Studies of adolescents who have attempted suicide usually focus on identifying how they differ from their nonsuicidal peers. Researchers at Emory University, however, have begun to identify how adolescents who attempt suicide differ from one another.

Their work adds to previous studies that have reported personality subtypes within samples of adult suicide attempters and completers. “Overall, assessing adolescents’ risk of suicide attempt should include not only a list of risk factors but also a deeper understanding and consideration of personality,” wrote lead author Dorthie Cross, a doctoral student in the Department of Psychology at emory University, and her colleagues in the October Journal of Nervous and Mental Disease.

The researchers used a Q-factor analysis to identify subtypes based on the Shedler-Westen Assessment Procedure-II for Adolescents (SWAP-II-A), a 200-item measure of personality pathology administered by clinicians that was codeveloped by Drew Westen, one of the authors of the new study.

“Q-factor analysis is also called inverted factor analysis because it aggregates patients rather than variables, identifying people with similar profiles across a set of items instead of items with similar content across cases,” the researchers noted. “the goal of Q-factor analysis in this study is to identify groups of adolescent attempters with shared personality characteristics that distinguish them from other adolescents who have attempted suicide,” they explained.

The researchers recruited 950 psychiatrists and psychologists with at least three years’ postlicensure experience from membership rosters of the American Academy of Child and Adolescent Psychiatry and the American Psychological Association. Participating clinicians were asked to provide data on a single randomly selected adolescent patient currently in treatment of “enduring maladaptive patterns of thought, feeling, motivation, or behavior—that is, personality.” they were asked to select a patient without a particular diagnosis, yielding 267 patients with a history of suicide attempt. Publications detailing the sampling procedure and the rationale for using clinicians as informants in basic science research, also coauthored by westen and published in the American Journal of Psychiatry, were referenced.

Q-factor analysis of the 267 patients with a history of suicide attempt resulted in six subtypes: externalizing, internalizing, emotionally dysregulated, high functioning, narcissistic, and immature.

• The externalizing subtype  comprised the largest group of adolescents, characterized by substance abuse, attachment disruption, and being the victim of childhood physical abuse. It included a higher proportion of males and was a significantly younger age group. “We found that this subtype was related to less-lethal suicide attempts, which may be partially explained by the younger age of this group,” the researchers said.

• The internalizing subtype  “reflects a substantial body of literature, showing the risk of suicide associated with mood disorders, independent of other factors like substance abuse,” and was characterized by depression, avoidance, and hopelessness; the group was predominantly female.

• The emotionally dysregulated subtype  was characterized by childhood sexual abuse, school problems, borderline personality pathology, and substance abuse, as well as increased risk of comorbid mood and substance use disorders. Adolescents fitting this description were considered to be of highest risk.

• The high-functioning subtype  “seems healthy and does not experience most of the risk factors associated with suicide,” wrote Cross and colleagues. They may be perfectionistic, experiencing guilt and anxiety, and their attempts tend to be less lethal. “It is possible … (they) may have less access to lethal means of suicide, or they may have a diminished desire to actually die relative to other suicidal adolescents whose functioning may be more impaired.”

• The narcissistic subtype  showed high adaptive functioning and school performance and low levels of substance use disorders and conduct disorder.

• The immature subtype  displayed marked social isolation and anxiety, as well as significant association with schizoid personality disorder. “although immature adolescents seem similar to socially phobic adolescents who are at an increased risk of suicide attempt, they may represent adolescents at risk for the development of psychotic disorders,” wrote Cross and colleagues.

To validate each subtype, the researchers compared the study subjects across relevant Axis I and Axis II pathology, adaptive functioning, and etiology. “Depression, substance abuse, physical and sexual abuse, attachment disruption, and particular personality characteristics (e.g., impulsivity or neuroticism) are all markers for risk; however, these markers do not apply, nor can they apply, to all suicidal adolescents,” they explained.

In discussing the limitations of their study, Cross and colleagues noted that only adolescents who attempted, rather than completed, suicide were able to be evaluated, potentially omitting meaningful subtypes, though they cited increasing evidence that suicide attempt and completion constitute a single spectrum of risk.

“Despite its limitations, this study shows that the path to suicide may be quite different depending on what type of person is making the decision,” concluded Cross and her colleagues. “Understanding those differences is vital to the prediction and prevention of future suicide attempts.”

The research was funded by a grant from the National Institute of Mental Health.

An abstract of “Personality Subtypes of Adolescents Who Attempt Suicide” is posted at <www.ncbi.nlm.nih.gov/pubmed/21964268>.