Personality disorders are characterized by the presence of inflexible and maladaptive patterns of perceiving oneself and relating to the environment that result in psychosocial impairment or subjective distress. The enduring nature of the behaviors, their impact on social functioning, the lack of clear boundaries between normality and illness, and the patient's perception of the symptoms as not being foreign make this group of conditions more difficult to conceptualize than the more typical, episodic mental disorders.

Personality disorders are both frequent and difficult to treat. Individuals with these problems consume substantial resources and seriously burden society. A survey of the work of Australian psychiatrists showed that although patients with personality disorders amounted to only 6 percent of the people in treatment, they exhausted 13 percent of the psychiatrists' treatment time (Andrews and Hadzi-Pavlovic).

The term personality disorders includes diverse conditions that originated in very different clinical, theoretical and research settings, as Rutter pointed out. Some conditions, such as the concept of borderline personality disorder, are mainly the result of psychodynamic thinking. Others, such as schizotypal personality disorder, have their origin predominantly in genetics. Irrespective of the history of each construct, most experts agree that personality disorders emanate from childhood.

Despite this, little empirical evidence is available on the developmental pathways that underpin these disorders, with the exception of antisocial personality disorder (Robins; Zoccolillo and others). This evidentiary lack may be the consequence of ambiguities surrounding the concepts of temperament, character and personality (Rutter) or the result of child and adolescent psychiatrists being reluctant to make a diagnosis of personality disorder. The recent surge of research reports dealing with personality disorders in adolescent groups suggests that this reluctance may be waning.

Because personality is shaped by experiences during childhood and adolescence, it is likely that mental disorders occurring during these years may have an influence on personality development. Childhood mental disorders may increase the risk of affected children developing a personality disorder when they grow up. This can happen in a variety of ways. The disorder itself, for example, depression, may directly influence personality development. Alternatively, symptoms of the condition (e.g., disruptive disorder) may elicit environmental responses (e.g., increased control or likelihood of abuse by caregivers) that in turn may alter personality development. It may be that the disorder observed in childhood is an earlier manifestation of the same underlying pathology that results in the development of personality disorder in adulthood. Consequently, the study of the continuities between child or adolescent disorders and adult personality disorders is likely to be a fertile ground that will facilitate our understanding of these conditions.

Referred Adolescents

Seeking to clarify some of these issues, my colleagues and I followed up a group of adolescents who had been referred for assessment to an adolescent unit in Sydney, Australia (Rey and others). Follow-up consisted of a lengthy interview during which a variety of diagnostic instruments and questionnaires was administered. These included the Personality Disorders Examination (Loranger). At the time of initial assessment, the average age was 14 years, while at follow-up it was 20 years. Of the 205 subjects who were located, 145 were fully interviewed. About half of these (44 percent) were female.

During the ensuing six years, four of the subjects had died. One female, initially diagnosed as having attention-deficit disorder with hyperactivity, died of a heroin overdose following a period of severe disturbance during which she probably met criteria for conduct disorder. Two males suffered from conduct disorder. One committed suicide; the other died of multiple organ failure caused by hepatitis one day after being released from prison. One male had an adjustment disorder with disturbance of conduct. Reports from relatives at the time of follow-up suggest he was well-adjusted. He died in a car accident. There were 114 (56 percent) individuals with a disruptive disorder diagnosis among the 205 subjects located. Although numbers are too small to draw conclusions, these findings suggest that mortality (3.5 percent) among adolescents with these conditions is likely to be high.

Young Adults

The main findings from the examination of the relationships between adolescent diagnosis, developmental variables and adult personality disorder were:

• Personality disorders were frequent (28 percent), particularly among those with disruptive diagnoses (40 percent). Further, of those young adults with a personality disorder, one-third had more than one personality disorder diagnosis.
• Females were more likely than males to have a Cluster C personality disorder. However, males were not more likely to have a Cluster B disorder once the effect of adolescent diagnosis was taken into account.
• Adolescents with disruptive disorders were more likely to have a Cluster B personality disorder than those with an emotional disorder but not less likely to have a Cluster C disorder.
• Significant associations existed between conduct disorder and antisocial personality disorder and between attention-deficit/hyperactivity disorder .i.attention-deficit/ hyperactivity disorder;(ADHD) and borderline personality disorder.
• Quality of the family environment before the age of 12 years was the most robust developmental predictor of personality disorder in young adults. Externalizing symptoms between the ages of 4 and 7 years was also predictive of Cluster B personality disorders (unpublished data).

ADHD and Borderline