Discriminant validity, diagnostic utility, and parent-child agreement on the Screen for Child Anxiety Related Emotional Disorders (SCARED) in treatment- and non-treatment-seeking youth.

The Screen for Child Anxiety and Related Emotional Disorder (SCARED) may be differentially sensitive to detecting specific or comorbid anxiety diagnoses in treatment-seeking and non-treatment-seeking youth. We assessed the SCARED's discriminant validity, diagnostic utility, and informant agreement using parent- and self-report from healthy and treatment-seeking anxious youth (Study 1, N=585) and from non-treatment-seeking anxious youth (Study 2, N=331) diagnosed with generalized anxiety disorder (GAD), social anxiety disorder (SAD), or comorbid GAD+SAD. Among treatment-seeking youth, the SCARED showed good diagnostic utility and specificity, differentiating healthy, comorbid, and non-comorbid anxious youth. Child-parent agreement was modest: healthy child self-reports were higher than parent-reports whereas anxious child self-reports were similar or lower than parent-reports. Less consistent results emerged for diagnostic utility, specificity, and informant agreement among non-treatment-seeking youth. Given the number of non-treatment seeking anxious youth (N=33), generalizability of these findings may be limited. Together, results suggest informants may provide distinct information about children's anxiety symptoms.

Social support buffers the effect of interpersonal life stress on suicidal ideation and self-injury during adolescence.

BACKGROUND: The effect of life stress on suicidal symptoms during adolescence is well documented. Stressful life events can trigger suicidality, but most adolescents are resilient and it is unclear which factors protect against the deleterious impact of stress. Social support is thought to be one such factor. Therefore, we investigated the buffering effect of specific sources of social support (parental and peer) on life stress (interpersonal and non-interpersonal) in predicting suicidal symptoms during adolescence. In order to test the specificity of this stress buffering, we also examined it with regard to dysphoric mood. METHOD: Data come from the Adolescent Development of Emotions and Personality Traits (ADEPT) Project, a cohort of 550 adolescent females aged 13.5-15.5 recruited from Long Island. Self-reported social support, suicidality, and dysphoria were assessed at baseline and suicidality and dysphoria were assessed again at 9-month follow-up. Life stress was assessed by interview at the follow-up. RESULTS: High levels of parental support protected adolescent girls from developing suicidal symptoms following a stressor. This effect was less pronounced for peer support. Also, social support did not buffer the pathogenic effects of non-interpersonal stress. Finally, social support did not buffer the effect of life stress on dysphoric symptoms. CONCLUSIONS: Altogether, our results highlight a distinct developmental pathway for the development of suicidal symptoms involving parental support that differs from the development of dysphoria, and signifies the importance and specificity of social support in protecting against suicidality in adolescent girls.

Disruptive mood dysregulation disorder at the age of 6 years and clinical and functional outcomes 3 years later.

BACKGROUND: Little is known about the predictive validity of disruptive mood dysregulation disorder (DMDD). This longitudinal, community-based study examined associations of DMDD at the age of 6 years with psychiatric disorders, functional impairment, peer functioning and service use at the age of 9 years. METHOD: A total of 473 children were assessed at the ages of 6 and 9 years. Child psychopathology and functional impairment were assessed at the age of 6 years with the Preschool Age Psychiatric Assessment with parents and at the age of 9 years with the Kiddie-Schedule of Affective Disorders and Schizophrenia (K-SADS) with parents and children. At the age of 9 years, mothers, fathers and youth completed the Child Depression Inventory (CDI) and the Screen for Child Anxiety Related Disorders, and teachers and K-SADS interviewers completed measures of peer functioning. Significant demographic covariates were included in all models. RESULTS: DMDD at the age of 6 years predicted a current diagnosis of DMDD at the age of 9 years. DMDD at the age of 6 years also predicted current and lifetime depressive disorder and attention-deficit/hyperactivity disorder (ADHD) at the age of 9 years, after controlling for all age 6 years psychiatric disorders. In addition, DMDD predicted depressive, ADHD and disruptive behavior disorder symptoms on the K-SADS, and maternal and paternal reports of depressive symptoms on the CDI, after controlling for the corresponding symptom scale at the age of 6 years. Last, DMDD at the age of 6 years predicted greater functional impairment, peer problems and educational support service use at the age of 9 years, after controlling for all psychiatric disorders at the age of 6 years. CONCLUSIONS: Children with DMDD are at high risk for impaired functioning across childhood, and this risk is not accounted for by co-morbid conditions.

Personality diatheses and Hurricane Sandy: effects on post-disaster depression.

BACKGROUND: According to diathesis-stress models, personality traits, such as negative emotionality (NE) and positive emotionality (PE), may moderate the effects of stressors on the development of depression. However, relatively little empirical research has directly examined whether NE and PE act as diatheses in the presence of stressful life events, and no research has examined whether they moderate the effect of disaster exposure on depressive symptoms. Hurricane Sandy, the second costliest hurricane in US history, offers a unique opportunity to address these gaps. METHOD: A total of 318 women completed measures of NE and PE 5 years prior to Hurricane Sandy. They were also assessed for lifetime depressive disorders on two occasions, the latter occurring an average of 1 year before the hurricane. Approximately 8 weeks after the disaster (mean = 8.40, s.d. = 1.48 weeks), participants completed a hurricane stress exposure questionnaire and a measure of current depressive symptoms. RESULTS: Adjusting for lifetime history of depressive disorders, higher levels of stress from Hurricane Sandy predicted elevated levels of depressive symptoms, but only in participants with high levels of NE or low levels of PE. CONCLUSIONS: These findings support the role of personality in the development of depression and suggest that personality traits can be useful in identifying those most vulnerable to major stressors, including natural disasters.

Affective modulation of the startle response among children at high and low risk for anxiety disorders.

BACKGROUND: Identifying early markers of risk for anxiety disorders in children may aid in understanding underlying mechanisms and informing prevention efforts. Affective modulation of the startle response indexes sensitivity to pleasant and unpleasant environmental contexts and has been shown to relate to anxiety, yet the extent to which abnormalities in affect-modulated startle reflect vulnerability for anxiety disorders in children has yet to be examined. The current study assessed the effects of parental psychopathology on affective modulation of startle in offspring. METHOD: Nine-year-old children (n = 144) with no history of anxiety or depressive disorders completed a passive picture viewing task in which eye-blink startle responses were measured during the presentation of pleasant, neutral, and unpleasant images. RESULTS: Maternal anxiety was associated with distinct patterns of affective modulation of startle in offspring, such that children with maternal histories of anxiety showed potentiation of the startle response while viewing unpleasant images, but not attenuation during pleasant images, whereas children with no maternal history of anxiety exhibited attenuation of the startle response during pleasant images, but did not exhibit unpleasant potentiation - even when controlling for child symptoms of anxiety and depression. No effects of maternal depression or paternal psychopathology were observed. CONCLUSIONS: These findings suggest that both enhanced startle responses in unpleasant conditions and failure to inhibit startle responses in pleasant conditions may reflect early emerging vulnerabilities that contribute to the later development of anxiety disorders.

Familial risk for distress and fear disorders and emotional reactivity in adolescence: an event-related potential investigation.

BACKGROUND: The late positive potential (LPP) is an event-related potential component that is sensitive to the motivational salience of stimuli. Children with a parental history of depression, an indicator of risk, have been found to exhibit an attenuated LPP to emotional stimuli. Research on depressive and anxiety disorders has organized these conditions into two empirical classes: distress and fear disorders. The present study examined whether parental history of distress and fear disorders was associated with the LPP to emotional stimuli in a large sample of adolescent girls. METHOD: The sample of 550 girls (ages 13.5-15.5 years) with no lifetime history of depression completed an emotional picture-viewing task and the LPP was measured in response to neutral, pleasant and unpleasant pictures. Parental lifetime history of psychopathology was determined via a semi-structured diagnostic interview with a biological parent, and confirmatory factor analysis was used to model distress and fear dimensions. RESULTS: Parental distress risk was associated with an attenuated LPP to all stimuli. In contrast, parental fear risk was associated with an enhanced LPP to unpleasant pictures but was unrelated to the LPP to neutral and pleasant pictures. Furthermore, these results were independent of the adolescent girls' current depression and anxiety symptoms and pubertal status. CONCLUSIONS: The present study demonstrates that familial risk for distress and fear disorders may have unique profiles in terms of electrocortical measures of emotional information processing. This study is also one of the first to investigate emotional/motivational processes underlying the distress and fear disorder dimensions.

Natural course of cannabis use disorders.

BACKGROUND: Despite its importance as a public health concern, relatively little is known about the natural course of cannabis use disorders (CUDs). The primary objective of this research was to provide descriptive data on the onset, recovery and recurrence functions of CUDs during the high-risk periods of adolescence, emerging adulthood and young adulthood based on data from a large prospective community sample. METHOD: Probands (n = 816) from the Oregon Adolescent Depression Project (OADP) participated in four diagnostic assessments (T1-T4) between the ages of 16 and 30 years, during which current and past CUDs were assessed. RESULTS: The weighted lifetime prevalence of CUDs was 19.1% with an average onset age of 18.6 years. Although gender was not significantly related to the age of initial CUD onset, men were more likely to be diagnosed with a lifetime CUD. Of those diagnosed with a CUD episode, 81.8% eventually achieved recovery during the study period. Women achieved recovery significantly more quickly than men. The recurrence rate (27.7%) was relatively modest, and most likely to occur within the first 36 months following the offset of the first CUD episode. CUD recurrence was uncommon after 72 months of remission and recovery. CONCLUSIONS: CUDs are relatively common, affecting about one out of five persons in the OADP sample prior to the age of 30 years. Eventual recovery from index CUD episodes is the norm, although about 30% of those with a CUD exhibit a generally persistent pattern of problematic use extending 7 years or longer.

DSM-5 disruptive mood dysregulation disorder: correlates and predictors in young children.

BACKGROUND: Despite the inclusion of disruptive mood dysregulation disorder (DMDD) in DSM-5, little empirical data exist on the disorder. We estimated rates, co-morbidity, correlates and early childhood predictors of DMDD in a community sample of 6-year-olds. METHOD: DMDD was assessed in 6-year-old children (n = 462) using a parent-reported structured clinical interview. Age 6 years correlates and age 3 years predictors were drawn from six domains: demographics; child psychopathology, functioning, and temperament; parental psychopathology; and the psychosocial environment. RESULTS: The 3-month prevalence rate for DMDD was 8.2% (n = 38). DMDD occurred with an emotional or behavioral disorder in 60.5% of these children. At age 6 years, concurrent bivariate analyses revealed associations between DMDD and depression, oppositional defiant disorder, the Child Behavior Checklist - Dysregulation Profile, functional impairment, poorer peer functioning, child temperament (higher surgency and negative emotional intensity and lower effortful control), and lower parental support and marital satisfaction. The age 3 years predictors of DMDD at age 6 years included child attention deficit hyperactivity disorder, oppositional defiant disorder, the Child Behavior Checklist - Dysregulation Profile, poorer peer functioning, child temperament (higher child surgency and negative emotional intensity and lower effortful control), parental lifetime substance use disorder and higher parental hostility. CONCLUSIONS: A number of children met DSM-5 criteria for DMDD, and the diagnosis was associated with numerous concurrent and predictive indicators of emotional and behavioral dysregulation and poor functioning.

Additive effects of the dopamine D2 receptor and dopamine transporter genes on the error-related negativity in young children.

The error-related negativity (ERN) is a negative deflection in the event-related potential that occurs approximately 50 ms following the commission of an error at fronto-central electrode sites. Previous models suggest dopamine plays a role in the generation of the ERN. We recorded event-related potentials (ERPs) while 279 children aged 5-7 years completed a simple Go/No-Go task; the ERN was examined in relation to the dopamine D2 receptor (DRD2) and dopamine transporter (DAT1) genes. Results suggest an additive effect of the DRD2 and DAT1 genotype on ERN magnitude such that children with at least one DRD2 A1 allele and children with at least one DAT1 9 allele have an increased (i.e. more negative) ERN. These results provide further support for the involvement of dopamine in the generation of the ERN.

Dyadic discord at baseline is associated with lack of remission in the acute treatment of chronic depression.

BACKGROUND: Dyadic discord, while common in depression, has not been specifically evaluated as an outcome predictor in chronic major depressive disorder. This study investigated pretreatment dyadic discord as a predictor of non-remission and its relationship to depressive symptom change during acute treatment for chronic depression. METHOD: Out-patients with chronic depression were randomized to 12 weeks of treatment with nefazodone, the Cognitive Behavioral Analysis System of Psychotherapy or their combination. Measures included the Marital Adjustment Scale (MAS) and the Inventory of Depressive Symptomatology - Self Report (IDS-SR30). Of 681 original patients, 316 were partnered and 171 of these completed a baseline and exit MAS, and at least one post-baseline IDS-SR30. MAS scores were analysed as continuous and categorical variables ('dyadic discord' v. 'no dyadic discord' defined as an MAS score >2.36. Remission was defined as an IDS-SR30 of 14 at exit (equivalent to a 17-item Hamilton Rating Scale for Depression of 7). RESULTS: Patients with dyadic discord at baseline had lower remission rates (34.1%) than those without dyadic discord (61.2%) (all three treatment groups) (chi2=12.6, df=1, p=0.0004). MAS scores improved significantly with each of the treatments, although the change was reduced by controlling for improvement in depression. Depression remission at exit was associated with less dyadic discord at exit than non-remission for all three groups [for total sample, 1.8 v. 2.4, t(169)=7.3, p<0.0001]. CONCLUSIONS: Dyadic discord in chronically depressed patients is predictive of a lower likelihood of remission of depression. Couple therapy for those with dyadic discord may increase remission rates.

Family study of subthreshold psychopathology in a community sample.

BACKGROUND: There has been increasing interest in the validity and familial transmission of subthreshold psychiatric conditions and the relationship between subthreshold conditions and full syndrome (FS) disorders. However, most of these studies examined a single subthreshold condition and thus fail to take into account the high co-morbidity among subthreshold conditions and between subthreshold conditions and FS disorders. METHOD: A family study of subthreshold psychiatric conditions was conducted with 739 community-drawn young adults and their 1744 relatives. We examined (1) whether relatives of probands with subthreshold major depression, bipolar disorder, anxiety disorders, alcohol use, substance use, and/or conduct disorder exhibited an increased rate of the corresponding (homotypic) FS disorder; (2) whether subthreshold disorders were associated with increased familial rates of other (heterotypic) FS disorders; (3) whether subthreshold and FS conditions are associated with similar familial liabilities; and (4) whether these homotypic and heterotypic associations persisted after controlling for co-morbidity. RESULTS: Significant homotypic associations were observed for subthreshold anxiety, alcohol, conduct, and a trend was observed for major depression. Only the homotypic association for alcohol and conduct remained after controlling for co-morbid subthreshold and FS conditions. Many heterotypic associations were observed and most remained after controlling for co-morbidity. CONCLUSIONS: It is important to broaden the study of subthreshold psychopathology to multiple disorders. In particular cases, controlling for co-morbidity with other subthreshold and FS conditions altered the patterns of familial aggregation. Etiological processes that are common to particular disorders and subthreshold conditions are discussed.

Bipolar disorders during adolescence.

OBJECTIVE: To examine the incidence, correlates, course and family history of bipolar disorder (BD) and subthreshold BD in adolescents. METHOD: Structured diagnostic interviews were conducted with a large community sample of adolescents and their first-degree relatives, and the adolescents were re-evaluated as young adults. RESULTS: The first lifetime onset of BD and subthreshold BD almost always occurred in adolescence. Adolescent BD and subthreshold BD were associated with elevated impairment, comorbidity, and suicide attempts. Adolescents with BD were at increased risk for BD, and adolescents with subthreshold BD were at increased risk for major depressive disorder (MDD) in young adulthood. Relatives of BD adolescents had elevated rates of subthreshold BD and MDD, and relatives of subthreshold BD adolescents had elevated rates of BD and MDD. CONCLUSION: 'Classical' BD clearly exists in adolescence, but there is also a spectrum of milder bipolar conditions. Remediating and preventing BD in adolescents should be a high public health priority.

Patient's therapeutic skill acquisition and response to psychotherapy, alone or in combination with medication.

BACKGROUND: We tested the hypotheses that the addition of medication to psychotherapy enhances participation in the latter by: (1) speeding the acquisition of the psychotherapy's targeted skill; and (2) facilitating higher skill level acquisition. METHOD: Participants were 431 chronically depressed patients who received Cognitive Behavioral Analysis System of Psychotherapy (CBASP), alone (N=214) or in combination with nefazodone (N=217), as part of a randomized chronic depression study (Keller et al. 2000). CBASP, developed specifically to treat chronic depression, uses a specific procedure, 'situational analysis' to help patients engage in more effective goal-oriented interpersonal behaviours. At the end of each session, therapists rated patients on their performance of situational analysis. Outcome on depressive symptoms was assessed with the 24-item Hamilton Rating Scale for Depression. RESULTS: Although reductions in depression were significantly greater in combined treatment compared to CBASP alone, there were no between-group differences in either the rate of skill acquisition or overall skill level at the end of treatment. Proficiency in the use of the main skill taught in psychotherapy at treatment midpoint predicted outcome independently of medication status and of baseline depressive severity. CONCLUSIONS: Effective participation in CBASP, as reflected by proficiency in the compensatory skill taught in psychotherapy, is not enhanced by the addition of medication and does not mediate the between-group difference in depression outcome.

Family study of co-morbidity between major depressive disorder and anxiety disorders.

BACKGROUND: Numerous studies have documented high rates of co-morbidity between major depressive disorder (MDD) and the anxiety disorders (ANX). However, the reason for this is unclear. Family studies provide one potentially useful approach for addressing this issue. METHOD: We explored six explanations of the co-morbidity between MDD and ANX using a family study of a large community sample of young adults and their first-degree relatives. Participants included 112 probands with a lifetime history of both MDD and one or more ANX, 290 probands with a history of MDD but no ANX, 43 probands with a history of one or more ANX but no MDD. 352 probands with no lifetime history of either MDD or ANX, and the probands' 2608 first-degree relatives. Probands were assessed using semi-structured diagnostic interviews on two occasions in adolescence and a third time at age 24. Diagnostic data on relatives were collected using both direct and family history interviews. RESULTS: Compared with controls, MDD aggregated in the families of probands with MDD, whether or not they had co-morbid ANX; ANX aggregated in the families of probands with ANX, regardless of whether they had co-morbid MDD; and co-morbid MDD/ANX aggregated only in the families of probands with both MDD and ANX. The relatives of probands with ANX alone had a significantly higher rate of ANX than the relatives of probands with MDD alone, although none of the other comparisons between the depressed and anxious groups were significant. CONCLUSIONS: This pattern of findings is largely, although not completely, consistent with the view that MDD and ANX are transmitted independently within families, and suggests that the comorbidity between MDD and ANX is caused by non-familial aetiological factors.

Outcome of dysthymic disorder at 5-year follow-up: the effect of familial psychopathology, early adversity, personality, comorbidity, and chronic stress.

OBJECTIVE: This study sought to identify predictors of course and outcome in dysthymic disorder. METHOD: Eighty-six outpatients with early-onset dysthymic disorder (before age 21) participated in a prospective 5-year follow-up study. Family history of psychopathology, early home environment, axis I and II comorbidity, social support, and chronic stress were assessed at baseline. The Longitudinal Interval Follow-up Evaluation and the Hamilton Depression Rating Scale were used in the follow-up assessments conducted at 30 and 60 months. RESULTS: Comorbid anxiety disorder, cluster C and depressive personality features, and chronic stress were associated with a lower rate of recovery from dysthymic disorder, while family history of bipolar disorder was associated with a higher probability of recovery. Family history of dysthymic disorder, poor childhood maternal and paternal relationships, childhood sexual abuse, cluster C features, neuroticism, a history of anxiety and eating disorders, and chronic stress predicted higher levels of depression at follow-up. Multivariate models indicated that almost all domains contributed to the prediction of course and outcome. CONCLUSIONS: The course and outcome of dysthymic disorder is best conceptualized within a multifactorial framework, with family history of psychopathology, early adversity, axis I and II comorbidity, and chronic stress all making important contributions.

A construct validation study of the Response Styles Questionnaire Rumination Scale in participants with a recent-onset major depressive episode.

This study examined the construct validity and clinical utility of S. Nolen-Hoeksema's (1991) Response Styles Questionnaire (RSQ) Rumination scale. Eighty-eight participants with recent-onset major depressive episodes were assessed and followed for 6 months, using semistructured interviews and self-report inventories. The RSQ Rumination scale exhibited poor 6-month stability and appeared to be closely linked to participants' clinical status-mood state. The scale was significantly correlated with conceptually related constructs such as emotion-focused coping, negative affectivity-temperament, and self-criticism. However, baseline negative temperament and self-criticism predicted key aspects of the 6-month course and outcome of major depressive episodes, whereas baseline rumination did not. Finally, rumination appeared to be closely associated with the severity of the depressive episode, rather than defining a distinct clinical subtype.

A family study of major depressive disorder in a community sample of adolescents.

BACKGROUND: Family studies provide a useful approach to exploring the continuities and discontinuities between major depressive disorder (MDD) in children and adolescents and MDD in adults. We report a family study of MDD in a large community sample of adolescents. METHODS: Probands included 268 adolescents with a history of MDD, 110 adolescents with a history of nonmood disorders but no history of MDD through age 18 years, and 291 adolescents with no history of psychopathology through age 18 years. Psychopathology in their 2202 first-degree relatives was assessed with semistructured direct and family history interviews, and best-estimate diagnoses were derived with the use of all available data. RESULTS: The relatives of adolescents with MDD exhibited significantly elevated rates of MDD (hazard ratio [HR], 1.77; 95% confidence interval [CI], 1.46-2.31), dysthymia (HR, 1.79; 95% CI, 1. 11-2.87), and alcohol abuse or dependence (HR, 1.29; 95% CI, 1.05-1. 53), but not anxiety disorders, drug abuse or dependence, or antisocial and borderline personality disorder. In contrast, anxiety, substance use, and disruptive behavior disorders in adolescents were not associated with elevated rates of MDD in relatives. However, the relatives of probands with anxiety and substance use disorders exhibited elevated rates of anxiety and substance use disorders, respectively. CONCLUSIONS: The results provide evidence of the familial aggregation of adolescent MDD, and also indicate that there is a considerable specificity in the pattern of familial transmission. In addition, we found preliminary evidence of the familial aggregation of adolescent anxiety and substance use disorders.

Comparison of DSM-III-R chronic major depression and major depression superimposed on dysthymia (double depression): validity of the distinction.

The nosology of chronic depression has become increasingly complex since the publication of the revised third edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-III-R; American Psychiatric Association, 1987), but there are few data available to evaluate the validity of the distinctions between the subtypes of chronic depression. The validity of the distinction between DSM-III-R chronic major depression (CMD) and major depression superimposed on dysthymia (double depression, DD) was examined. Participants were 635 patients with chronic depression in a 12-week trial of antidepressant medications. Patients with CMD, DD, and a 3rd group with a chronic major depressive episode superimposed on dysthymia (DD/CMD) were compared on demographic and clinical characteristics, family history, and response to treatment. Few differences were evident, although the depression of patients with DD/CMD tended to be more severe.

A family study of outpatients with borderline personality disorder and no history of mood disorder.

While several studies have examined psychiatric disorders in the relatives of individuals with borderline personality disorder, many of these studies have not employed a family study methodology and suffer from other methodological shortcomings. Thus, the conclusions from family data addressing the validity of borderline personality disorder, its relation to other conditions, and its distinction from mood disorders, continue to be debated. The present investigation employed a family study design with direct interviews with relatives, structured diagnostic interviews with both probands and relatives, and blind assessment of relatives. Rates of psychiatric disorders were examined in 563 relatives of outpatients with mood disorders (n = 119), 54 relatives of outpatients with borderline personality disorder and no history of mood disorder (n = 11), and 229 relatives of never psychiatrically ill controls (n = 45). Results indicate increased rates of mood disorders and personality disorders in the relatives of borderline probands compared with never psychiatrically ill controls. Familial aggregation of psychiatric disorders was generally similar for borderline personality and the mood disorder comparison group. The results suggest there may be common etiological factors between borderline personality disorder and mood disorders.

Natural course of adolescent major depressive disorder in a community sample: predictors of recurrence in young adults.

OBJECTIVE: The primary purpose was to identify factors related to the recurrence of major depressive disorder during young adulthood (19-23 years of age) in a community sample of formerly depressed adolescents. METHOD: A total of 274 participants with adolescent-onset major depressive disorder were assessed twice during adolescence and again after their 24th birthday. Lifetime psychiatric information was obtained from their first-degree relatives. Adolescent predictor variables included demographic characteristics, psychosocial variables, characteristics of adolescent major depressive disorder, comorbidity, family history of major depressive disorder and nonmood disorder, and antisocial and borderline personality disorder symptoms. RESULTS: Low levels of excessive emotional reliance, a single episode of major depressive disorder in adolescence, low proportion of family members with recurrent major depressive disorder, low levels of antisocial and borderline personality disorder symptoms, and a positive attributional style (males only) independently predicted which formerly depressed adolescents would remain free of future psychopathology. Female gender, multiple major depressive disorder episodes in adolescence, higher proportion of family members with recurrent major depressive disorder, elevated borderline personality disorder symptoms, and conflict with parents (females only) independently predicted recurrent major depressive disorder. Comorbid anxiety and substance use disorders in adolescence and elevated antisocial personality disorder symptoms independently distinguished adolescents who developed recurrent major depressive disorder comorbid with nonmood disorder from those who developed pure major depressive disorder. CONCLUSIONS: Formerly depressed adolescents with the risk factors identified in this study are at elevated risk for recurrence of major depressive disorder during young adulthood and therefore warrant continued monitoring and preventive or prophylactic treatment.