Five of the articles in this month’s issue of the Journal touch on aspects of developmental psychopathology. Traditionally the province of child psychologists and child psychiatrists, developmental psychopathology, with its emphasis on gene-­environment interactions during the course of brain development, is emerging as a key discipline for our field as a whole. Despite the complexity inherent in the multiplicity of genes interacting within cellular domains to guide CNS morphogenesis and the subsequent self-organizing connectivity of neural structures, there are a few simple conclusions about the continuities and discontinuities of psychopathology that echo through the paragraphs of these reports.

First, it is now well established that early-onset axis I psychopathology is an important risk factor for the development and persistence of axis I psychopathology in adolescence and adulthood (1–3). This continuity is best documented in the case of disruptive disorders (1) and depression (2). Two of the articles in this issue highlight selected aspects of these continuities in community-based groups of adolescent subjects. In one report, Disney and colleagues usefully explore the relative importance of conduct disorder versus attention deficit hyperactivity disorder (ADHD) as predisposing factors for later substance abuse among older adolescents. In the other report, Pilowsky and colleagues point to an association between presumptive panic attacks and suicidal ideation and suicide attempts in 13- to 14-year-olds. Of interest in this study group is that the association between panic attacks and suicidal ideation and behavior held even after controlling for the effects of major depression and the use of illicit drugs and alcohol.

Second, early-onset axis I psychopathology is an important risk factor for the development and persistence of axis II psychopathology in adulthood (4–6). Building on earlier results from long-term longitudinal studies from the New York State Psychiatric Institute (4, 5) and the Oregon Research Institute (6), Kasen and colleagues report that a range of early-onset disorders significantly increased the risk of developing personality disorders in adulthood. The increase in risk was nearly monotonic: as the number of axis I diagnoses increased, so did the risk of having a personality disorder. For example, the risk of being diagnosed with a cluster B (antisocial, borderline, histrionic, and narcissistic) DSM-III-R personality disorder increased from 11.6% for subjects with no axis I diagnosis, to 21.9% for one axis I diagnosis, to 33.9% for two or three axis I diagnoses, to 50.0% for four or more axis I diagnoses. Similar increases were observed for both cluster A (paranoid, schizoid, and schizotypal) and cluster C (avoidant, dependent, and passive-aggressive) personality disorders. The range of prior axis I diagnoses that contributed to these findings included ADHD, conduct disorder, oppositional defiant disorder, overanxious disorder, separation anxiety disorder, social phobia, and major depressive disorder.

The association between adult personality disorders and early-onset axis I disorders, if confirmed in future studies, is heuristically and clinically significant. Conceptually, these findings may point toward more precise models of pathogenesis for personality disorders. They also point to the far-reaching and potentially devastating effects of psychopathology in childhood, since psychological troubles early on have enduring adverse effects on self-perceptions, the capacity to form close interpersonal relationships, and academic performance.

A third conclusion is that the presence of personality disorder symptoms in adolescence is likely to be a significant risk factor for subsequent psychiatric morbidity in adulthood (6). The report by Levy and colleagues provides limited evidence that hospitalized adolescents with borderline personality disorder are more likely to require subsequent inpatient hospitalizations and to use illicit drugs than hospitalized adolescents without a personality disorder diagnosis. Although the stability of axis II disorders diagnosed in adolescence remains in doubt (7, 8), it is clear that individuals diagnosed in adolescence with a personality disorder remain at higher risk for axis II disorders in adulthood (reference 7 and the report by Kasen et al.).

Fourth, dimensional, as opposed to strictly categorical, approaches appear to provide the best means to examine the continuities between mental disorders in childhood and adulthood (6, 9–12). Parsing the phenomenology of axis I and axis II disorders into more elemental symptom dimensions—often continuous with normal behavioral traits—may help to resolve the role of specific genes and neurobiological systems (13, 14). Thus far, the best example is dyslexia, where the use of specific phenotypic dimensions resolved long-standing controversies concerning the presence of susceptibility genes in particular genomic regions (15). Although none of these articles addresses directly the role of specific susceptibility genes, the collection of DNA specimens from the participants in the population-based prospective longitudinal studies should be encouraged (reference 6 and the studies by Disney et al., Pilowsky et al., and Kasen et al). As specific susceptibility genes are identified and characterized, the extensive phenotypic data on these subjects should prove to be an invaluable resource in clarifying genotype-phenotype associations.

Fifth, in addition to the action of specific susceptibility alleles, the role of early epigenetic events in shaping an individual’s developmental course is a major factor. The article by Terr and colleagues describing the psychological sequelae of the Challenger spacecraft explosion provides a glimpse into this important domain. Although the minimal long-term impact of this event nicely illustrates the psychological resilience of normal children when faced with a single discrete adverse event with virtually no enduring threat, the results of this study belie the serious role adverse environmental events can play in shaping maladaptive outcomes. The likelihood of psychopathological outcomes depends on the number of childhood traumas, the severity of the trauma, the duration of the trauma, the relationship with the perpetrator, and in cases of abuse, the availability of supports, other stressors, and the child’s inherent vulnerability to developing psychopathology.

For example, child abuse is a pervasive social problem that functions as a nonspecific risk factor for multiple forms of psychopathology, including posttraumatic stress disorder, personality disorders, major depression, self-injurious behaviors, eating disorders, ADHD, conduct disorder, and drug and alcohol abuse (16, 17). Remarkably, the rates of physical and sexual abuse among children and adolescents referred for treatment range from 30% to 60% (17). Not surprisingly, there is mounting evidence that the chronic adversity and threat associated with early abuse and neglect have enduring neurobiological as well as psychological consequences (18).

Sixth, too many of America’s children have diagnosable mental disorders or are at increased risk of developing them (6, 19, 20, and the studies by Disney et al., Pilowsky et al., and Kasen et al.). Within this complex, multidimensional landscape, genes and environments often conspire to derail normal developmental pathways. Individuals with dreadful personal histories of deprivation and abuse and an accumulating set of psychiatric comorbidities require a lot of care and often do not obtain all the psychiatric care they need, especially when they are in foster care or in the criminal justice system. While early detection and intervention hold some promise, primary prevention must be a priority (21, 22). Adverse prenatal and perinatal events can greatly influence an individual’s susceptibility to a range of conditions, including ADHD, tic disorders, and schizophrenia (23–25). The right time to intervene is as early as possible—in the first years of life, during the prenatal period, and with prospective parents in their adolescence.

While these articles help us to understand more about the development of various forms of psychopathology, they also highlight our need to better understand the mechanisms by which early psychopathology promotes the persistence of certain disorders and the onset of others. The need for more research is clear. The level of funding for research on these early-onset disorders remains far lower than it is for comparable medical conditions (19–21). Advances in the developmental neurosciences make this an ideal time to move into this area of investigation—to identify specific vulnerability genes and epigenetic risk factors and to develop animal models. There is a pressing need to recruit and sustain centers of excellence and the next generation of clinical investigators. Finally, community-wide support is needed to develop an integrated system of care that can provide high-quality services tailored to individual children and their families. Prevention and continuity of care is the right prescription.

Address reprint requests to Dr. Leckman, I-267 Sterling Hall of Medicine, Child Study Center, Yale University School of Medicine, 230 South Frontage Rd., New Haven, CT 06520-7900; james.leckman@yale.edu (e-mail).