Children for whom the differential diagnosis includes bipolar disorder and/or attention deficit hyperactivity disorder (ADHD; combined type) are usually complex and present thorny assessment and treatment problems. The cases of “Seth” and “Eric” illustrate these issues. They are presented in tandem to illustrate how information might unfold and how the flow of information directs treatment in situations where the question of bipolar disorder and/or ADHD is raised.

Seth and Eric highlight the “bipolar disorder versus ADHD” or “bipolar disorder and ADHD” controversy in preadolescent children. They both had symptoms of ADHD, but they also had symptoms of severe mood lability, inadequate response to ADHD treatment (or any other medication for that matter), and family histories of mood disorders. Youngstrom et al. (1) attribute some of the bipolar/ADHD controversy to researchers’ use of different conceptualizations of bipolar disorder, different diagnostic interviews, and different criteria to define study samples. They also note that researchers’ definitions may alter the DSM-IV criteria and do not necessarily reflect how the diagnosis is used in clinical practice.

The question of the prevalence, pathophysiology, and treatment of bipolar disorder in children (versus adolescents) will obviously depend on how one diagnoses it. Literature review is unhelpful since most authors combine child and adolescent data or data on bipolar I disorder, bipolar II disorder, and bipolar disorder not otherwise specified. Whether research groups have used more liberal or more conservative definitions of episodes, euphoria, and grandiosity, they can marshal data to validate their approach.

Besides the question of how broadly or narrowly to define mania in youths, there is the related question of how to diagnose children with explosive, aggressive behavior. Previously these symptoms in children were included within the broad definition of hyperkinesis, the predecessor of ADHD. Laufer and Denhoff (2) described “behavior of almost volcanic intensity” and changeability such that the child could be “sometimes good and sometimes bad.” In DSM-III, this emotion component of hyperkinesis was split off from the “core” attention and hyperactivity symptoms, and the explosive, aggressive behaviors were absorbed into the “associated symptoms” of ADHD, oppositional disorder/oppositional defiant disorder, and conduct disorder. Investigators of aggression have developed their own terms—“affective,” “impulsive,” or “hot” aggression—which appear to identify the explosive behavior manifested by children usually with symptoms of ADHD (3). Leibenluft et al. (4), exploring differences between children with episodic mania and those with chronic explosive behavior, called the latter “severe mood dysregulation.” This is distinguished from mania by the absence of episodes, euphoria, and grandiosity and the presence of prominent and chronic (at least 1 year) anger/irritability or sadness, with severe tantrums occurring several times a week in multiple settings. Other symptoms that may be seen in both mania and ADHD—insomnia, distractibility, flight of ideas, pressured speech, and intrusiveness—are also present. More than 80% of these children have comorbid ADHD and oppositional defiant disorder (5).

Although debate rages about whether severe ADHD symptoms, fluctuating behavior, and short temper with significant aggression represent a virulent form of ADHD or a juvenile subtype of bipolar disorder, there is an evidence base, albeit a small one, for treating both conceptualizations—that is, mania with or without ADHD as well as ADHD with explosive aggression. Thus, practice parameters (6) and consensus documents (7) suggest that in bipolar disorder mood and/or mania should be treated first, and if ADHD symptoms remain, they should be addressed with evidence-based treatments for ADHD. Most of the data for effective treatments for mania in children (down to age 10) and adolescents come from FDA-requested, industry-sponsored studies of medications approved for mania in adults. Data have been published or presented demonstrating that for acute or mixed mania, about 50% of the patients treated with olanzapine, risperidone, quetiapine, or aripiprazole improve about 50%, compared with a response of about 25% with placebo (8, 9). Results for divalproex are mixed (10, 11), and results for other mood stabilizers (e.g., oxcarbazepine) are disappointingly negative (12). A large-scale placebo-controlled study of lithium is under way, but at this time the only similar study that has been conducted on acute mania in children and adolescents with lithium was negative (13).

Treatment for ADHD includes stimulant medications or atomoxetine, behavior modification, and academic accommodations if needed (14); data also suggest that in cases where ADHD is accompanied by extreme aggression in the absence of a mood disorder, stimulants are somewhat beneficial (15–17). There is, in fact, a mandate to begin controlled studies of children whose aggression is not satisfactorily addressed by ADHD treatments alone (3).

There have been three small but systematic studies indicating that the addition of ADHD medications to antimanic medications in children with bipolar disorder and ADHD does not worsen symptoms and may improve them (18–20). There have been no studies of atomoxetine in comorbid ADHD and bipolar disorder.

Circumstantial evidence indicates that even for children with manic-like symptoms, stimulants alone generally do not cause children to develop bipolar disorder (21–23), and where there is evidence of poor response or worsening of symptoms, the implications are nonspecific in that such responses may occur in any number of conditions (24–26). There are anecdotal reports of atomoxetine-induced mania in children with ADHD, but to date, placebo-controlled trials of children with ADHD and depression have not provided evidence of a placebo-drug difference (27). The phenomenon of switching and drug-induced disinhibition is difficult to study (28), and nonstimulant treatment of ADHD needs such attention.

Finally, evidence suggests that atypical antipsychotics, lithium, stimulants, and valproate are effective for treating aggression (29). The ADHD practice parameter thus recommends the addition of these medications to ADHD treatments for patients who have ADHD plus aggression (14).

Diagnosis Expectations

Until we truly understand early-onset bipolar disorder, it will be important to acknowledge the different viewpoints about the condition and to determine both what parents’ understanding is and why they want to know whether the diagnosis pertains to their child. For instance, Seth’s mother had been told that his rages had prompted the bipolar disorder diagnosis. She had not described manic episodes—that is, distinct periods when Seth’s mood was clearly different from usual, lasting at least several days with concurrent elation/irritability, grandiosity, and accelerated verbal and physical activity. Rather, she noted that he became incredibly enraged for up to an hour when he did not get his way, was disappointed, or felt insulted. Some investigators (30) would concur that this presentation is likely bipolar disorder; others (4) would not diagnose bipolar disorder in the absence of discrete manic episodes. Seth’s mother wanted to know if he really had bipolar disorder in order to “find the right medication.”

Eric’s bipolar disorder diagnosis was made by his referring clinician after the apparent worsening of his symptoms on stimulant medication, and it was additionally supported by his positive family history. Unlike with Seth, in Eric’s case there appeared to be a distinct period, after he began treatment with a stimulant, in which his mood was markedly worse than usual, although it was not clear whether this period met other DSM-IV criteria for a manic episode. Regardless, some investigators view increased irritability on stimulants and other medications as evidence of a manic switch, which is especially portentous in the context of a family history of bipolar disorder (31). Unlike Seth’s mother, Eric’s parents wanted to know their son’s trajectory. Current treatment was also important, but the long-term implications concerned them more.

Comprehensive Assessment

Comprehensive, standardized parent and teacher rating scales are an important preliminary part of an ADHD evaluation (14). Screening measures have likewise been advocated for bipolar disorder (6). The best screens cover the important comorbidities and “rule-outs,” including ADHD, oppositional defiant disorder, conduct disorder, anxiety and depressive disorders, psychosis, tic disorders, autism, and, of course, mania. These measures do not make a diagnosis but, accurately completed, alert the clinician to important problem areas to be pursued further.

Seth’s ratings showed parent and teacher concordance for high levels of hyperactivity, impulsivity and distractibility, oppositional defiance, conduct disorder, generalized anxiety, and deviant language items (rapid, excessive, off-topic speech). There were differences in mania ratings on the Child Mania Rating Scale (32), however. A parent score of 16 was made up of items reflecting irritability, explosive behavior, distractibility, and rapid speech. (Often it is helpful to know not only a score but what items were rated to achieve the score.) The teacher score was 24, reflecting, in addition to irritability and explosiveness, periods when Seth was “wound up and excited” or had higher than usual energy. Like Jensen et al. (33), we find that the reasons ratings are discordant, in this case between parent and teacher, are as informative as specific scores. For instance, Seth was explosive both at home and at school, but his teacher noted additional symptoms. We need to know why there is a discrepancy between his mother’s and his teacher’s ratings.

Seth’s school report underscored a notable disability in math and written expression, which had been present since first grade. He had a full-scale IQ of 84, with especially poor performance in working memory and processing speed.

Eric’s ratings evidenced different inconsistencies between parents and teachers. His parents noted manic and depressive symptoms in addition to severe ADHD and oppositional defiant behavior, whereas his teachers endorsed only ADHD and oppositional defiance. Eric, previously at grade level, experienced a drop in grades over the past several years. It will be important to try to understand why.

Parent and Child Interview

Although interviews developed for the study of mood disorders in children have good reliability, what is rarely discussed is the fact that reporting on one’s own behavior may pose a difficult cognitive task, and both parent and child reports can be profoundly influenced by question wording, format, and context (34).

Seth’s Mood Disorder History and Mental Status

Eric’s Mood Disorder History and Mental Status

Seth’s Treatment and Follow-Up

Eric’s Treatment and Follow-Up

There are important reasons why mania and severe ADHD should be understood as different conditions. However, from a therapeutic standpoint, the difference between mania and severe ADHD (plus aggression) is not in the use of atypical antipsychotics and mood stabilizers, both of which are supported by a substantial evidence base for use in both disorders. Nor is it in the need to provide psychoeducation and specific parenting help for families. There is a robust literature on behavioral treatment for ADHD and a growing literature on psychosocial treatments for bipolar disorder (39). Although the terms used to describe interventions to control dysregulated mood and aggressive behavior are different, many of the actual interventions are similar. Two of the most substantial differences in treatment are whether or when to treat ADHD and how to advise parents and child about future treatment.

As we have seen, both Seth and Eric had a long history of ADHD, a volatile temper, and problems with mood regulation, overwhelming parents and teachers alike. Both were significantly challenged academically and had more than ADHD. The diagnosis of bipolar disorder rests on demonstrating an episode, a sustained period of behavior that differs from the person’s usual self, in which elated/expansive/irritable mood co-occurs with other symptoms that operationalize the accelerated energy, thinking, and hyperhedonic activity that underpin mania. Seth was actually observed to have such a period lasting several weeks during which he appeared different—that is, excessively elated, even more explosive than previously, grandiose, and “constructively” energetic (in contrast to his background hyperactivity), with changes in sleep behavior. Once these features were pointed out, his teachers and his mother confirmed that they had occurred previously.

A case might be made that Eric also had an episode, starting around third grade, when his behavior worsened. His aggression and frustration intolerance increased, but he did not experience a simultaneous co-occurrence of other manic symptoms. His ADHD was inadequately treated, but his psychiatrist, concerned about his family history of bipolar disorder, chose to initiate treatment for that disorder.

Both Seth and Eric, even when treated with the best medication and parent and school interventions, remained somewhat symptomatic, which illustrates the fact that we simply do not yet have completely effective treatments for many children with this constellation of symptoms and behaviors.

In diagnosing bipolar disorder in children, it is necessary to keep an open mind, and to continue to do so after the first interview; ongoing observation is critical. A family history of bipolar disorder is important but does not, in and of itself, make a diagnosis. A full assessment, using multiple informants, is needed to address differential diagnosis, including learning and language disorders. Ultimately, it is important to understand the child, not just to reach a diagnosis.