Treatment of oppositional defiant and conduct disorders

INTRODUCTION — Oppositional defiant disorder and/or conduct disorder (ODD/CD) are syndromes defined by patterns of behaviors that lead to conflict with adults and/or peers and are associated with impairment in home, school, and/or community settings [1]. Although both ODD/CD can be diagnosed in adults, symptoms typically begin during childhood or adolescence.

The symptoms of ODD include chronic irritability (eg, low frustration tolerance, frequent temper outbursts) and behavior problems (eg, persistent defiance and obstinance beyond what it is expected for the developmental age of the individual). Symptoms of ODD typically first appear during the preschool and childhood years and may precede the development of CD [2].

The symptoms of CD include severe chronic behavior problems including insensitivity to other people’s needs or feelings. Typical behaviors associated with CD include serious rule violations and infringing on the rights of others (eg, theft, truancy, physical fights). Symptoms of CD most often begin during adolescence and may precede a diagnosis of antisocial personality disorder in adulthood [2,3]. More serious symptoms often appear between middle childhood and middle adolescence. Adolescents with severe CD often encounter the juvenile justice system and might receive treatment during and/or following a stay at a residential treatment facility or juvenile detention center.

Treatment for ODD/CD as well as subthreshold symptoms of the disorders that cause psychosocial disruption in childhood or adolescence are reviewed here and in algorithms (algorithm 1 and algorithm 2). Clinical manifestations, diagnosis, and other aspects of ODD/CD are reviewed separately. (See "Conduct disorder: Epidemiology, clinical manifestations, course, and diagnosis" and "Oppositional defiant disorder: Epidemiology, clinical manifestations, course, and diagnosis".)

ESTABLISHING TREATMENT — We prefer to treat all children or adolescents that present with or are referred for behavior or conduct problems that cause psychosocial or academic distress or impairment. We do this whether they meet full criteria for either disorder. Treatment outcome trials often include participants who meet diagnostic criteria for either disorder or have subthreshold symptoms.

Setting goals — We work closely with the patient, referral source, and family to individualize treatment goals. The overarching goal for treatment of children and adolescents with diagnosed or subsyndromal oppositional defiant disorder and/or conduct disorder (ODD/CD) is to help develop alternative, more adaptive behaviors to cope with stressors or stressful situations, increase frustration tolerance, or decrease severity of temper outbursts.

In the treatment of ODD or its symptoms, we focus on diminishing a persistent and sometimes severe irritable mood.

In the treatment of CD or its symptoms, we focus on diminishing maladaptive behaviors and developing empathy.

We address comorbid disorders such as mood disorders or substance use disorders concurrently. Lack of identification or treatment of comorbid disorders may interfere with the treatment of ODD/CD or their symptoms or worsen their prognosis.

Children or adolescents with either of these disorders may not be seeking treatment. In many cases, they are referred by schools, teachers, social services, or the court system.

Maintaining an alliance — We maintain and enhance our therapeutic alliance through continuous support of the individual and family, offering easy availability of services especially during periods of crisis, education about symptoms and symptom management, nonjudgmental listening, medication when needed, and referral to outside community support as appropriate.

As improvements may be difficult to measure and may not be seen for several weeks or longer maintaining a strong alliance may encourage adherence to treatment. Poor alliance with the patient and family often leads to early termination. (See "Overview of the therapeutic relationship in psychiatric practice".)

Monitor progress — We use the Vanderbilt Assessment scale to monitor response to treatment [4]. We monitor treatment progress at regular intervals (eg, monthly) to track clinical change and guide treatment planning [5,6].

We consider a clinically meaningful reduction in symptoms as reported by parents, relatives, teachers, or patient report (in those with good insight) to indicate response to treatment.

PSYCHOSOCIAL INTERVENTION AS FIRST LINE — Our preference is to treat most children or adolescents with symptoms of oppositional defiant disorder and/or conduct disorder (ODD/CD) with an evidence-based psychosocial intervention as first-line treatment (algorithm 1) [7].

Modality and setting — Our choice of setting or treatment modality is based on the availability of the particular treatment and patient preference.

Psychosocial interventions for ODD/CD are routinely delivered in both individual, family, and group formats and in a wide variety of settings including traditional outpatient clinics, homes, schools, and primary care settings. Data from research indicate that evidence-based psychosocial interventions are at least modestly effective for children with disruptive behavioral problems regardless of treatment modality and setting [8-10].

As an example, a meta-analysis of 158 studies investigating psychosocial treatments to reduce conduct problems in children and adolescents reported an overall effect size (standardized mean difference at posttreatment = 0.46) between treatment and control conditions [11]. These differences were generally maintained over follow-up periods of variable lengths (average 44 weeks).

Group settings — Psychosocial treatments for ODD/CD can be effectively delivered in group settings [12-15]. Additionally, behavioral parent training programs can be delivered in group sessions with parents or caregivers. (See 'For preschool-age children' below.) As examples:

●In a review of 13 trials including 1078 participants of group parenting programs for conduct problems in children (age 3 to 12), reductions in child behavior problems were reported by parents (standardized mean difference -0.53, 95% CI -0.72 to -0.34) or independent assessment (standardized mean difference -0.44, 95% CI -0.77 to -0.11) for those subjects in the intervention group as compared with control condition including waitlist, treatment as usual, and no treatment [12].

●In a randomized clinical trial conducted in 11 pediatric practices, 150 parents of children age two to four with disruptive behavior problems were randomly assigned to receive a 10-week parent-training group or to a waitlist condition [8]. Greater improvement in child disruptive behavior was reported at treatment end (standardized mean difference -0.43, 95% CI -0.79 to -0.06) in parent training group as compared with the waitlist condition. Additional improvements in other measures such as negative parenting and negative parent-child interactions were reported.

Concerns have been raised over possible iatrogenic effects of group interventions for adolescents with delinquent behaviors. The perspective is that delinquent youth might learn from and/or reinforce one another’s delinquent behaviors through a process termed “deviancy training” [16]. However, the empirical evidence for such iatrogenic affects is weak and inconsistent. In a systematic review, only 1 of 18 meta-analytics tests supported the presence of iatrogenic effects [17].

Residential treatments — We do not routinely treat individuals with ODD/CD in residential treatment facilities as a first-line treatment. Treatment in a residential facility requires separating the child from their family and is extremely costly. We consider residential care to be a treatment of last resort when therapeutic services cannot be provided in a family/outpatient environment. This recommendation is consistent with a 2014 consensus statement by the American Orthopsychiatric Association [18].

Choosing initial psychosocial intervention — Our preference the specific first-line psychosocial intervention depends on the age of the child and the symptoms to be addressed. Effective treatments are typically skills-focused and usually involve the participation of both the child and parent. These treatments have overlap in content and technique and can be categorized into behavioral parent management training (for parents), cognitive-behavioral skills (for children) and systemic family therapy. (See 'For preschool-age children' below and 'For school-age children (age 6 to 12)' below and 'Moderate or severe symptoms' below.)

In general, the evidence points to greater efficacy for interventions with younger children (eg, preadolescent) compared with older children [19-21].

For preschool-age children — For preschool-age children with disruptive behavior problems or other symptoms of ODD/CD, we suggest treatment with behavioral parent management training for parents. One form of behavioral parent management training is parent-child interaction training. (See "Attention deficit hyperactivity disorder in children and adolescents: Overview of treatment and prognosis", section on 'Psychosocial interventions' and "Psychotherapy for anxiety disorders in children and adolescents".)

Behavioral parent management training — Behavioral parent management training is a broad category of treatment that involves teaching parents skills and techniques such as labeled praise, positive attention, tangible rewards, effective commands, response cost, time-outs, problem-solving, loss of privileges, consequences, and reward systems. This treatment has the most evidence for efficacy compared with other psychosocial interventions for ODD/CD and subthreshold disruptive behavior problems, particularly in younger children (preschool and school-age children) [7,8,12,19]. As an example, in a meta-analysis including 36 trials and 3042 children with disruptive behavior problems, treatment with behavior parent management training programs led to robust effects on reducing disruptive behavior problems as compared with control condition [19].

For school-age children (age 6 to 12) — For school-age children (age 6 to 12) with disruptive behavior problems or other symptoms of ODD/CD, we suggest first-line treatment with a combination of behavioral parent management training plus cognitive-behavioral skills training for youth. (See 'Behavioral parent management training' above.)

Cognitive-behavioral skills training for youth — Cognitive-behavioral skills training for youths involves teaching a variety of cognitive and/or behavioral skills to children and adolescents (often in groups) with conduct problems. These skills training programs focus on one or more skill deficits that are believed to be contributing to the pathogenesis and maintenance of the conduct problems. Skill deficits that are frequently targeted include social skills, anger coping skills, and problem-solving skills. Cognitive-behavioral skills training for youth has been studied as part of multicomponent treatment only. Our recommendation is based on clinical experience.

For adolescents (age 13 to 17) — Parenting interventions may be less effective for older children than younger children with symptoms of ODD/CD [21].

Mild symptoms — We typically address mild symptoms (eg, symptoms that cause only minor disruption in psychosocial functioning) of ODD/CD in adolescents with a combination of behavioral parent management training plus cognitive-behavioral skills training for youth. (See 'For preschool-age children' above and 'For school-age children (age 6 to 12)' above.)

Moderate or severe symptoms — We typically address moderate to severe symptoms of ODD/CD in adolescents with a systemic therapy. One type of systemic therapy we commonly use is multisystemic therapy.

Systemic therapy (including family therapy) — We involve family members in the treatment of children and adolescents with problematic conduct. We believe that the most important social context for children and adolescents is the immediate family. Involvement of family is essential for the successful management of children with problematic behaviors.

Systemic therapy is a broad term for psychotherapies that view the social context as key to understanding a patient’s problems and the focus of intervention. A review of 47 trials involving youths with behavior problems including ODD/CD found systemic therapy to be efficacious in 42 of the trials [22]. (See "Substance use disorder in adolescents: Psychosocial management", section on 'Family-based treatment'.)

●Multisystemic therapy – Our first choice of systemic therapy is multisystemic therapy. Multisystemic therapy is a particularly intensive form of systemic therapy developed for older youth (middle childhood and adolescence) with more severe forms of conduct problems than are typically addressed in parenting groups or traditional family therapy. The treatment is typically delivered in the home setting and includes components of cognitive-behavioral therapy and parent management training as described above (see 'For preschool-age children' above and 'For school-age children (age 6 to 12)' above). Antisocial behaviors, delinquency, and substance problems are seen within the larger context of multiple systems of influence including individuals, family, peer, school, and community. Treatment in multisystemic family therapy uses multiple sessions and hours per week.

Clinical trials on the efficacy of multisystemic therapy for older children (age 10 to 17) with behavioral problems have been inconclusive [23,24]. Although several individual studies point to the efficacy of multisystemic therapy, critical reviews with attention to methodologic limitations and potential biases point to mixed results. In a meta-analysis including eight randomized trials, multisystemic therapy was compared with usual services [23]. Treatment with multisystemic therapy was found to lead to similar rates for of out of home placement (33 versus 32 percent), arrest, or conviction (47 versus 42 percent) as compared with control condition. Pooled results include studies with data of varying quality tend to favor multisystemic therapy; however, small sample sizes and heterogeneity limit interpretation of data. Despite limited support for this treatment, there are few evidence-based alternatives and we have found it to be efficacious in some patients.

ADJUNCTIVE PHARMACOTHERAPY FOR SPECIFIC POPULATIONS — We use pharmacotherapy concurrently or adjunctively for individuals with irritability or aggressive behavior that interfere with implementation of the therapy or causes psychosocial disruption. We also use adjunctive pharmacotherapy to address co-occurring disorders such as attention deficit hyperactivity disorder (ADHD) or mood disorders. We agree with the American Academy of Child and Adolescent Psychiatry, which concluded that medication may be a useful adjunct to psychosocial interventions for severe mood symptoms of irritability and aggression when present in oppositional defiant disorder and/or conduct disorder (ODD/CD) with or without comorbidity (algorithm 2) [25,26].

Irritability and aggressive behavior — For children and adolescents with ODD/CD with severe and frequent verbal and/or physical aggression and/or irritability that affects psychosocial functioning or interferes with psychosocial interventions, we typically begin treatment with combined psychosocial therapy and pharmacologic management. In some cases, the patient may require prompt control of the aggression/irritability with medications before the therapy is effective.

Our preference is treatment with a second-generation antipsychotic (SGA). SGAs have been found to rapidly improve aggression and irritability in children and adolescents with a variety of disorders [27,28]. Although these medications do not treat the behavior problems per se, diminishing the aggression and/or irritability and regulation of mood can indirectly improve the behavior problems.

Among the SGAs, risperidone is the agent with the most evidence supporting its use in youth for the treatment of symptoms of ODD/CD (eg, aggression, irritability) [28-33]. However, many clinicians choose alternative choices with more favorable side effect profiles such as aripiprazole (lower likelihood extrapyramidal symptoms or metabolic dysregulation) or quetiapine (lower likelihood extrapyramidal symptoms) (table 1).

In an adolescent with increased irritability or aggression, when starting risperidone we begin at 0.5 mg orally given at night (if nighttime sedation is desired) or in divided dose (if daytime effect needed). We increase by 0.5 mg orally every three to five days to a total daily dose of 3 to 6 mg, typically given in divided doses. If using quetiapine, we begin with a dose of 25 mg twice daily and increase by 25 mg twice daily every three to five days depending on response and side effects up to doses of that range between 50 to 600 mg/day. Once at therapeutic range we monitor for four weeks to assess effectiveness prior to further increases and side effects (eg, metabolic syndrome). Dose and titration of antipsychotic medications for children and adolescents with schizophrenia is found on the table (table 2).

If an acceptable response is not achieved or if side effects limit further dose increases, we try another SGA. (See 'Poor or partial response' below.)

Studies investigating the effects of pharmacotherapy for the management of impulsive aggression and/or irritability in youth with ADHD (which often is comorbid with ODD/CD), intellectual disabilities, or autism spectrum disorders are informative for the treatment of ODD/CD because aggression and irritability are key symptoms associated with these disorders. Almost all studies examining the use of antipsychotics for aggression and/or irritability as symptoms of various disorders or as part of the behavior disorders have been carried out using SGAs in youths with ADHD, autism spectrum disorder, or intellectual disabilities [27,28].

Evidence from trials suggests that risperidone is effective for treating aggression and conduct problems in youth with symptoms ODD/CD who present with or without ADHD [28-33].

●In one meta-analysis of three trials including 238 youths (age up to 18 years) with aggression in the context of disruptive behavior disorders including ADHD, individuals treated with risperidone showed reduced aggression on the Aberrant Behavior Checklist, irritability subscale (a 45-point scale measuring irritability level) as compared with those treated with placebo (mean difference -6.49, 95% CI -8.8 to -4.2) [28]. Additionally, risperidone appeared to improve problems of conduct as measured by the Nisonger Child Behavior Rating scale, conduct problem subscale (a 48-point measure of conduct in children) as compared with placebo (two trials, 225 participants, mean score -8.6, 95% CI -11.5 to -5.7) [28].

●An earlier meta-analysis also concluded that risperidone was effective in the management of aggression in children with ODD/CD or another disorder associated with aggression [34]. However, interpretation of these studies is limited by small samples, heterogenous outcome measurements, treatment settings and populations. Additionally, use of risperidone was associated with metabolic side effects including increased weight, glucose and lipid elevations.

Limited data support the use of first generation antipsychotic medications (eg, haloperidol, thioridazine) targeting aggressive behavior in youths with aggression and a primary diagnosis of ADHD, CD, autism spectrum disorder, or subaverage intelligence quotient (IQ) [34]. Due to risks of extrapyramidal symptoms and tardive dyskinesia we prefer to use SGAs for treatment of these individuals. (See "Second-generation antipsychotic medications: Pharmacology, administration, and side effects" and "First-generation antipsychotic medications: Pharmacology, administration, and comparative side effects".)

Co-occurring disorders — The presence of comorbid disorders such as ADHD, depression, or anxiety interfere with the implementation of the therapy and sometimes the pharmacologic treatment of these disorders may significantly ameliorate the symptoms of the behavior disorders.

●ADHD – Our preference is to treat individuals with co-occurring ODD/CD and ADHD with a long-acting stimulant in addition to psychosocial treatment. We use the same initial dose and titration rate as is recommended for treatment of ADHD in children or adolescents. For example, we would start amphetamine at 5 mg daily (in a child over six years) and titrate by 5 to 10 mg weekly to a total of 40 mg per day. Treatment of ADHD in children and adolescents is discussed elsewhere. (See "Attention deficit hyperactivity disorder in children and adolescents: Treatment with medications".)

Evidence of efficacy of stimulant medications in the treatment of impulsive aggression or irritability in youths with co-occurring ADHD and ODD/CD include [27,29,30,34-37]:

•In a meta-analysis of nine studies including 1344 youths with aggression, oppositional behavior or conduct problems, subjects with ADHD treated with psychostimulants showed greater improvement on measures of oppositional behavior or conduct problems than those treated with placebo. The standardized mean difference in teacher-rated oppositional behavior between treatment group and placebo group was -0.84 (95% CI -1.1 to -0.59). The standardized mean difference between the treatment versus placebo groups on parent-rated oppositional behavior conduct problems or aggression was -0.55 (95% CI -0.73 to -0.36) [30].

•In a meta-analysis of 18 trials including 1057 children and adolescents with primary diagnoses of ADHD, autism spectrum disorders, intellectual disabilities, or disruptive behavior disorders (ODD/CD), treatment with stimulant medications (eg, methylphenidate, methylphenidate/amphetamine, methylphenidate dextroamphetamine) showed a medium to large effect on aggression (mean effect size = 0.78) [34].

●Co-occurring mood or anxiety disorders – Our preference is to treat mood or anxiety symptoms that co-occur with ODD/CD with an evidence-based treatment for that particular disorder in addition to psychosocial intervention for ODD/CD. Our first choice is typically a selective serotonin reuptake inhibitor (SSRI) such as citalopram. If the treatment with the antidepressant is ineffective, we typically try a second SSRI. (See 'Poor or partial response' below.)

Small randomized controlled trials and open reports suggest that antidepressants may help some children and adolescents with aggression and/or irritability. A study of children with disruptive dysregulation disruptive disorder and ADHD showed that the combination of stimulants plus citalopram was better than the stimulants alone [38]. Most evidence that medications that improve mood and anxiety also target aggression and irritability derive from secondary analyses of randomized controlled trials that primarily evaluated the efficacy of these medications for mood or anxiety disorders [34-36,38]. (See "Pharmacotherapy for anxiety disorders in children and adolescents" and "Psychotherapy for anxiety disorders in children and adolescents" and "Overview of prevention and treatment for pediatric depression" and "Substance use disorder in adolescents: Treatment overview".)

SUBSEQUENT TREATMENT

Robust response — If the treatment response to psychosocial treatment is robust, we follow up within the first two months and then in three to six months to assess the maintenance of treatment gains.

If treatment response to medication is robust, unless there are significant side effects, we typically continue medications for a minimum of one year and then try to taper the medication. We typically taper medication in the summer months when the academic stressors are less. We monitor these individuals with face to face appointments every six to eight weeks. For individuals on antipsychotic medications, we monitor metabolic parameters including weight on a regular basis (table 3).

Poor or partial response — For individuals with poor or partial response to the initial treatment, our first steps are to confirm treatment adherence, review the history to confirm the diagnosis, and reassess for overlooked comorbid disorders and environmental factors (eg, conflicts, abuse) that warrant management. For those treated with antipsychotic medications, we make sure that a full eight-week trial has been completed at maximum tolerated dose within the therapeutic range.

Our next steps are:

●For those managed with psychosocial intervention – For those with a poor or partial response to initial psychosocial intervention, we suggest trying a second evidence-based psychosocial treatment. As an example, for school-age children (age 6 to 12) who do not respond to a first-line combination of behavioral parent management training and skill training for children we would switch to a multicomponent treatment such as multisystemic therapy or, if unavailable, functional family therapy. Functional family therapy uses behavioral approaches, such as modifying family member behaviors using contingency management techniques and behavioral contracts. (See "Substance use disorder in adolescents: Psychosocial management", section on 'Family-based treatment'.)

●For those managed with pharmacologic intervention – For children/adolescents who have not responded to pharmacologic management to this point (ie, two antipsychotics, antidepressant) our next choice of medication depends on patient preference, history of prior treatment with medications, family history of treatment with medication and clinical comfort and familiarity with the specific agent.

We prefer to try sequential trials of mood stabilizers or other agents; however, limited evidence supports the effectiveness of these medications [29-31]. Options include:

•Mood stabilizers – Most studies evaluating the use of mood stabilizers in the treatment of aggressive behavior are for short periods of time with small sample sizes [31,39].

-In a study, 40 children (median age 12.5 years) with conduct disorder (CD) who were hospitalized due to severe and chronic aggression were randomized to receive lithium or placebo [39]. After four weeks of treatment, 80 percent of individuals in the treatment group (16/20) responded to treatment (eg, Overt Aggression Scale) compared with 30 percent (6/20) of individuals in the placebo group.

-In an analysis of two studies including 50 youth with aggressive behavior due to oppositional defiant disorder (ODD), individuals treated with divalproex had a higher likelihood of response than those treated with placebo (odds ratio 14.6, 95% CI 3.25-65.6) [31].

•Others – Clonidine [30], guanfacine [30], and atomoxetine [30,40] appear to have a limited role in the treatment of symptoms of ODD/CD, but secondary analyses of studies of the treatment of other disorders such as major depression, anxiety, bipolar disorder, and tic disorders, suggest that they may help to control aggression [27,30,34,35,40-42].

Atomoxetine is the most extensively studied. In a meta-analysis of 25 trials including 3928 children with attention deficit hyperactivity disorder with or without comorbid ODD/CD, treatment with atomoxetine appeared to have a small effect on oppositional behaviors (effect size 0.33 95% CI 0.24-0.43) [40]. Low-quality evidence limits interpretation of efficacy of these treatments.

SUMMARY AND RECOMMENDATIONS

●Establishing treatment – We treat all children or adolescents with behavior problems that cause psychosocial distress or impairment whether they meet full criteria for either oppositional defiant disorder and/or conduct disorder (ODD/CD). (See 'Establishing treatment' above.)

•Setting goals – The overarching goal for treatment of children or adolescents with syndromal or subsyndromal ODD/CD is to develop more adaptive behaviors to cope with stressors, increase frustration tolerance, or decrease severity of temper outbursts (See 'Setting goals' above.)

•Maintaining an alliance – We maintain and enhance our therapeutic alliance through continuous support of the individual and family, offering easy availability of services, psychoeducation, and nonjudgmental listening. (See 'Maintaining an alliance' above.)

●Psychosocial intervention for most – We treat most children with disruptive behaviors with an evidence-based psychosocial interventions as first-line treatment (algorithm 1).

•Choosing psychosocial intervention – Our first-line treatment recommendations depend on the age of the child and symptoms present. (See 'Choosing initial psychosocial intervention' above.)

-For preschool-age children (eg, 2 to 5 years) – We suggest first-line treatment with behavioral parent management training for parents versus other psychosocial treatments (Grade 2C). (See 'For preschool-age children' above.)

-For school-age children (eg, 6 to 12 years) – We suggest first-line treatment with cognitive-behavioral training skills for youth combined with behavioral parent management training, rather than other psychosocial treatments (Grade 2C). (See 'For school-age children (age 6 to 12)' above and 'For preschool-age children' above.)

-For adolescents (eg, 13 to 17 years) with mild symptoms – We suggest first-line treatment with a combination of cognitive-behavioral skills training for youth and behavioral parent management training, rather than other psychosocial interventions (Grade 2C). (See 'Mild symptoms' above.)

-For adolescents with moderate to severe symptoms – We suggest first-line treatment with a systemic therapy, such as multisystemic therapy (Grade 2C). (See 'Moderate or severe symptoms' above.)

●Adjunctive pharmacotherapy for aggression or comorbid disorders – An algorithm describes pharmacologic management of ODD/CD (algorithm 2).

•Aggressive behavior – We suggest augmentation of the psychosocial intervention with a second-generation antipsychotic rather than other agents for children or adolescents with symptoms of aggression/irritability that are affecting their psychosocial functioning (Grade 2C). (See 'Irritability and aggressive behavior' above.)

•Co-occurring attention deficit hyperactivity disorder (ADHD) – We treat children and adolescents with symptoms of ODD/CD and co-occurring ADHD with adjunctive stimulant medication in addition to psychosocial treatment. (See 'Co-occurring disorders' above.)

•Co-occurring mood or anxiety disorder – We prefer to treat children and adolescents with comorbid mood or anxiety disorders with a selective serotonin reuptake inhibitor in addition to psychosocial treatment. (See 'Co-occurring disorders' above.)

●Subsequent treatment

•Robust response – If treatment response to psychosocial treatment is robust we follow-up every three to six months and as needed to assess the maintenance of treatment gains. (See 'Robust response' above.)

If treatment response to medication is robust, we typically continue medications for a minimum of one year prior to taper. (See 'Robust response' above.)

•Poor or partial response – For those with a poor or partial response to initial psychosocial intervention, after ruling out other factors, our preference is to try a second evidence-based psychosocial treatment. (See 'Poor or partial response' above.)

For those with a poor response to initial pharmacologic management of severe irritability or aggression we prefer subsequent trials of mood stabilizers, atomoxetine, guanfacine, or clonidine. (See 'Poor or partial response' above.)