Adult attention-deficit hyperactivity disorder in patients with bipolar I disorder in remission: Preliminary study
Lut Tamam, MD, Cukurova University Tip Fakultesi Psikiyatri Anabilim Dali, 01330, Balcali, Adana, Turkey. Email: Ltamam@yahoo.com
Abstract Attention-deficit hyperactivity disorder (ADHD), a syndrome that typically first appears in early childhood, can occur in individuals of all ages. Prospective studies have demonstrated that at least half of children diagnosed as having ADHD continue to suffer the symptoms of this disorder in their adult life with significant impacts on their social status, achievement level and sense of well-being. The purpose of this preliminary study was to determine the rate of ADHD in patients with bipolar disorder (BD) and to examine the effects of comorbid ADHD on several clinical and sociodemographic variables of bipolar patients. Forty-four BD-I patients followed up in psychiatric outpatient clinics in two university hospitals, were assessed for the presence of adult ADHD according to DSM-IV. All patients also completed the Wender Utah Rating Scale for objective evaluation of ADHD. Of 44 patients with BD-I, only seven (15.9%) fulfilled criteria for a diagnosis of adult ADHD. Bipolar disorder-I patients with comorbid ADHD were more likely to be female, and have more affective episodes (especially depressive episodes) than bipolar patients without comorbid ADHD. Age at onset of affective illness was not significantly different between the two groups. In line with results of several previous reports, the present study also showed higher prevalence of ADHD in patients with BD-I than in normal population. A higher number of affective episode in patients with comorbid ADHD may suggest a more severe clinical course of BD in these patients. A larger group of samples is required to clarify the exact association and interaction between these two clinical entities.
Attention-deficit hyperactivity disorder (ADHD), a clinical syndrome that typically first appears in early childhood, can occur in individuals of all ages.1 Prospective studies have found that 10–60% of children diagnosed as having ADHD continue to suffer the symptoms of this disorder in their adult life with significant influence on their social status, achievement level and sense of well-being.2–4 It is estimated that the prevalence of adult ADHD is 4–5%, which correlates with a prevalence of 5–9% in school-aged children, further confirming an approximate 50% persistence into adulthood.4
The relationship between bipolar disorder (BD) and ADHD also draws significant interest due to extensive symptom overlap of these two disorders in all age groups.5–14 A number of studies have consistently reported high rates of ADHD in prepubertal children with BD, ranging from 57 to 100%.11 In a pioneer study, Wozniak et al. found that all manic prepubertal subjects included in the study except one (98%) also fulfilled diagnostic criteria for ADHD.15 The ADHD comorbidity among patients with BD, decreases steadily as the studied population grows older. Comorbid ADHD prevalence decreases in BD populations from a rate of approximate 100% during prepuberty to 70% during young adolescence, and to 30% among older adolescents.1,14,15
Although the research on the overlap and comorbidity of ADHD and BD in adults is still in its infancy, reported rates for comorbidity of ADHD among BD populations are apparently lower (i.e. 9–35%) than the rates for other younger age groups.5 For instance, in a pioneering study, Winokur et al. using Schedule for Affective Disorders and Schizophrenia (SADS), found the rate of childhood ADHD among 189 adult BD probands to be 21.3% as compared to 4% of normal controls.6
Adult BD patients with comorbid ADHD have different phenomenology, presentation and disease course when compared to adult BD patients without comorbid ADHD. The BD patients with comorbid ADHD are reported to have higher affective dysregulation, an early onset disease, and more depressive and mixed episodes.7 Moreover, a recent study reported an association between childhood history of ADHD and attenuated response to mood stabilizers during primary treatment for BD, indicating the possible unfavorable effect of ADHD on the course of BD.9
Currently, due to the paucity of studies on adult populations, it is not possible to draw a definite conclusion regarding the prevalence, phenomenology and exact influence of this comorbidity. The majority of samples in these previous studies investigating the comorbidity of adult ADHD and BD usually consisted of patients with a diagnosis of BD-II, a less severe type of BD. Several authors questioned the applicability of these data to adults with BD-I because studies focusing on BD-I patients are relatively rare.7
The aims of the present preliminary study were to determine the rate of adult ADHD in a group of patients with BD-I who were in remission; to examine and compare several clinical and sociodemographic variables of BD-I patients with and without comorbid ADHD; and to test the hypothesis of whether early onset BD-I is a harbinger of BD-adult ADHD comorbidity in later life.
The subjects consisted of 44 consecutive adult patients between 19 and 58 years of age who were recruited from outpatient clinics and a BD follow-up program in two university hospitals (Cukurova University, Adana and Trakya University, Edirne, Turkey) located in southern and north-west regions of Turkey, between January 2003 and July 2003. Both hospitals are tertiary-level hospitals serving large populations. Because of the study’s preliminary nature, only a limited number of subjects (19 male, 25 female; total, n = 44) were enrolled. There were no differences between sites with regard to gender, age, and age at onset of the disorder. Of 78 patients approached for the study in two sites, 34 patients were eliminated due to exclusion criteria given in the following section (in particular, for not meeting the criteria for remission and diagnosis of BD-I).
All patients had already received a diagnosis of BD-I according to DSM-IV and were in remission for at least 3 months at the time of inclusion in the study, as corroborated with the Young Mania Rating Scale (<12) and Hamilton Depression Scale (<7) scores. Potential subjects were excluded from the study if they had any clinically significant chronic medical conditions, organic brain disorders or current substance abuse or dependence (in the last 6 months), or if they had any affective episode during the last 3 months. Patients with other BD diagnoses (i.e. BD-II, BD not otherwise specified [BD-NOS], schizoaffective disorder–bipolar type) were excluded. All patients provided informed consent before participating into the study. Local institutional review boards also approved the study.
The diagnosis of BD-I was based on clinical psychiatric interview and further confirmed using the affective disorder module of Structured Clinical Interview for DSM-IV (SCID-I), a semistructured diagnostic interview.16 The SCID–Clinical Version (SCID-CV) was administered by the senior authors (LT and CT) in each site, who were trained and certificated for its use. The other modules of the SCID-CV other than the affective disorder module were not used due to the preliminary nature of the study. Thus comorbidity with other psychiatric disorders could not be identified. For the diagnosis of BD-I, patients had a clinical course characterized by the occurrence of a least one full manic or mixed episode not attributable exclusively to substance abuse, medical disorders or another psychiatric illness. To avoid any other confounding factor due to the nature of the active phase of BD-I (i.e. inability to cooperate with interviewer, increased deficits in cognitive functioning due to illness or use of antipsychotics that could cause problems in recalling childhood histories), we dismissed patients within acute manic or mixed episodes and included only patients in remission for 3 months.
All subjects underwent a standard clinical evaluation including a detailed psychiatric assessment and a detailed medical history. While reviewing the past psychiatric histories, special emphasis was put on age at onset of the disorder (BD or ADHD if present), number and type of affective episodes in the subjects.
All patients were also assessed for the presence of ADHD according to DSM-IV. To reach a full diagnosis of adult ADHD, the following criteria7 were used: (i) full diagnosis of childhood ADHD according to DSM-IV by the age of 7; (ii) chronic course of ADHD symptomatology from childhood to adulthood that existed aside from manic episodes; (iii) some level of impairment due to ADHD symptoms; and (iv) all symptoms of ADHD observed independent of the presence of any bipolar episode within the last month.
The evaluation of the diagnosis of childhood ADHD was based mainly on the self-report of patients and was augmented by reports from other sources if available (i.e. parents, spouses and medical files belonging to childhood if followed up since childhood). Due to lack of such data, we were not able to obtain confirmation or proof from primary schools of the patients retrospectively.
All patients also completed the Turkish version of the Wender Utah Rating Scale-25 (WURS) for evaluation of possible ADHD levels.17 Total WURS scores have been determined.
The χ2 test and Fisher’s exact test were used to analyze categorical variables. Non-parametric Mann–Whitney U-test was performed for continuous variables because the data were not normally distributed. All P were two-tailed and statistical significance was set at P < 0.05. All analyses were conducted using SPSS version 10.0 for Windows (SPSS, Chicago, IL, USA).
Of the 44 patients with a diagnosis of BD-I included in the study, only seven (15.9%) fulfilled the proposed diagnostic criteria for adult ADHD. The majority of patients (5/7) with both diagnoses (adult ADHD and BD) were female. Only two male subjects (2/19; 10.5%) had received adult ADHD diagnosis whereas 20% of female BD-I patients (5/25) received comorbid adult ADHD diagnosis. On the other hand, eight more patients (four male, four female) received a diagnosis of childhood ADHD that did not extend into adulthood. Among all patients 15 (34.1%) had a diagnosis of childhood ADHD.
The first affective episode in the majority of cases was manic episode (n = 31; 70%), the remainder of the patients had depressive episode (n = 13; 30%) as their initial episode. This was also valid for the predominant affective episode during the course of the disease, in which 85% had manic episode as the predominant episode whereas the remainder (n = 7, 15%) had predominantly depressive episode. Among seven patients with both diagnoses (ADHD-BD group) depressive episode was the predominant affective episode in four cases (58%); in contrast, this rate was very low in patients with only BD diagnosis (BD group; n = 3, 8%). The difference between ADHD-BD and BD-only groups was statistically significant (Fisher exact test, P = 0.007). Similarly, depression as the first affective episode was statistically significantly more common in the ADHD-BD group (5/7; 71%) in contrast with the low number of such patients in the BD group (8/37, 22%; Fisher’s exact test, P = 0.0017).
The ADHD-BD group had a statistically significantly higher number of depressive episodes than the BD group (P < 0.05). The difference between these two groups with regard to total number of affective episodes did not reach, but very close to, statistical significance (P = 0.05). No differences were found in terms of the number of manic, hypomanic and mixed episodes between two groups.
Comparison of some of the clinical and sociodemographic variables for ADHD-BD and BD groups is given in Table 1.
Table 1. Clinical and sociodemographic variables for ADHD-BD and BD-only groups
|Female, n (%)||25 (61)||5 (71)||20 (55)||0.680†|
|Current age (years)||33.7 ± 10.6||27.8 ± 9.6||34.9 ± 10.6||U = 70.5, P = 0.058|
|Age at onset (years)||22.7 ± 6.6||20.3 ± 4.8||23.2 ± 6.8||U = 102.5 P = 0.39|
|Early onset BD (<18 years), n (%)||12 (27)||3 (43)|| 9 (24)||0.369†|
|Late-onset BD (≥18 years), n (%)||32 (73)||4 (57)||28 (76)|| |
|First affective episode, n (%)|
| Manic episode||31 (70)||2 (29)||29 (78)||0.017†|
| Depressive episode||13 (30)||5 (71)|| 8 (22)|| |
|Previous episodes||4.2 ± 2.5||7.6 ± 4.2||3.8 ± 1.6||U = 70, P = 0.05|
|Manic episodes||2.8 ± 1.8||3.3 ± 2.9||2.8 ± 1.6||U = 109.5 P = 0.5|
|Depressive episodes||1.0 ± 1.7||4.3 ± 2.6||0.5 ± 0.6||U = 11, P < 0.0001|
|Hypomanic episodes||0.5 ± 0.6||0.8 ± 0.7||0.4 ± 0.6||U = 98.5, P = 0.24|
|Mixed episodes||0.1 ± 0.2||0.2 ± 0.4||0.1 ± 0.2||U = 114.5, P = 0.63|
|WURS-25 score||18.5 ± 15.7||41.3 ± 11.7||12.8 ± 10.7||U = 5, P < 0.0001|
The age, and age at onset of affective illness, were not significantly different between the ADHD-BD and BD groups (age at onset: 20.3 years and 23.2 years, respectively). Grouping of patients according to age at onset of the first episode of BD (age at onset ≤18 years, early onset; >18, late onset) found that three of 12 early onset BD-I patients (25%) and four of 32 late-onset BD-I patients (14%) also received adult ADHD diagnosis, with no statistically significant difference between the two (Fisher’s exact test, P = 0.369). There were also no significant differences for childhood ADHD presence in both groups (50%, early onset BD; 29%, late-onset BD; Fisher’s exact test, P = 0.284).
As expected, mean WURS-25 scores for the ADHD-BD group were significantly higher than the BD group (39.8 ± 11.3, 12.2 ± 10.1, respectively; P < 0.0001). The BD patients with a childhood ADHD diagnosis also had significantly higher mean WURS-25 scores than BD patients without a childhood ADHD diagnosis (31.2 ± 13.2, 9.1 ± 17.8, respectively; P < 0.0001). No significant differences in terms of WURS scores were observed between late-onset and early onset BD groups and between gender groups (P > 0.05).
In several studies, it was estimated that 1–6% of adults have ADHD.7,18 Thus the comorbidity of ADHD among the present sample of patients with BD-I (15%) is higher than the community rate of adult ADHD, which might be a reflection of bidirectional symptom overlap between these two disorders. However, our finding is close to the rates (21%, 9.5%, respectively) reported by a pioneering clinical study in the field and a recent large cohort study of 1000 BD patients,6,13 and is similar to the approximate rates (15%) declared for adult patients with BD in a recent authoritative review.18
In the literature a number of hypotheses have been put forward to explain the high comorbidity of adult ADHD and BD patients ranging from 9 to 35%.5,13 One hypothesis states that symptoms of ADHD that precede the onset of BD represent a prepubertal expression of illness antecedent to the development of full affective episode.5 Testing this hypothesis in a small sample of BD patients, Sachs et al. suggested that co-occurring ADHD may be a marker for early onset (juvenile-onset) BD.5 Nierenberg et al. also confirmed the validity of this hypothesis and determined that comorbid ADHD seems to be linked to age of onset of BD.13 In that study, of all patients with early onset BD, 13% met ADHD criteria, in contrast to 5% of those ADHD-BD patients with an onset at age ≥18. The results of the present study did not confirm the possibility that childhood ADHD is an expression of the early onset of affective illness or a marker for early onset BD. Although the percentage of cases of early onset BD in the ADHD-BD group was higher than that in the BD only group (42% vs 24%), the difference within groups was not significant (P > 0.05), as it should be in order to confirm this suggestion. However, the present results should be considered with caution due to the small number of ADHD-BD cases (n = 7) in the sample.
A higher number of affective episodes in patients with comorbid ADHD, suggests a more severe clinical course of BD in these patients. This might be an expected finding because ADHD and BD are individually associated with high morbidity and disability, and thus their co-occurrence could result in a more severe clinical picture of high morbidity. Also, a combination of attention symptoms and hyperactive/impulsive symptoms, which might result in non-compliance to pharmacological treatment, may also cause these patients to have a heightened risk for developing new affective episodes. Similarly Nierenberg et al., in a large group of patients, found that BD with a lifetime comorbid ADHD was associated with more episodes of depression and mania, more days irritable and a more severe disease course.13 Another significant finding in the present study was the higher number of depressive episodes (including the first episode) in the ADHD-BD group. Several previous studies also reported increased risk for developing depressive episodes (unipolar or bipolar) in ADHD patients,19,20 in support of our findings. According to some authors, ADHD and depression may possibly represent different expressions of a common etiology that requires further extensive research.21 The verification of such an association between these two disorders in future studies, would most likely influence and add new alternatives to our current diagnostic and therapeutic approaches to adult depressive patients, particularly treatment-resistant ones.
Interestingly, the majority of the subjects in the ADHD-BD group were female, but the difference between the ADHD-BD and BD-only groups was not statistically significant. In all studies exploring childhood ADHD, the disorder is found to be at least 2–3-fold more common in boys than girls.18 However the male preponderance in ADHD among pediatric cases is often not observed in the adult ADHD population. The gender ratio in adults is estimated to be 1:1.18 One plausible explanation for this difference is that female adult patients are more likely than men to seek treatment for ADHD symptoms.12 However, the applicability of this suggestion to the present sample is questionable because the sample is too small to allow comment on differences in gender distribution specifically for ADHD-BD patients. In contrast to our findings of female dominance in comorbid ADHD-BD, in a recent multicenter study conducted among 919 treatment-seeking adults with BD, an important male gender dominance was found to exist, with 14.7% of male patients and 5.8% of female patients having comorbid lifetime ADHD.13 Considering the discrepancies with our study and other previous studies, further studies on homogenous BD subgroups is necessary.
There are several methodological limitations of this preliminary study that should be mentioned and borne in mind during the interpretation of the results. This is a relatively small patient group from a BD outpatient unit of two university hospitals, therefore it is not possible to generalize the results to the whole community. The study also did not include a control group apart from the BD-I patients, therefore comparison of the study group with a normal population or another psychiatrically ill group was unable to be made. Another limitation is the exclusion of all other forms of BD from the study, and enrolling patients only with BD-I. This might have an influence on some data because ADHD-BD has been suggested to be a more common presentation in BD-II patients. Tight exclusion criteria that included only recovered patients, could have weaned out sicker subjects with comorbid ADHD and caused the ADHD rates in the present groups to be lower than expected. Retrospective recollection of childhood symptoms by adult subjects and the absence of any valid measures of psychopathology in childhood are other limitations that should be mentioned. Keeping in mind their impact on our findings, we tried to eliminate these limitations by obtaining additional information from close relatives (if available) and previous medical files. Also, previous studies show that adults with ADHD are acceptable reporters of their own conditions even for their childhood, and self-report of ADHD symptoms is a reliable and valid method of assessing ADHD in adults.22,23
In conclusion, despite the growing number of studies indicating high comorbidity of ADHD and BD including the present one, the nature of the comorbidity between ADHD and BD has not as yet been clearly defined. The phenomenological overlap between these two disorders, and changes in the courses of these disorders over time should always be considered in the diagnosis of each disorder at all ages.21 A larger group of samples with long follow-up periods are required to determine the exact association and interaction between these two clinical entities, which would have important clinical implications especially for the treatment of comorbidity of ADHD and BD.