Prevalence Rate and Course of Symptoms of Disruptive Mood Dysregulation Disorder (DMDD)

Introduction

Disruptive Mood Dysregulation Disorder (DMDD) was introduced as a new diagnosis in the 5^th edition of the Diagnostic and Statistical Manual (DSM-5) (American Psychiatric Association, 2013). It is located in the section concerning depressive disorders. Core features of DMDD include irritable mood and temper outbursts that are grossly out of proportion in intensity and duration. Temper outbursts occur, on average, three or more times per week. Furthermore, the individual’s mood between temper outbursts is persistently irritable or angry most of the time during most days. The diagnosis is restricted to persons in the age group between 6 and 18 years, however symptom-onset should generally be established before the age of 10 years (American Psychiatric Association, 2013).

Before inclusion of DMDD in DSM-5, the co-occurrence of irritable mood and temper outbursts in children posed a major classificatory problem. Therefore, children were diagnosed with childhood bipolar disorder, leading to increased prevalence rates during the mid-1990s particularly in the United States (Blader & Carlson, 2007; Moreno et al., 2007) and to treatment with mood stabilizers and antipsychotic drugs (Rao, 2014). This, in turn, sparked controversial discussions about whether the aforementioned symptoms can be subsumed under the diagnosis of bipolar disorders, or whether there is a need for a completely new diagnostic category (Parens & Johnston, 2010) such as “severe mood dysregulation” (Leibenluft, Charney, Towbin, Bhangoo, & Pine, 2003).

Because it is a relatively new diagnostic category, little is known about the prevalence rates of DMDD in the general population as well as in clinical samples. Because of a lack of diagnostic instruments in earlier studies, the dysregulation profile (DP) of the Child Behavior Checklist (CBCL) (Althoff, Rettew, Ayer, & Hudziak, 2010) was used to assess prevalence rates of DMDD. Results indicated 6-month prevalence rates for the CBCL-DP-phenotype of 1 % for children from a nonclinical sample (Hudziak, Althoff, Derks, Faraone, & Boomsma, 2005) and of 6 % for a German child psychiatric sample (Holtmann, Goth, Woeckel, Poustka, & Boelte, 2008). Copeland et al. (2013) reported 3-month prevalence rates for DMDD between 0.8 % and 3.3 % by evaluating data from community samples of youth (2–17 years) using items from a structured psychiatric interview (Child and Adolescent Psychiatric Assessment, Preschool Age Psychiatric Assessment (PAPA)) (Copeland et al., 2013). Dougherty et al. (2014) found a 3-month prevalence rate of DMDD of 8.2 % in 6-year-olds of a community sample using the PAPA as well.

According to DSM-5, DMDD symptoms should be present for at least 1 year or longer (American Psychiatric Association, 2013), although there is little knowledge about the course of symptoms over time. Two studies revealed low rates with respect to stability of prevalence rates from childhood to adolescence (Axelson et al., 2012; Mayes et al., 2015): Only 29 % of the subjects who showed irritable-angry mood and temper outbursts during childhood revealed such symptoms during adolescence as well (Mayes et al., 2015). Axelson et al. (2012) assessed DMDD symptoms by means of a structured clinical interview in child-psychiatric outpatients aged 6–12 years. Results indicated that only 40 % of the patients who met the diagnostic criteria of DMDD at baseline fulfilled the requirements for DMDD at follow-up 2 years later. Because of the limited stability of symptoms, some authors have raised concerns about the diagnostic utility of the DMDD diagnosis (Mayes et al., 2015).

Only a few studies assessed associations between DMDD in childhood and the development of psychiatric disorders as well as psychosocial factors in later life. In a community sample, a history of DMDD in childhood and adolescence was associated with elevated rates of anxiety and depression, adverse health outcomes, being impoverished, having police contacts, and a low educational level compared to psychiatric or healthy controls in young adulthood (Dougherty et al., 2014). Furthermore Althoff et al. (2010) found that a CBCL-DP profile in children was associated with increased rates of anxiety disorders, mood disorders, and drug abuse 14 years later. Holtmann et al. (2011) studied 325 subjects from childhood to young adulthood and found that young adults who had had a CBCL-DP profile during childhood revealed an increased risk for substance use disorder, suicidality, and poorer overall functioning at age 19. Further, Baroni et al. (2009) showed that children who met criteria for severe mood dysregulation at age 10.6 revealed an elevated risk for major depressive disorder at age 18.2. Also, Stringaris et al. (2009) assessed 631 participants and found that irritability in early adolescence was a specific predictor for depressive and anxiety disorder 20 years later. Additionally, Biederman et al. (2004) found a relationship between disruptive disorder and childhood-onset psychosis. Taken together these findings suggest that early disruptive symptoms constitute a risk marker of deficit of self-regulation of affect and behavior in adulthood.

Because the concept of DMDD is relatively new, studies assessing the stability of symptoms from childhood to adulthood as well as the relationship between DMDD in childhood and the onset of psychiatric disorders as well as its influence on socioeconomic variables in later adulthood are missing. The present study investigates the point prevalence rate of DMDD symptoms among adults in a representative German nationwide sample, the prevalence rate of DMDD symptoms in primary school age retrospectively as well as its stability from primary school age to adulthood. Furthermore, we compared subjects with symptoms of DMDD with subjects without symptoms of DMDD with respect to demographic, socioeconomic variables as well as regarding depressive and anxious symptoms.

Methods

Diagnostic Instruments

Symptoms of DMDD in accordance with DSM-5, demographic and socioeconomic variables as well as symptoms of depression and anxiety were assessed by means of self-report using paper and pencil format. The assessment of DMDD was done in two blocks, consisting of six identical questions each referring to current severe temper outbursts (block 1) and severe temper outburst during primary school (block 2): (1) Participants were asked whether they were presently suffering from severe recurrent temper outbursts that manifest verbally and/or behaviorally, and that are grossly out of proportion in intensity or duration to the situation. If this question was answered with yes, the remaining five questions (2–6) of the first block were also posed (see Table 1). If they answered this question with no, they were asked to proceed with the second block with the first question referring to the existence of severe temper outbursts during primary school.

Table 1 DMDD criteria 2–6 of those that affirmed to suffer from severe temper outbursts

Time point/frame	Item	Yes		No		Not applicable		Total
		N	%	N	%	N	%	N
During primary school (N = 138)
	2) Outbursts occur three or more times per week	18	13.0	43	31.2	77	55.8	138
	3) Mood between outbursts is persistently irritable	20	14.5	41	29.7	77	55.8	138
	4) Criteria are present for at least 1 year	13	9.4	7	5.1	118	85.5	138
	5) Criteria are observable by others	17	12.3	3	2.2	118	85.5	138
	6) Criteria are present in two or more settings	10	7.2	10	7.2	118	85.5	138
	7) mania symptoms	67	48.6	71	51.4	–	–	138
	8) psychiatric disorder	41	29.7	97	70.3	–	–	138
Adulthood (N = 162)
	2) Outbursts occur three or more times per week	37	22.8	125	77.2	0	0	162
	3) Mood between outbursts is persistently irritable	51	31.5	111	68.5	0	0	162
	4) Criteria are present for at least 1 year	32	19.8	18	11.1	112	69.1	162
	5) Criteria are observable by others	44	27.2	7	4.3	111	68.5	162
	6) Criteria are present in two or more settings	38	23.5	13	8.0	111	68.5	162
	7) Mania symptoms	79	48.8	82	50.6	1	0.6	162
	8) Psychiatric disorder	65	40.1	97	59.9	–	–	162

Table 1 DMDD criteria 2–6 of those that affirmed to suffer from severe temper outbursts

View as image HTML

Symptoms of depression and anxiety were assessed by means of the 4-item Patient Health Questionnaire (PHQ-4), which is an ultrabrief self-report questionnaire consisting of two items for the assessment of depressive and anxious symptoms in accordance with DSM-IV. Two symptoms of anxiety disorder “anxiety” and “unable to stop sorrows” and two symptoms of depression “depressive mood” and “reduced interest in activities” were answered on a Likert scale ranging from 1 (not at all) to 4 (nearly every day). The PHQ-4 has previously been shown to be reliable and valid in assessing depressive and anxious symptomatology (Khubchandani, Brey, Kotecki, Kleinfelder, & Anderson, 2016; Loewe et al., 2010).

Results

Of all participants, 138 (5.72 % of total population) affirmed suffering from severe temper outbursts, and 19 (0.79 % of total population) retrospectively reported symptoms of DMDD during primary school (see Table 1, upper part for DMDD criteria 2–6). Of the latter, 12 (63.3 %) also reported symptoms of DMDD during adulthood. Of all participants who reported symptoms of DMDD during primary school age, 6 (31.60 %) also fulfilled criteria for elevated symptoms of mania during primary school, and 17 (89.29 %) were diagnosed with one (N = 4) or more (N = 13) lifetime psychiatric disorders (depression: N = 10; posttraumatic stress disorder: N = 1; bipolar disorder: N = 1; autism: N = 1; anxiety disorder: N = 9; ADHD: N = 7; conduct disorder: N = 4) according to self-report.

Out of all participants, 162 reported severe and recurrent temper outbursts (6.71 %) in adulthood. In turn, 12 participants (0.50 % of the total population) also fulfilled criteria 2–6 for DMDD (see Table 1 lower part). Thereof, 8 (66.67 %) also affirmed that symptoms of mania and 11 (91.74 %) had been diagnosed with one (N = 1) or more (N = 10) lifetime psychiatric disorders (depression: N = 9; bipolar disorder: N = 1; autism: N = 1; anxiety disorder: N = 5; ADHD: N = 4; conduct disorder: N = 3) according to self-report. Of those who reported symptoms of DMDD during adulthood (N = 12), all reported symptoms of DMDD during primary school as well.

With respect to demographic and socioeconomic variables, subjects reporting symptoms of DMDD during primary school age were more often single or divorced, had fewer own children, and were more frequently unemployed and diagnosed with a lifetime psychiatric disorder than subjects without symptoms of lifetime DMDD (see Table 2). Subjects who revealed symptoms of DMDD during adulthood had also been diagnosed more frequently with a lifetime psychiatric disorder than the non-DMDD group. Because of low cell numbers, results could not be calculated with respect to number of own children and unemployment for the adult DMDD population.

Table 2 Demographic and socioeconomic variables

	DMDD during primary school		DMDD during adulthood		No DMDD, neither during adulthood nor during primary school		DMDD during adulthood versus no DMDD
Demographic and socioeconomic variables	N	%	N	%	N (missing)	%	Significance	Significance
Total	19	100	121	100	2394	100
Female	10	52.6	5	41.7	1335	55.8	p > .05	p > .05
Married	2	10.5	2	16.7	1091	45.7	–	–
Single, divorced	15	78.9	8	66.7	979 (9)	40.9	p = .002	p = .10
Widowed	2	10.5	2	16.7	262	11.0	–	–
Own children	7	36.9	4	33.3	1593	66.5	p = .01	–
German nationality	17	89.5	11	91.7	2312	96.6	p = .09	p = .35
Religious	14	73.7	9	75	677	28.3	p = .87	p = .81
Student	1	5.3	–	–	3	0.1	–	–
No graduation	1	5.3	1	8.3	58	2.4	–	–
“Hauptschule”	7	36.8	5	41.7	774	32.3	p > .05	p > .05
“Realschule”/“POS”	8	42.1	6	50	982	41	p > .05	p > .05
“Gymnasium”	2	10.5	–	–	286	11.9	–	–
University degree	–	–	–	–	223	9.3	–	–
Unemployed	6	31.6	4	33.3	127	5.3	p < .001	–
Low-income earner2	9	47.4	5	41.6	831	33.2	p > .05	p > .05
Ever been diagnosed with PD3	17	89.3	11	91.7	326	13.6	p < .001	p < .001
Currently suffering from depression4	9	47.4	6	50.0	114	4.8	p < .001	p < .001
Currently suffering from anxiety4	9	47.4	7	58.3	124	5.2	p < .001	p < .001

Table 2 Demographic and socioeconomic variables

View as image HTML

Subjects with both symptoms of DMDD during primary school age, χ²(1) = 69.40, p < .001, and during adulthood, χ²(1) = 49.33, p < .001, suffered significantly more often from depression compared to non-DMDD subjects. The same was true concerning anxiety when subjects with symptoms of DMDD during primary school age, χ²(1) = 61.93, p < .001, and during adulthood, χ²(1) = 62.23, p < .001, were compared with non-DMDD subjects (for more details see Table 2, last two rows). On the symptom level, subjects with DMDD during primary school age and/or adulthood reported the symptoms “depressed mood,” “reduced interest in activities,” “anxiety,” and “being unable to stop sorrows” more often than subjects of the non-DMDD group (see Table 3).

Table 3 Comparison of symptoms of depression and anxiety between DMDD group in primary school and adulthood versus no DMDD group

	DMDD group in primary school	DMDD group in adulthood	No DMDD group	DMDD during primary school versus no DMDD		DMDD during adulthood versus no DMDD
Symptoms	M (SD)	M (SD)	M (SD)	t	p	t	p
Anxiety	2.53 (1.12)	2.67 (1.07)	1.42 (.61)	4.26	< .001	3.99	.002
Unable to stop sorrows	2.37 (1.10)	2.50 (1.09)	1.27 (.55)	4.72	< .001	3.91	.002
Reduced interest in activities	2.47 (1.07)	2.50 (1.00)	1.38 (.61)	4.43	< .001	3.86	.003
Depressive mood	2.47 (1.02)	2.50 (1.00)	1.26 (.54)	5.17	< .001	4.37	.001
Note. Symptoms can be answered on a Likert scale ranging from 1 = not at all to 4 = nearly every day.

Table 3 Comparison of symptoms of depression and anxiety between DMDD group in primary school and adulthood versus no DMDD group

View as image HTML

Discussion

We found a prevalence rate for elevated DMDD symptoms during primary school age of 0.79 % by retrospective self-report. This is in line with findings from Copeland et al. (2013), who reported a prevalence rate for DMDD in children and adolescents of approximately 1 % when applying the full DMDD criteria (Copeland et al., 2013). Contrary to those findings, Mayes et al. (2015) found prevalence rates of 9.3 % in 6–12-year-olds, and Dougherty et al. (2014) found a 3-month prevalence rate of 8.2 % in 6-year-old children. The fact that DMDD was assessed in the study of Mayes et al. (2015) solely by means of the two major symptoms of DMDD, and Dougherty et al.’s (2014) study focused on children that were 6 years old only might explain comparatively high prevalence rates, compared to the ones of the present study. However, we assessed prevalence rates during primary school age, and it has been shown that prevalence rates seem to decrease as children grow older (from a 3-month-prevalence rate of 3.3 % in preschoolers to a prevalence rate of 0.8 % in 9–17-year-olds) (Copeland et al., 2013).

Our results indicate a prevalence rate for elevated DMDD symptoms during adulthood of 0.50 %. To the best of our knowledge, no previous study has assessed prevalence rates of elevated DMDD symptoms in adults, probably because of the fact that DSM-5 limits the diagnosis of DMDD to the age group below 18 years. The lower prevalence rate in our sample could also be explained by rigorous use of DSM-5 diagnostic criteria. Althoff et al. (2016) showed that, depending on a broader or more narrow definition of DMDD, prevalence rates vary between 0.12–5.26 %. Therefore, the rate found in our study lies within the range reported from the US National Comorbidity Survey – Adolescent Supplement. Thus, rates from the US and Germany seem to be comparable.

Of those who reported elevated DMDD symptoms during adulthood, all retrospectively reported elevated symptoms of DMDD during primary school age as well, and of those who reported elevated symptoms of DMDD during primary school age, 63.2 % reported elevated symptoms of DMDD during adulthood as well.

Only a few studies have assessed the stability of DMDD symptoms over time. In the study of Mayes et al. (2015), mothers rated DMDD symptoms of their 6–12-year-old children (baseline) and again 8 years later (follow-up). Results revealed that of those who fulfilled criteria at baseline, 29 % still had DMDD symptoms at follow-up. Of those who fulfilled criteria at follow-up, 55 % retrospectively fulfilled criteria at baseline (Mayes et al., 2015). Axelson et al. (2012) found that, of those with DMDD symptoms at baseline (6–12 years), only 53 % continued to have DMDD 12 months later, and only 19 % both 12 and 24 months later. The instability of DMDD symptoms discovered by Mayes et al. (2015) and Axelson et al. (2012) contrast the relatively stable findings in the study at hand. This might be explained by the fact that we assessed the stability of symptoms over a longer timeframe than other authors did (decades versus years). Furthermore, recall bias might play a role, in that sense that subjects currently suffering from severe temper outbursts might remember temper outbursts during primary school age better than subjects who do not suffer from temper outburst today.

The DSM-5 states that symptoms of DMDD often exist next to symptoms of other disorders. With respect to our findings, 89.3 % and 91.7 % of those subjects who reported elevated symptoms of DMDD during primary school age and adulthood, respectively, were diagnosed with one or more lifetime psychiatric disorder(s). Those high comorbidity rates are in line with the findings of Mayes et al. (2015), who noted that only one child with DMDD (0.4 %) fulfilled no further criteria for other psychiatric disorder(s). Subjects of the present study with elevated symptoms of DMDD during primary school compared to subjects without DMDD showed elevated symptoms of depression and anxiety disorder today – findings also reported by others (Althoff et al., 2010; Copeland, Shanahan, Egger, Angold, & Costello, 2014; Stringaris et al., 2009). This underlines the validity of including DMDD in the affective disorder section of the DSM-5. However Axelson et al. (2012) did not describe elevated rates of depression and anxiety in DMDD patients versus no DMDD controls in their study. This might be explained by the young age of the children at both the baseline (6–12 years) as well as the follow-up measurement (1 and 2 years later, respectively).

One explanation for high numbers of lifetime prevalence rates of further psychiatric disorders besides DMDD is that the occurrence of DMDD during childhood might pose a risk factor for the development of comorbid psychiatric disorders later on, as Masi et al. (2015) showed. On the other hand, high comorbidity rates might question the validity of DMDD as a unique and independent diagnosis, and questions whether the fact that DMDD symptoms often co-occur or even overlap with other disorders should be emphasized (e. g., in the sense of a specifier) until more is known about the validity of the diagnosis. Especially the overlap with oppositional-defiant disorders (ODD) seems to be high. Therefore, it may be a valid approach to discuss the addition of chronic irritability and anger as a specifier for ODD than establishing a new diagnostic entity of DMDD (Birkle, Legenbauer, Grasmann, & Holtmann, 2017).

In accordance with our results, adults reporting elevated symptoms of DMDD in primary school age were more often single or divorced, had less often children, and were more often unemployed in adulthood than subjects not reporting these symptoms. Also, Copeland et al. (2013) reported that youths with DMDD experienced higher levels of social impairments and elevated levels of suspension probably affecting their occupational perspectives in the long run than either psychiatric or non-case controls. Alatupa et al. (2013) found associations between high childhood aggression and low educational level in adulthood and between poor childhood social adjustment and low occupational status in adulthood, supporting our data. However, our findings might also be partially explained by the fact that subjects from the DMDD groups were of a lower mean age than subjects without DMDD. Therefore, future studies should investigate the influence of DMDD on demographic aspects in a prospective manner.

Because of the lifelong impact of DMDD on social functioning, there is a great need for its effective treatment. Existing psychosocial treatments are effective only by improving irritability but not by enhancing mood (Waxmonsky et al., 2016). A review of the medical treatment of DMDD concluded that clinical trials are warranted to identify effective and safe treatment modalities (Tourian et al., 2015).

Several limitations must be discussed in the interpretation of the present data. First, concerning the assessment of DMDD, we simply asked whether patients fulfilled symptoms according to the DSM-5. Yet the psychometric properties of this approach have not been determined before. Second, because subjects were asked retrospectively about their DMDD symptoms during primary school age, their information might have been influenced by recall bias. For example, subjects still suffering from DMDD-related symptoms in adulthood might remember similar difficulties in childhood more precisely than subjects who had already overcome those symptoms. This might overestimate our relatively high chronicity rate and could explain our comparatively low prevalence rate of reported DMDD symptoms in childhood. Interestingly, subjects who reported elevated symptoms of DMDD were younger than those who did not report symptoms of DMDD, which might be due to recall bias increasing with age (Takayanagi et al., 2015). Third, diagnosing DMDD in adults is not based on a solid validity since diagnostic criteria in accordance with DSM-5 were constituted for childhood only. This approach was chosen in order to judge the persistence of symptomatology. Fourth, third parties might be better informants than the patients themselves in regards of assessing externalizing symptoms (Stieglitz, 2008). Therefore, additional information might have been valuable with respect to DMDD symptoms from third parties (e. g., partners or parents). Fifth, in accordance with DSM-5, the exclusion criteria (a) existence of mania or (b) comorbid psychiatric disorder should normally be taken into account. For this reason, participants were asked whether they had ever fulfilled symptoms of mania/hypomania and whether DMDD symptoms could be better explained by a different psychiatric disorder – and whether such a mental disorder has ever been diagnosed by a psychiatrist, psychologist, or psychotherapist. If we take both exclusion criteria for DMDD into account, prevalence rates are drastically reduced in both samples to 0.04 % each. It should be pointed out, however, that (1) the question about the existence of psychiatric disorders refers only to lifetime diagnoses (and might have not been present during the referential diagnostic time frame of primary school age or within the last year), (2) some symptoms of DMDD are difficult to distinguish from other psychiatric disorders, and (3) the exploration of symptoms of mania (during primary school age or the last year) consisted of only one question.

Conclusions

In light of the limitations of a retrospective analysis, our results still revealed a strong chronicity of DMDD symptoms persisting from childhood to adulthood. This questions the restriction of the diagnosis to minors only and emphasizes the need for further – ideally prospective – investigations on DMDD symptoms in adult age. Consequently, specific diagnostic tools to assess DMDD symptoms in adults still have to be developed.

Because DMDD is accompanied by further comorbid psychiatric disorders, especially depression and anxiety and seems to affect familial and occupational status in adulthood, early detection and subsequent effective treatment of DMDD symptoms are paramount.

Conflicts of interests: None