- Top of page
Aim: The aim of this research was to clarify the development of depression among boys with attention deficit hyperactivity disorder (ADHD) by examining the correlation between depressive mood, oppositional defiant behavior, and age for each ADHD subtype.
Methods: The Birleson Depression Self-Rating Scale (DSRS) was used to evaluate depressive mood while the Oppositional Defiant Behavior Inventory (ODBI) was used to evaluate oppositional defiant behavior. The 90 subjects were divided into three groups: 22 boys (mean age, 12.4 ± 1.9 years) were placed in the ADHD predominantly inattentive type (ADHD-I) group; 45 boys (mean age, 10.4 ± 2.0 years) were placed in the ADHD combined type (ADHD-C) group; and 23 boys (mean age, 12.7 ± 2.4 years) were placed in the depressive disorder (DD) group. The DD group was included to highlight characteristics of depressive mood among boys with ADHD.
Results: The DSRS score was significantly higher in the DD group compared to the ADHD-I and ADHD-C groups. The ODBI score was significantly higher in the ADHD-C group compared to the ADHD-I (P = 0.043) and DD (P = 0.013) groups. In the DD group, ODBI was seen to decrease with increasing age. A certain degree of oppositional defiant behavior was seen in each ADHD subtype. The DSRS score correlated with the ODBI score in the ADHD-C group, while the DSRS score correlated with age in the ADHD-I group.
Conclusion: The characteristics of developing depressive mood in childhood ADHD appeared to differ between subtypes and also differed from depression without ADHD.
ATTENTION DEFICIT HYPERACTIVITY disorder (ADHD) is characterized by impaired attention, deficient impulse control, and hyperactive behavior and affects 5–10% of people, usually arising during childhood.1 Based on elevations of two symptom dimensions, those of inattention (IA) and hyperactivity/impulsivity (H/I), diagnostic criteria classify ADHD into three subtypes.1 The predominantly inattentive subtype (ADHD-I) includes six or more symptoms of inattention but fewer than six symptoms of H/I. The predominantly hyperactive/impulsive subtype (ADHD-H) includes six or more symptoms of H/I but fewer than six symptoms of inattention. The combined subtype (ADHD-C) includes six or more symptoms on both dimensions.
It has been claimed that the symptoms of IA and H/I improve or show remission with increasing age; however, many cases with ADHD show persistent symptoms that continue into adulthood and lead to diverse conditions and other comorbid mental disorders.2,3
It is believed that the internalization of aggression in the course of ADHD is indicative of mood problems, such as depression and anxiety disorders, while the externalization of aggression is indicative of behavioral problems, such as oppositional defiant disorder (ODD) and conduct disorder (CD). Such internalization and externalization do not develop independently but rather develop concurrently and influence each other.3–6
Comorbid depression was reportedly found in 20–30% of ADHD cases.7,8 However, the evaluation of these symptoms is often difficult and clinicians tend not to diagnose depression in cases of ADHD if patients have no apparent symptoms, such as suicide attempts.4 The reasons for difficulty in the diagnosis of depression include poor verbalization of symptoms, which are easily overshadowed by behavioral problems in children, non-standardization of evaluations for such condition, and changes in ADHD symptoms with increasing age. However, depression, ODD, CD, and substance abuse are common comorbid conditions with ADHD and ADHD is the most common comorbid diagnosis in those with a primary mood disorder who have committed suicide.9–14
As noted above, comorbid depression seriously affects ADHD prognosis. Therefore, understanding the conditions related to the development of depression in ADHD has great significance from the perspective of both prevention and early intervention.
The purpose of this study was to clarify the characteristics of depressive mood as a key internalization symptom, the characteristics of oppositional defiant behavior as a key externalization symptom among elementary and junior high school students for each ADHD subtype and to examine the conditions related to the development of depression in ADHD.
- Top of page
Table 1 shows the mean age and the mean scores of DSRS and ODBI in the three groups. In terms of age, the ADHD-C group was significantly younger than the ADHD-I (P = 0.004) and DD (P < 0.001) groups. The DSRS score was significantly higher in the DD group compared to the ADHD-I (P < 0.001) and ADHD-C (P < 0.001) groups. The ODBI value was significantly higher in the ADHD-C group compared to the ADHD-I (P = 0.043) and DD (P = 0.013) groups.
Table 1. Mean scores of age, DSRS, and ODBI
| ||ADHD-I||ADHD-C||DD|| || |
|All subjects||Mean ± SD ||Mean ± SD||Mean ± SD||Kruskal–Wallis test||Statistical analysis|
| ||n = 22||n = 45||n = 23|| || |
|Age||12.4 ± 1.9||10.4 ± 2.0||12.7 ± 2.4||χ2 = 18.911, d.f. = 2, P < 0.001||1 > 2 (P = 0.004), 2 < 3 (P < 0.001)|
|DSRS||9.7 ± 5.8||10.2 ± 4.2||17.7 ± 6.7||χ2 = 21.069, d.f. = 2, P < 0.001||1 < 3 (P < 0.001), 2 < 3 (P < 0.001)|
|ODBI||19.5 ± 9.5||29.3 ± 13.5||18.5 ± 14.8||χ2 = 11.3552, d.f. = 2, P = 0.003||1 < 2 (P = 0.043), 2 > 3 (P = 0.013)|
|Elementary school|| || || || || |
| ||n = 7||n = 35||n = 5|| || |
|Age||10.1 ± 1.1||9.6 ± 1.9||8.6 ± 0.9||χ2 = 4.6331, d.f. = 2, P = 0.099|| |
|DSRS||4.9 ± 4.6||10.1 ± 4.4||13.6 ± 5.9||χ2 = 7.7634, d.f. = 2, P = 0.021||1 < 3 (P = 0.033)|
|ODBI||19.9 ± 10.8||29.0 ± 13.7||35.4 ± 9.0||χ2 = 4.3058, d.f. = 2, P = 0.12|| |
|Junior high school|| || || || || |
| ||n = 15||n = 10||n = 18|| || |
|Age||13.5 ± 1.0||13.3 ± 1.1||13.8 ± 1.0||χ2 = 1.5711, d.f. = 2, P = 0.46|| |
|DSRS||12.0 ± 4.9||10.6 ± 3.8||18.8 ± 6.6||χ2 = 13.5222, d.f. = 2, P = 0.0012||1 < 3 (P = 0.014), 2 < 3 (P = 0.006)|
|ODBI||19.3 ± 9.5||30.3 ± 13.1||13.8 ± 12.6||χ2 = 9.5306, d.f. = 2, P = 0.0085||2 > 3 (P = 0.009)|
We evaluated elementary and junior high school students, comparing their age, DSRS, and ODBI scores. Among the elementary school students, no significant difference was found between the three groups with regard to age. The DSRS score was significantly lower in the ADHD-I group compared to the DD group (P = 0.0033); however, no significant difference was found between either the ADHD-C and ADHD-I groups or the ADHD-C and DD groups. In addition, no significant difference was noted in the ODBI score in any of the three groups. Among the junior high school students, no significant differences were found with regards to age among the three groups. The DSRS value was significantly higher in the DD group compared to the ADHD-I (P = 0.014) and ADHD-C (P = 0.006) groups. The ODBI value was significantly higher (P = 0.009) in the ADHD-C group compared to the DD group; however, no significant difference was found between either the ADHD-I and DD groups or the ADHD-I and ADHD-C groups.
Table 2 shows the numbers of high DSRS and high ODBI scorers in each ADHD subtype. Subjects with high DSRS scores were 10 of 67 (14.9%) from the total ADHD subjects, four of 42 (9.5%) among elementary school students, and six of 25 (24.0%) among junior high school students. In terms of ADHD subgroups, subjects with high DSRS scores were five of 22 (22.7%) (zero of seven [0%] elementary school students and five of 15 [33.3%] junior high school students) in the ADHD-I group and five of 45 (11.1%) (four of 35 [11.4%] elementary school students and one of 10 [10%] junior high school students) in the ADHD-C group. Subjects with high ODBI scores were 41 of 67 (61.2%) from the total ADHD subjects, 28 of 42 (66.7%) among elementary school students, and 13 of 25 (52%) junior high school students. Subjects with high ODBI scores were eight of 22 (36.4%) (three of seven [42.9%] elementary school students and five of 15 [33.3%] junior high school students) in the ADHD-I group and 33 of 45 (73.3%) (25 of 35 [71.4%] elementary school students and eight of 10 [80%] junior high school students) in the ADHD-C group. Comparing the rates of subjects with high DSRS scores in terms of ADHD subtype, no significant difference was found among elementary school students and junior high school students (Fisher's exact test, P = 1.00 for elementary students, P = 0.34 for junior high students). Similarly, when comparing the rate of subjects with high ODBI scores in terms of ADHD subtype, although no significant difference was found in elementary school students (Fisher's exact test, P = 0.196), a significantly higher rate was noted in the ADHD-C group for junior high school students (Fisher's exact test, P = 0.0414).
Table 2. Number of high DSRS/ODBI scores in each grade
|Junior high students||ADHD-I||10||5|
|Junior high students||ADHD-I||10||5|
The correlation between DSRS, ODBI, and age is presented in Table 3. In the ADHD-I group, a positive correlation was found between age and DSRS (Spearman's correlation coefficient [ρ] = 0.502, P = 0.017), but no correlation was found between the DSRS and ODBI scores or the ODBI and age. In the ADHD-C group, no correlation was found between age and DSRS or age and ODBI, but a positive correlation (ρ = 0.449, P = 0.002) was noted between the DSRS and ODBI scores. In addition, no apparent differences were noted after the variable of age was controlled (ρ = 0.435, P = 0.003). In the DD group, no correlation was found between age and the DSRS or the DSRS and ODBI scores, but a negative correlation was found between age and the ODBI score (ρ = −0.583, P = 0.004).
Table 3. Correlation between age, DSRS, and ODBI
|ADHD-I|| || || |
| Age||•|| || |
| DSRS||0.502*||•|| |
|ADHD-C|| || || |
| Age||•|| || |
| DSRS||0.08||•|| |
|DD|| || || |
| Age||•|| || |
| DSRS||0.090||•|| |
- Top of page
Based on the present findings, the rate of subjects with a high degree of depressive mood was similar between ADHD-I and ADHD-C; however, the developmental characteristics of depressive mood differed between ADHD subtypes. In the ADHD-C group, the severity of depressive mood correlated with an increasing degree of oppositional defiant behavior regardless of age. Meanwhile, in the ADHD-I group, oppositional defiant behavior did not weaken with age but was maintained at a certain degree while the severity of depressive mood correlated with age.
Similar levels of depressive mood and oppositional defiant behavior were seen in the ADHD-C group and the DD group among elementary school students. The reason for this similarity was believed to be that depression in children tends to be expressed through an irritable mood rather than a depressive mood.1 Conversely, the difference between the ADHD-C and DD groups became apparent among junior high school students. Subjects in the DD group had more depressive mood than those in the ADHD subtype groups while the DD group subjects showed less oppositional defiant behavior than the ADHD-C group subjects. In summary, the symptomatologic differences between ADHD and depression became obvious after the elementary school years and we must bear this point in mind when diagnosing depression that appears concurrently with ADHD.
In terms of ADHD subtypes, the rate of subjects who showed high depressive mood was relatively low across each subtype among elementary and junior high school students. On the other hand, the rate of subjects who showed high oppositional defiant behavior was higher in the ADHD-C group than in the ADHD-I group, especially among junior high school students. In a previous study of the general Japanese population,20 the mean score of DSRS was 7.97 ± 5.02 in elementary school and 11.15 ± 6.79 in junior high school. The mean scores of DSRS in the ADHD-C group and in the ADHD-I group were almost the same as those in the general population. We therefore believe that subjects in each group and the general population might tend to show the same level of depressive mood throughout elementary and junior high school, but subjects in the ADHD-C group might simultaneously display oppositional defiant behavior in a more conspicuous manner than those in the ADHD-I group once they reached junior high school. In studies from Western countries, subjects in the ADHD-C group tended to acquire oppositional defiant disorder more easily than subjects in the ADHD-I group; however, the prevalence of depression was not significantly different between the subtypes.23–25These previous works targeted adult subjects and studied the long-term prognosis, while we studied the development of the symptoms in childhood. Therefore, it is difficult to compare directly the results of the current study with those of previous works. To make an effective cross-cultural comparison, we believe that it would be necessary to follow the course of each case and study the incidence of comorbid disorders as well.
In terms of the correlation between oppositional defiant behavior and depressive mood, previous studies have indicated that the appearance of ODD and CD with ADHD was a strong predictor of later depression.26–28 In this research, it was found that the severity of depressive mood in the ADHD-C group correlated with the severity of oppositional defiant behavior regardless of age, which was similar to the results reported in previous studies. It might be noted that increases of depressive mood and oppositional behavior occurred in parallel regardless of increasing age in the ADHD-C group. Therefore, to ascertain the development of depression occurring in the ADHD-C group, it is important to pay attention not only to depressive mood but also to oppositional defiant behavior as key symptoms of developing depression among both elementary and junior high school students.
In the ADHD-I group, the severity of depressive mood was correlated with increasing age. In addition, no correlation was observed between a decrease in oppositional defiant behavior and increasing age. Instead, oppositional defiant behavior of a certain degree was found to persist even after the junior high school years. When considering the course of ADHD symptoms, however, we must consider that the ADHD-I group could consist of subjects transitioning from the ADHD-C group.29–34 This transition was designated as improvement of hyperactivity and impulsivity; however, Biederman and collegues31,35have pointed out that improvement in hyperactivity and impulsivity does not occur in proportion to the improvement in social functions and that the remission of ADHD did not necessarily correlate with a decrease in depressive mood. So we should understand that even if an increase in oppositional defiant behavior was not apparent, an increase in depressive mood could be observed with an increase in age in ADHD-I. Such characteristics were shared common traits with the ‘failure model,’ which states that an experience of failure created by behavioral problems gives rise to later depression.36,37
Several limitations should be noted with respect to the current study. First, our study contains a relatively small sample size. ADHD-H subjects were especially too few to evaluate. We should organize the comparative study among ADHD subtypes, including ADHD-H in future. Second, we adopted a cross-sectional approach in the present study. We need to conduct a longitudinal study to reveal the developing depression, ODD, and CD in a greater number of subjects with ADHD. Third, our study evaluated the depressive mood using only the self-completion method and did not systematically evaluate comorbid disorders by semi-structured diagnostic interview at the time of first diagnosis. This could be one of the limitations of this study. Depressive mood is an episodic state while behavioral problems are stable, and therefore their correlation could not be accurate at the time of evaluation.38 Fourth, we could not precisely evaluate the neuropsychological profile, especially the intelligence quotient, in several subjects although we could identify subjects with mental retardation by information from their teachers and report cards. Despite these limitations, the present findings are considered quite important because relatively few studies have focused on characteristics of depressive mood in patients with ADHD and comorbid conditions in ADHD subtypes.