Practical Clinical Concerns Regarding the DMDD Phenotype
The following is a section of the JBRF response to the proposal of the new classification DMDD. The response was submitted to the American Psychiatric Association in 2010. This reprint has been edited from the original form. At the time, DMDD was still referred to by its original name; Temper Dysregulation Disorder (TDD). References herein have been changed to DMDD. While JBRF research has distanced itself from bipolar disorder, at the time of this writing it was not as clear a separation.
We have serious concerns that both the intended and unintended consequences of the Disruptive Mood Dysregulation Disorder (DMDD) diagnosis will obstruct the identification and appropriate treatment of many children who are affected by a bipolar condition. For some of them, this could come with life threatening consequences.
Investigators who defined DMDD found a lifetime prevalence of SMD (the precursor to DMDD) of 3.3% in youth 9-19 years of age (Brotman et al., 2006). Having thus identified the significant presence of a previously un-delineated condition, one which had heretofore been associated with bipolar disorder not-otherwise-specified (BP-NOS), it seems clear that investigators and the Work Group members expect operationalization of the DMDD diagnosis to re-assign a great number of children from a BP-NOS diagnosis to DMDD, a non-bipolar condition, and that the presence of this new diagnosis is likely to decrease the number of BP-NOS diagnoses in the future.
Alternatively, research supported by the JBRF found that two-thirds of children identified from a study of youth at risk for, or who were diagnosed in the community with bipolar disorder (n = 2,795) exhibit some level of the Fear of Harm (FOH) trait (Papolos et al., 2009). This heritable and defining feature of a novel subphenotype of juvenile-onset bipolar disorder is associated with the most severe forms of mania and depression, as well as the highest rates of hospitalization compared to non-FOH subjects. These findings inform us that, despite the alarm over increases in the rates of pediatric bipolar diagnoses, it is likely that approximately two-thirds of these children do indeed belong to a bipolar spectrum.
The above two expectations:
- that many children who currently receive a BP-NOS diagnosis, or who would have received one in the future, will instead, be diagnosed with DMDD, a non-bipolar classification, and,
- that approximately two-thirds of the children at risk for or diagnosed with bipolar disorder (BD) are appropriately included in the BD population,
cannot both be true.
If the FOH phenotype is a well-conceived phenotype, and if its prevalence is as substantial as our research shows it to be, then we fear that DSM inclusion of the proposed DMDD classification, with its overlapping symptoms of severe and chronic irritability marked by disproportionate frustration, will cause a large number of children to be erroneously assigned to a DMDD diagnosis when in fact, they struggle with a bipolar condition.
We believe that this will be the case. As explained elsewhere, we have serious concerns with both the concepts and methodology used to arrive at the DMDD classification and we suspect it to be a cleaved-off fragment of the FOH condition.
However, even if we were to disregard our criticisms and accept the DMDD classification and prevalence at face value, its operationalization would still leave children in the BP-NOS classification and we are likewise concerned that the same misdiagnosis will occur but only with a relatively smaller number of children.
It could be argued that both of these concerns are unwarranted; that the exclusionary criteria of DMDD would prevent children with FOH from receiving a DMDD diagnosis.
Since symptoms of FOH include (Papolos et al., 2009) :
Has elated or silly, goofy, giddy mood states,
Has exaggerated ideas about self or abilities, and
Tells tall tales; embellishes or exaggerates,
and Criterion H of the DMDD Draft Criteria (APA DSM5 Development, 2010a) states:
In the past year, there has never been a distinct period lasting more than one day during which abnormally elevated or expansive mood was present most of the day for most days, and the abnormally elevated or expansive mood was accompanied by the onset, or worsening, of three of the “B” criteria of mania,
then technically, children who fit the FOH phenotype would be excluded from a DMDD diagnosis and would continue to be diagnosed under the BP-NOS classification.
However, due to practical clinical realities, we believe that this important differentiation is not what will happen. The combined effects of the following considerations will strongly bias towards a DMDD classification:
- DMDD is introduced as a correction to a highly controversial and confusing illness. Professionals have faced a diagnostic conundrum when asked to assign a bipolar diagnosis to a presentation which is often so dissimilar from defined criteria of adult BD I or II. Families are exhausted and desperate for an explanation as to why their children are not finding effective relief under the current assumptions and treatment options. The absence of any further articulation of a developmental aspect of bipolar disorder makes migration to this classification a default logical next action.
- The fact that the DMDD classification brings with it a less-imposing lifetime prognosis and the potential for lower stigma makes DMDD a comparatively desirable diagnosis to BD. While these issues should have nothing to do with assessing a child’s condition, in a real world devoid of biological markers, they assert tremendous pressure.
- The severe chronic irritability punctuated with rage which defines the DMDD phenotype presents with highly observable, recognizable and problematic behavior. It undermines, and comes crashing into, the child’s and family’s life with much notice. On the other hand, hypomania, which would exclude a child from a DMDD diagnosis, is much more difficult to discern and rarely prompts clinical attention as it is often perceived as normal well-being (Angst, 2003, 2007).
Identification of hypomania in children is even more challenging than its identification in adults given its overlap with developmental behaviors. It is well established that “there is a generally low parent-child agreement in reporting symptoms of mania” (Stringaris et al., in review). Parents may not be as tuned in to pick up changes from baseline, the differences that would distinguish the behavior from developmental norms, or they may even regard the changed energy state as a relief from the irritable mood (Youngstrom et al., 2008). All of these diminish the likelihood of the hypomania receiving notice.
Further, the increasing recognition of BD Spectrum Disorder (Ghaemi et al, 2002) and the importance of subsyndromal symptoms bring BD more into line with the often subsyndromal, or non-classic, presentation of BD in children (Danner et al., 2009). Judd et al. (2003) found that BD II adults spent less than 1% of their total time in syndrome-level hypomania. Given this frequency for an adult who meets DSM criteria, how likely can we expect the presentation of DSM hypomania to be present in a child with BP-NOS?
When the presence of (hypo)mania is the only thing that sends the diagnosis back in to a bipolar realm, then there will undoubtedly be many cases in which the wrong call is made. We would suggest that this call could be made more assuredly if the presence of additional important features associated with (hypo)mania were also determined. However the DMDD criteria do not provide any.
- The above criticism notwithstanding, an ideal clinical setting should be able to detect the presence of hypomania. A capable clinician using thorough diagnostic practices, given sufficient clinical opportunity, and reliable input from a person who is able to observe the child’s behavior and symptoms outside of the clinical setting, should be able to pick up the mood symptoms of hypomania and identify the simultaneous change from baseline in order to make the “right call” between DMDD and BP-NOS. But rarely is this ideal possible.
- Case loads in some of the most vulnerable population centers; schools, youth correctional facilities, and under-supported communities, are simply not able to provide the type of individual, well considered attention that would allow for such observation and analysis. In these cases we cannot help but think that the more obvious and circumstantially problematic behaviors will dictate the diagnosis.
- Even in clinical settings in which the time and the attention of the clinician are not so constrained, there is a bias against careful differentiation. In her article Dictionary of Disorder, (Spiegel, 2005) Alix Spiegel writes,
Reliability is probably lowest in the place where the most diagnoses are made: the therapist’s office. As Tom Widiger, who served as head of research for the DSM-IV, points out, “There are lots of studies which show that clinicians diagnose most of their patients with one particular disorder and really don’t systematically assess for other disorders. They have a bias in reference to the disorder that they are especially interested in treating and believe that most of their patients have.
A fresh unbiased examination of all the inclusionary and exclusionary criteria simply gives way to practical experience. Given all the above comments that push a DMDD diagnosis into the foreground, it is reasonable to be concerned that many clinicians will come to apply the DMDD diagnosis more often than would be justified.
- By not simultaneously announcing a proposal which would provide greater guidance and clarity for non-classic BD, a need which is now greater than ever given the potential for confusion with DMDD, professionals and the public alike cannot help but interpret that the Work Groups do not confer upon it equal importance to operationalization of DMDD. The vast media coverage of the DMDD proposal announcement, including a feature story on NPR’s All Things Considered and supportive comments by Dr. Alan Schatzberg, APA President in the March 5, 2010 Psychiatric News, has already affected clinical decisions and research grant requirements.
We fear that the above mentioned concerns will strongly bias the diagnosis of many mood impaired children towards DMDD. For those children who do have a bipolar condition, the effects of misdiagnosis could be tragic. The following excerpt from the Justification for DMDD (APA DSM5, 2010b) makes quite clear how divergent the treatment protocol would be for a child who receives one diagnosis or the other.
The question of whether youth with the DMDD phenotype have a form of BD, vs. a syndrome on a pathophysiological continuum with anxiety disorders, unipolar depression, and ADHD, has profound treatment implications. That is, if DMDD is a form of BD, first-line treatment would consist of atypical antipsychotic medication and/or mood stabilizers. On the other hand, if DMDD is on a continuum with unipolar depressive disorders, anxiety disorders, and ADHD, first-line treatment would consist of serotonergic reuptake inhibitor antidepressants (SSRI’s) and stimulants. Importantly, the latter two medication classes are considered relatively contraindicated in BD. Thus, nosological questions regarding DMDD have significant treatment implications, and it is essential that clinical trials to address this question be launched. (italics ours)
While this may be re-assuring for those who have confidence in the DMDD classification and the clinical capability to differentiate it from BP-NOS, it causes great worry for us. Treatment with SSRI’s and stimulants without prior mood stabilization could be catastrophic if a bipolar condition really exists and is not diagnosed (Ghaemi et al., 2003). As Faedda et al. (2004) point out:
Of 82 juvenile BD patients, 57 (69%) had been given a mood-elevating agent at least once; 33/57 (58%) so-exposed met criteria for [treatment induced episodic mania] TEM, with median latency of 14 days; TEM was observed twice as often with antidepressants as stimulants (44% vs. 18%). TEM led to first-recognition of BPD in 14 cases (17%), and some drug-exposed children (4-9%) had prominent suicidal, homicidal or psychotic behavior. In addition to recent exposure to a mood-elevating agent, TEM was associated with early-onset anxiety and female gender.
Given the “significant treatment implications”, we reiterate our request for field trials that would clarify the specific clinical and physiological characteristics of children with DMDD in comparison to those with FOH.