Depression

Edited by Shane R. Jimerson, Ph.D.
Contributed to by the Graduate Students in the Counseling, Clinical, and School Psychology Program at the University of California, Santa Barbara.

Ongoing design and publication of this site is completed by Shane R. Jimerson, Jeff R. Klein and Angela D. Whipple.  Please forward comments regarding this site to Shane R. Jimerson.  This page was last updated 1.11.02.  © 2002

Symptoms
Epidemiology
Etiology
Assessment
Treatment

Authors
Shane R. Jimerson, Ashley Duggan, Angela Whipple and Jeffrey K. Ellens
University of California, Santa Barbara

The authors also note the important contributions of Traci Marx, Sharon Noble and  Stacey Peerson who provided information critical to the development of this website

DEPRESSION

     During the past decade, as depression is becoming increasingly evident at younger ages, a great deal of attention has been directed towards the diagnosis and treatment of depression in children and adolescents (Birmaher, Ryan, Williamson, & Brent, 1996).  There is a consensus that children and adolescents experience depressed mood, however, controversy continues to exist when differentiating childhood from adult depressive disorders.  It also must be acknowledged that a fundamental difference between adults and children is the ongoing developmental changes that children experience.  Because of difficulties in school settings for depressed children and the high comorbidity with other pathologies, understanding depression in children is important for school professionals. Consequently, a clear need exists for an understanding of normative behaviors as well as the development of diagnostic criteria which are specific for every childhood and adolescent developmental stage.
     Numerous outcome studies have documented several negative effects of depression on children (Hammen & Compas, 1994; Birmaher et al., 1996; Kaslow, Deering & Racusin, 1994).  Childhood depression is commonly associated with high incidences of sympomatic comorbidity, (Angold & Costello, 1993; Birmaher et al., 1996; Hammen & Compas, 1994), wherein, features associated with other disorders are present.  For example, depression and features of anxiety have been found to be correlated (Weems, Hammond-Laurence, Silverman, & Ferguson, 1997).   Other disorders often observed with depression include conduct/ behavioral disorders, and substance abuse.  Symptomatic comorbidity complicates both the detection and treatment of depession by “masking” its presence.  The earlier the onset of depression, the poorer the prognosis for the child (Birmaher et al., 1996; Garrison, Waller, Cuffe, & McKeown, 1997); remaining undetected (and untreated), childhood depression increases a child’s risk for substance abuse, suicidal behavior, and poor psychological, social, and academic functioning (Birmaher et al., 1996; Kaslow et al., 1994).  Additional factors may influence depression as well, including gender, ethnicity, and social class (Cicchetti & Toth, 1998; Siegel, Aneshensel, Taub, Cantwell, & others, 1998).  In order to more clearly understand the multiple risk factors of childhood depression, and in order to better differentiate normative from non-normative developmental trajectories, the ongoing study of youth depression is vital.
 
 


Return to Top
 

SYMPTOMS

     Although the criteria for depression are the same for children and adolescents as they are for adults, there are unique developmental challenges and considerations in diagnosing and treating youth depression.  For example, children often have difficulty expressing or recalling information related to their disorder (Emslie & Mayes, 1999); Due to this complication, it is often critical for purposes of diagnosis that corroborating information be obtained from parents, school teachers, and other adults in the child’s life.  Another factor that complicates diagnosis and treatment is the high rate of comorbidity that is characteristic of childhood depression.  Often times depression is not the only diagnosis warranting clinical attention, and it is important for clinicians to be sensitive to all mental health needs with which a child presents.
     Major Depressive Episode and Dysthymic Disorder are two distinct classifications for depressive symptoms in the DSM-IV (1994).  The severity and duration of the symptoms distinguish the two, as noted in Table 1.  The core symptoms of a Major Depressive Episode, as reported in the DSM-IV (1994), are basically the same for children, adolescents, and adults, despite indications that some characteristic symptoms may vary with age. The criteria do allow a 1-year duration for children and adolescents (as opposed to at least 2 years for adults); and permit the substitution of “irritability” for “depressed mood,” as irritability, somatic complaints, and social withdrawal are more common symptoms of depression for children and adolescents.

DSM-IV Criteria for Depressive Disorders

     Major Depressive Episode and Dysthymic Disorder are two distinct classifications for depressive symptoms in the DSM-IV (1994).  The severity and duration of the symptoms distinguish the two, note below.

     The core symptoms of a Major Depressive Episode, as reported in the DSM-IV (1994), are basically the same for children, adolescents, and adults, despite indications that some characteristic symptoms may vary with age. The criteria do allow a 1-year duration for children and adolescents (as opposed to at least 2 years for adults); and permit the substitution of “irritability” for “depressed mood,” as irritability, somatic complaints, and social withdrawal are more common symptoms of depression for children and adolescents.

 DSM-IV (1994)  Criteria for Major Depressive Episode

 A. Five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure.
 

     (1) depressed mood most of the day, nearly every day.
         Note:  In children and adolescents, can be irritable mood.

     (2) markedly diminished interest or pleasure in all, or  almost all activities most of the day, nearly every day.

     (3) significant weight loss when not dieting or weight  gain or decrease or increase in appetite nearly every day.
     Note:  In children, consider failure to make expected weight gains.

     (4) insomnia or hypersomnia nearly every day.

     (5) psychomotor agitation or retardation nearly every day  (observable by others, not merely subjective  feelings of  restlessness or being slowed down).

     (6) fatigue or loss of energy nearly every day.

     (7)  feelings of worthlessness or excessive or  inappropriate guilt nearly every day (not merely self reproach or guilt about being sick).

     (8) diminished ability to think or concentrate, or  indecisiveness, nearly every day.

     (9) recurrent thoughts of death (not just fear of  dying), recurrent suicidal ideation without a specific plan,  or a suicide attempt or a specific plan for committing  suicide.

     Source.  American Psychiatric Association.  (1994).  Diagnostic and statistical manual of mental disorders (4th ed., p. 327).  Washington  DC:  Author.
 

 DSM-IV (1994)  Criteria for Dysthymic Disorder

 A. Depressed mood for most of the day, for more days than not, as indicated either by subjective account or observation by others, for at least 2 years.  Note:  In children and adolescents, mood can be irritable and duration must be at least 1 year.

 B. Presence, while depressed, of two (or more) of the  following:

 (1) poor appetite or overeating
 (2) insomnia or hypersomnia
 (3) low energy or fatigue
 (4) low self-esteem
 (5) poor concentration or difficulty making decisions
 (6) feelings of hopelessness

 C. During the 2-year period (1 year for children or adolescents) of the disturbance, the person has never been without the symptoms in Criteria A and B for more than 2 months at a time.

 D. No Major Depressive Episode has been present during the first 2 years of the disturbance (1 year for children and adolescents).

 E. There has never been a Manic Episode, a Mixed Episode, or a Hypomanic Episode, and criteria have never been met for Cyclothymic Disorder.

 F. The disturbance does not occur exclusively during the course of a chronic Psychotic Disorder, such as Schizophrenia or Delusional Disorder.

 G. The symptoms are not due to direct physiological effects of a substance or a general medical condition.

 H. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Source.  American Psychiatric Association.  (1994).  Diagnostic and statistical manual of mental disorders (4th ed., p. 349).  Washington, DC:
Author.
 


Return to Top
 

EPIDEMIOLOGY

     Researchers have found it difficult to document the prevalence of depression in children and adolescents due to controversy about the definition and diagnostic criteria, including unclear distinctions between symptoms, disorders, and syndromes.  The various types of assessments are not equal in diagnosing depression in all subjects, and considerable differences have been found in prevalence rates depending upon the population studied (Hammen & Compas, 1994; Kaslow et al., 1994).  Considering the above, the prevalence of depression in prepubertal children is estimated at approximately 2% (Carlson, 2000; Institute of Medicine, 1989).  During adolescence the prevalence of depression increases substantially, with 5% to 10% of adolescents manifesting a major depressive disorder (Cicchetti, & Toth, 1995).  At any given time, approximately 10% to 15% of children in the general population will report moderate to severe levels of depressive symptoms (Nolen-Hoeksema et al., 1992).  Prevalence figures are much higher for various groups of referred and clinical samples of children (Hammen & Compas, 1994).  Although 90% of major depressive episodes remit after 1.5 to 2 years, studies have shown that depression recurs in adulthood for 60% to 70% of depressed children (Birmaher et al., 1996).  Depression is surprisingly common in adolescence, with recent estimations that approximately 28% of adolescents will have experienced a Major Depressive Episode by age 19 (35% of young women and 19% of young men) (Lewinsohn, Rohde, & Seeley, 1998).  It should be noted that these prevalence rates compare to the rates in the National Comorbidity Study (Kessler, McGonagle, Nelson, Hughes, Schwartz, & Glazer, 1994).
     The mean age of onset of first depressive episodes was approximately 15 years of age in an epidemiological study of over 1,500 adolescents (Lewinsohn, Clark, Seeley & Rohde, 1994).  It is interesting to note that children who are diagnosed with a comorbid anxiety disorder have an earlier age of onset (9 to 11 years) of depression (Hammen & Compas, 1994).  Usually gender differences are not reported in children ages 6 to 12; however, some researchers have found that prior to puberty (12 years), boys are more likely to be diagnosed with depression than girls (Angold & Rutter, 1992).  Some research suggests (Suesser, 1998) that these prepubescent gender differences may be the result of boys’ greater tendency to utilize externalizing coping styles (i.e. fighting) when dealing with depression.  Externalizing children are more likely to be targeted, assessed, and consequently diagnosed with depression than children who utilize an internalizing coping strategy (i.e. social withdrawal).  Additionally, the methods clinicians utilize in their diagnosis may help determine whether or not a child (especially a young child) is inevitably diagnosed with depression.  Regardless of what the true gender differences of early childhood depression are, there appears to be a sudden shift in the aforementioned trend beginning in adolescence (following puberty) in which the gender ratio shifts, with depression becoming at least twice as common among females.  One study reported even more dramatic gender differences for adolescent depression finding that girls were four times more likely to suffer from depression than boys (base rates were 13% and 4%, respectively) (Kashani, Carlson, Beck, & Hoeper, 1987).  Clearly, factors associated with puberty warrant further attention in understanding the formation of adolescent depression (Clarizio, 1989).
     With childhood depression, comorbidity is the rule rather than the exception.  Comorbidity, or the co-occurrence of two independent diagnoses, has been found to exist in 40% to 70% of depressed children and adolescents (Birmaher et al., 1996; Hammen & Compas, 1994).  Anxiety, dysthymic and conduct/behavioral disorders, and substance abuse are the most frequent comorbid diagnoses (Birmaher et al., 1996; Hammen & Compas, 1994).

Current Research Regarding Depression Epidemiology

     Angold and Costello (1993) discuss depressive comorbidity in children and adolescents.  Their article reviewed a study that examined comorbidity in children and adolescents experiencing depression across several recent epidemiological studies. Comorbidity, in the context of child and adolescent depression, was examined with three of the common psychiatric disorders of childhood and adolescence:  anxiety disorders, conduct and oppositional disorders, and attention deficit hyperactivity disorders.   The authors addressed four empirical questions:  1) Is depressive comorbidity more common than expected by chance in studies of child and adolescent psychopathology.  In the majority of cases, the disorders (anxiety disorders, conduct and oppositional disorders, and attention deficit hyperactivity) were more common in depressed children than expected by chance.  2) Is depression associated with a greater risk of other disorders in general, or do some diagnoses accompany depression more frequently than do others?  Depression co-occurs with anxiety, conduct and oppositional, and attention deficit disorders more frequently than with other disorders in general.  3) Are comorbid disorders found more often in children with depression than depression is found in children with other diagnoses?  There was a high rate of comorbidity in children and adolescents with major depressive disorders or dysthymia.  Comorbidity with conduct disorder/oppositional defiant disorder ranged from 21% to 83%; comorbidity with anxiety disorders ranged from 30% to 75%; and comorbidity with attention deficit disorder ranged from 0% to 57.1%; and 4) Is comorbidity more common in clinical than in epidemiological studies of depressed children? Rates of depressive comorbidity found in community studies were similar to rates found in clinical studies. While the authors conclude that the specific mechanisms by which comorbidity occurs are currently obscure they offer several possible substantive explanations that are worth exploring.
     Birmaher, Ryan, Williamson, and Brent (1996) review the literature published over the last decade on issues pertaining to early onset depression. These authors note that prevalence rates of depression in children range from .4% to 2.5%, with a wider range (.4% to 8.3%) indicated for adolescents.  They contend that less information is available for those with dysthymic disorder, and suggest prevalence rates of .6% to 1.7% in children, and 1.6% to 8.0% in adolescents.  Although boys and girls manifest symptoms of Major Depressive Disorder (MDD) equally in childhood, MDD is twice as common in females than in males during their adolescent years. The mean length of MDD is approximately seven to nine months in duration. Though virtually 90% of major depressive episodes remit in 1.5 to 2 years from time of onset, studies consistently find that MDD is a recurrent condition. The interaction of genetics and environment are strongly implicated in the onset of MDD. An individual with a genetic vulnerability is likely to be at increased risk for MDD when experiencing a stressful environment.  Many environmental factors have been associated with MDD in youths.  Parental psychopathology, poor social support, and stressful life events such as divorce or bereavement are merely a few of the factors suggested to influence the onset and natural course of MDD.  Beck associates depression with "low self-esteem, high self-criticism, significant cognitive distortions, and a feeling of lack of control over negative events." (as cited in Birmaher etal., 1996).
     A surprisingly large number of individuals struggle with difficulties associated with additional psychiatric problems.  Forty to seventy percent of depressed youths have comorbid psychiatric disorders, with 20 to 50% experiencing two or more comorbid disorders. Comorbidity is problematic in that it appears to contribute to a more negative life course.  Those with comorbid psychiatric diagnoses that include MDD are at greater risk for recurrent depression, more enduring depressive episodes, suicidal behaviors, poorer response to treatment and underutilization of psychological services.  Children and adolescents with depression frequently suffer in their school performance, as well as in their relationships with others.  Social skills deficits have also shown a significant correlation with depression, although the direction of relationship is not as clear (Segrin, 2000).  According to Birmaher, Ryan, Williamson, & Brent (1996), "depression in children and adolescents is also associated with an increased risk of suicidal behaviors, homicidal ideation, tobacco use, and abuse of alcohol and other substances during later adolescence." (p. 1431). Several studies cited by Birmaher et al. suggest that suicide accounts for a staggering 12% of the total mortality of adolescents.  These authors emphasize the heterogeneous nature of depression.  They imply that its pathogenesis may be complex and varied from one affected individual to the next.
     Cicchetti and Toth (1998) provide a thorough article on the development of depression in children and adolescents.  Their article presents a developmental psychopathology view of childhood and adolescent depression, where a multi-level approach is used in examining the diverse pathways that may lead to depressive disorder.  Cicchetti and Toth (1998) include research that supports current views on the definitions, epidemiology, clinical characteristics, and gender differences in childhood depression.  The possible relationship between depressed mood and that of the child’s affect and attachment to a caregiver are also explored.  As such, the ontogenic development of depression is evaluated, including homeostatic and physiological regulation, the role of affect, attention, and arousal, the attachment relationship and the self-system.  A transactional model incorporating a social ecological framework is introduced, including an overview of the microsystem(s), exosystem(s), and macrosystem(s) that may be involved.  These systems include the family, home environment, neighborhood, school, community-at-large, culture, and societal support.  Prevention, child competence, and therapeutic strategies are promoted, with an acknowledgement of the larger developmental context as key.
     Garrison, Waller, Cuffe, and McKeown (1997) give evidence for the incidence of major depressive disorder and dysthymia in young adolescents.  They conducted a longitudinal study examining depression and dysthymia in a sample of junior high and high school students from one public school district in the southeastern United States.  Data collection included a school screening and diagnostic interview process. The former procedures consisted of a self-administered questionnaire that obtained demographic information; the Center for Epidemiologic Studies Depression Scale (CES-D), with three additional items assessing suicide intent; a modified Coddington Life Event Schedule for Adolescents (LES-A); and the Family Adaptability and Cohesion Evaluation Scales (FACES-II). These instruments evaluated depressive symptomatology/suicidality, stress from undesirable life events, and level of perceived emotional bonding/individual autonomy in the family system.  Students selected to participate in the diagnostic interview included individuals whose scores on the CES-D indicated higher risk for major depressive disorder (MDD), and a random sample of remaining students from phase one screening.  During the diagnostic interview, an adolescent and his/her parent (typically the mother) completed separate semi-structured interviews.  These included the Present Episode version of the Schedule for Affective Disorders and Schizophrenia for School-Age Children (K-SADS), the Children's Global Assessment Scale (CGAS), and the Hollingshead Two Factor Index of Social Position.  These instruments assessed for psychiatric disorders during the previous year, level of functional impairment, and household socioeconomic level, respectively.
     Lewinsohn, Rohde, and Seeley, (1998) summarize current understanding of major depressive disorders (MSD) in adolescents 14 to 18 years old.  Results are part of a larger program of research under the Oregon Adolescent Depression Project.  Pheomenology of major depressive disorder is described in terms of frequency of symptoms of MDD, cases observed in clinics as compared to the community, and age and gender difference in symptom presentation.  Epidemiology is described in terms of MDD onset, duration of depressive episodes and time to recurrence, and the impact of age, gender, and pubertal timing.  Occurrence of suicidal behavior linked to suicide is also discussed.  Comorbidity of depression is described as occurring with dysthemia and other mental disorders, and implications for high comorbidity are presented.  A psychosocial link with becoming, being, and having been depressed is described in some detail.  An etiologic model of depression is presented, and assessment and screening options are described.  Finally, treatment and prevention options are compared with regard to efficacy, meeting wants and needs of depressed adolescents, and responses to different types of treatment.  Conclusions and recommendations covering a broad array of treatment and intervention options are presented.
     Moses (1999) presented information regarding exposure to violence, depression, and hostility in inner city high school youth.  This article is presented in response to previous findings that severe stress can have a harmful impact on adolescent emotional adjustment and has been associated with the development of various forms of psychological distress.  The study examines the prevalence of violence in one population of students (N = 337) who attend inner city schools.  Furthermore, the relationship between exposure to violence and depression and hostility among those adolescents was examined.  The method included an anonymous survey where students counted their experiences with traumatic violence, including witnessing shooting or stabbing of a family member, friend, or stranger, being raped, being shot or stabbed, and being beaten up/jumped.  Sixty-two percent of students had been exposed to three of the six types of violence (mean = 3.41).  Exposure to different types of violence showed high multicolinearity.  Males had been exposed to more violence on the whole than females.  Exposure to violence was predictive of hositlity for both males and females, and predictive of depression for females.  Adolescent development and depression within violent exposure are discussed.
     Nolen-Hoeksmea, Girgus, and Seligman (1992) reviewed a longitudinal study in which the interrelationships among children’s experiences of depressive symptoms, negative life events, explanatory style, and helplessness behaviors in social and achievement situations were examined.  The 5-year longitudinal study followed 352 third-grade children assessing several variable at six month intervals over 5 years.  At six month intervals, the participants were administered the Child Depression Inventory (CDI), the Children’s Attributional Style Questionnaire (CASQ), and the Life Events Questionnaire (LEQ). Additionally, the teachers of the students completed a Student Behavior Checklist (SBC) rating students’ helplessness behaviors in a variety of situations.
     The results of this study revealed differences in how children in early childhood and middle childhood experienced depressive symptoms. In early childhood, a pessimistic explanatory style emerged as a significant predictor of depressive symptoms, while in middle childhood, a pessimistic explanatory style emerged as a significant predictor of depressive symptoms, alone and in conjunction with negative events.  The study found that some children repeatedly showed elevated levels of depressive symptoms, even across the 5-year span of childhood.  Additionally, these children tended to show a constellation of pessimistic thinking and helplessness behaviors in the classroom and in peer interactions. The article concluded with suggestions for interventions and prevention programs for children at risk for developing depression.  For example, “family therapy is the therapy of choice for depressed children” in order to stabilize the home environment and facilitate the children’s ability to cope with the remaining stressors. Social skills training was also emphasized to help children overcome “passivity and helplessness in peer interactions.”

Conclusions

     The magnitude of child and adolescent depression is clearly a major mental health problem.  Current research supports a multifaceted view of childhood and adolescent depression, including social, family, psychological, developmental, and medical concerns.  Children continue to be diagnosed with depression at increasing levels, and at younger ages, than ever before, and comorbidity with other disorders is extensive. Early onset depression interferes with a child’s psychological, social, and academic functioning, placing him or her at greater risk for problems such as substance abuse and suicidal behavior (Birmaher et al., 1996; Hammen & Compas, 1994).  A high recurrence rate provides evidence for a need to help children and adolescents learn prevention strategies for future use, as well as providing tools for reducing the risk of subsequent episodes.  Moreover, depression in childhood more often than not co-occurs with other disorders, primarily anxiety, conduct and behavioral disorders. This comorbidity can further impede normative child development, which may result in a more negative life course (Hammen & Compas, 1994; Birmaher et al., 1996; Angold & Costello, 1993).  Attempting to account for the biological, environmental, and developmental influences on the pathogenesis of depression is a complex and challenging task.  To facilitate this understanding, distinctions are necessary in evaluating variables that cause, maintain, or are merely associated with childhood depression, so that appropriate prevention and intervention measures may be implemented with this high-risk population.
 
 
 


Return to Top
 

ETIOLOGY

     What one sees as the etiology, the cause or origin, of any disorder depends on the lens through which one views psychopathology.  From a medical orientation, disorder result from organic dysfunction; from a behavior model, abnormal behavior has been learned; from a psychodynamic perspective, interpersonal forces are key; and from a social framework, a person's relationship within the social system is examined (Wicks-Nelson & Israel 1996).  For depression, single risk factors can rarely be conceived as resulting in depressive outcomes.  Instead, the biological, psychosocial, and social systems may be considered within a larger framework for explaining the etiology of depression.  This section presents explanations as they are summarized in research articles and web sites and offers a developmental framework to include more than one etiological perspective.
     Although the medical model is prominent today, many consider the etiology of various psychopathologies to be multidimensional, commanding the consideration of the multiple factors that influence a person's life and development.  Since many different factors can lead to psychopathology for different individuals, the etiology of a given disorder is perhaps best understood by looking at [but etiology may be] the interaction or transaction between these multiple variables over time (Cicchetti, Nurcombe & Garber, 1992). The following discussion presents various explanations of the etiology of childhood depression; however, we urge the reader to begin thinking about how these variables may interact with one another throughout development, and the ways in which they put a child at risk for depression.
     Although the explanations provided below help researchers and practitioners  understand the etiology of depression as it relates to specific etiological models, a developmental perspective suggests: 1) the manifestation of depression varies with age [across ages as it is displayed]; 2) how individuals experience depression internally varies greatly from one individual to the next [that depression varies across individuals as it is experienced]; and 3) the context of ongoing development changes must be considered and the causes of depression understood within the child’s developmental level.  With this in mind, aspects unique to childhood should also be considered within a developmental perspective in order to understand the etiology of depression, and to insure that etiology is not be limited to a single perspective.

Biological Influences

     Reynolds (1992) conducted an extensive review of the literature and found several persuasive biological models in the etiological structure of depression.  Many of these models focus on brain neurochemistry, specifically on the role neurotransmitters play in the etiology of depression.  Research has discovered that insufficient levels of chemicals such as norepinephrine and serotonin in the brain lead to a depressed state.  More recently, however, attention has been directed toward growth hormone secretion and sleep disturbance in children and adolescents.
     Neuroendocrine research claims that a relationship exists between hypothalamic-pituitary-adrenal axis function and childhood depression. Literature suggests that individuals with depression have higher levels of a hydrocortisone, cortisol, in their system, a hormone manufactured by the adrenal gland.  This controversial model has spawned a proliferation of research in the study of adolescent and childhood depression.  Both the American Psychiatric Association Task Force on Laboratory Tests in Psychiatry and the World Health Organization state the inconclusive nature of these hypotheses at this time.
 Field, Fox, Pickens, & Nawarocki (1995) provide evidence of brain activity related to depression.   The brain electrical activity of adolescent mothers and their infants were studied using an electroencephalogram (EEG) technique.  A sample of 32 infants were studied, evenly split by gender, with a mean age of 4.8 months.  All were full-term at birth and without medical complications.  The mothers were right-handed, single, had lower socio-economic status.  This sample included only two ethnicities: Hispanic (35%) and African-American (65%).  All mothers were evaluated for depression both neonatally and when their infants were between three and six months of age.  It was found that depressed mothers and their infants displayed right frontal asymmetry.  These findings parallel those of inhibited infants and children, and chronically depressed adults who exhibit depressed affect as well.  The authors suggest that further research is needed to determine if these EEG patterns are indicative of current or chronic mood states.
     Steingard (2000) recently conducted a review of research on the neuroscience of childhood and adolescent depression that included scientific studies spanning more than a decade.  According to Steingard, recent endocrine studies show inconsistent results; one of the studies under review found higher rates of depression among children with high levels of nocturnal growth hormone secretion, however, these results were not replicated by subsequent researchers.  Recently, sleep studies have gained much attention in the childhood and adolescent depression literature.  It has been noted by several authors that children with a tendency toward reduced REM latency will be more likely to show signs of depression in subsequent years than normal controls.  But as the author warns, such REM sleep patterns are also seen in children exposed to stressful life events and so results should be interpreted with caution.  Neuroimaging studies, especially magnetic resonance imaging (MRI) studies show promise in helping us understand the etiology of childhood depression.
     One reason for the current popularity of MRI’s is that they do not require radionucleotide exposure common of other neuroimaging techniques and, thus, repeated scanning using MRI’s appears to be safe for use with developing children.  Results of MRI neuroimaging show that children suffering from depression have decreased frontal lobe volume when compared with normal age mates.  The author states that future MRI studies will more clearly identify the psychophisiological markers of depression; this will, in turn, facilitate our ability to identify individuals at risk for developing depression and allow us to predict patient response to treatment.
     Apter et al. (1999) provide evidence for increased cholesterol as a possible biologic risk factor in depression.  Their study was undertaken to examine the relationship between serum cholesterol levels and suicidal behaviors in adolescent psychiatric inpatients.  Any association between serum cholesterol and measures of suicidal behavior, impulsivity, aggression, anxiety, and depression was also examined.  Individuals admitted to an adolescent inpatient unit were assessed for their suicidal behavior, violence, impulsivity, and depression.  Serum cholesterol for individuals admitted for suicidal tendencies were compared with serum cholesterol for individuals admitted for other reasons.  Serum cholesterol levels were significantly higher among patients who were suicidal than those who were admitted for other reasons.  Within the suicidal group, serum cholesterol was found to have a negative correlation with the degree of suicidal behavior.  Implications for associations between cholesterol and suicidal tendencies are presented, and the usefulness of understanding serum cholesterol as a possible risk factor in depression is presented.

Genetic Influences

     Recently, the focus of empirical studies and literature reviews on child and adolescent phychopathology has moved beyond documenting the prevalence of, and risk factors for individual disorders, as these data are reasonably well documented.  O’Connor, McGuire, Reiss, Hetherington, and Plomin (1998) extend previous research documenting the prevalence of depression by studying the patterns of co-occurring disorders.  Specifically, the highlight the consistently high rates of co-occurring dimension of phychopathology, particuarly between internalizing and externalizing disorders.  Using a sample of same-sex adolescent siblings between 10 and 18 years of age consisting of monozygotic and dizygotic twins, full, siblings, half siblings, and unrelated siblings.  Their research indicates a significant proportion of variability in depressive symptoms and antisocial behavior is attributed to genetic influence.  The authors also discuss the role of environmental influences as predictors.  Results are discussed in terms of the genetic influence of child and adolescent phychopathology.
     Cichetti and Toth (1995) provide an extensive review that covers a broad span of issues, including the role of genetics on depression.  Timed biological events, which are genetically released, create challenges at each and every developmental phase.  Therefore, genetic factors "contribute sources of vulnerability" as well as "resilience to the probabilistic unfolding of unipolar and bipolar illnesses".  There has been a tremendous body of literature that has demonstrated that mood disorders occur more commonly among the relatives of depressed persons than in the general population.  The empirical literature reveals that the greater the percentage of genes shared with the proband (i.e., the affected individual), the greater the chance that the relative will be similarly affected by a mood disorder.  Additionally, there is an increased rate of affective disorder in the biological relatives and not in the adopted relatives of the adoptees that have been found in on-going adoption studies.  Following this, twin studies of affective illness have consistently revealed the concordance rates for monozygotic (MZ) twins to be substantially higher than in dizygotic (DZ) twins, leading one to conclude that a genetic "trigger" may exist for clinical depression.
     Genetic markers can be viewed as risk traits. Although genetic research has provided strong evidence for the existence of specific genes that may increase the probability of developing a mental disorder, such as major depression, these genetic markers are not guarantees that depression will develop.  What must be included in this equation is the protective, or resilience, factors that each person may possess.  Therefore, a multifactorial model of risk is supported; such that environmental factors play in either protecting or triggering a depression. That is, environmental experiences may be more critical than genes in increasing an individual's liability to depression. Collaboration between the fields of molecular genetics and developmental psychopathologist would allow much needed research in this arena.  Future work for developmental geneticist would be necessary to uncover the ontogenetic processes whereby "genetic and environmental factors conspire to orchestrate normal and abnormal patterns of adaptation".  It would be especially helpful to identify the factors that turn particular genes on and off during phases of development.
     Similarly, Eley & Stevenson present a primarily genetic explanation for risk of depression.   Their research utilizes self-reported anxiety and depression symptoms in children and adolescents have been shown to be heritable, and are also highly correlated.  This study set out to ascertain to what extent the genetic and environmental factors that influence anxiety symptoms also influence depression symptoms, and whether these are the same in children and adolescents, and males and females. Twins age 8-16 years old completed the measures of depression. There were significant effects of age and sex on the variance and on the variance and covariance between these two types of symptoms.  The authors conclude that genetic influences on anxiety and depression were shared for all 4 groups, a finding that has been consistently demonstrated for adults.
     Thapar and McGuffin (1998) provide evidence for genetic influence on depressive symptoms.  Their study utilized monozygotic and dyzygotic twin subjects to test the hypothesis that genetic factors influence the same life events that influence depression.  Also, these genetic factors are thought to mediate the association between the life events and the depression.  Mothers rated depression and life events for 270 twin pairs 8- to 17-years-old.  It was found that depression and some life events, such as number of events and their negative impact, share a common genetic influence.  A portion of the association between depression and life events was best explained by both genes and the environment.  Bivariate genetic model fitting indicated that the covariation between life events and depression was best explained by a common environmental factor.  Due to the limitations of the study, directionality of the results is impossible to determine.  As such, it is difficult to evaluate if life events predated the depression or vice versa.  These results point out the need for longitudinal studies as well as the necessity of larger sample sizes.

Environmental Influences

     With regard to the family, an important environmental influence on children and adolescents, Chasin, Pitts, DeLucia, and Todd (1999) suggest the importance of family influences on depression.  Their study tested the specificity of parent
alcoholism effects on young adult alcohol and drug abuse/dependence, anxiety, and depression, and tested whether adolescent symptoms and substance use mediated the effects of living with an alcoholic parent. Using structured interviews, the researchers suggest unique effects of parent alcoholism on young adult substance abuse/dependence diagnoses over and above the effects of other parental psychopathology. There was some evidence of parent alcoholism effects on young adult depression and of maternal alcoholism effects on young adult anxiety, although these were not found consistently across subsamples. Mediational models suggested that parent alcoholism effects could be partially (but not totally) explained by adolescent externalizing symptoms.
     Whether the context is a child with depression, or a child with depressed parents, the family environment plays a role in both the onset and maintenance of childhood depression.  Kaslow, Deering, and Rucasin (1994) present an extensive literature review that hightlights numerous family variables and patterns associated with, but not causing, childhood depression, as well as the ways in which these factors may interact to sustain a child's depression.  Family variables associated with childhood depression are parental psychopathology, divorce, low SES, negative life circumstances including loss, abuse, or neglect, and low levels of social support.  Families with depressed children have been found to be less cohesive and supportive, less able to communicate effectively, have less secure parental attachment, and more conflictual relationships, more controlling and critical parents (leading to feelings of helplessness in depressed children), and higher levels of expressed emotion (hostility, criticism, and overinvolvement) than do the families of nondepressed peers.  Siblings play a role in the development of depression, as problematic sibling relationships have been associated with greater depression, and a positive sibling relationship may mediate depression.
     Children of depressed parents are at high risk for childhood depression.  Parental depression may be transferred to children by way of genetic predisposition, maladaptive parent-child interactions, and marital conflicts.  Specifically, maternal depression negatively affects a child's functioning, parenting styles, and interactional patterns.  Taking into account all of the above mentioned factors, family environments may interact with childhood depression in various ways that contribute to the onset and maintenance of childhood depression.  From a systems perspective, families reinforce the child's depression.  A family dynamics perspective suggests a "subtle reciprocity" between childhood depression and family dynamics; for example, a certain family environment may not be a 'good fit' for a child predisposed to depression.  The authors do not contend that family variables and interaction patterns are the only or primary contributors in the etiology and development of childhood depression; but rather that it is one variable that should be considered in conjunction with developmental, social, and/or biological factors.
     Hamilton et. al., (1997) studied 49 children with either depression, schizophrenia, or no clinical diagnosis, to see if differences existed in their social, academic and/or behavioral functioning.  In addition, they investigated the influence of transactional family behavior on child competence in these domains.  Statistical analyses showed that the children with depression and their non-clinical cohort functioned comparably in academic performance.  No impaired cognitive functioning was evidenced by the students who were depressed.  However, a significant difference was noted in the social domain, with the depressed children demonstrating poorer social functioning than their non-clinical peers.  No distinctions could be made between the children with depression and the children with schizophrenia in this domain.  This finding is important, when we consider that the problem only compounds as the child reaches adolescence and adulthood. It is important to note, however, that the children who participated in this study were inpatients, and that the severity of their psychiatric illness may not allow for generalization to children with depression in the community.  In other words, although a child may experience depression, s/he will not necessarily have dysfunctional interpersonal interactions.
     Another important finding in the study showed a relationship between family transactional patterns and child adjustment.  They found that children from parents who use more negative interactional styles demonstrated poorer social competence and more behavioral problems than those whose parents interact with them in a more benign manner.  Conversely, children with parents who engaged them in a more positive manner showed better social and behavioral competence.  This study found a relationship between parenting behavior and child social and behavioral competence.  It does not reveal whether the parents' negative interactional style caused the child's depression, whether the parents' negative interactional style evolved in response to the child's depressive characteristics, or whether the two components of the family system interact to maintain and perpetuate the child's depressive symptoms.  Even without the certain identification of the causal agent, it seems evident that the family system needs to be altered in the treatment of childhood depression.
     Segrin (2000) provides an in-depth conceptualization of the role that social skills deficits play in childhood and adolescent depression.  His perspective, which draws from the work of such authors as Coyne and Lewinsohn, posits that there are [at least] three ways to understand the relationship between depression and social skills: 1) poor social skills as the cause of depression; 2) depression as the cause of poor social skills; and 3) poor social skills as a risk factor to the development of depression.  While much of the available literature has focused on adult populations, there appears to be an emerging body of literature that demonstrates the relationship between poor social skills and depression in children as well.  However, Segrin’s evaluation of the literature failed to determine the direction of the correlation between the development of depression and social skills deficits.  While some evidence supports the notion that social skills deficits are a risk factor in the development of depression, most studies provide inconclusive evidence as to directionality.  More longitudinal studies of childhood and adolescent depression would facilitate our understanding of the temporal ordering of social skills deficits and depression.  Additionally, utilizing an approach that considers the interaction between social skills and other risk factors might enhance our ability to predict the onset of depression.  In the meantime, it appears that the question “Do social skills deficits cause depression or vice-versa” remains – at least for the moment – unanswered.

Parent Influences

     Downey and Coyne (1990) provide an integrative review of children of depressed parents.  Among women of child-bearing age, depression, is a highly prevalent disorder, with approximately 8%-12% of mothers exhibiting clinical depression at any given time. A child's adjustment is likely to reflect the type of depression, the degree of family stress and marital discord experienced, and the parent's level of symptomology and social impairment. This article is an extensive, integrative review on various literatures on the adjustment of children of depressed parents, difficulties in the parent-child interaction in these families and contextual factors that may play a role in child adjustment. In earlier studies, children of depressed parents were used as the controls in high-risk research on children of schizophrenic parents. This earlier work led to the serendipitous finding that children of depressed parents were equally at risk for childhood disturbance (especially mood disorders) as the children of schizophrenic parents. Subsequent studies have found that children of depressed parents are at risk for a full range of adjustment problems and at specific risk for clinical depression.
     Not surprisingly, maternal depression is associated with difficulties in parenting; depressed mothers view the role of parenting less positively than do control mothers.  In addition to experiencing greater feelings of depression and hostility towards the parenting role, a depressed mother's behavior and affective expression is constricted and her speech is flat as compared to control group mothers.  Depressed parents respond less positively, less frequently, and less quickly to their children's efforts to engage their attention. Difficulties are found beginning in infancy in the delayed speech patterns of infants of depressed parents.  In school-aged children, the adjustment problems emerge in peer, teacher and observer reports demonstrating children experiencing difficulties across social and academic settings.  Furthermore, parents who are clinically depressed show heightened levels of child-directed hostility and negatively, and their attempts to control child behavior are marked by coercion rather than by negotiation. The authors conclude by stating, "there is a distinct and consistent mother-bashing quality to much of this body of work."  While depression in parents is clearly associated with adjustment and mood disorders in children, the specificity and origin of this association is not clearly established.
     Parent influence may vary by social class, and the types of social support available for parents, according to Mathiesen, Tambs, & Dalgard (1999).  This study identified risk and protective factors for anxiety and depression among mothers of toddlers.  Mothers with 18-mo-old children completed a questionnaire designed to examine the affect of socioeconomic and demographic factors, somatic health problems, negative life events, chronic strain and social support on symptoms of anxiety and depression. There was a moderate negative effect of negative life events and chronic strain and a moderate protective effect of social support on the symptom level, but no interaction effects were found between the risk and protective factors. Behavior problems among the children clearly seemed to affect the mothers' symptom level. The symptom level varied with background factors. The largest effect of background factors seemed to be indirect, mediated through their effect on the risk and protective factors. Although problems with children's behavior and child care arrangements were observed to have a strong impact on the mothers' symptom level, the frequencies of such problems appeared to be less dependent on socioeconomic conditions than did other types of strain.

Environmental Influences

     Monroe, Rohde, Seeley, and Lewinsohn examine life events and depression among adolescents.  Specifically, risk as the result of a recent romantic break-up was examined as a predictor of 1st onset versus recurrence of MDD. Results indicated a heightened likelihood of 1st onset of MDD during adolescence if a recent break-up had been reported; in contrast, a recent break-up did not predict recurrence of depression. These results held for both genders and remained significant after controlling for gender. Implications of these findings and subsequent research directions are discussed.
     Tulisalo and Aro (2000) conducted a study to determine the effects of parental remarriage on depression in young adults.  Considering the vast number of children and adolescents that will go through a divorce, and considering that many of these individuals’ parents will enter a second marriage, exploring the ways in which parental remarriage moderates depression is a pertinent research question.  The authors distributed questionnaires to 2,194 9th grade students (mean age 15.9 years) in a small Finnish city.  Six years after the original questionnaires had been completed, another questionnaire was sent to the same cohort of students.  Now 22-years-old, there were 210 females and 146 males who had been through a divorce, half of which also experienced the remarriage of one or both of their parents.  Contrary to expectations, results of the study indicated that mothers’ remarriage did not result in an increase in depressive symptomatology.  There was, however, a significant relationship between fathers’ remarriage and depressive symptoms of both boys and girls; boys displayed a decrease in signs of depression, while girls harbored an increasing number of depressive symptoms.  The authors discuss the limitations of the present study and highlight the fact that the original design was not intended to study the impact of remarriage.  Additionally, considering that the results tended to conflict preexisting research in the area, the reader is encouraged to exercise caution in interpretation.

Abuse
     Pearce and Pezzot-Pearce (1997) provides a comprehensive description of individual difficulties and psychotherapy strategies that can be used to assess and treat children who have suffered various types of abuse and neglect. Child maltreatment can be regarded as one of several variables that may contribute to childhood depression. Pearce & Pezzot-Pearce outline the various methodological difficulties in this type of research demonstrating how difficult it is to define and characterize children exposed to different types of maltreatment; not to mention, problems with subject selection, research design and control groups.  The chapter provides an overview of attachment theory including a discussion of the organizations of maltreated children following some of the work of the Minnesota Mother-Child Interaction Project, which is a prospective, longitudinal study of the development of a sample of high-risk children (see Egeland, 1991).  This chapter further reviews the emotional and behavioral self-regulation of maltreated children including a thorough discussion of externalizing problems (i.e., physical aggression & sexualized behaviors) as well as internalizing problems (i.e., depression & posttraumatic stress disorder).
     Maltreated children are at a significant risk for the development of a number of problems, including insecure attachment, poor emotional and behavioral self-regulatory skills, lowered cognitive functioning, poorer adaptation to school, and language delays. Depression, viewed as a manifestation of disturbances in self-regulation, is associated with various types of maltreatment.  The authors review a number of studies in which depressive symptomology is correlated with sexually abused children.  All of the reviewed studies found higher rates of depression in sexually abused children and also higher rates of suicide.  Furthermore, not only is depression the most common symptom reported in children and adolescents who have been molested, but these elevated depressive symptoms have been found to carry well into adulthood (see Briere, 1992).  Therefore, depression is an important risk factor to consider for children who have been maltreated in any way.
     Kaufman, Birmaher, Dahl, Bridge, & Ryan (1998) evaluated the first- and second-degree relatives of 26 depressed adolescents (13 of which were abused) compared to the relatives of 27 normal control children.  The first-degree relatives totaled 104 (25 depressed-abused, 29 depressed-nonabused, and 50 normal control) and the second-degree relatives totaled 503 (127 depressed-abused, 117 depressed-nonabused, and 259 normal control).  It was found that the first-degree relatives of depressed-abused adolescents had nine times the odds for depression over the first-degree relatives of the normal control adolescents.   In addition, the odds for other disorders, such as alcoholism and anti-social personality, were found to be three to nine times that for the control group.  A similar pattern was found for the second-degree relatives.  No differences were found between the abused/nonabused groups.  These findings are consistent with previous research and suggest that depression, and related disorders, have a biological component.  Further research is needed to replicate and extend these findings.
     Brown, Cohen, Johnson, and Smailes (1999) provide additional evidence for the role of abuse in youth depression.  Investigated the magnitude and independence of the effects of childhood neglect, physical abuse, and sexual abuse on adolescent and adult depression and suicidal behavior. 776 randomly selected children was studied from a mean age of 5 years to adulthood in 1975, 1983, 1986, and 1992 during a 17-yr period. Assessments included a range of child, family, and environmental risks and psychiatric disorders. A history of abuse was determined by official abuse records and by retrospective self-report in early adulthood on 639 youths. Attrition rate since 1983 has been less than 5%. Adolescents and young adults with a history of childhood maltreatment were 3 times more likely to become depressed or suicidal compared with individuals without such a history.  Adverse contextual factors, including family environment, parent and child characteristics, accounted for much of the increased risk for depressive disorders and suicide attempts in adolescence but not in adulthood. The effects of childhood sexual abuse were largest and most independent of associated factors. Risk of repeated suicide attempts was 8 times greater for youths with a sexual abuse history.

Social Identity

    Safren and Heimberg (1999) examined factors related to depression, hopelessness, and suicide risk in gay, lesbian, and bisexual adolescents, compared with demographically similar heterosexual adolescents. Sexual minority adolescents reported greater depression, hopelessness, and past and present suicide risk than did heterosexual adolescents. However, when controlling for other psychological predictors of present distress, significant differences between the 2 samples disappeared. For past suicide risk scores, the effects of sexual orientation were reduced, but still significant, when accounting for the other predictor variables. These results suggest that environmental factors associated with sexual orientation, which can be targeted and changed through prevention and intervention efforts, play a major role in predicting distress in this population.

Transactional Model Explanation

     A transactional perspective suggests that depression as a pathology does not occur in isolation, but rather as the result of various factors operating together.  Cicchetti and Toth (1995) provide an exhaustive review of affective disorders and a comprehensive understanding of clinical depression from a transactional approach.  A transactional model states that an illness does not occur in isolation, by one process alone, but rather is the result of various factors operating together.  A transactional model specifies that the interrelationships between the organization of developmental domains (biological, socioemotional, cognitive, representational) and the environment in which the individual resides (family, school, community) exert a bi-directional influence on the individual.  Understanding an individual child's risk and protective factors may be the appropriate lens through which to understand a transactional model of psychopathology. Rutter (1990) has cautioned that risk, vulnerability, and protective factors do not cause (emphasis added) pathological outcomes per se, but rather indicate more complex processes and mechanisms that will affect an individual's adaptation.  The ways in which risk and protective factors exert themselves depend on the developmental period within which the individual is developing.
     Viewing childhood depression from a transactional model allows us to see how the child and the environment are mutually influenced as the child evolves.  The bi-directional transactions will continue to influence the child, and the environment in which the child resides, allowing for continued variations in the organization of the child as development unfolds.  For a child who is experiencing clinical depression, an understanding of the child's current world in terms of his/her protective and risk factors is an important beginning.  For example, a child who is the product of depressed parents may possess a unique set of risk and protective factors, while another child who has undergone chemotherapy may face a completely different set of factors; each child may warrant a diagnosis of clinical depression but the etiologies leading to each child’s depression differ drastically.  A transactional model would allow each child's unique set of circumstances to explain the his/her current functioning.
     Consistent with the transactional model described above, the developmental perspective of psychopathology maintains that no single cause determines pathology.  Rather, many risks and protective factors interact with one another to influence a person’s development over time.  Due to these complex interactions, there is no singular pathway common to all individuals with negative outcomes.  Maladaptation is an outcome that can unfold differently for everyone; in the same way that competence does, through successive interactions with the environment (Sroufe, 1997).  A disorder also changes throughout development, manifesting itself differently at various stages of development.  A child’s age and (more importantly) developmental level, then, becomes a very important question to a developmental psychopathologist; because whether a behavior is regarded as normal or pathological depends on where it occurs in the developmental sequence (Cicchetti, Nurcombe, & Garber, 1992).
     The developmental perspective provides a comprehensive and integrative way of conceptualizing the etiology and course of childhood depression.  Unlike other models, this view accounts for the various genetic, biological and environmental factors that have been identified as differentially and interactively putting a child at risk for depression  (Sroufe, 1997). Depression is viewed as a state that develops over time, not as an intractable trait [something] that a child is simply inherits at birth.  For example, an adolescent with insecure attachment (one known risk factor for depression) may react maladaptively to a life stress situation, thereby learning poor problem skills that will negatively affect subsequent interactions with the environment. This pattern may eventually lead this individual to the experience of depression (Kovak, Sudler, & Gamble, 1992). When viewed as a "snowball effect" (i.e. transactional) it is easy to see how detrimental early onset depression can be to a child's development.  Since early childhood is such a critical period in development, the onset of depression is especially detrimental.  Psychopathology at such a young age can result in the formation of negative cognitive styles and in decreases in cognitive flexibility.  Such a combination can be catastrophic to a child’s subsequent development.
    In addition to acknowledging the various factors that influence depression, the developmental perspective also acknowledges the ongoing developmental changes characteristic of childhood. Clearly a child's depressive symptoms will vary with his/her developmental level of cognitive ability, psychosocial functioning, emotional awareness and psychosexual maturity (e.g. puberty) (Cicchetti, Nurcombe & Garber, 1992).  For example, depression during childhood and adolescence may be manifested as "irritable" mood rather than the depressed mood of an adult (DSM-IV). Such developmental considerations are especially essential for diagnosing youth depression.  From a developmental perspective, pathology is defined as the extent to which children's' behaviors reflect deviations from normal developmental processes at different ages. Crying hourly, for example, at age one has different implications than this same behavior at age thirteen.

Conclusion

     As with many disorders, the etiology of depression is not well understood.  Although some adhere to a specific orientation, it seems that the etiology of depression is multidimensional.  A plethora of factors have been identified as putting a child at risk for depression, however, none of these factors alone seem to lead to depression for all individuals.  Studies of biological factors, such as neurotransmitter and hormone levels, as causes of depression have led to inconclusive results; genetic explanations cannot account for depression without the effect of the environment; environmental factors including family variables and interaction patterns, parent variables including maternal depression, and abuse all play a part in the development of depression, but no single one of these factors fully explains the development of childhood depression.
 The developmental perspective, consistent with the transactional model and multifactorial orientation of depression is integrative, comprehensive, and likely to better explain the etiology of depression than could any single model in isolation.  The developmental perspective also provides a more complex understanding of the etiology and considers numerous risk factors that interact over time to create multiple pathways to depression.  It is also the most comprehensive  model in that it recognizes developmental changes that will undoubtedly affect the experience of depression for children.  It is important to consider that depression may manifest itself differently at different developmental stages and the extent to which behaviors deviate from normal developmental processes at any given age.  In viewing depression as an outcome that develops, the developmental perspective takes an important first step in eliminating the notion that pathology exists solely within the child; such a perspective emphasizes the synergistic impact that biology, genetics, and the environment can have on the etiology and manifestation of childhood depression.

 


 Return to Top
 

Assessment

 The proper and objective assessment of childhood and adolescent depression is the first step in moving toward a comprehensive treatment process (McConaughy, Milling & Martin, 1992).  Only trained professionals or clinicians should perform assessments as these individuals have an understanding of normative and non-normative behaviors (Cicchetti & Toth, 1998).  For instance, it is common for children and adolescent to experience short episodes of sadness and disappointment.  However, these episodes generally do not meet the criteria contained in the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV).  From a developmental perspective, multiple assessments that account for developmental concerns are crucial in the diagnosis of depression (Milling & Martin, 1992).  This not only allows for a more complete picture of the child’s behavior within multiple domains, but also includes information from the child, the child’s parent(s), clinician, and possibly even teachers and peers (McConaughy, 1992).  Even if a youth does not fully meet the criteria for depression, professional attention may still be warranted in order to prevent clinical depression or possible suicide (Cicchetti & Toth, 1998).
 Assessments for depression primarily include interviews and questionnaires.  More specifically, assessments include child self-report measures, parent report measures, teacher scales, peer nominations, clinical interviews, and semi-structured interviews (Kaslow & Rehm, 1991).  Essau, Hamik-Larson, Crocker, and Petermann (1999) state, “When choosing an assessment approach for measuring depressive disorder, important considerations are the reliability and validity of the instruments.” The measurements for depression listed below are empirically-based and are capable of being verified or disproven through observation or experiment (McConaughy, 1992).  All scales contain explicit instructions, multiple items, and result in a score that is compared against clinically established cutoffs.  Each measurement has been tested for validity and reliability (Kaslow & Rehm, 1991).
Table of Assessment Instruments and References:

ASSESSMENT INSTRUMENT REFERENCE
Children’s Depression Inventory  (CDI; Kovacs & Beck, 1977)
Children’s Depression Scale  (CDS; Tisher & Lang, 1983)
Depression Adjective Checklist  (C-DACL; Sokologg & Lubin, 1983)
Center for Epidemiological Depression Studies’ Depression Scale Modified for Children (CES-DC; Weissman, Orvaschel, & Padian, 1980)
Reynolds Adolescent Depression Scale  (RADS; Reynolds, 1986)
Children’s Depression Scale Revised  (CDS-R; Reynolds, Anderson, & Bartell, 1985)
Modified Zung  (M-Zung; Lefkowitz & Tesiny, 1980).
Kiddie-SADS Epidemiologic Version  (K-SADS-E; Orvaschel & Puig-Antich, 1987)
Kiddie- SADS Present Episode  (K-SADS-P; revised by Puig-Antich & Ryan, 1986)
NIMH Diagnostic Interview Schedule for Children  (DISC; National Institute of Mental Health, 1992)
Bellevue Index of Depression  (BID; Petti, 1978)
Interview Schedule for Children  (ISC; Kovacs, 1981)
Children’s Affective Rating Scale  (CARS; Kovacs, 1981)
Children’s Depression Rating Scale - Revised (CDRS-R; Poznanski et al., 1984)
Hamilton Depression Rating Scale  (HDRS; Hamilton, 1960, 1967)
School Aged Depression Listed Interview  (SADLI; Petti & Law, 1982)
 Dysthymic Check List  (DCL; Fine, Moretti, Haley, & Marriage, 1984)
Diagnostic Interview Schedule for Children  (DISC; Costello, Edelbrock, Dulcan, Kales, & Klavic, 1984)
Diagnostic Interview for Children and Adolescents  (DICA; Herjanic, Herjanic, Brown, & Wheatt, 1975)
Child Assessment Schedule  (CAS; Hodges, McKnew, Cytryn, Stern, & Kline, 1982)
Child Behavior Checklist  (CBCL; CBCL/4-18; Achenbach & Edelbrock, 1983)
Personality Inventory for Children Depression Scale    (PIC; Wirt, Lachar, Klinedinst, & Seat, 1977)
Peer Nomination Inventory for Depression  (Lefkowitz & Tesiny, 1980)

Major Methods of Assessment
 Self-report measures are useful for evaluating the severity of depressive symptoms, but do not provide a complete assessment of depression.  Self-report measures of depression for children and adolescents include:  Children’s Depression Inventory (CDI; Kovacs & Beck, 1977), Children’s Depression Scale (CDS; Tisher & Lang, 1983), Depression Self-Rating Scale (DSRS; Birleson, 1981), Depression Adjective Checklist (C-DACL; Sokologg & Lubin, 1983), Center for Epidemiological Depression Studies’ Depression Scale Modified for Children, (CES-DC; Weissman, Orvaschel, & Padian, 1980), Reynolds Adolescent Depression Scale (RADS; Reynolds, 1986), Children’s Depression Scale Revised (CDS-R; Reynolds, Anderson, & Bartell, 1985), and the Modified Zung (M-Zung; Lefkowitz & Tesiny, 1980).  Benefits of using self-report measures include their ability [Self-report measures allow children] to describe [provide] subjective internal experiences that only the child him/herself has access to, and the relative ease with which self-report measures can be administered and scored.  However, the limitations of self-report measures include assumed cognitive, language, and reading skills that may exceed a given child’s abilities.  Thus, care should be taken to consider the child’s level of development and age when administering self-report measures of childhood depression.
 Structured interviews are commonly used in the diagnosis of depressive syndromes.  Such interviews include:  Kiddie-SADS Epidemiologic Version (K-SADS-E; Orvaschel & Puig-Antich, 1987), Kiddie- SADS Present Episode (K-SADS-P; revised by Puig-Antich & Ryan, 1986), NIMH Diagnostic Interview Schedule for Children (DISC; National Institute of Mental Health, 1992), Bellevue Index of Depression (BID; Petti, 1978), and the Interview Schedule for Children (ISC; Kovacs, 1981).  Structured interviews provide a consensus of multiple perspectives and can allow the interviewer to observe behavior in addition to asking questions, providing a rich description of the depression.  However, interviews may be time consuming, and therefore cumbersome in terms of resources.
 Similarly, clinical interviews are also used to evaluate depressive symptoms.  Clinical interviews include the Children’s Affective Rating Scale (CARS; Kovacs, 1981), Children’s Depression Rating Scale Revised (CDRS-R; Poznanski et al., 1984), Hamilton Depression Rating Scale (HDRS; Hamilton, 1960, 1967), School Aged Depression Listed Interview (SADLI; Petti & Law, 1982), Dysthymic Check List (DCL; Fine, Moretti, Haley, & Marriage, 1984), Diagnostic Interview Schedule for Children (DISC; Costello, Edelbrock, Dulcan, Kales, & Klavic, 1984), Diagnostic Interview for Children and Adolescents (DICA; Herjanic, Herjanic, Brown, & Wheatt, 1975), and the Child Assessment Schedule (CAS; Hodges, McKnew, Cytryn, Stern, & Kline, 1982).
 A parent report or another informant important to the child or adolescent’s world may provide information to assess depression.  The most commonly used methods employing informants include the Child Behavior Checklist (CBCL; CBCL/4-18; Achenbach & Edelbrock, 1983), and the Personality Inventory for Children Depression Scale (PIC; Wirt, Lachar, Klinedinst, & Seat, 1977).  When peers are the informants, the assessment instrument would likely include the Peer Nomination Inventory for Depression (Lefkowitz & Tesiny, 1980).  Although informants can provide valid descriptions of a child’s behavior in social and school settings, informants are not able to provide information about the child or adolescent’s internal state.

Current Research
 Current research supports the importance of multiple assessment in depression.  A developmental perspective highlights the relevance of various perspectives accounting for growth within the child or adolescent.  Articles presented provide evidence of self-report, interviews, and parent reports on symptoms of depression.

Questionnaires
 Reynolds, & Mazza, J.  (1998) examine the validity and reliability of a relatively new self-report depression assessment, the Reynolds Adolescent Depression Scale (RADS; Reynolds, 1986).  This scale is intended for use with adolescents from 12 to 18 years of age and consists of 30 self-report items geared to identify depressive symptoms according to the Diagnostic and Statistical Manual for Mental Disorders, Third Edition (DSM-III; American Psychiatric Association, 1980).  It was noted within the article that few changes have been made for depression symptomology in subsequent editions of the DSM.
 The participants (N=89) were in 6th, 7th, or 8th grade at a parochial school in Brooklyn, NY.  The age and gender breakdown were as follows: 36 males and 53 females with a total mean age of 12.53 years.  The ethnicity of the sample was 71% African-American, 20% Hispanic American, 1% Caucasian, and 8% other.  The RADS was administered to the participants on the same day.  They were interviewed separately within one to five weeks later with the Hamilton Depression Rating Scale (HDRS; Hamilton, 1960, 1967), which is a clinical interview.  A second administration of the RADS was given just prior to the interview to evaluate test-retest reliability.  The participants were then assigned to one of two groups (“clinically depressed” and “not depressed”) based on the assessment results.  Statistical analyses were then performed to evaluate validity and reliability.  Researchers found that the internal consistency reliability was high (.91) as was the test-retest reliability (.87).  The criterion-related validity was measured by the correlation coefficient between the HDRS and the second administration of the RADS and was considered strong.  The authors conclude that overall the validity and reliability of the RADS is high.  This self-report measure is relatively easy to use and can be completed in about 10 minutes by most adolescents.
 From a more broad measurement perspective, Lonigan, Hooe, David, and Kistner (1999) provide a model of positive affect (PA) and negative affect (NA) model of anxiety and depression as an investigation of the measurement of PA and NA and the relation between PA and NA and levels of adjustment in children and youth. A confirmatory factor analysis was used in this study to examine the structure of self-reported affect and its relation to depressive and anxious symptoms in school children (4th to 11th grade). Results supported a 2-factor orthogonal model that was invariant across age and sex. Support for the expected pattern of relations between NA and PA with symptoms of depression and anxiety was strong for the older sample (M = 14.2 years) but weaker for the younger sample (M = 10.3 years). Results also provide preliminary support for the reliability and validity of the Positive and Negative Affect Schedule for children.
 Chorpita, Yim, Moffitt, Umemoto, and Francis (2000) revised the Spence Children’s Anxiety Scale (SCAS) so that the measure included items designed to assess child and adolescent depression.  The authors believed, due to the high rates of comorbidity between anxiety and depression, that incorporating items to assess depression made good intuitive sense.  Additionally, since items added corresponded to DSM-IV criteria for major depression, the Revised Child Anxiety and Depression Scale (RCADS) also enhanced the clinical utility of the original measure.  The revised scale was administered to a sample of 1641 children and adolescents ranging in age from 3rd to 12th grade.  Results of the study indicate that the RCADS has high structural validity, reliability, retest reliability, internal consistency, and convergent and discriminant validity.  The RCADS also showed good convergence with the Children’s Depression Inventory (CDI), a widely accepted assessment instrument for child and adolescent depression.  The authors believe that the RCADS has several advantages over the SCAS including its close association with DSM-IV criteria for childhood disorders and its ability to assess both anxiety and depression in children and adolescents.  Other advantages of the RCADS, including its suitability for use in clinical and research settings are discussed.
With regard to self-report, Prince, Reischies, Beekman, Fuhrer, Jonker, Kivela, Lawlor, Lobo, Magnusson, Fichter, Van Oyen, Roelands, Skoog, Turrina, and Copeland, describe the development of the Euro-D.  This study aimed to derive from these instruments a common depression symptoms scale, the EURO-D, to allow comparison of risk factor profiles between centers. Common items were identified from the instruments. Algorithms for fitting items to the GMS were derived by observation of item correspondence or expert opinion. The resulting 12-item scale was checked for internal consistency, criterion validity and uniformity of factor-analytic profile. The EURO-D is internally consistent, capturing the essence of its parent instrument. A two-factor solution seemed appropriate: depression, fearfulness and wishing to die loaded on the first factor (affective suffering), and loss of interest, poor concentration and lack of enjoyment on the second (motivation). It is concluded that the EURO-D scale should permit valid comparison of risk-factor associations between centres, even if between-centre
variation remains difficult to attribute.
 Mathiesen, Tambs, Dalgard (1999) identified risk and protective factors for anxiety and depression among mothers of toddlers. A population-based sample of 92 mothers (aged 19-46 yrs) with 18-mo-old children completed a questionnaire designed to examine the impact of socioeconomic and demographic factors, somatic health problems, negative life events, chronic strain and social support on symptoms of anxiety and depression. There was a moderate aversive effect of negative life events and chronic strain and a moderate protective effect of social support on the symptom level, but no interaction effects were found between the risk and protective factors. Behavior problems among the children clearly seemed to affect the mothers' symptom level. The symptom level varied with background factors like the mothers' education, employment status, and age, even after controlling for the effect of strain and social support. The largest effect of background factors seemed to be indirect, mediated through their effect on the risk and protective factors. Although problems with children's behavior and child care arrangements were observed to have a strong impact on the mothers' symptom level, the frequencies of such problems appeared to be less dependent on socioeconomic conditions than did other types of strain.
 Winter, Steer, Jones-Hicks, and Beck (1999) investigated the psychometric characteristics of the Beck Depression Inventory for Primary Care (BDI-PC) among adolescents (aged 12-17 yrs) scheduled for health maintenance examinations, and determined the effectiveness of the BDI-PC in screening the adolescents for Mental Disorders-IV (DSM-IV) major depression disorders (MDD). The BDI-PC was administered to 50 male and 50 female adolescents who received pediatric health maintenance examinations. The diagnosis of MDD was established with the Mood Module from the Primary Care Evaluation of Mental Disorders. Results showed that the internal consistency of the BDI-PC was high (Cronbach alpha = .88), and it was not significantly associated with gender, ethnicity, age, or having a medical disorder. A cutoff score of >=4 had both 91% sensitivity and specificity rates for identifying adolescents with and without MDD. Results show the BDI-PC to be a useful instrument for screening for clinical depression in adolescents receiving routine medical examinations.

Interviews
 From a pharmacology perspective, Emslie and  Mayes (1999) discuss the epidemiology, diagnosis, and treatment of major depressive disorder (MDD) in childhood and adolescence. Although the criteria for MDD are the same for children and adolescents as for adults, some challenges exist in ascertaining the diagnosis. Children often have difficulty in expressing or recalling information regarding their disorder; therefore, multiple informants must often be used to obtain this information. Additionally, comorbid diagnoses are common in early onset depression, making diagnosis more difficult. The treatment of depression in this population is multi-modal, including the patient, parents, and school, and is aimed at shortening the episode of depression. Treatment, which is individualized based on need, may include psychotherapy, family therapy or education, and pharmacological treatment.
Pine, Cohen, Cohen, and Brook (1999) examined the relationship between sub-clinical depressive symptoms in adolescence and major depressive episodes in adulthood. A sample of 776 young people (aged 9-26 yrs) received psychiatric assessments in 1983, 1985, and 1992. Among adolescents not meeting criteria for majordepression, the authors estimated the magnitude of the association between subclinical adolescent depressive symptoms and adult major depression.  Symptoms of major depression in adolescence strongly predicted an adult episode of major depression: having depressive symptoms more than two-standard-deviations above the mean in number predicted a twofold to three-fold greater risk for an adult major depressive episode. Symptoms of depression in adolescence strongly predict an episode of major depression in adulthood, even among adolescents without major depression.
 Parker, Gladstone, Mitchell, Wilhelm, and Roy (2000) conducted a study to examine the relationship between early adverse events and depression in adulthood.  The authors hypothesized that exposure to events that mirrored traumatic events of the individual’s childhood would result in higher levels of depression in adult subjects.  That is, early adverse experiences create a vulnerability to depression when the individual is exposed to similar circumstances in adulthood.  The authors also posit that this vulnerability may be cognitively mediated; negative cognitive schemas that are formed in childhood may re-surface in the face of similar events in adulthood.
A sample of 96 clinically depressed adults were given a detailed clinical interview as well as questionnaires to assess a link between early adverse experiences and depression in adulthood.  The clinical interview focused on life events immediately preceding the current episode of depression and subjects were asked to identify the most relevant/pertinent “key” that led to their current depression.  The questionnaire, which was incorporated into the clinical interview, asked subjects to reveal the extent to which their parent or parents were critical, physically violent, verbally abusive, overprotective, avoidant and a host of other negative attributes.  Finally, the subjects were asked to identify how the most detrimental attribute of their parents “affected” them (i.e. made them feel rejected, criticized, abused, etc).  Results indicate that early adverse experiences do tend to influence the development of negative cognitive schemas that can be reactivated in adulthood.  However, the results did not support the hypothesis that exposure to events that “mirror” the traumatic experiences of their childhood reactivates negative schemas that consequently lead to an episode of depression.  The authors discuss the methodological shortcomings of the present study in an attempt to explain the lack of significant findings and make suggestions for future research.

Comparison between questionnaires and interviews
 Boyle, Offord, Racine, Szatmari, Sanford, and Fleming (1997) evaluated a structured interview and a checklist against criteria from the Diagnostic and Statistical Manual for Mental Disorders, Third Edition - Revised (DSM-III-R; American Psychiatric Association, 1987) in order to determine the reliability and validity of the measures.  The initial sample of children (N=1751) and their parents and teachers were given the Ontario Child Health Study scales (OCHS; Boyle et al., 1987).  A stratified random sample of the participants were then selected for an intensive follow-up.  This group of children (N=251) filled out the OCHS scales a second time for test-retest reliability.  The revised version of the NIMH Diagnostic Interview Schedule for Children and Adolescents (DICA-R; Shaffer et al., 1996) was administered to parent-child dyads.  The order of the interview and scale completion was randomized.
 Researchers found that reliability was relatively high (.70) for major depressive disorder  (MDD) from the structured interview.  Reliability was lower for the OCHS (.27).  Convergent and discriminant validity were comparable for both assessments.  As the results would indicate, a multiple assessment approach is supported as the interview was more reliable than the self-report scale.  Again, surveys or checklists may be a good starting point, however, a clinical or structured interview remains important to adequately assess depression.
 Kasius, Ferdinand, van den Berg, & Verhulst (1997) explain and compare parent report and structured interviews as two diagnostic approaches within the fields of psychology and psychiatry.  The clinical-diagnostic approach primarily utilizes critieria from the Diagnostic and Statistical Manual of Mental Disorder (DSM; American Psychiatric Association, 1994) and can be described as “top-down,” where decisions are made about what should be included based on consensus among experts.  This framework primarily utilizes the clinical interview for diagnostic purposes.  On the other hand, the empirical-quantitative approach uses psychometric measures and procedures to diagnose mental disorders.  This approach can be thought of as “ground up” due to the use of rating scales that are based on norms and distributions of scores within populations, with attention given to reliability and validity issues.
 The purpose of this study was to evaluate the relationship between the two approaches.  Participants (N=231) were obtained from a sample of families who were referred to one outpatient facility in the Netherlands.  The gender and mean age of this sample was split as follows: 148 boys (mean age = 10.4 years) and 83 girls (mean age = 10.6 years).  The parents, usually the mother (78%), filled out the Child Behavior Checklist (CBCL; Achenbach, 1991).  This measure achieved good reliability and validity in the Dutch translation (Verhulst, Berden, & Sanders-Wondstra, 1985), similar to that found by Achenbach (1991).  The parent then participated in a structured interview (NIMH Diagnostic Interview Schedule for Children and Parent; DISC-P; NIMH, 1992).
 Researchers found that most DSM-III-R diagnoses were predicted by more than one CBCL scale independent from other scales.  In particular, the Anxious/Depressed Scale (CBCL) was most strongly related with the Generalized Anxiety Disorder and Overanxious Disorder (DSM-III-R).  The authors conclude that the clinical-diagnostic approach and the empirical-quantitative approach converge, but not to the extent that one can replace the other.  It was suggested that both should be used in combination to better capture the complexity of behaviors presented by children and adolescents.
 Fristad, Emery, and Beck, (1997) describe the ways self-report alone may lead to lack of generalizability and predictive validity.  The Children’s Depression Inventory (CDI; Kovacs & Beck, 1977) is frequently used as a self-report diagnostic tool for depression in children.  It is inexpensive and easy to administer, score, and interpret.  However, this rating scale was intended to be used in conjunction wit other assessments, such as structured or clinical interviews (Fristad, Emery, & Beck, 1997).  With this in mind, the authors evaluated published studies from May 1993 to May 1995 in order to determine how the CDI was utilized.
 A total of 133 articles were found which involved the diagnosing of childhood depression.  It was found that the CDI was used in 75% of the studies that used a self-report assessment.  Of these studies, 34% (46) used the CDI without a clinical or structured interview to determine the diagnosis.  In 44% of the studies in which the CDI was used in isolation it was also found that the participants who scored highly were referred to as “depressed,” with no clear cautionary statement about the limitations of the CDI scale.  Generalizability of the findings was also not clearly addressed.  The authors conclude that the CDI should be used within an assessment strategy that includes multiple measures, including a structured or clinical interview.  A “multiple gating” procedure was also noted, where the self-report measures could potentially be used to screen large pools of potential participants (Kendall, Cantwell, & Kazdin, 1989).  The measure would then be given again to the high-scoring participants in addition to utilizing a structured or clinical interview.  This “multiple gating” sequence would most likely prevent misuse of self-report measures, including the CDI.
 Stark (1990) also recommends a “Multiple Gate Assessment Model” in assessing child and adolescent depression.  The first step in such a model is to use the Children’s Depression Inventory (CDI) as a screening instrument.  Benefits of using the CDI as the initial screening device include: the ease with which the CDI can be administered to large groups of children simultaneously; the short amount of time involved in completing this paper-and-pencil, self-report measure; and the ease with which the CDI can be scored and interpreted.  Children who score higher than 19 (the recommended cutoff score) move on to the next stage, or “gate,” of multiple gate assessment, while those who score below 19 receive no further assessment.  Additionally, any child who endorses a suicide question  (i.e. “I want to kill myself”) should be interviewed immediately following the administration and should also remain in the group of children participating in the second assessment phase.  The next step in the multiple gate model is to re-administer the CDI to children who scored above 19 on the first administration and to children who endorsed a suicide question.  Again, children who score above 19 move on to the next phase of the assessment while those scoring under 19 are excluded.  The third phase involves interviewing each remaining child with an instrument such as the K-SADS.  At this point in the model, it is up to the clinician to determine the severity of the child’s depression and to make treatment recommendations.  In one of Stark’s research studies, just 43% of children who scored above 19 on both administrations of the CDI exhibited the full criteria necessary for a diagnosis of depression.  However, in spite of the fact that the CDI does result in a large number of false positives, utilizing 2 administrations of the measure has been shown to reduce the number of children administered the K-SADS by more than 37%; thus, the multiple gate assessment model can greatly reduce the time spent on more lengthy assessment measures like the K-SADS without sacrificing accuracy in diagnosis.
 Self-report measures suggest differences among cultures and youth assesssment, as Manson, Ackerson, Dick, Baron (1999) report using psychometric characteristics of the Center for Epidemiologic Studies Depression Scale (CESD) as investigated with American Indian boarding school students (N = 188; Grades 9-22). The CESD showed good internal consistency (alpha = .82). Its dimensional structure was different from that described for adults, both Indian and non-Indian. Of the Indian students, 58% were classified as depressed, using the standard cutoff score of 16 or more, consistent with past studies of this age group. Alternative scoring methods were considered, based on persistence of symptoms as well as derived Diagnostic and Statistical Manual of Mental Disorders (DSM-III) and Research Diagnostic Criteria algorithms. Major gender differences were observed in the prevalence of reported symptoms as well as patterns of endorsement for certain CESD items. The CESD should be used with caution with American Indian adolescents, given the observed variation in its dimensional structure and uncertainty regarding appropriate cutoff scores.
 Often, child and informant reports are provided concurrently, as Mulhern, Fairclough, Smith, and Douglas (1999) discuss.  They investigated the incidence of depressive symptoms and their covariates in 99 child cancer patients (aged 8-26 yrs) and their mothers.  Classification of depressed children was highly dependent on the informant and instrument used. Separate multiple regression analyses of the mother's and nurse's ratings of the child's level of depression, the child's self-report on the Child Depression Inventory, and the mother's responses to the ChildBehavior Checklist depression scales revealed different statistical models for each method of assessment. However, increased severity of the mother's self-report of depressive symptoms on the Beck Depression Inventory, which was predicted by low perceived social support and hospitalization of her child, was associated with higher levels of child depression on all child- and parent-report measures.

Developmental Perspective
 Unfortunately, developmental concerns are not always addressed well in assessments, regardless of approach (clinical-diagnostic or empirical-quantitative).  Often distinctions are generally made between the categories of “children” and “adolescent,” however both categories contain large age ranges, often overlap, and are used inconsistently from measurement to measurement.  For instance, the developmental trajectory of a 12-year-old is often very different from that of a 16-year-old, yet both are considered to be adolescents.  This hold true for the “child” category as well, as a six-year-old experiences a much different world than that of a ten-year-old.  Many developmental theorists, such as Piaget, would categorize these two children in different cognitive stages, yet as with the adolescents, the two would be assessed with identical assessments and measured against the same diagnostic criteria.
 The developmental issues set forth here may be the result of a “developmental continuity myth,” where it is thought that the symptoms of psychopathology are similar regardless of age (Shirk, 1988).  This can be actively seen in the many adaptations of adult criteria, assessments, and measures that are then applied to children and adolescents.  In the end, it is critical that old lenses and frameworks are challenged with new research so that developmental changes across the lifetime are accurately and adequately considered.
 Another dangerous assumption that is commonly made by professionals in the field is that children of various racial and ethnic identifications can be adequately, validly, and reliably assessed by standard assessment measures.  There are many reasons that this assumption is quite ill-founded.  Garcia Coll and Garrido (2000) contend that minority children growing up in this country have a qualitatively different experience than their White counterparts, and that such divergent experiences predisposes them to more risks of developing a psychopathology.  For example, the authors state, “It is our position that the systematic exclusion from critical resources and power experienced by many minority populations places these children and their families on less favorable developmental pathways from the very beginning.  By not having access to good prenatal care, by being exposed to environmental toxins in utero and thereafter, and by experiencing on a day-to-day basis the cumulative effects of other poverty-related factors (such as limited nutrition, exposure to violence in local communities, parental un- or underemployment, inadequate housing, etc.), these children are placed at a higher risk for developmental psychopathology.”  The result of these factors tends to be the over-pathologizing of minority youth in this country using instruments that are considered to be standard measures by most.  The additional fact that the vast majority of assessment measures have been constructed using relatively homogenous samples of middle class adults should make professionals extremely cautious when employing adaptations of such adult measures in working with any child; hence, the validity of such measures in assessing children from ethnic minority groups is even more questionable.  Symptoms of depression (or any other psychopathology for that matter) may have very different meaning depending on the context within which they are displayed.   A child who grows despondent and loses interest in things he/she used to take pleasure in (i.e. playing with friends at a nearby park) may be displaying an adaptive response to an abusive parent who beats the child when he/she returns from play, not suffering from Major Depressive Disorder in the conventional sense.  What we as professionals have labeled as the syndrome of depression may be just as adaptive given one set of circumstances as it is harmful in another context.  This logic highlights the importance of multiple assessment procedures in assessment and diagnosis of child and adolescent depression.  We have more to gain from understanding the context within which a given child’s depression developed than we do from the diagnosis of depression alone.  Until future research allows for the development of assessment devices that are sensitive to racial and ethnic diversity, there is no substitute for analysis of pertinent contextual factors of childhood depression.
Children may not have the vocabulary to talk about negative feelings and, as such, may express their feelings through behavior. Younger individuals with depression are may show phobias, separation anxiety disorder, somatic complaints and behavior problems. With psychotic depression, children are more likely to report hallucinations. Older adolescents and adults with psychotic depression may have delusions.  An additional developmental consideration is that children may display irritable symptoms rather than typical “sadness” associated with adult depression.  This is an important consideration in the assessment and diagnosis of childhood depression.

Optimal Assessment Strategy
 As was seen in the literature and current research, multiple assessments are key for diagnosing psychopathology, particularly that of depression.  Because of developmental concerns and the range of symptoms, it may take more time to diagnose depression in a child than it does to diagnose an adult. The diagnostic process from a developmental perspective includes interviews of both parents the child.  Parents are more likely to report outward signs of depression, but the child may be more aware of experiences the parents do not recognize.  Additionally, school and other outside reports are useful, as they are less subject to the parents’ agendas.  A report of the developmental history and information about the existence of other conditions would be helpful in assessment as well.
 Self-report scales, in addition to a clinical or structured interview, would provide the cornerstone of a proper assessment strategy.  One advantage of self report, quantitative assessment is that they can be graded in the same way every time for all patients and situations, so the information is useful for making comparisons, and therefore for assessing the effect of treatment on the patient.  However, as previously stated, there is danger in making broad-based comparisons between groups of children and adolescents in that such an approach assumes homogeneity across ethnic and racial groups.  Thus, quantitative self-report measures, although helpful under some circumstances, is not indicated in all situations.  No scale can yield as much information as a frank personal interview.  The clinical interview [would] takes advantage of at least [the] two assessment approaches, clinical-diagnostic and empirical-quantitative, in order to converge on a more informed diagnosis.  In the clinical environment, an initial screening with self-report scales may be indicated in order to assess large groups of referred clients in a less time-consuming or expensive manner.  A structured or clinical interview can then be used to confirm the initial diagnosis or give one pause to review assessment strategies.  However, professionals are cautioned to err on the side of producing false positives so that children and adolescents have access to the professional assistance that they need in order to cope with clinical depression.  It is more ethical to over-diagnose and, thus, offer treatment to children with sub-clinical levels of depression than it is to under-diagnose and fail to treat youth suffering from the disorder.  Moreover, such a strategy may serve as an effective intervention for children who may be on a developmental pathway leading to depression.
 
 


Return to Top
 

I. TREATMENT

     Because it has only been in the past two decades that the concept of childhood depression has come to be accepted, most research on treatment of depressed children is recent.  The narrow range of treatment options that include developmental considerations is likely a product of the complexity of depressive symptoms within child and adolescent life changes.  It should also be noted that methods of assessing child and adolescent depression are evolving to include ways the expression of depression differ with developmental level.
     Just as treatment for a headache will be different if one goes to an acupuncturist, a pediatrician, or an herbalist, so too will the treatment of childhood depression differ depending on the professional consulted. Treatments logically go hand in hand with a professional's model for viewing psychopathology and conceptualization of the etiology of a disorder.  If one chooses to see a psychiatrist, for example, where the medical model is the perspective and etiology is viewed as biological, pharmacology (drugs) are likely be a component of the treatment.  A cognitive-behaviorist, perceiving that depression is learned, will likely want to change the behavior by replacing faulty cognitions.  In addition, certain cultural and/or religious beliefs may lead one to seek treatment from someone other than a psychologist or psychiatrist, and again, the treatment will reflect that person's framework for viewing depression.
     In Western psychology, the two most common treatments for childhood depression are pharmacology and psychosocial treatments, namely cognitive-behavioral interventions (Stark, Sommer, Bowen, Goetz, Doxey, & Vaughn, 1997).  Approaches to pharmacology and psychosocial treatments are reviewed below.  Since most professionals currently agree that the etiology of childhood depression is multifactorial, optimal treatments must be multimodal (considering and incorporating aspects from various models of depression).

Pharmacology Treatments

     According to Stark, Bronik, Wong, Wells, and Ostrander (2000), there is literally no evidence that antidepressant medications work better than a placebo in the treatment of child and adolescent depression.  However, as the authors point out, the use of antidepressant medications for youth depression has gained widespread acceptance in the field.  In fact, Stark, Laurent, Livingston, Boswell, and Swearer (1999) stated, “Despite the lack of empirical evidence for the efficacy of antidepressants, they are so commonly used with depressed youngsters that many consider them to be the standard of care.”  In spite of the lack of empirically supported research documenting the efficacy of pharmacological treatments for youth depression, the authors outline a current trend called polymedicating that may show promise.  According to Stark and his colleagues, “Proponents of this approach believe that one medication may potentiate the effectiveness of the other medication.  There is some literature that suggests that adolescents who are non-responsive to an antidepressant may respond favorably to augmentation of the antidepressant with lithium.”  The authors discuss alternative augmentation strategies and provide and outline of the considerations a psychiatrists might make in determining whether a given child was appropriate for augmentation procedures.
     A literature review conducted by Hammen, Rudolph, Weisz, Rao, and Burge (1999) revealed a significant lack of research focusing on treatments specific to childhood depression.  Although studies examining child and adolescent depression were plentiful, the authors were hard-pressed to locate research focusing on children alone.  In fact, only 2 controlled studies using clinical samples of depressed children were identified – neither of which presented results specifically for children.  The authors’ review of the literature on pharmacological treatments revealed that tricyclic antidepressant (TCAs) treatment of child and adolescent depression show mostly insignificant results.  In all studies reviewed, the authors reported that placebo treatments were as effective as TCAs treatments of youth depression.  The authors remarked that one study examining the effectiveness of flouxetine in reducing depressive symptomatology demonstrated modest advantages over placebo, with just 31% of patients demonstrating symptom remission.  The authors discuss the role of methodological limitations including small sample size, brevity of treatment, and sample heterogeneity and provide an overview of a developmental model that they recommend as an alternative to current treatment approaches.
     Bostic, Wilens, Spencer, and Biederman, (1999) provide a literature and research review of depression and related treatments.  Specifically, they describe depression in children and adolescents in terms of how these groups have responded to conventional tricyclic antidepressants less robustly than depressed adults. Emerging research suggests that juvenile depression may respond better to serotonergic and atypical pharmacologic agents, so guidelines for selection and administration of these agents are provided.
     Ray-Sanchez, and Gutierrez-Casares (1997) provide an open trial of paroxetine in children with major depressive disorder.  This study included 45 participants under the age of 14.  Each was assessed with the Clinical Global Severity (CGS) scale and a semi-structured interview by an experienced child psychiatrist.  Assessments were done at the beginning of the protocol, at 1 month, at 3 months, and when the protocol terminated.  It should be noted that all children were treated until their depressive episode was resolved, the longest of which was 6 months.
     Each participant was given paroxetine, a selective serotonin reuptake inhibitor (SSRI), in specified and progressive doses.  Any side effects were mild and disappeared with dosage adjustment.  The researchers concluded that the drug was well tolerated within this population.  In addition, it was noted that paroxetine may serve as an alternative to tricyclic antidepressants for children and adolescents.  However, more research was suggested, particularly incorporating a double-blinded design that would include a placebo group comparison.
     Rao, Lutchmansingh, and Poland (1999) assessed REM sleep responses to scopolamine in 5 normal 14-18 year olds and 7 normal adults in order to assess the influence of development on the regulation of REM sleep by cholinergic systems. Subjects were studied on two separate occasions for 3 consecutive nights. They received placebo or scopolamine (1.5 mug/kg) on night 2. As expected scopolamine delayed REM latency and suppressed REM sleep on night 2 in both age groups. Subtle differences were seen, with scopolamine having a tendency to suppress REM sleep less effectively in younger Subjects. On night 3 (recovery), REM latency was shortened and REM sleep was increased to comparable extent in both the adolescents and adults. The comparable REM sleep responses to scopolamine between normal adolescents and adults, particularly on night 3, are discussed in relation to the age-related expression of REM sleep abnormalities in depression.
     James (1999) describes 4 cases where children and adolescents with major depressive disorder fail to respond to 1st-line psychological therapies.  The author suggests potential treatments and a protocol for more resistant depression. The treatment of depression requires a multimodal approach, starting with a comprehensive assessment and psychological treatments such as cognitive-behavioural therapy, family therapy and interpersonal therapy. Physical treatments are necessary for severe depression and can be prescribed in a logical manner starting with serotonin reuptake inhibitor antidepressants and should include the possibility of augmentation of antidepressant action with either lithium or T3, followed by, or alternatively, a combined serotonin-noradrenergic blocking antidepressant, such as venlafaxine or mirtazapine. ECT should be reserved for the very few resistant cases of endogenous, psychotic depression and those with a risk of suicide.
     Birmaher, Waterman, Ryan, Perel,  and others (1998) describe a randomized, controlled trial of amitriptyline versus placebo for adolescents with “treatment-resistant” major depression.  27 participants in this study were referred to an in-patient hospital for long-term care and were aged 12- to 18-years-old.  Each was diagnosed with major depressive disorder by a trained nurse using the Schedule for Affective Disorders and Schizophrenia-Present Episode (K-SADS-P) and the Hamilton Depression Rating Scale (HDRS).  Any potential participant was excluded if there was an accompanying diagnosis of bipolar disorder, psychosis, substance abuse, pregnancy, or medical or neurological problems.  It was determined that all participants were taking some form of medication, therefore within four weeks of admission, all medications were progressively limited until no longer necessary. A multimodal approach to depression was recommended by the researchers, with further study deemed necessary for children and adolescents with major depressive disorder.

Psychosocial Interventions

     Gladstone and Beardslee (2000) outline their preventative intervention approach for child and adolescent depression.  Their intervention strategy focuses on increasing children’s resilience to depression, especially with children who may be at high risk for developing the disorder (i.e., children of depressed parents).  Four main principles guided Gladstone and Beardslee in the design of their intervention.  First, the fact that the intervention follows a developmental perspective necessitates that age-appropriate children be identified.  Due to the cognitive component of the intervention strategies, the authors targeted children between the ages of 8 and 15.  Second, since research shows that the majority of individuals suffering from depression will be treated by their general practitioner, the intervention was designed to be compatible with all types of health care practitioners.  Third, since depression often has a strong impact on family functioning, the family was the level of intervention and the focus was on strengthening the role of the parent in supporting their children.  Finally, the focus of the approach was on prevention, thus, children not meeting DSM-IV criteria for depression were targeted.  The preventive intervention project consisted of two components, clinician-facilitated intervention (i.e., 6-10 sessions in which the clinician meets with both parents and children individually and as a group) and lecture (group format meetings without children present).  A total of 100 families participated in the present longitudinal study.  Results indicate that both intervention approaches helped to improve family communication, reduce levels of parent guilt regarding depression, and increase children’s understanding of their parent’s illness.  According to the authors, “Children whose parents reported a positive response to intervention correspondingly reported better outcomes, in terms of depressive symptoms and current global functioning.  The reported findings indicate that providing parents with factual information regarding risk and resiliency in children can result in changes in illness-related behaviors and attitudes.”  The authors end by discussing related prevention projects and make recommendations for future directions of research in the area of preventive intervention.
     Kendall (2000) provides a brief review of empirically supported treatments for various childhood disorders including child and adolescent depression.  Of the seven treatment outcome studies of children with depressive symptomatology, “…only the self-control behavioral treatment program developed by Stark and colleagues approached the criteria for probably efficacious status, as defined by the 1995 task force.  No treatments met the criteria for well-established.
     Lewinsohn and Clarke (1999) provided an overview of psychosocial treatments of adolescent depression.  Meta-analysis of CBT therapies that have been utilized in the treatment of adolescent depression revealed that 63% of individuals receiving some form of cognitive-behavioral therapy showed significant improvements.  Thus, while CBT methods vary drastically from one clinician to the next, it appears that the overall efficacy of CBT methods is well documented.  The authors provide an overview of the commonalities found across CB therapies as well as examples of the various formats within which CBT can be administered (i.e., group therapy, individual therapy, educational settings, intensity and duration considerations, etc.).  The authors conclude by calling attention to the absence of research examining cognitive-behavioral treatments in working with African American, Native American, Hispanic, and other ethnic groups.
     Bandura, Pastorelli, Barbaranelli, and Caprara, (1999) provide a longitudinal study of self-efficacy and childhood depression.  Their research analyzed how different facets of perceived self-efficacy operate in concert within a network of sociocognitive influences in childhood depression. Perceived social and academic inefficacy contributed to concurrent and subsequent depression both directly and through their impact on academic achievement, prosocialness, and problem behaviors.  In the shorter run, children were depressed over beliefs in their academic inefficacy rather than over their actual academic performances.  In the longer run, the impact of a low sense of academic efficacy on depression was mediated through academic achievement, problem behavior, and prior depression. Perceived social inefficacy had a heavier impact on depression in girls than in boys in the longer term. Depression was also more strongly linked over time for girls than for boys.
     Stark, Laurent, Livingston, Boswell, and Swearer (1999) provide a conceptualization that integrates components of cognitive theory with those of attachment theory in an attempt to better understand and treat childhood and adolescent depression.  The authors posit that depressive schemas (i.e., negative self-schema) proposed by cognitive theory may be understood in children via their attachment relationships.  For example, a child with an overly critical parent who consistently demeans and rejects him/her may fail to develop a positive self-schema that is common of other children his/her age.  In the authors words, “the diathesis of depressive disorder a lack of a positive sense of self which is formed through early interactions within the family and maintained by interactions within the family and with peers as well as through other day-to-day experiences.”  Furthermore, it is hypothesized that the lack of a positive sense of self will, over time, result in the formation of negative sense of self and in negative self-schemas that are associated with depression in adults.  This critical interplay between a child’s attachment with his/her primary caregiver and the subsequent development of negative self-schemas is the crux of the integrative theory of attachment theory and cognitive theory.  Although the article is theoretical in nature, not treatment focused, the empirical evidence the authors do provide within each theory individually yields important treatment recommendations.  Specifically, the authors argue that prevention and intervention of childhood depression needs to incorporate both cognitive (negative self-schemas, negative cognitions) and interpersonal (i.e., parent-child interaction patterns) domains.
     Harrington, Whittaker, Shoebridge, and Campbell, (1998) approach depression treatment with a systematic review of efficacy of cognitive behavioral therapies in childhood and adolescent depressive disorder.  Their article reviewed six studies that utilized cognitive behavior therapy with children and adolescents.  The age of the participants varied within each study, with an overall range of 8- to 18-years-old, with all participants diagnosed with mild or moderate depressive disorder.  The cognitive behavioral therapies used included a depression coping course, structured learning therapy, and specific depression treatment programs.  The therapy consisted of eight to 12 weekly sessions.  A comparison condition was part of each research design, with some control groups consisting of participants on waiting lists or involved in placebo conditions.
      Five out of the six studies assessed outcomes using the Schedule for Affective Disorders and Schizophrenia Child (K-SADS).  The remaining study utilized a global rating of improvement using the Child Depression Inventory (CDI) and other assessments not noted.  It was found that the rate of remission was significantly higher in the cognitive behavior therapy group (62%) versus the comparison groups (36%).  It was concluded that cognitive behavioral therapy was effective in treating mild to moderate depressive disorder.  However, further research is necessary due to the limited number of studies within this report, as well as the need for the study of children and adolescents with severe depression.
     Within a more specific population, McArt, E.W., Shulman, D.A., Gajary, E. (1999) describe the development of an educational workshop on teen depression and suicide.  Community outreach activities showed that adolescents and parents in Monroe County, New York, had difficulty identifying and accessing mental health crisis services for youths. The need to address this deficit led to the development of an educational workshop on teen depression and suicide. Local, national, and international trends in teen suicide led the authors to suggest a proa