|Home | About | Journals | Submit | Contact Us | Français|
“Jordan”, who is 8 years old, has been referred psychiatrically because of his restless, distractible, impulsive, unpredictable, oppositional and defiant behavior. He has trouble following directions; sometimes he doesn’t wish to do the task, other times he doesn’t seem to understand it. His frequent frustration often results in prolonged outbursts at home. In school, he has to be removed from class because his behavior becomes dangerous. When he must be restrained by his parents or classroom aide, he becomes even more agitated.
Temper outbursts, sometimes called “rages”, are a major reason for outpatient and inpatient referral. These behaviors have also been a focus of assessment in child psychology and psychiatry since rating scales were developed. In fact, items consistently recorded on the same factors in frequently used behavior rating scales for children reflect negative mood (mood changes quickly/explosive, easily angered/stubborn, sullen, irritable), oppositionality (being demanding, uncooperative and disobedient) and aggression (argumentativeness, temper tantrums) 1.
Although the dimensions of disagreeable mood/oppositionality/aggression have been long recognized, the field has had difficulty agreeing on an appropriate diagnostic label. Rages have been associated with extreme irritability or mania 2, Tourettes Disorder 3, intermittent explosive disorder and conduct disorder 4–5, autism/Aspergers disorder 6 and other conditions. They are part of a syndrome termed severe mood dysregulation [SMD7] defined by markedly increased and frequent reactivity to negative emotional stimuli (e.g. response to frustration with extended temper tantrums, verbal rage and/or aggression toward people or property) that occurs at least 3 times/week in the context of chronic anger or sadness. This can co-occur with other symptoms of e.g., ADHD or anxiety (i.e. hyperarousal, distractibility, rapid speech/racing thoughts, insomnia).
The term “rages” implies that these outbursts consist solely of high intensity anger, but we actually know little about their content or structure. It has been suggested that outburst content and/or structure might vary with psychiatric conditions like oppositional defiant disorder (ODD), depression and mania 2.
Therapeutically, when strategies to avoid rage outbursts fail, safety of the child and surroundings often requires selecting among alternatives such as seclusion, restraint and/or medication, each with its own advantages and disadvantages 8. Moreover, medication may need to address the diagnostic condition that is felt to underlie the rage or reduce the arousal/agitation of the rage itself (e.g. by sedation)9.
Our team at Stony Brook has studied rages by direct observation (rather than asking a parent about them), attending to the duration of outbursts, the kinds of behaviors that occur in them, and if they might be diagnostically specific 10. We have examined rages on a children’s psychiatric inpatient service, which cares for children from ages 4–12, because these outbursts precipitate admission in over half the children. In addition, since the unit functions as both a short and intermediate term facility, the median length of stay of about a month gives ample opportunity to observe behavior. Finally, because of our long-standing interest in bipolar disorder, and because rages have become synonymous with bipolar disorder in some views 2, we are interested in whether rages reflect a manic rapid cycle and/or occur disproportionately in children with mania.
In order to understand rages better, we collected diagnostic and observational data over 18 months on 130 children (age 9.7 + 2.1), one fifth of whom were female, for 151 hospitalizations 11. Examining first admissions of 130 children, we found 71 (54.6%) were admitted for rages though only half (n=37, 52.1%) subsequently had a rage outburst while hospitalized. (Seven other children had rages, too, but had been hospitalized for other reasons.) Of the 44 children with rages observed in hospital, 23 had just the one; the remaining 21 had 2 to 9 outbursts. In other words, half of the children with serious dysregulation at home or school could maintain self control on a structured unit with clear expectations, positive support and/or when they were out of a very stressful environment. No significant gender differences were found.
A rage was deemed to have started when the child became loudly verbally defiant and out of control when asked to do, or stop doing, something by staff. Rages were then observed at 5, 15, 30, 45, 60, 90 and 120 minutes after onset by nursing staff trained to identify as present/absent behaviors on a list of the most commonly observed and operationally defined outburst constituents. Behaviors coded during each rage included verbal acts (whining, verbal threats, cursing, yelling/screaming), discrete physical acts (stamping, pushing/pulling, throwing things, biting/scratching, punching the wall, hitting others, kicking others or objects), and expressive, “psychomotor” behaviors (looking tearful/sad, anxious/fearful, or withdrawn/unresponsive). Notably, manic symptoms were never observed.
Over 18 months of data collection, there were 117 outbursts in 49 hospitalizations. Of the coded behaviors observed, anger behaviors (yelling/screaming/cursing/violent threats and stamping/kicking/hitting/throwing objects) were the most common, occurring in 93% of rages; being tearful/sad or anxious/fearful behavior occurred in less than half (46%) of rages. A rage generally lasted over 45 minutes but varied widely in duration with 19% of outbursts lasting less than 30 minutes, and 19% lasting more than 60 minutes. Rage outbursts usually occurred early in the course of hospitalization. For instance, 44% of children with outbursts had their first or only one within 2 days of admission. The remainder was between 2 days and discharge. Half of the children with rages after admission, then, seemed to “test the waters” and quickly desisted. In addition to the question of whether outburst content and/or structure might vary with psychiatric condition, we were mindful of frequent parental comments to the effect that their rage-prone child had “never outgrown the terrible twos”. This raised the hypothesis that rages might resemble childhood temper tantrums. A set of factor analyses identified 5 groups of behaviors: three were interpretable as progressive levels of anger intensity; two others as levels of “distress” (i.e., sadness/anxiety)11. This model of outburst organization accounted for 54% of total variance. An independent cluster analysis of behavior slope (change over the rage) revealed that anger behaviors peaked early and declined relatively rapidly. Distress behaviors were more evenly distributed. Thus, rages closely resembled the observed tantrums of typically-developing preschoolers with respect to behavior types and time course (early peaking anger, more evenly distributed distress)12,13. However, in terms of severity and duration, they were like parent-reported tantrums of preschoolers with depression and disruptive behavior disorders which lasted more than 20 minutes 14. Rages of our inpatients lasted at least twice as long, and because the children were bigger, were more destructive.
How did children with observed rage outbursts in hospital differ from their hospitalized peers? Younger children were marginally more likely to have rages (r = −0.155.) IQ, history of abuse, and living status (with/without a parent) had no significant effect. Best estimate diagnoses encompassing parent and school history, child mental status and hospital course were made after discharge. Considering single psychiatric diagnoses, children having outbursts were almost 5 times more likely to have a discharge diagnosis of ADHD, more than 5 times more likely to have a learning or language disorder but were 3 times less likely to have an anxiety disorder than children without rages in hospital. However, children with rages were almost 3 times more likely to have 3 or more concurrent diagnoses than those without. The most common combination was ADHD, oppositional defiant disorder, and learning/language disorder, seen in 70% of children with rages, vs. 34% of children without. However, when overlaps between clinical features were readdressed by ordinal regression of number of outbursts (categorized as 0, 1, or > 1) only two features significantly predicted in-hospital rages: a prehospitalization history of outbursts and, secondarily, a learning/language disorder. All other psychiatric diagnoses dropped out. The association between language disorder and outbursts is consistent with the more general connection between language and behavior 15; however it is almost never studied in research based solely on structured interviews.
Given the interest in rages as a symptom of mania and SMD 2,16, we examined the frequency of those conditions based both on referral information and subsequent best estimate diagnosis for all hospitalizations (n=151) because a child might have been admitted for mania on one admission but not another. Children with rages in hospital had, in fact, been referred with a diagnosis of mania more often than those without (34.7% Vs 15.7%, OR 2.8, CI 1.3, 6.3). However, of the 33 admissions where a referral diagnosis of mania had been given, it was observed and confirmed in only 5 children in hospital. Only 9% of 44 children with rages had observed mania or manic symptoms (i.e. bipolar NOS), compared to 4.7% of 86 without, and always with comorbid other disorders. Current mania, at least as defined by DSM IV, then, did not account for children whose rages occurred in hospital, nor were their outbursts examples of rapid cycles, at least as defined by symptoms of mania.
We also tried to determine how many children might have met criteria for SMD. Two thirds (65.9%) of children with rages were defined as having this condition based on behavior at home, compared to only one quarter (25.6) of children without rages in hospital, (O.R. 5.6, CI 2.55, 12.39). One child had rages 3 times a week once hospitalized. He was diagnosed with childhood-onset schizophrenia.
A published treatment algorithm has compiled the evidence base for treatment of ADHD and aggression 17, basically evidence-based ADHD medication and behavior modification. If that doesn’t suffice, an atypical antipsychotic is added, then lithium or divalproex. While treatment makes some impact, the often unstated result is that children are rarely normalized. For instance, in the Multimodal Treatment of ADHD study (MTA), rigorous treatment with stimulant medication with or without behavior modification significantly improved behavior in children with comorbid ADHD and children with the CBCL “bipolar” phenotype (T scores >67 on the ADHD, aggression, anxiety/depression scales), but at the end of 14 months, children were still considerably more impaired than those with uncomplicated ADHD 18. A study of outpatient children with ADHD in a summer treatment program found that the combination of stimulant medication and rigorous behavior modification reduced symptoms on the Young Mania Rating Scale (YMRS19) from 23.7 ± 3.5 to 15.4 ± 6.1 20. These data specifically addressing comorbidity suggest that the combination of ADHD and irritability/mood symptoms improves but doesn’t completely remit with strategies directed at ADHD. Specifically, the two YMRS aggression and irritability items dropped almost 45%. ADHD symptoms improved, but less so. Four “manic” symptoms (elevated mood, sexual interest, sleep, and thought disorder) decreased from mild to even less impairing. There is no evidence that children with SMD respond to lithium alone 21.
Children in our “rages” sample had received the gamut of treatment as outpatients including ADHD treatments, atypical antipsychotics, and mood stabilizers. Also, 88% of children with rage outbursts had been in special education; almost half had been hospitalized previously.
Occurrence of rages added at least 2 weeks to length of stay (median 36 versus 22 days with and without rages); number of rages correlated positively with length of stay (r=0.32, p<0.001). Most children improved with a combination of behavior modification, family treatment, appropriate academic intervention, and medication.
The thrust of non-medication inpatient treatment is teaching the child self-control, and helping parents learn to respond to the child appropriately – basically a behavior modification approach 22. Hospitalized children earn points toward fun activities and increasing time home. Especially for children with a history of rages, learning to take a good “time out” (i.e. “time out from positive reinforcement” which requires that the child sit in a chair quietly for 10 minutes and subsequently talking about alternative ways of better handling whatever issue got them there) 23 is central. As noted earlier, many children respond to the unit structure and “time out” from the time of admission. Staff observe children starting to lose control and will teach them to “chill out” (i.e. take their own time out) before total control is lost. This places self-control in the child’s hands.
Children who are unable to take a “time-out” are escorted to the quiet room where the door remains open (as long as they stay in the room and don’t try to hurt themselves) and a nurse observes unobtrusively, until the child has remained quiet for 10 minutes and can subsequently talk about alternatives. The availability of this option has cut the use of closed seclusion and physical restraint dramatically.
Some children, in spite of being offered these less restrictive options, simply cannot calm themselves down and require immediate medication intervention over and above whatever medication they are taking. An oral alternative (e.g. liquid risperidone or a rapidly dissolving oral atypical antipsychotic) is offered first 24. An injection of diphenhydramine is used in those instances when oral medication is unsuccessful. Even then, children seem to learn from experience though they usually need repeated experiences.
Finally, there is a small group of children who, when angry or distressed, become so unreachable, that all corrective interventions have proven to be unsuccessful. Alternate strategies (like collaborative problem solving 25) might provide an alternative but have not been used systematically. These youngsters present the most management difficulty, require the longest lengths of stay, and, depending on their psychosocial circumstances, require out of home placement.
Work with parents to maintain the child’s gains begins immediately on admission. To enlist parents’ cooperation and reduce their defensiveness, the therapist helps them understand that while they didn’t create the child’s problems, these problems nevertheless require a particular approach. For parents who feel sorry for the child and think any form of consequence is unfair, the social learning paradigm (which underlies behavior modification) 26 is reframed so that behavior that causes failure requires a negative consequence; behavior that ensures success earns a positive consequence. Most parents want their child to succeed.
Parents are also taught shortly after admission how to use the “time out” procedure 26 when the child is not following directions after warnings, or for verbal and physical aggression. In optimum circumstances (which sometimes requires a length of stay prohibited by managed care), parents/caregivers learn the paradigm and can execute a “time out” before taking the child off the unit for a pass and certainly before discharge. Like the children themselves, parents have different learning curves in understanding and consistently responding appropriately to their child.
Time on therapeutic passes provides plenty of opportunity for parents to practice skills with the option of bringing the child back from pass if s/he is unable to “time out”. This step is vital in supporting parents’ obtaining the child’s cooperation. Parents and children are instructed in the basic premise of “I will meet your needs when you meet my expectations”. If the child refuses to time out, parents are taught not to give in until the child does as expected. A confident, reasonable and well trained parent ultimately helps the child over the hurdles after discharge, though some children will need a re-admission before s/he truly understands that coercing caregivers into capitulating is not an option.
Discharge planning for children included in our study included smaller, special education class placements (for 75%), and 6 needed residential placement. With regard to medication, 90% of children with rages received either or both an ADHD medication or an atypical antipsychotic, and were 4 times more likely to receive both than children without rages.
In summary, rage behaviors may represent a disease modifier (like psychosis) that complicates a number of disorders. These behaviors clearly need a precise and accurate label in order to be studied further. To date, treatment requires combined medical and nonmedical strategies. Although the majority of children stopped having rages once hospitalized, or perhaps had only one, children who had continued outburst episodes remained difficult to treat. Conceptualizing children with rages as having mania or bipolar disorder was not consistent with inpatient observations. That is, few children were observed in an episode of mania. What was observed was severe language processing problems that needed to be understood by teachers and caregivers. Multidisciplinary treatment was partially successful for the majority of children, but children continued to need many services and polypharmacy. Successful parenting was necessary but rarely sufficient in and of itself to manage rage behaviors. Some children required readmission and sometimes residential placement. For a condition that is as common and disabling as rage outbursts, a more consistently successful series of options is sorely needed.
Gabrielle A. Carlson, Professor of Psychiatry and Pediatrics, Director, Child and Adolescent Psychiatry, Stony Brook University School of Medicine, Putnam Hall-SUNY Stony Brook, Stony Brook, NY 11794-8790, Phone 631-632-8840, Fax 631-632-8953, Email: ude.koorBynotS@noslraC.elleirbaG.
Michael Potegal, University of Minnesota, School of Medicine.
Paul J. Grover, Stony Brook University Hospital.