Disruptive mood dysregulation disorder (DMDD), as introduced in
DSM-5 (
1), ostensibly developed in response to the
DSM-5 committee’s concerns about the apparent increase in the diagnosis of bipolar disorder between the mid-1990s and mid-2000s (
2,
3) (
Table 1). This increase evolved for a number of reasons. First,
DSM-III eliminated emotion dysregulation, which had been an intrinsic part of the “hyperkinetic child” (
4), from the ADHD criteria. This meant that children with ADHD (i.e., hyperactivity, impulsivity, and distractibility) and severe irritability were classified as having bipolar disorder because mania criteria include all of those symptoms. Second, changes in mental health care delivery decreased the amount of time clinicians were able to spend with patients, eroding accurate data gathering. Third, a satisfying and reimbursable diagnosis was lacking for children with explosive temper outbursts, which created the “perfect storm” for increased diagnosis of bipolar disorder. Fourth, mania criteria lacked a developmental perspective, including how to handle multiple informants and comorbid ADHD symptoms and what to elicit from the child. Fifth, episode definition lacked clarity regarding the duration of symptoms over the day and having an offset as well as an onset, such that behaviors constituted a difference from the person’s usual self. Sixth, for diagnosis of adults, irritability was removed from depression criteria but left in mania, increasing rates of “mixed episodes.” Finally, availability of medications to treat mania of adults increased and medications became available to older children and adolescents because of the Best Pharmaceuticals Act for Children, which requires that medications approved for adults be tested with youths for safety and efficacy if the drugs are likely to be prescribed to young people. For a review, see Carlson and Klein (
5).
As these situations were unfolding, the question arose regarding whether there was a special juvenile bipolar phenotype characterized by severe irritability/explosiveness and ADHD/manic-like symptoms (
6). Leibenluft (
7) developed a line of research to compare classic, episodic mania with the chronic, severely irritable phenotype postulated to be juvenile mania. The latter condition was labeled as severe mood dysregulation (
7) and was characterized by chronic irritability (feeling “crabby” and angry most of the day, every day), symptoms of hyperarousal (basically the “B” symptoms of mania, which encompass many ADHD symptoms), and “markedly increased reactivity to negative emotional stimuli that is manifest verbally or behaviorally” (e.g., response to frustration with extended temper tantrums, verbal rage, and/or aggression toward people or property). Outbursts had to occur on average more than three times per week for the last 4 weeks. The severe mood dysregulation label has enabled genetic, neuroscience, and treatment studies, most of which have concluded that severe mood dysregulation describes a seriously impaired population of children, but the condition is different from bipolar disorder (
7).
Diagnostic and Assessment Issues
The evaluation of DMDD rests on ascertaining irritability, outbursts, and other conditions that present with irritability (e.g., mania, depression, anxiety, posttraumatic stress disorder) to ensure that these conditions are not accounting for the symptoms. However, there was relatively poor test-retest reliability in the
DSM field trials, in which kappa values ranged from 0.06 in outpatient settings to 0.49 in inpatient settings (
8). Even in an inpatient sample in which observations within informants were good (κ=0.72 and 1.0 for irritability and explosiveness, respectively), agreement between parent report and staff observation was poor. Of a sample of 82 hospitalized children, parents said that 68% of the patients were severely irritable and explosive (which is why they were hospitalized); however, this rate was only 39% when the patients were observed (
9).
In the most frequently used semistructured interviews, “irritability” is elicited as having feelings of anger or crankiness (in the section on depression) or as being easily annoyed (in the oppositional defiant disorder section). For a review, see Carlson et al. (
10). The severity of irritability is rated by frequency, such as irritable several times a week, irritable more than 50% of the time, or irritable at least one-half the day, most days for at least 12 months. In some interviews, the definition of irritability becomes confounded with outburst behavior, which is one of the results of getting very angry. For instance, in the Washington University in St. Louis Kiddie Schedule for Affective Disorders and Schizophrenia (
11), a rating of 4 is defined as being irritable more than 50% of the time or the patient shouts, loses his or her temper, or has tantrums. A rating of 6 requires extremely angry feelings most of the time or frequent, uncontrollable tantrums (
11).
The module developed by Leibenluft et al. (
12) to study severe mood dysregulation separates irritability from outbursts but rates the latter not by the severity of the actual outburst but rather as “reactions out of proportion to stressors, verbal rages and/or aggression toward people or property at least 3 times a week.” (A frequency of three times a week was purely arbitrary. Distinctions between three verbal outbursts per week versus one physical outburst, for instance, have not been drawn.) Otherwise, the “outburst” criterion is taken from the “often loses temper” question in the oppositional defiant disorder section. However, the ratings available from the most commonly used interviews and rating scales do not make the distinction between temper loss and tantrums. The Child Behavior Checklist item, for instance, is “temper tantrums or hot temper” (
13).
Irritability and temper problems are common with young children. In a report on a community sample of 462 children aged 6 years, 9% of parents described their child as frequently irritable, although they did not have a temper loss of three times a week or more. Another 19% of children were said to lose their temper at least three times a week, although only 11% of those who lost their temper actually had tantrums. Interestingly, not all of the children were described as chronically irritable. Their tantrums were generally verbal rather than physical (
10). In other words, although irritability, losing temper, and having tantrums are related, they are not synonymous and need to be ascertained separately, not rated as one entity. There are major public health implications between becoming angry with relatively little provocation (irritability) and expressing that anger by threatening to kill, throwing things, or attacking someone.
Differential Diagnosis
Although irritability characterizes a number of disorders, the two conditions most easily confused with DMDD are oppositional defiant disorder and intermittent explosive disorder. DMDD and oppositional defiant disorder differ in that the former is supposed to have had at least a year’s duration, must be present in more than one setting, and requires “outbursts” several times a week. Where both occur, DMDD takes precedence. The difference between DMDD and intermittent explosive disorder is that the former is characterized by chronic irritability rather than just impulsive and explosive overreactions to stressors.
Epidemiology
Rates of DMDD in published community studies have extrapolated criteria onto data because there have been no specific interview materials for either DMDD or its predecessor, severe mood dysregulation . Given that caveat, rates of DMDD in the community may depend on the method of assessment (interview versus rating scale) and the age of the sample. DMDD rates derived from responses to interviews were between 3% and 8% for young children and less than 1% for older children (
14,
15). In another study using a rating scale (
16), rates of 9.3% and 5.9% were reported for children at a mean age of 8.7 years and for those same children 8 years later at age 16.4 years, respectively.
Comorbidity
Most psychiatric conditions among children co-occur with other disorders, and DMDD is no exception. In both community studies and clinical studies, comorbidity of DMDD and oppositional defiant disorder is extremely high (
Table 2). In fact, some researchers question the difference between the two conditions (
17). The overlap is apparent in both community and clinical samples. Odds ratios are lower in clinical samples because rates of the comorbid diagnosis are higher in the comparison group. In clinical samples, the combination of ADHD and oppositional defiant disorder occurs with more than 75% of children.
Diagnostic Stability and Outcome
Most longitudinal studies have shown that the diagnosis of DMDD decreases over time. In a study from Deveney et al. (
18), clinically ill children with severe mood dysregulation/DMDD (age 11 years) were seen 2 and 4 years later, at which point rates declined from 100% to 49% and 40%, respectively. This was not to say that the children were well—only that they no longer met full criteria. Findings were similar in the Longitudinal Assessment of Manic Symptoms (LAMS) study, in which 52% of the sample met DMDD criteria at one assessment, 29% at two assessments, and 19% at all three follow-up assessments (
17). As cited earlier, in one community sample, rates declined from 9.3% to 5.9% from childhood to adolescence. The lifetime prevalence of severe mood dysregulation dropped from 3.3% to 0.4% if the diagnosis persisted over two follow-up waves in the Great Smoky Mountains study (
19). Whether these observations reflected the developmental course of the disorder or inherent instability of diagnosis is not yet clear.
Two follow-up studies have been done with the Great Smoky Mountains epidemiologic data and extrapolated criteria. The first analysis compared children with a diagnosis of severe mood dysregulation at age 10.6±1.4 years with diagnoses again at age 18.3±2.1 years and found significant odds ratios only for depressive disorders (7.2; 95% confidence interval=1.3–38.8). Controlling for baseline diagnoses, analyses showed that none of the externalizing disorders emerged as significant at age 18 (
19).
Using the same sample, Copeland et al. (
20) used follow-up data from three follow-up assessments in young adulthood (between ages 19 and 26 years). Adults with childhood DMDD differed from a control group without psychiatric diagnoses on many variables. However, using a psychiatric comparison group, only rates of anxiety and depression were significantly higher among those with DMDD. However, there were no measures of either adult ADHD or oppositional defiant disorder. Bipolar disorder rates were negligible. Other outcomes, however, suggest that externalizing disorders were still contributing to impairment, with both police contact and poor educational outcomes (
Table 3).
Impairment
Where it has been measured, impairment as a result of DMDD is substantial. Clinical Global Assessment Scale (CGAS) scores in a community sample of 6-year-olds were significantly lower than the rest of the sample (64.6±11.1 versus 76.3±10.7) (
15). Even within a psychiatrically ill sample, consistent with the study by Copeland et al. (
20), impairment was over and above that of other psychiatrically ill children. Thus, in a clinic sample of 6-year-old children with and without DMDD, CGAS scores were 39.9±4.1 versus 46.4±9.4 (
10). Participants in a National Institute of Mental Health (NIMH) sample of children with severe mood dysregulation had a CGAS score of 45.8±6.9. In the LAMS study (
17), children with DMDD were lower functioning than the psychiatric comparison group (50.7±9.1 versus 56.0±10.3, p<0.001).
The DMDD diagnosis was developed to prevent children with irritable or explosive behavior from being misdiagnosed as having bipolar disorder. In an attempt to determine whether the strategy might be successful, Margulies et al. (
9) used scores greater than 20 on the Child Mania Rating Scale (CMRS) (
20,
21)—identifying being at risk of a bipolar diagnosis—to see how many might meet criteria for DMDD. More than one-half of inpatient children (56%) had parent-reported manic symptoms on the CMRS. That is not to say that they had mania, only that they were at risk of receiving the diagnosis. In fact, 45.7% met criteria for DMDD by parent report, and only 17.4% did so by the end of their inpatient stay (
9).
Treatment
Although treatment studies of DMDD have not yet been reported, there have been a few treatment studies of severe mood dysregulation. For instance, a placebo-controlled study of lithium with a sample of 25 children and adolescents (lithium, N=14; placebo, N=11) hospitalized in an NIMH research unit showed no improvement on any clinical outcome measures (
22).
Another study compared children in a summer treatment program for children with ADHD, in which 33 children with severe mood dysregulation and ADHD were compared with 68 children who received methylphenidate and behavior modification. Children with severe mood dysregulation responded as well to methylphenidate and an intensive behavior modification as did children with ADHD without severe mood dysregulation. Higher doses of both aspects of the intervention produced greater improvement in both groups. Nevertheless, only 6% of parents of children with severe mood dysregulation and 27% of parents of children with ADHD without severe mood dysregulation considered their child “unimpaired” at the end of treatment (p<0.03)—a finding that suggests that even with the best ADHD and oppositional defiant disorder treatment, children do not achieve complete remission in either category, especially children with severe mood dysregulation (
23,
24). Finally, in an open-label study of 21 children with severe mood dysregulation, all of the conditions measured (ADHD, irritability, depression, and global functioning) improved (
14).
Roy and Klein (
25) outlined treatment approaches in the absence of specific data, noting that parent training and behavior modification, stimulants, valproate, and antipsychotic medications for children have been useful in reducing affective aggression with robust effect sizes.
Rates of combined ADHD and oppositional defiant disorder in clinical samples of children with DMDD or severe mood dysregulation are approximately 75% (
7,
9,
10), which has suggested a strategy of stepped pharmacotherapy (
26). This involves maximizing ADHD treatment and then adding medication for “mood” or reactive aggression. In two similar studies, ADHD was treated with parent training and optimized stimulant medication. In the first study (
26), extended-release divalproex was subsequently added in a double-blind fashion compared with placebo. Using remission of reactive aggression as the primary end point, the authors found that 57% of children achieved remission while taking extended-release divalproex compared with 15% for those taking placebo (
26). Using a similar approach, Aman et al. (
27) found that adding risperidone, compared with placebo, provided added benefit in treating disruptive, aggressive behavior. In both studies, however, most of the treatment effect occurred with the stimulant phase, and a secondary analysis showed that both mood and aggression improved with optimized stimulants (
28,
29).
There has been speculation that because follow-up data point to depression and anxiety as important outcomes for children with severe mood dysregulation or DMDD, using mood dysregulation psychotherapeutics such as dialectical behavior therapy or interpersonal psychotherapy or adding antidepressant medications like citalopram to stimulant medication might be helpful. ClinicalTrials.gov notes that a number of these studies are currently recruiting patients.
Future Directions
The criteria of DMDD combine emotional (irritability) and behavioral (temper outbursts) symptoms and thus encompass comorbidity within a single entity. Up to now, DMDD has been studied with post hoc extrapolation of criteria. It is not clear whether DMDD is the same as severe mood dysregulation, on which there are at least some data, albeit not much in regular clinical settings. How DMDD relates to bipolar disorder and oppositional defiant disorder or combined ADHD and oppositional defiant disorder remains to be seen. There have been a plethora of studies of reactive, hot, affective aggression, but this has not been a target of treatment allowed by the U.S. Food and Drug Administration (
28). Insofar as a diagnosis of DMDD provides a way to describe children with explosive outbursts and to provide enhanced treatment for this significantly impairing behavior, use of the DMDD diagnosis will have clinical benefit.