyright: © 2006 Savitz et al. This is an open-access [15] argue that the variation in the article distributed under the terms of the Creative suicide rate across European countries attempters are at an increased risk for Commons Attribution License, which permits (7–43 per 100,000 inhabitants per year) suicidal behaviour [19,20]. Twin studies unrestricted use, distribution, and reproduction in indicate that this familial clustering of any medium, provided the original author and source cannot be explained by sociocultural are credited. factors alone and is probably due to suicidal behaviour has a partly genetic basis with heritability estimates of 17%– Abbreviations: COMT, catechol-O-methyl- shared genetic vulnerability. A case transferase; EPQ, Eysenck Personality Questionnaire; in point is the high suicide rate in 55% for suicidal behaviour [21,22] FFM, Five-Factor Model inventory; MAO-A, Hungary and Finland, two populations and 20% for suicide [23] reported. monoamine oxidase A; SERT, serotonin transporter; The only adoption study we are aware TCI, Cloninger’s Temperament and Character with a common genetic origin but Inventory; TPH, tryptophan hydroxylase with divergent cultural and political of suggested that as far as suicidal trajectories [15]. behaviour is concerned, adoptees Jonathan B. Savitz, Cinda-Lee Cupido, and Rajkumar resemble their biological parents more S. Ramesar are at the UCT/MRC Human Genetics Research Unit, Institute of Infectious Disease and than the adoptive family [24]. Molecular Medicine, University of Cape Town, Cape Research in Translation discusses health interventions These data have catalysed the Town, South Africa. in the context of translation from basic to clinical search for genes that predispose to research, or from clinical evidence to practice. * To whom correspondence should be addressed: suicidal behaviour, with more than E-mail: [email protected]

PLoS Medicine | www.plosmedicine.org 0619 May 2006 | Volume 3 | Issue 5 | e107 A rich body of data details the Table 1. Association Analyses of Genes Predisposing to Suicide or Suicidal Behaviour association between suicide attempts Gene OMIM Locus Association Evidence and impulsivity. Impulsivity has been Serotonin transporter 182138 17q11.2 7 + and 7 − studies, 2 + meta- shown to be a risk factor for suicidal analyses behaviour in a variety of adults Tryptophan hydroxylase 191060 11p15-14 5 + and 7 − studies, 2 + meta- [7,8,28–30] and adolescents [11,31,32] analyses and 1 − meta-analysis with psychiatric illness, in a forensic Monoamine oxidase A 309850 Xp11.23 3 + and 2 − studies psychiatric population [33], and in the Serotonin receptor 1A (5-HT1A) 109760 5q11.2-q13 1 + and 2 − studies general population [34]. Serotonin receptor 1B (5-HT1B) 182131 6q13 1 + and 5 − studies Serotonin receptor 2A (5-HT2A) 182135 13q14.2 3 + and 9 − studies, 1 − meta- Freud regarded suicide as an analysis aggressive act, a view borne out to Catechol-O-methyltransferase (COMT) 116790 22q11.21 3 + and 2 − studies some extent by modern data. Simon Dopamine receptor 2 (DRD2) 126450 11q23 1 + study and 1 − study et al. [35] found that involvement Dopamine receptor 4 (DRD4) 126452 11p15.5 2 − studies in physical fi ghts was associated with violent suicide attempts, and, according + and − indicate positive and negative studies published. DOI: 10.1371/journal.pmed.0030107.t001 to Appleby et al. [36], people who committed suicide were more likely 100 studies now published [25]. environmental factors—then the to have been arrested in the previous Post-mortem studies of people who endophenotype is a more elementary six months. Cavanagh et al. [37] committed suicide have led to the trait that is tightly correlated with confi rmed that a criminal record is general consensus that a disturbance suicide. The genetic architecture of a risk factor for subsequent suicide. of the serotonergic system is associated the endophenotype is assumed to Fulwiler et al. [38] detected a higher with suicidal behaviour. Therefore, it echo its relative phenotypic simplicity, rate of suicide attempts in a group is no surprise that most of the genes presenting a more tractable target for of violent compared with nonviolent implicated in suicidal behaviour—the geneticists. An understanding of the patients with psychiatric illness, and serotonin transporter (SERT), molecular basis of the endophenotype in South Africa conduct disorder was tryptophan hydroxylase (TPH), should theoretically be the fi rst step found to be the most frequently made monoamine oxidase A (MAO-A), and towards the larger prize: uncovering the diagnosis in suicidal adolescents [39]. the serotonin receptors, 5-HTR1A, molecular basis of the phenotype itself. Negative affect or anxiety is most 5-HTR2A, and 5-HTR1B—modulate Gottesman and Gould [27] commonly measured in the form central serotonergic function (see assert that for a biological marker of neuroticism from the Eysenck Table 1). As is the case with most to be classifi ed as a bona fi de Personality Questionnaire (EPQ) complex traits, however, success tends endophenotype, it must meet the or the Five Factor Model (FFM) to plateau at a point where good following conditions: 1) be associated inventory (see Glossary). Neuroticism candidate genes are identifi ed but with the illness in the relevant is a higher order trait that measures conclusive causal inferences remain population; 2) be largely state- a predisposition to experience anger, elusive because of replication failures. independent, manifesting in the anxiety, depression, guilt, and other The unequivocal identifi cation of individual during both periods of negative emotions or cognitions. genes related to psychiatric disorder health and illness; 3) be heritable; 4) A signifi cant relationship between is retarded by a complex interplay within families, the endophenotype neuroticism and suicidal ideation of latent environmental infl uences and illness should co-segregate; 5) if has been regularly demonstrated or gene–environment interactions; found in affected individuals should [11,22,40,41], and the same holds true genetically and phenotypically also be found in nonaffected family for neuroticism and attempted suicide heterogeneous samples; the possible members at a higher rate than in the [28,42–46]. High levels of anxiety or effects of numerous loci of small effect general population. negative affectivity have been equally size; and the diffi culty of adequately In the following sections we use implicated in completed suicide [47– correcting for multiple testing. Faced 50]. Gottesman and Gould’s [27] fi ve with these frustrations, the use of A number of studies have reported criteria above to evaluate the latest endophenotypes as an aid to molecular introversion to be associated with trend in suicide research: the merits genetic investigations has become suicidal thoughts and behaviour of using an endophenotype such almost de rigueur. In the case of [42,51–53]. In two psychological as personality to identify suicide suicide, a number of researchers have autopsies of completed suicides, susceptibility genes. advocated the use of personality traits Duberstein et al. [50] detected lower as endophenotypes [12,26]. Are Specifi c Personality Traits levels of extraversion in their cohort, and Maser et al. [53] found that Associated with Suicide What Is an Endophenotype? people who committed suicide were An endophenotype is an intermediate Behaviour? shyer and less optimistic than other trait that lies somewhere on the Four main constellations of personality affectively ill patients. Extroversion has developmental pathway from genes characteristics are associated with been shown to predict levels of social to phenotype [27]. If suicide is the suicidal behaviour: impulsivity, support [54,55], perhaps explaining phenotype of interest—the fi nal hostility–aggression, introversion, and the association between introversion product of different genetic and anxiety-neuroticism. and suicide.

PLoS Medicine | www.plosmedicine.org 0620 May 2006 | Volume 3 | Issue 5 | e107 Are the Suicide-Associated Temperament and Character in suicidal acts show increased anxiety, Personality Traits State- Inventory’s (TCI) [60]. hostility–aggression, impulsiveness, Independent? Impulsivity and aggression, traits and introversion, some combination which are slightly peripheral to the of these traits or just one of these Suicide attempts may occur in the “traditional” dimensions of personality characteristics? How many different context of an acute psychiatric described by the FFM and the TCI, also personality traits are enough to indicate crisis, biasing self-report measures of display signifi cant heritability values. a predisposition to suicidal behaviour? personality. One way of mitigating this These have been estimated to be At this time, no pertinent evidence problem is to use longitudinal data. In approximately 40% [64,65], 47% [66], exists for this question to be accurately a study of 921 three-year-old children, 72% [67], and 80% [68]. Molecular answered. Caspi et al. [52] found that impulsive genetic studies have lent support to toddlers were at an increased risk for the notion that personality is at least Are Particular Personality Traits suicide attempts at age 21. Fergusson Found in Unaffected Family et al. [11] followed a birth cohort partially heritable. Functional variants of the mono- Members More Frequently than for 21 years and similarly reported in the General Population? that high neuroticism and novelty- amine oxidase A (MAO-A) and seeking were risk factors for suicidal catechol-O-methyl-transferase (COMT) To the best of our knowledge only behaviour. Similarly, a prospective genes have been associated with two papers have touched on this study of affectively ill patients by Maser violence and aggression [69–72], issue. Brent et al. [75] found that the et al. [53] identifi ed impulsivity as a while Lesch et al. [73] found evidence offspring of siblings concordant for key risk factor for both completed and indicating that the short allele of a suicidal behaviour displayed higher attempted suicide, and Epstein et al. functional insertion–deletion variant levels of impulsive aggression than [48] found an association between in the SERT gene was associated with the offspring of siblings discordant high impulsivity and completed anxiety-related personality traits. A for suicide attempts and an unrelated suicide. plethora of replication attempts have control group. Kim et al. [76] Epstein et al. [48] also found that met with mixed success, although reported an increased prevalence physicians who committed suicide had a recent review suggested that the of aggression—defi ned as verbal or shown increased levels of hostility. association is genuine [74]. physical aggression towards others—in Berglund [56] detected more the fi rst-degree relatives of suicide Do Personality and Suicidality irritability or aggression in people with completers compared with a control alcoholic dependence who completed Co-Segregate in Families? group. suicide. Among children who received This question is immediately Personality as an Endophenotype treatment at a child guidance clinic, problematic to answer because not one aggressive feelings and acts were but many personality characteristics for Suicidality: An Evaluation predictive of later suicide [57], while have been implicated in suicidality. The same functional polymorphisms of among adult men both spontaneous Should an individual who has engaged SERT, MAO-A, and COMT that have and reactive aggression scores were been implicated in personality traits elevated in later suicides [58]. such as negative affectivity, aggression, While depressed affect certainly and impulsiveness have been has the potential to bias personality Five Key Papers on Gene statistically associated with suicide or data, there is enough evidence from Studies and Suicide suicide attempts. This association raises longitudinal studies to support the Brent et al., 2003 [75]. The authors the possibility that the relevant genetic hypothesis that certain personality demonstrate that the familial loading variants contribute to the risk of suicide traits, in particular hostility–aggression for suicidal behaviour is mediated by via their effect on personality. and impulsiveness, predispose to impulsive–aggressive personality traits. Nevertheless, the endophenotypic suicidal acts. In the next section, we Caspi et al., 1996 [52]. The authors approach to gene identifi cation has examine the genetic basis of some of showed that undercontrolled (impulsive) perhaps been too easily accepted as a these personality traits. and inhibited three-year-olds were more panacea for the ills of modern genetics. Some critical questions need to be Is Personality Heritable? likely than comparison children to have attempted suicide by the age of 21. raised. As with the diathesis for suicide, First, does the status of SERT, Gottesman and Gould, 2003 [27]. A personality or temperament is known MAO-A, COMT, and TPH as prime detailed discussion of the concept of the to be at least partly heritable. The two candidates for involvement in suicide endophenotype, including its origins and most widely used personality models owe anything to their putative role in its future use in psychiatry. in genetic investigations of personality anxiety-related, hostile, or impulsive are the FFM [59] and Cloninger and Lesch et al., 1996 [73]. The fi rst paper to behaviour? The answer is most colleagues’ Biopsychosocial Model reveal an association between a variant probably no. The link between suicidal [60]. Twin studies have suggested of the serotonin transporter gene and behaviour and abnormalities of the heritability estimates for the FFM anxiety-related personality traits. serotonergic system was recognised personality traits of between 40% and Roy et al., 1991 [23]. One of the fi rst at least as far back as 1976 [77,78]. 60% [61–63], and comparable results studies to demonstrate that genetic Endophenotypes should facilitate the have been obtained for Cloninger’s factors impact suicidal behaviour. identifi cation of candidate genes, but

PLoS Medicine | www.plosmedicine.org 0621 May 2006 | Volume 3 | Issue 5 | e107 Glossary measures of impulsivity. The diversity References of phenotypic assessment tools makes Eysenck Personality Questionnaire 1. Brent DA, Perper JA, Goldstein CE, et al. comparisons across studies diffi cult (1988) Risk factors for adolescent suicide: A (EPQ): Measures three different comparison of adolescent suicide victims with and raises the spectre of false positive dimensions hypothesised to underpin suicidal inpatients. Arch Gen Psychiatry 45: results from multiple testing. 581–588. personality—neuroticism, psychoticism, 2. Lewinsohn PM, Clarke GN, Seeley JR, Rohde and extraversion [82]. P (1994) Major depression in community Conclusion adolescents: Age at onset, episode duration, Five Factor Model (FFM): Reduces A deeper understanding of the biology and time to recurrence. J Am Acad Child personality to variation along fi ve Adolesc Psychiatry 33: 809–818. of suicidal behaviour may facilitate the 3. Goodwin F, Jamison K (1990) Manic–depressive dimensions— extraversion, neuroticism, development of new pharmacological illness. New York: Oxford University Press. 938 p. conscientiousness, agreeableness, and interventions that could be targeted 4. Blair-West GW, Cantor CH, Mellsop GW, openness to experience [59]. Eyeson-Annan ML (1999) Lifetime suicide risk at vulnerable individuals, potentially in major depression: Sex and age determinants. Biopsychosocial Model: Cloninger’s saving thousands of lives. Identifying J Affect Disord 55: 171–178. 5. Kessler RC, Berglund P, Demler O, Jin R, model of personality [60] that the genes that contribute to the risk Koretz D, et al. (2003). The epidemiology of differentiates between temperament for suicidal behaviour is an integral major depressive disorder: Results from the factors which have a greater genetic part of this process. While tantalising National Comorbidity Survey Replication (NCS-R). JAMA 289: 3095–3105. loading, and character traits which are genetic clues are beginning to emerge 6. Diekstra RFW (1996) The epidemiology of heavily infl uenced by the environment. from the mass of data, replication is suicide and parasuicide. Arch Suicide Res 2: a problem and new strategies such as 1–29. Temperament and Character Inventory 7. Oquendo MA, Galfalvy H, Russo S, Ellis SP, (TCI): The questionnaire used to measure endophenotyping are thus evolving. Grunebaum MF, Burke A, Mann JJ (2004) In principle, the use of Prospective study of clinical predictors of the seven dimensions of personality suicidal acts after a major depressive episode proposed in the Biopsychosocial endophenotypes is an excellent in patients with major depressive disorder or model. The four temperaments are strategy for parsing complex biological bipolar disorder. Am J Psychiatry 161: 1433– systems into their genetically simpler 1441. harm avoidance, novelty seeking, 8. Mann JJ, Waterneux C, Haas GL, Malone KM reward dependence, and persistence. components and holds out much (1999) Toward a clinical model of suicidal promise for bridging the chasm behavior in psychiatric patients. Am J Psychiatry Character traits are self-directedness, 156: 181–189. cooperativeness, and self-transcendence between research and clinical 9. Beautrais AL, Joyce PR, Mulder RT (1996). [60]. practice. The choice of the correct Access to fi rearms and the risk of suicide: A endophenotype is, however, of case control study. Aust N Z J Med 30: 741–748. 10. Breslau N, Schultz LR, Johnson EO, Peterson fundamental importance. Given EL, Davis GC (2005). Smoking and the risk of the complexity of personality and suicidal behaviour: A prospective study of a in this case it seems as though analyses the myriad ways in which it can be community sample. Arch Gen Psychiatry 62: 328–334. of candidate genes are helping to quantifi ed, we have reservations about 11. Fergusson DM, Woodward LJ, Horwood LJ identify endophenotypes. the fruitfulness of this approach. (2000) Risk factors and life processes associated with the onset of suicidal behaviour during Second, are anxiety-related, Future research should aim to identify adolescence and early adulthood. Psychol Med aggressive, and impulsive traits a simpler endophenotype—an 30: 23–29. really genetically simpler than endophenotype of personality—that 12. Baud P (2005) Personality traits as intermediary phenotypes in suicidal behavior: Genetic issues. suicidal behaviour? Perhaps they will introduce less noise into genetic Am J Med Genet C133: 34–42. are less complex in neurobiological analyses. One emerging possibility 13. Gould MS, Fisher P, Parides M, Flory M, Shaffer D (1996) Psychosocial risk factors of terms but are they simple enough is the extensive body of work child and adolescent completed suicide. Arch to signifi cantly enhance success? describing asymmetrical prefrontal Gen Psychiatry 53: 1155–1162. For example, the potential for cortical activation as evinced by 14. Cheng AT, Chen THH, Chen CC, Jenkins R (2000) Psychosocial and psychiatric risk factors aggression may be an evolved trait electroencephalograph (EEG) and for suicide: Case-control psychological autopsy that is characteristic of all mammalian its relationship to positive (appetitive study. Br J Psychiatry 177: 360–365. species [79]. The key variable might drive or behavioural activation) 15. Marusic A, Farmer A (2001) Genetic risk factors as possible causes of the variation in therefore be the threshold at which and negative (inhibitory) affective European suicide rates. Br J Psychiatry 179: aggression is elicited rather than style; see [80,81] for reviews. Other 194–196. 16. Garfi nkel BD, Froese A, Hood J (1982) Suicide self-reported aggression per se. The neuroimaging modalities such as attempts in children and adolescents. Am J proximate environmental factors functional magnetic resonance Psychiatry 139: 1257–1261. that elicit this behaviour and the imaging (fMRI) may also facilitate 17. Linkowski P, de Maertelaer J, Mendlewicz J (1985) Suicidal behaviour in major depressive historical childhood environment the identifi cation of the neural illness. Acta Psychiatr Scand 72: 233–238. that potentially contributes to this correlates of impulsive and aggressive 18. Marusic A, Roskar S, Hughes RH (2004) Familial study of suicide behaviour among threshold are thus variables that dispositions. adolescents in Slovenia. Crisis: 74–77. should not be ignored. The challenge Genetic knowledge is without peer 19. Brent DA, Bridge J, Johnson BA, Connolly J of modelling these effects in genetic in its potential for radically altering the (1996) Suicidal behaviour runs in families: A controlled family study of adolescent suicide investigations is signifi cant. practice of medicine, but the obstacles, victims. Arch Gen Psychiatry 53: 1145–1152. Third, can a consensus be reached especially in the fi eld of psychiatric 20. Johnson BA, Brent DA, Bridge J, Connolly J on the optimum way of measuring traits genetics, are formidable, and care (1998) The familial aggregation of adolescent suicide attempts. Acta Psychiatr Scand 97: such as impulsivity and aggression? should be taken not to overestimate 18–24. In our review of the literature we the probability of short-term success. 21. Fu Q, Heath AC, Bucholz KK, Nelson EC, Glowinski AL, et al. (2002) A twin study of encountered at least seven measures Advances are likely to be incremental genetic and environmental infl uences on of hostility–aggression and four but invaluable. suicidality in men. Psychol Med 32: 11–24.

PLoS Medicine | www.plosmedicine.org 0622 May 2006 | Volume 3 | Issue 5 | e107 22. Statham DJ, Heath AC, Madden PAF, Bucholz 43. Pallis DJ, Birtchnell J (1976) Personality and 64. Tellegen A, Lykken DT, Bouchard TJ, Wilcox KK, Bierut L, et al. (1998) Suicidal behaviour: suicidal history in psychiatric patients. J Clin K, Seagal N, et al. (1988) Personality similarity An epidemiological and genetic study. Psychol Psychol 32: 246–253. in twins reared apart and together. J Pers Soc Med 28: 839–855. 44. Isherwood J, Adam KS, Hornblow AR (1982) Psychol 54: 1031–1039. 23. Roy A, Segal NL, Centerwall BS (1991) Suicide Life event stress, psychosocial factors, suicide 65. Coccaro EF, Bergeman CS, McClearn GE in twins. Arch Gen Psychiatry 48: 28–32. attempt and auto-accident proclivity. J (1993) Heritability of irritable impulsiveness: 24. Wender PH, Kety SS, Rosenthal D, Schulsinger Psychosom Res 26: 371–383. A study of twins reared together and apart. F, Ortmann J, et al. (1986) Psychiatric disorders 45. Ashton CH, Marshall EF, Hassanyeh F, Marsh Psychiatry Res 48, 229-242. in the biological and adoptive families of VR, Wright-Honari S (1994) Biological 66. Seroczynski, A.D., Bergeman, C.S., Coccaro, adopted individuals with affective disorders. correlates of deliberate self-harm behaviour: A E.F. (1999). Etiology of the impulsivity Arch Gen Psychiatry 43: 923–929. study of electroencephalographic, biochemical aggression relationship: Genes or 25. Balazic J, Marusic A (2005) The completed and psychological variables in parasuicide. Acta environment? Psychiatry Res 86: 41–57. suicide as interplay of genes and environment. Psychiatr Scand 90: 316–323. 67. Rushton JP, Fulker DW, Neale MC, Nias DKB, Forensic Science Int 147: S1–S3. 46. Chance SE, Kaslow NJ, Baldwin K (1994) Eysenck HJ (1986. Altruism and aggression: 26. Roy A (2002) Family history of suicide and Anxiety and other predictors of severity of The heritability of individual differences. J Pers neuroticism: A preliminary study. Psychiatry suicidal intent in urban psychiatric inpatients. Soc Psychol 50: 1192–1198. Res 110: 87–90. Hosp Community Psych 45: 716–718. 68. O’Conner M, Foch TT, Sherry T, Plomin R 27. Gottesman II, Gould TD (2003) The 47. Thomas CB (1971) Suicide among us: (1980) A twin study of specifi c behavioral endophenotype concept in psychiatry: II. Habits of nervous tension as potential problems of socialization as viewed by parents. Etymology and strategic intentions. Am J predictors. Hopkins Med J 129: 190–201. J Abnorm Child Psychol 8: 189–199. Psychiatry 160: 636–645. 48. Epstein LC, Thomas CB, Shaffer JW, Perlin S 69. Caspi A, McClay J, Moffi tt TE, Mill J, Martin J, 28. Beautrais AL, Joyce PR, Mulder RT (1999). (1973) Clinical prediction of physician suicide et al. (2002) Role of genotype in the cycle of Personality traits and cognitive styles as risk based on medical student data. J Nerv Ment violence in maltreated children. Science 297: factors for serious suicide attempts among Disease 156: 19–28. 851–854. young people. Suicide Life Threat Behav 29: 49. Fawcett J, Scheftner WA, Fogg L, Clark 70. Strous RD, Bark N, Parsia SS et al. (1997) 37–47. DC, Young MA, et al. (1990) Time-related Analysis of a functional catechol-O- 29. Corruble E, Damy C, Guelfi JD (1999) predictors of suicide in major affective methyltransferase gene polymorphism in Impulsivity: A relevant dimension in depression disorder. Am J Psychiatry 147: 1189–1194. schizophrenia: Evidence for associations with regarding suicide attempts? J Affect Disord 53: 50. Duberstein PR, Conwell Y, Caine ED aggressive and antisocial behaviour. Psychiatry 211–215. (1994) Age differences in the personality Res 69: 71–77. 30. Michaelis BH, Goldberg JF, Davis GP, Singer characteristics of suicide completers: 71. Lachman H, Nolan KA, Mohr P et al. (1998) TM, Garno JL, et al. (2004) Dimensions of Preliminary fi ndings from a psychological Association between catechol-O-transferase impulsivity and aggression associated with autopsy study. Psychiatry 57: 213–224. genotype and violence in schizophrenia and suicide attempts among bipolar patients: A 51. Benjaminsen S, Krarup G, Lauritsen R (1990) schizoaffective disorder. Am J Psychiatry 155: preliminary study. Suicide Life Threat Behav Personality, parental rearing behaviour and 835–837. 34: 172–176. parental loss in attempted suicide: A comparative 72. Kotler M, Barak P, Cohen H et al. (1999) 31. Kingsbury S, Hawton K, Steinhardt K, James study. Acta Psychiatr Scand 82: 389–397. Homicidal behaviour in schizophrenia A (1999) Do adolescents who take overdoses 52. Caspi A, Moffi tt TE, Newman DL, Silva PA associated with genetic polymorphism have specifi c psychological characteristics? (1996) Behavioral observations at age 3 years determining low catechol-O-methytransferase A comparative study with psychiatric and predict adult psychiatric disorders. Arch Gen (COMT) activity. Am J Med Genet 88B: 628– community controls. J Am Acad Child Adolesc Psychiatry 53: 1033–1039. 633. Psychiatry 38: 1125–1131. 53. Maser JD, Akiskal, HS, Schettler P, Scheftner 73. Lesch KP, Bengel D, Heils A, Sabol SZ, 32. Askenazy FL, Sorci K, Benoit M, Lestideau W, Mueller T, et al. (2002) Can temperament Greenberg BD, et al. (1996) Association of K, Myquel M, et al. (2003) J Affect Disord 74: identify affectively ill patients who engage in anxiety-related traits with a polymorphism 219–227. lethal or near-lethal suicidal behavior? A 14- in the serotonin transporter gene regulatory 33. Stalenheim EG (2001) Relationships between year prospective study. Suicide Life Threat region. Science 274: 1527–1530. attempted suicide, temperamental vulnerability Behav 32: 10–32. 74. Savitz JB, Ramesar RS (2003) Genetic variants and violent criminality in a Swedish forensic 54. Kendler R (1997) Social support: A genetic- implicated in personality: A review of the more psychiatric population. Eur Psychiatry 16: epidemiological analysis. Am J Psychiatry 50: promising candidates. Am J Med Genet 131B: 386–394. 853–862. 20–32. 34. Grucza RA, Przybeck TR, Spitznagel EL, 55. Roy, A. (2003) Distal risk factors for suicidal 75. Brent DA, Oquendo M, Birmaher B, Greenhill Cloninger CR (2003) Personality and behaviour in alcoholics: Replications and new L, Kolko D, et al. (2003) Peripubertal suicide depressive symptoms: A multi-dimensional fi ndings. J Affect Disord 77: 267–271. attempts in offspring of suicide attempters with analysis. J Affect Disord 74: 123–130. 56. Berglund M (1984) Suicide in alcoholism. Arch siblings concordant for suicidal behavior. Am J 35. Simon TR, Swann AC, Powell KE, Potter LB, Gen Psychiatry 41: 888–891. Psychiatry 160: 1486–1493. Kresnow M, et al. (2001) Characteristics of 57. de Chateau P (1990) Mortality and 76. Kim CD, Seguin M, Therrien N, Riopel G, impulsive suicide attempts and attempters. aggressiveness in a 30-year follow-up study Chawky N, et al. (2005) Familial aggregation Suicide Life Threat Behav 32: 49–59. in child guidance clinics in Stockholm. Acta of suicidal behaviour: A family study of 36. Appleby L, Cooper J, Amos T, Faragher B Psychiatr Scand 81: 472–476. male suicide completers from the general (1999) Psychological autopsy study of suicides 58. Angst J, Clayton PJ (1998) Personality, smoking population. Am J Psychiatry 162: 1017–1019. by people aged under 35. Br J Psychiatry 175: and suicide: A prospective study. J Affect 77. Asberg M, Traskman L, Thoren P (1976) 5- 168–174. Disord 51: 55–62. HIAA in the cerebrospinal fl uid. A biochemical 37. Cavanagh JTO, Owens DGC, Johnstone EC 59. Costa PT, McCrae RR (1989) NEO personality suicide predictor? Arch Gen Psychiatry 33: (1999) Suicide and undetermined death in inventory. Odessa: Psychological Assessment 1193–1197. south east Scotland. A case-control study using Resources. 101 p. 78. Stanley M, Vigilio J, Gershon S (1982) Tritiated the psychological autopsy method. Psychol Med 60. Cloninger CR, Pryzbeck TR, Svrakic DM, imipramine binding sites are decreased in the 29: 1141–1149. Wetzel RD (1994) The Temperament and frontal cortex of suicides. Science 216: 1337– 38. Fulwiler C, Grossman H, Forbes C, Ruthazer Character Inventory (TCI): A guide to its 1339. R (1997) Early-onset substance abuse and development and use. St. Louis (Missouri): 79. Panksepp J (1998) Affective neuroscience: community violence by outpatients with chronic Center for Psychobiology of Personality, The foundations of human and animal mental illness. Psychiatr Serv 48: 1181–1185. Washington University of St. Louis. emotions. New York: Oxford University Press. 39. Schlebusch L, Minnaar GK (1980) The 61. Loehlin JC (1992) Genes and environment 480 p. management of parasuicide in adolescents. S in personality development. Newbury Park 80. Davidson RJ (1998) Affective style and Afr Med J 57: 81–84. (California): Sage. 160 p. affective disorders: Perspectives from affective 40. Velting D (1999) Suicidal ideation and the fi ve- 62. Jang K, Livesley WJ, Vernon PA (1996) neuroscience. Cognition Emotion 12: 307–330. factor model of personality. Pers Indiv Diff 27: Heritability of the big fi ve personality 81. Demaree HA, Everhart DE, Youngstrom EA, 943–952. dimensions and their facets: A twin study. J Harrison DW (2005) Brain lateralization of 41. Kerby DS (2003) CART analysis with unit- Personality 64: 577–591. emotional processing: Historical roots and a weighted regression to predict suicidal ideation 63. Riemann R, Angleitner A, Strelau J (1997) future incorporating “dominance.” Behav Cogn from Big Five traits. Pers Indiv Diff 35: 249–261. Genetic and environmental infl uences on Neurosci Rev 4: 3–20. 42. Colson CE (1972) Neuroticism, extraversion personality: A study of twins reared together 82. Eysenck HJ, Eysenck SBG (1964) Manual of and repression-sensitization in suicidal college using the self and peer report NEO–FFI scales. the Eysenck Personality Inventory. London: students. Brit J Soc Clin Psych 11: 88–89. J Personality 65: 449–475. University Press. 26 p.

PLoS Medicine | www.plosmedicine.org 0623 May 2006 | Volume 3 | Issue 5 | e107