|Home | About | Journals | Submit | Contact Us | Français|
Diagnosis and treatment of mood disorders in youth are still problematic because in this age the clinical presentation is atypical, and the diagnostic tools and the therapies are the same as that used for the adults. Mood disorders are categorically divided into unipolar disorders (major depressive disorder and dysthymic disorder) and bipolar disorder in Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition, Text Revision), but mood symptoms are also comprised in the diagnostic criteria of the adjustment disorder (AD), which occur in many different psychiatric disorders, and may also be found in some physical conditions. The differential diagnosis is not much addressed in the midst of clinical investigation and so remains the major problem in the clinical practice.
The associations between some variables and the depressive disorder and AD were analyzed to make considerations about differential diagnosis.
We reported a retrospective study of 60 patients affected by depressive disorder and AD. The analysis has evaluated the association between some variables and the single diagnostic categories. We have considered 10 variables, of which 6 are specific to the disorders, and 4 have been considered related problems.
The statistical analysis showed significant results for the associations of 3 variables (prevalent symptoms, treatment, and family history) with the single diagnostic categories.
The discriminate analysis resulted in statistically significant differences between patients with depressive disorders and those with AD on 3 variables, of which 2 are specific to the disorders, and 1 is included in the related problems. The other variables were weakly associated with the single diagnostic categories without any statistically significant differences. The 3 variables that were associated with the single diagnostic categories support the distinct construct validity of the 2 diagnostic categories, but, to date, it is difficult to establish if these variables can be considered diagnostic predictors. On the other hand, the other variables did not support the distinct construct validity of the 2 diagnostic categories, which suggest an overlapping and dimensional concept. The spectrum approach could unify categorical classification that is essential with a dimensional view. Combination of dimensional and categorical principles for classifying mood disorders may help to reduce the problems of underdiagnosis and undertreatment.
Despite the great attention directed toward the diagnosis and treatment of depression in children and adolescents in the past years, underdiagnosis and undertreatment remain major problems in mood disorders of childhood because the clinical presentation is atypical, and the diagnostic tools and the therapies are the same as that used for the adults.1 The World Health Organization predicts that depression will be the world’s leading cause of disability by 2020, in the whole age range.2 Mood disorders in childhood and adolescence cause significant cognitive, academic, social, and emotional impairment and increase the risk of other negative health outcomes, so precocious diagnosis and treatment are needed.
Mood disorders are categorically divided into unipolar disorders (major depressive disorder [MDD] and dysthymic disorder [DD]) and bipolar disorder (BD) in the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition, Text Revision) (DSM-IV-TR),3 but mood symptoms are also comprised in the diagnostic criteria of the adjustment disorder (AD), which occur in many different psychiatric disorders, and may also be found in some physical conditions.
In childhood and adolescence, MDD and DD are characterized by irritable mood, associated with several depressive symptoms and feelings; however, expression of symptom varies with developmental stage, and some children and adolescents may have difficulty identifying and describing internal mood states.4–21 The prevalence of MDD ranges from 1.8% to 2.5% for children and from 2.9% to 4.7% for adolescents and the prevalence of DD ranges from 0.6% to 4.6% for children and from 1.6% to 8% for adolescents.2,22–24 BD is not a single disorder but a category of mood disorders and is defined as the presence of 1 or more episodes of abnormally elevated mood, which is clinically referred to as mania. Individuals who experience manic episodes also commonly experience depressive episodes or symptoms or mixed episodes that present with features of both mania and depression.25–29 There is evidence for an increased risk of developing BD in young people with depression, so precocious depression seems to be a possible precursor of BD.11,25 The prevalence of juvenile BD ranges from 0.2% to 0.4% for prepubertal populations and is about 1% for adolescents, usually with a depressive episode as the first episode in 40%.11,25–30
AD is characterized by emotional and behavioral symptoms, which occur in close temporal relationship to stressful events, and is time-limited.31–38 The distress is more than what would be expected, given the nature of the stressors, and causes significant impairment in social or academic functioning. On the basis of predominant symptoms, the categories of DSM- IV-TR include AD with depressed mood, anxiety, mixed anxiety and depressed mood, disturbance of conduct, and mixed disturbance of emotions and conduct. The prevalence of AD ranges from 2% to 8% for children and from 12.5% to 34% for adolescents.3,39–41
Mood symptoms occur in many different psychiatric disorders, such as anxiety, disruptive behavioral, eating, personality, schizoaffective disorders, and learning disabilities. Moreover, some physical conditions (in particular, endocrine and neurologic diseases) may also express them with mood symptoms.
So the differential diagnosis is the major problem in clinical practice. Although there is an extensive body of literature regarding depression in adults, to date, there have been a relatively few previous population-based studies that examine the problems of differential diagnosis in mood disorders. The distinction between unipolar disorders and BD is probably the best established of the subdivisions between the mood disorders, which have been supported by several different lines of evidence. However, mania is rare in children. Moreover, some symptoms of depression could be considered predictor of BD.30,42 So, the differential diagnosis between these 2 disorders is still difficult in childhood. Moreover, at no age it is clear where the line should be drawn between clinical depressive disorder and feelings of sadness, which is based on adversity, such as in AD.
We have retrospectively studied a childhood population affected by mood disorder and AD. The aim of this study is to consider differential diagnosis between these diagnostic categories. In order to do this, several groups of variables, some of these specific to the disorders and others being related problems, were analyzed.
Data are collected from a case sheets study of a cohort of inpatient sample of 60 children and adolescents (28 male, 32 female), aged between 5 and 17 years (mean age 11.75; SD = 3.1), who were referred to the Child Neuropsychiatric Unit, University of Bari, from January 2004 to December 2008. According to DSM-IV-TR criteria, all patients were categorized into 2 groups: depressive disorders (MDD and DD) and AD.
During the first neuropsychiatric visit, a clinical interview with patients and their parents provided a developmental history exploring premorbid development, onset of symptoms, symptoms, stressors, social context, and dysfunction. Then, all patients were tested through an individualized assessment, comprising psychodiagnostic and clinical and instrumental procedures.
For all patients, the psychodiagnostic assessment included the following: self-interviews, observer semistructured interviews, rating scales, administered individually to the children and adolescents and to their parents, some of these exploring general psychopathology and other specific for mood disorders (schedule for affective disorders, schizophrenia for school-age children – present and lifetime version, child behavior checklist, children depression rating scale, children depression inventory, screening for child’s anxiety- related disorders); projective tests (spontaneous drawing, familial test, children’s apperception test, and thematic apperception test). In the cases suspected for mental retardation and/or learning disabilities, intelligence (Wechsler Preschool and Primary Scale of Intelligence, Wechsler Intelligence Scale for Children- III) and academic evaluations were performed to assess comorbidity.
The clinical and instrumental assessment comprised physical and neurological examinations, general laboratory investigations, and on the basis of individual needs, electroencephalogram and magnetic resonance imaging.
The analysis involved 10 variables of the disorders, collected during the enrolment evaluation: 6 considered specific to the disorders (characteristics of the sample at first consultation, reasons of first consultations, duration of symptoms, severity, treatment, and prevalent symptoms); 4 considered related problems of the disorders (family history, sender, life stressors, comorbidity). In order to detect possible specific associations between the variables and the single disorder and to make considerations about differential diagnosis, the statistical analysis was performed using 2 × 2 contingency tables. Chi square (χ2) (with the Fisher exact test when indicated), odds ratio (OR), and 95% confidence interval (CI) were estimated. The P < 0.05 value has been considered significant. The data elaboration was performed by the software Epi-Info 6.00 (public domain software: CDC, Atlanta, Georgia, USA; WHO, Geneva, Switzerland).
The mean age at the first consultation was 12.9 years (SD = 3) for MDD, 10.9 years (SD = 2) for DD, and 10.9 years (SD = 3.1) for AD groups.
We have grouped those patients in 5 categories (Table 1):
Clinical interview revealed that 48% of MDD, 75% of DD, and 55.5% of AD groups had symptoms many years before the diagnosis (Table 1).
The severity was assessed on the basis of symptoms and impairment of global functioning, using Children’s Global Assessment Scale. In the all groups, the moderate to severe forms were prevalent [moderate: 36% MDD, 50% DD, and 44% AD; severe: 52% MDD, 37.5% DD, and 26% AD (Table 1)].
Twenty-four percent of MDD, 100% of DD, and 74% of AD patients underwent psychological therapy (χ2 = 7.06, P = 0.0007). The psychological therapy was associated with the diagnosis of AD (OR = 3.39; 95% CI, 1.03–11.5; χ2 = 5.17, P = 0.02) (Table 1).
Seventy-six percent of MDD and 15% of AD patients underwent combination therapy (psychological and psychopharmacological therapy); 11% of AD patients were admitted to the social services.
The combination therapy was associated with the diagnosis of MDD (OR = 22.96; 95% CI, 4.86–122.97; χ2 = 24.25, P < 0.0001).
The reported and observed prevalent symptoms, at the first observation, were different in all groups and they varied with respect to the reasons of the first consultation. These prevalent symptoms were depressed or irritable mood, loss of enjoyment, eating disturbances, sleep disturbances, psychomotor abnormalities, fatigue or loss of energy, low self-esteem, feelings of worthlessness, inappropriate guilt, or hopelessness, reduced ability to concentrate, recurrent thoughts of death or suicidal behaviors, mental over activity (flights of ideas and racing or crowded thoughts), vague physical complaints, psychotic symptoms, poor performances in school, disengagement from peer play, anxiety, and behavioral problems. The following symptoms were more frequent in the MDD groups than in other groups: eating disturbances (χ2 = 4.00, P = 0.04), psychomotor abnormalities (χ2 = 5.19, P = 0.02), fatigue or loss of energy (χ2 = 4.76, P = 0.04), and feelings of worthlessness, inappropriate guilt, or hopelessness (χ2 = 5.68, P = 0.02) (Table 2).
The loss of enjoyment was more frequent (χ2 = 7.33; P = 0.009) in the DD group than in other groups. The feelings of worthlessness, inappropriate guilt, or hopelessness were less frequent in the AD group than in other groups (χ2 = 8.66; P = 0.002).
Statistical analysis showed that MDD was associated with vague physical complaints (OR = 12.67; 95% CI, 3.19–54.33; χ2 = 18.62, P < 0.0001), while AD was associated with poor performances in school (OR = 6.73; 95% CI, 1.62–30.34; χ2 = 9.49, P = 0.002) and behavioral problems (OR = 13.47; 95% CI, 1.48–310.12; χ2 = 8.24, P = 0.005).
We have considered the family history of generic neuropsychiatric illness and more specifically, mood disorders. Fifty-six percent of MDD, 37% of DD, 18% of AD patients showed familiarity for mood disorders.
Familiarity for mood disorders was associated with the diagnosis of MDD (OR = 4.30; 95% CI, 1.23–15.45; χ2 = 6.9, P = 0.008).
Familiarity for generic neuropsychiatric disorders was more frequent in patients with AD (55%) than in patients with MDD (32%) and DD (24%; χ2 = 3.9, P = 0.04).
We have considered the following senders who sent the patients to the child neuropsychiatrist: clinicians (48% MDD, 50% DD, and 18% AD), families (20% MDD, 37.5% DD, and 41% AD), teachers (12.5% DD and 26% AD), first aid station (24% MDD and 4% AD), and social services and/or juvenile court (8% MDD and 11% AD). The first aid station was more frequent in the MDD group than in other groups (χ2 = 6.33; P = 0.01) and teachers were more frequent (χ2 = 6.74; P = 0.01) in the AD group, while clinicians were less frequent, than in other groups (χ2 = 5.80; P = 0.01) (Table 3).
One or more stressful life events were found in 88% of MDD group, 87% of DD group, and 100% of AD group (Table 3).
We have grouped them in 6 categories: familial problems (such as divorce, dysfunctional family relationships, poor communication style, parental or relative’s illness, and adoption), bereavement, bullying, abuse, illness, school problems (such as learning disabilities and leaving school). The familial problems were more frequent in the DD group than in other groups (χ2 = 5.19; P = 0.02).
The comorbidities were found in 64% of MDD, 87.5% of DD, and 70% of AD groups. The personality disorder was more frequent in the MDD group (χ2 = 4.76; P = 0.04), while learning disorders were more frequent in the AD group than in other groups (χ2 = 3.94; P = 0.04) (Table 3).
Although there is an extensive body of literature regarding depression in adults, to date, there have been a relatively few previous population-based studies that examine the problems of the differential diagnosis of mood disorders in childhood.32,33,36,37,43–46
Although the criteria for all mood disorders are the same for children and adolescents as they are for adults, there are unique developmental challenges and considerations in diagnosing and treating disorders in youth.4–17,19,20,28
Although there is evidence that distinguishes unipolar disorder and BD, indeed very few studies have reported on the problems related to the distinction between depressive disorder and AD and on the role of predictors for these 2 pathologies. Some studies have found differences between patients with AD and those with other mood disorders, which include severity of symptoms, psychosocial adaptation, and number and intensity of stressors.32,34–36,38
In our sample, the depressive disorder and AD are fairly distributed. Although the statistical analysis was performed in a small group of patients, it is possible to make some considerations about the differential diagnosis.
The discriminate analysis revealed statistically significant differences between patients with MDD and DD and those with AD on 3 variables, of which 2 are specific to the disorders (prevalent symptoms and treatment) and 1 is included in the related problems (family history).
Although the majority of the prevalent symptoms were common among the 2 diagnostic categories, the unique symptoms that were more specific for MDD were the vague physical complaints (P < 0.0001), while that for AD were the poor performances in school (P = 0.002) and behavioral problems (P = 0.005). These findings are in accordance with the literature. For the depressive disorders, some symptoms (more the affective) are atypical, subtle, and not visible by the family, and are present in children who have limited ability to communicate emotions and thoughts, with consequent tendency toward somatization; moreover, depressive symptoms vary on the basis of the age and of the cognitive and social levels.4–13,19,28,47 Indeed, children express the emotional problems due to the stress as externalized symptoms (such as poor performances in school and behavioral problems) as in AD.32,40,43,48 The prevalent symptoms observed in our patients, such as vague physical complaints, poor performances in school, disengagement from peer play, anxiety symptoms, and behavioral problems, are not included in the diagnostic criteria of DSM-IV, confirming the literature data on the atypical presentation and the difficulties in diagnosing mood disorders in childhood.4–9,36,37,49–51
The second variable that was statistically significantly associated with the disorders was the treatment resulting from the psychological therapy specific for the AD group (P = 0.02) and the combination therapy for the MDD group (P = 0.0001); however, the psychological therapy emerged more frequently without statistical significance in the DD group (P = 0.0007).
With regard to the role of the family history, in our sample, 56% of MDD patients showed familiarity for mood disorders (P = 0.008), whereas 55% of AD patients showed familiarity for generic neuropsychiatric disorders (P = 0.04). Our findings confirm the literature data by supporting the principal role of the genetic factors in depressive disorders and of the environmental factors in AD.4–7,51–56 The role of the genetic factors in AD is still unclear.35,36,44
Seven variables, of which 4 are specific to the disorders and 3 are included in the related problems, are weakly associated with the single diagnostic categories, without statistically significant differences. These variables are characteristics of sample, duration of symptoms, reasons of first consultations, severity, sender, life stressors, and comorbidity.
With regard to the characteristics of the samples, considering the range of the age at the first observation, it seems that the depressive disorders affect older people more than patients with AD. Although the statistical analysis did not show significant results, some questions emerged. We have hypothesized that the MDD and DD have been recognized later for the atypical presentation. However, in our cases, we have been unable to estimate the exact age at which the symptoms began. The clinical interview revealed that the disorders really appeared early with respect to the time of diagnosis (many years before in 48% of MDD, 75% of DD, and 55.5% of AD groups). Only few parents present a child for evaluation after a 2-week period, as defined in DSM-IV-TR. Moreover, the onset of depression is often gradual, and only by looking previous months and sometimes years, a parent or even the child or adolescent can state that there has been a change in functioning. Perhaps, the ADs are recognized early compared with depressive disorders, which are always recognized later, because these disorders onset with externalized symptoms and also because, during a life stressor (such as bereavement and school problems), major attention has been directed toward the child. So, the precocious diagnosis remains a major problem in both the disorders.
Considering the reasons of first consultation, the affective symptoms were more frequent in the MDD group (P < 0.05 Fisher exact), while poor performances in school were more frequent in the AD group (P = 0.02). Even though the results are not statistically significant, they overlap with the variables prevalent symptoms and sender for the AD group.
The symptoms were severe in all groups because all of our patients were hospitalized. With regard to the treatment, some AD patients received medication because of the severity of their symptoms.
We have hypothesized that the severity of depression has brought the patients to the first aid station, which resulted the prevalent sender (P = 0.01). Instead, the externalized symptoms of AD are perhaps perceived more troubling by the teachers, who were the prevalent senders (P = 0.01).
Although the familial problems were more frequent in DD patients (P = 0.02), none of the life stressors was associated with the disorders with a statistical significance. All evidence pointed to strong familial and school effects and to other weak life stressors for both the groups, supporting the strict relation to the context. To date, it is difficult to distinguish if the life stressors antedate or complicate the outcome of the disorder. Considering the high genetic role for the depressive disorders, the life stressors seem to manifest a depressive vulnerability.57 For the AD, in accordance with the literature data, it is difficult to establish if the patients showed a major vulnerability to stress or if an excess of life stressors determined negative effects in all the subjects; moreover, even though children show more resistance to stress, the effects of the life stressors are more severe if psychiatric familiarity exists.32,35,58,59
With regard to comorbidity, pure disorders were rare in our sample. Although the associations between the single comorbidity and the diagnostic groups did not show statistically significant results, personality disorders were more frequent in MDD group (P = 0.04), and learning disorders were more frequent in the AD group (P = 0.04). Complications in diagnosis and treatment and in differential diagnosis can arise from high rates of comorbidity by “masking” its presence.32,60 Even in a situation where the mood disorder precedes the comorbidity, the differences in diagnostic approach required by both disorders should be considered.
Finally, there is evidence for an increased risk of developing BD in young people with depression. In fact, in 40% of children or adolescents with BD, the illness begins with a major depressive episode, and the switch rates to mania in adults were significantly lower (3.9%) than in children (31.7%).25,61–63 Moreover, considering that the recognition of the BD was delayed by as much as 8–10 years and that unipolar depression remains an uncertain diagnosis lifelong, it is impossible to establish how many of the our depressive patients will become bipolar. This question did not help us to distinguish these 2 disorders.
The 3 variables that were statistically associated with the single diagnostic categories support the distinct construct validity of the single diagnostic categories, but, to date, it is difficult to establish if these variables can be considered diagnostic predictors. On the other hand, the other variables did not support the distinct construct validity of the diagnostic categories, which suggests an overlapping and dimensional concept.
The diagnostic problems are certainly related to the features of the developmental psychopathology, which studies developmental transformations and reorganization occurring over the time. In fact, a particular stress or underlying mechanism may result in different behavioral difficulties at different times in the developmental process and in different contexts, and the developmental transformations and reorganizations occurring over the time modify the expression of the disorders or lead to new symptoms and difficulties.
Moreover, the difficulties in diagnosis and treatment of mood disorders in youth are also related to the interpretation of the interaction between genetic and environmental factors (personal vulnerability and life stressors) because specific types of stressful events and individual patterns of stress response appear to be differentially related to the onset of psychiatric syndromes.64–68
To alleviate the problems of underdiagnosis and undertreatment of mood disorders, it is possible to introduce the concept of spectrum in clinical practice. The term “spectrum” was first used in psychiatry by Kety et al69 for the schizophrenia spectrum and then proposed for the treatment of affective disorders during the past decade. Recently, the division between bipolar and unipolar has also been questioned by the continuity or spectrum concept of mood disorders, which includes the overlapping and dimensional disorders ranging from BD to MDD.63,70,71 Moreover, in the clinical practice, the distinction between depressive disorder and AD are unclear.
The actual diagnostic systems are limited for diagnosing childhood psychiatric disorders, especially in preschool age. In mood disorders of childhood and adolescence, an excessive focusing on the symptoms is limitative, and defining the syndrome of depression in terms of number of symptoms only may give rise to error. From a clinical practice point of view, findings of our study suggest not only a systematic assessment of the depressive symptoms but also the affective and individual significances of the symptoms in the context of the development. Health care professionals should also consider the age of onset, reasons of first consultation, duration of symptoms, severity, familiarity, sender, life stressors, comorbidity, and predictor factors of BD.
The spectrum approach could unify categorical classification that is essential with a dimensional view. Combination of dimensional and categorical principles for classifying mood disorders may help to reduce the problems of underdiagnosis and undertreatment. The development of a validated mood spectrum concept can provide a more differentiated research and treatment model for mood disorders and may help to reduce the underrecognition of bipolarity.
The small size of the sample represents a limitation, but it is possible that with a greater number of cases some of our data, actually resulted “border” at the statistical analysis, may become more specific.
It may take longer than expected to develop better-validated diagnostic criteria. The diffusion of guidelines and the forthcoming International Classification of Diseases, Eleventh Revision (ICD-11) and Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) (DSM-V) may give us new definitions.
The authors report no conflicts of interest in this work.