Twenty years of progress in paediatric psychopharmacology: accomplishments and unmet needs

Abstract

The systematic assessment of the efficacy and safety of psychiatric medications in children and adolescents started about 20 years ago. Since then, a considerable number of randomised clinical trials have been conducted, including also a series of publicly funded comparative effectiveness studies to evaluate the therapeutic benefit of medications relative to psychosocial interventions, alone or combined with medications. On the whole, these studies have been informative of the paediatric pharmacokinetics, efficacy and safety of the most commonly used psychotropics. As a consequence, a number of meta-analyses have been conducted that have documented both the benefits and harms of the most common medication groups, such as stimulants, antidepressants and antipsychotics. Evidence-based practice guidelines have been produced, and clinicians can now better estimate the therapeutic value and the risk of treatment, at least at the group mean level. However, most clinical trials have been conducted in research settings, and this limits the generalisability of the results. There is a need for evaluating treatment effects under usual practice conditions, through practical trials. The ongoing debate about the proper role of pharmacotherapy in child mental health can be advanced by comparative effectiveness research to assess the benefit/risk ratio of pharmacotherapy vis-à-vis alternative treatment modalities. In addition, analyses of large population databases can better inform on the impact of early treatment on important distal outcomes, such as interpersonal functioning, social and occupational status, quality of life and risk for disability or mortality. Thus far, paediatric psychopharmacology has been mostly the application to children of medications that were serendipitously discovered and developed for adults. By focusing on the neurobiological mechanisms of child psychopathology, it may be possible to identify more precise pharmacological targets and arrive at a truly developmental psychopharmacology.

Introduction

Paediatric psychopharmacology addresses the therapeutic use of psychotropic medications in children (here defined as individuals under age 18 years) who suffer from psychiatric disorders. Even though the first report on the use of amphetamines for treating child hyperkinesis goes back to 1937, the existence of paediatric psychopharmacology as a specific section of psychopharmacology is much more recent. It was only in the late 1990s that the need for a systematic evaluation of the efficacy and safety of psychotropic medications in children was recognised by the clinical and scientific community, patient advocacy groups and drug regulators.¹

At that time, the main impetus for paediatric psychopharmacology research came from the increasing clinical use of psychotropics in children, a use that was neither supported by scientific evidence nor approved by regulatory agencies.² Being medications available on the market for adults, it had been possible for practitioners to use them ‘off-label’ also in children.³ Thus, following a sequence contrary to the logical flow from research to clinical use, in the beginning was the clinical use, which later begot the research to evaluate its appropriateness.

Now, 20 years after, seems to be an appropriate time to examine the work done thus far, with respect to both accomplishments and shortcomings, from an evidence-based perspective.

Accomplishments

The concept of evidence-based child psychopharmacology

The first and, arguably, the most important accomplishment has been the successful introduction and general acceptance of concept itself of evidence-based child psychopharmacology. That pharmacological treatment should be considered a therapeutic option for children with emotional and behavioural disorders is a notion that had been either ignored by many as of little relevance and peripheral to clinical practice, or actively resisted by some, as inappropriate and invasive.⁴ The definition of the child as ‘therapeutic orphan’, which was applied to the unsatisfactory situation of paediatric pharmacology research in general, has been especially applicable to psychopharmacology. Therefore, the acceptance by the clinical, scientific and bioethical community of research on child psychopharmacology as both necessary and ethically justifiable can be seen as an essential step towards systematic research in this field.

Psychiatric disorders are major contributors to the global burden of illness in economically developed countries,⁵ The most common psychiatric disorders have onset in youth,⁶ and early onset disorders tend to have greater severity and worse outcome.⁷ Prompt treatment, including both psychosocial and pharmacological interventions, could lead to better functional outcomes.⁸ Given these premises, because children cannot be denied the highest standard of evidence-based care and because research in adults is insufficient for determining efficacy and safety of medications in children, the only logical and ethical conclusion can be that child psychopharmacology research in children becomes scientifically necessary and ethically appropriate.

This realisation happened at a time when increasing attention to the off-label use for most medications in paediatrics led to a series of legislative and regulatory actions that were introduced as a remedy, first in the USA, starting in 1997 and then, years later, in Europe.^{9 10} These initiatives provided both financial incentives to industry to conduct paediatric pharmacology studies and authority to regulatory agencies to request such studies when appropriate. Under these rules, the market exclusivity of a medication has been extended by an additional 6-month period in return for conducting specific paediatric studies. In parallel, pharmaceutical companies must implement a paediatric investigational plan (PIP) when developing a new compound for use in adults if a paediatric use is to be expected.¹¹ These provisions are aimed at correcting and preventing the off-label use of medication in children. The development of paediatric psychopharmacology research has been greatly helped by these initiatives, even though industry can use a PIP merely as a method to exploit patent extension, and this may lead to studies of lower quality.

Evaluation of efficacy

In order to document the extent of research in child psychopharmacology research during the past 20 years (1999–2018) as compared with previous years, a systematic research of randomised clinical trials of the most common types of psychotropics was conducted.

Methods

We searched for primary publications in English language of randomised clinical trials that tested the treatment effects of psychotropic medication in children . The control or comparison group could consist of placebo, another medication or a psychosocial intervention. In June 2018, systematic reviews and meta-analyses of stimulants, antidepressants, antipsychotics, lithium and mood stabilisers in children (aged <18 years) published since January 2015 were identified through searching the Medline databases (accessed through www.PubMed.gov). The search terms included: children and stimulants or antidepressants or antipsychotics or mood stabilisers or lithium, and meta-analysis or systematic review. Recent systematic reviews were identified for stimulants, antidepressants, antipsychotics and mood stabilisers.^12–21 These reviews were used to identify individual trial publications. In addition, for medications for which no recent systematic review or meta-analysis was available or for additional indications not covered by available reviews (eg, eating disorder), specific Medline searches were conducted to identify relevant clinical trials. Finally, Medline was searched for the years since the completion of each systematic review to cover the most recent period until June 2018. The identified studies were assessed independently by the two authors. The number of trials was computed for each type of psychotropic and for each indication, together with the number of subject randomised to study medication and placebo, respectively. Each clinical trial contributed to only one entry. Thus, a trial including more than one medication was assigned to one category and counted only once, based on evaluation of the main focus on the study. For within-subject (crossover) placebo-controlled trials, the number of study subjects contributed to both the medication and the placebo groups, as each subject was sequentially exposed to both.

Results and comment

Table 1 summarises the number of clinical trials, published before and after January 1999, which tested the effects of the major groups of psychotropic medications in children by clinical indication. It is evident that there has been a dramatic increase in the number of clinical trials in paediatric psychopharmacology in the last 20 years. Only few, mostly statistically underpowered, randomised clinical trials had been reported in paediatric psychopharmacology until the late 1990s. Since then, however, the scenario has rapidly changed. The acute efficacy of most psychotropics of relevance to child mental health has been tested in controlled clinical trials. The number of research participants increased even more, as a reflection of the larger sample sizes, which provide greater statistical power. Also, the crossover, within-subject design, which is typically limited to the assessment of acute and transient treatment effects, has become less frequently used (table 1).

Table 1.

Randomised clinical trials in paediatric psychopharmacology for common medication groups and therapeutic indications*

Medication group	Indication	Through 1998				1999–June 2018†
		RCTs (N)	Patients randomised (N)			RCTs (N)	Patients randomised (N)
			Total‡	To medication	To placebo		Total	To medication	To placebo
Stimulants	ADHD	66§	2306	2276	2190	130¶ (2.0)	12 239 (5.3)	9838 (4.3)	7838 (3.6)
Antidepressants	MDD	10	462	230	232	24 (2.4)	4956 (13.1)	2561 (11.1)	1937 (8.3)
	OCD	6**	357	208	138	6 (1.0)	636 (1.8)	336 (1.6)	268 (1.9)
	Anxiety disorders	1	15	6	9	12 (12.0)	2091 (139.4)	1064 (171.2)	855 (95.0)
	Other††	7‡‡	260	192	142	10§§ (1.4)	451 (1.7)	298 (1.6)	281 (2.0)
Antipsychotics	Schizophrenia spectrum	6	211	196	15	18 (3.0)	2764 (13.1)	2099 (10.7)	665 (44.3)
	Bipolar	0				12	2522	1591	931
	Aggression and other conduct disturbances	7	242	139	66	26 (3.1)	2113 (8.7)	1202 (8.6)	866 (13.1)
	Tic disorders	3¶¶	118	118	118	7***	317	201	43
Lithium	Bipolar disorder					5	496	208	61
	Aggression and other conduct disturbances	3***	118	52	52	1	40 (0.3)	20 (0.4)	20 (0.4)
	Mood dysregulation					1	25	14	11
Anticonvulsants	Bipolar disorder	0				8	791	339	245
	Aggression and other conduct disturbances†††	4§§	115	88	84	9*** (2.2)	285 (2.5)	175 (2.0)	129 (1.5)

*Based on searching the Medline database for primary English language publications of randomised clinical trials of psychotropic medications for children and adolescents. Each clinical trial contributes to only one entry in the table (eg, a trial including more than one medication was assigned to one category and counted only once). For within-subject (crossover) placebo-controlled trials, the number of study subjects contributes to both the medication and placebo group (see ’Methods' section in text).

†Relative changes vs ≤1998 are provided in parentheses.

‡Total number of children who were randomised to investigational medication and/or to placebo and/or comparator, according to study design.

§Sixty-one (92%) with crossover design.

¶Ninety-three (85%) with crossover design; lower proportion than in the <1999 group (Fisher’s exact test, p=0.0007).

**Two with crossover design.

††Other conditions, including ADHD, Tourette, autism, enuresis and smoke prevention.

‡‡Five with crossover design.

§§Four with crossover design.

¶¶Three with crossover design.

***One with crossover design.

†††Including conduct disorder, ADHD, irritability in autism spectrum disorder.

ADHD, attention deficit/hyperactivity disorder; CD, conduct disorder; MDD, major depressive disorder; OCD, obsessive-compulsive disorder; RCT,  randomised  clinical trial.

Some of these clinical trials were publicly funded, but most have been funded by industry, under the regulatory incentives that extend the period of market exclusivity. The accumulation of data has allowed systematic analyses and meta-analyses to be conducted on most types of medications. On antidepressants alone, >20 systematic reviews and meta-analyses have been published since 2010.²²

On the whole, this research activity has been informative, but not without limitations. Because the extension of the patent market exclusivity is granted to industry in return for the prompt completion of paediatric studies regardless of their outcome, some clinical trials have been conducted at a very large number of clinical sites (at times >60 sites in different countries and continents).²³ The experimental error brought by such a large number of sites, each often contributing only a small number of subjects, likely reduces the study assay sensitivity. In fact, the majority of the industry-funded clinical trials of antidepressants were unable to detect a statistically significant difference.^{15 16} It should be noted that the inclusion of a standard comparator in a number of these trials can be considered an important methodological improvement because it allows a distinction to be made between a ‘negative’ study (ie, no statistically significant difference between experimental medication and placebo, but superiority of the standard comparator over placebo) and a ‘failed’ study (ie, no statistically significant difference between either experimental medication or standard comparator and placebo).²³

Comparative effectiveness

Proving superiority over placebo is only the first step in evaluating the potential benefit of a medication. For informed decision making, clinicians, patient, families and health administrators need data on the effectiveness of treatments relative to ecologically valid alternatives. In the past 20 years, a number of randomised clinical trials have been completed that compared alternative treatments for attention deficit/hyperactivity disorder, major depression, anxiety disorders, obsessive-compulsive disorder, early onset schizophrenia, bipolar disorder and autism (table 2). In attention deficit/hyperactivity disorder (ADHD), pharmacotherapy, as a treatment strategy rather than a specific medication, was compared with the alternative strategies of using psychosocial intervention, or combining medication with psychosocial intervention.²⁴ Likewise, pharmacological, psychotherapeutic and combined treatments were compared with each other in youth with major depression, obsessive-compulsive disorder and anxiety disorders (table 2). The combination of pharmacotherapy and cognitive-behavioural psychotherapy has proven to be superior to each monotherapy for the treatment of anxiety disorders and obsessive-compulsive disorder,^{25 26} whereas for depression the results are mixed.^27–30 Many of these studies have been influential in informing clinical practice guidelines and regulatory decisions.³¹

Table 2.

Selection of randomised comparative effectiveness clinical trials in paediatric psychopharmacology

Study	Main research question	Sample	Setting	Randomisation to	Main findings
Multimodal Treatment Study of Children with ADHD24	How do different treatment strategies (pharmacological, behavioural and combined) compare with usual community care for decreasing ADHD symptoms and improving functioning?	n=579 Aged 7–9 years, with ADHD combined type	Outpatient university clinics	Medication (stimulant) management, behaviour therapy, their combination or usual care, for 14 months, followed by a 10-year naturalistic follow-up.	Greater improvement with medication management, either alone or in combination, with no difference between these two. Dissipation of treatment differences during naturalistic treatment.
Paediatric OCD Treatment Study25	Is SSRI combined with CBT more effective than either monotherapy in childhood OCD?	n=112 Aged 7–17 years, with OCD	Outpatient university clinics	Sertraline, CBT, their combination or placebo, for 12 weeks.	Combined treatment was more effective than monotherapy, which was better than placebo.
Treatment for Adolescents with Depression Study27 28	Is SSRI combined with CBT more effective than either monotherapy adolescent MDD?	n=439 Aged 12–17 years, with MDD	Outpatient university and community clinics	Fluoxetine, CBT, their combination or placebo for 12 months, followed by unblinded maintenance treatment for 6 months.	Fluoxetine, either alone or combined with CBT, was better than CBT, or placebo, in improving mood. Fluoxetine as monotherapy, but not when combined with CBT, increased the risk of suicidal events. No distal differences in outcome 6 months after randomisation.
Adolescent Depression Antidepressant and Psychotherapy Trial29	Is combined CBT and SSRI treatment more effective than SSRI monotherapy for adolescent depression?	n=208 Aged 11–17 years, with MDD	Outpatient practice settings	SSRI+routine care or CBT+SSRI+ routine care, for 12 weeks.	No difference between combined treatment and monotherapy.
Treatment of Resistant Depression in Adolescents30	After an unsuccessful treatment with SSRI, is switching to another antidepressant plus adding CBT more effective than switching to another antidepressant monotherapy?	n=326 Aged 12–18 years	Outpatient university clinics	SSRI or venlafaxine with or without CBT for 12 weeks.	Combined treatment was more effective than monotherapy.
Treatment of Early Onset Schizophrenia Spectrum Disorders 59	Are second-generation antipsychotics superior to first-generation antipsychotic in the treatment of early onset schizophrenia?	n=119 Aged 8–19 years, with schizophrenia or schizoaffective disorder	Outpatient university clinics	Risperidone, olanzapine or molindone for 8 weeks (acute treatment) followed by 10-month maintenance treatment.	No difference among medications in efficacy, but with important differences in safety outcomes. Most patients discontinued the randomly assigned treatment after a few months.
Child-Adolescent Anxiety Multimodal Study26	Is SSRI combined with CBT more effective than monotherapy in childhood anxiety disorders?	n=488 Aged 7–17 years, with separation anxiety disorder, generalised anxiety disorder or social phobia	Outpatient university clinics	Sertraline, CBT, their combination or placebo, for 12 weeks.	Combined treatment was the most effective intervention. Monotherapy with sertraline or CBT was better than placebo.
Treatment of Early Age Mania60	How effective are antidopaminergic vs anticonvulsant medications vs lithium for acute mania stabilisation in children?	n=290 Aged 6–15 years	Outpatient university clinics	Valproate, lithium or risperidone for 8 weeks.	Risperidone was the most effective intervention, with no significant difference between lithium and valproate.
Treatment of Serious Behaviour Problems in PDD61	Does the addition of parent training to pharmacotherapy result in better outcomes in PDD?	n=124 Aged 4–13 years, with PDD	Outpatient university clinics	Risperidone, as monotherapy or combined with behaviour therapy, for 24 weeks.	Medication plus parent training was more effective than medication alone at decreasing maladaptive behaviours.

ADHD, attention deficit/hyperactivity disorder; CBT, cognitive-behavioural  therapy; MDD,   major depressive disorder; OCD, obsessive-compulsive disorder; PDD, pervasive developmental disorders (autism spectrum disorder); SSRI, selective serotonin reuptake inhibitor.

Assessing safety

Significant progress has been made also in both documenting common adverse effects and estimating the risk for infrequent toxicities in children treated with stimulants, antidepressants and antipsychotics.

Based on anecdotal reports of sudden death in children treated with stimulants for ADHD, concerns were raised about the possible cardiovascular toxicity of these medications. Analyses of clinical trial databases have documented a small, but statistically and potentially clinically significant, increase in heart rate and blood pressure.³² Sudden death, however, being a rare event, cannot be evaluated with clinical trials databases. A case-control study found an increased risk, but this type of study design is limited by the relative small sample size and inability to control completely for confounders.³³ Analyses of large clinical and population databases have not found an increased risk of severe cardiac events with the use of stimulants.^{34 35} The integration of information from these methodologically different investigations has led to the current position that stimulants are contraindicated in children with known cardiovascular pathologies, such as cardiomyopathy or conduction defects, but can be used without major restrictions in otherwise healthy children with ADHD.

Another long-standing safety issue about stimulants has been regarding a possible increase in the risk for substance abuse through mechanisms of brain behavioural sensitisation consequent to dopaminergic stimulation. Analyses of large population databases, however, have not found evidence of an increased risk,³⁶ thus supporting the notion that, although stimulants have a potential for abuse and misuse, their therapeutic use is not causally linked to substance abuse or dependence.

A large, multinational, long-term observational study of children treated with stimulant medications for ADHD has been recently completed.³⁷ The forthcoming results will provide additional information on the safety on a comprehensive set of outcomes, including physical growth, pubertal development, cardiovascular function and risk for neurological and psychiatric adverse effects.

A major drug safety concern was raised in the early 2000 about antidepressants potentially increasing the suicide risk in youth. Unlike the issue of stimulants and cardiac adverse effects, a link between antidepressant use and increased suicide risk seems counterintuitive and paradoxical. Depression is a primary risk factor for suicide, and antidepressants should treat depression and consequently decrease the risk of suicide. Given its infrequency, suicide cannot be used as a safety outcome in randomised clinical trials, which, however, can detect differences in the incidence of severe suicidal ideation and suicidal attempts (‘suicidality’). Suicidality was used as an index of suicide risk in placebo-controlled clinical trials and meta-analyses, which found a higher risk of suicidality in patients randomised to antidepressant compared with placebo, but only for patients up to 25 years of age.27 28 38 39 These data suggest that the risk is age-dependent and point to an interesting interaction between development and medication safety. Analyses of patient population databases, on the other hand, have not found evidence of increased risk of suicide with antidepressant treatment,⁴⁰ thus raising an issue of the validity of the construct of ‘suicidality’ as an index of suicide risk.

A very different type of safety concern is regarding second-generation antipsychotics, which have become increasing used in children, often off-label for the management of aggressive behaviour and extreme irritability. While these medications generally represent an improvement over the first-generation antipsychotics because of the lower neurological adverse effects, they often cause marked metabolic adverse effects (ie, increased caloric intake, reduced insulin sensitivity and hyperlipidemia) to which children are more sensitive than adults, with increased risk for obesity and type 2 diabetes.⁴¹ These data have major implication for clinical practice with respect to treatment decision making and monitoring.

Unmet needs

Beyond efficacy research in academic setting: towards clinical trials in practice settings

Thus far, almost all paediatric psychopharmacology randomised trials, including also the comparative effectiveness trials, have been conducted by researchers in research settings (table 2), and this limits the generalisability of the results. When considering the burden of mental illness in childhood and the increasing use of psychotropics,^{5 42} there is a dearth of practical trials in both child and adult psychopharmacology.⁴³ Practical trials are needed to evaluate the comparative effectiveness of pharmacological treatments under usual practice conditions. Measures should include, in addition to symptom scores, developmentally appropriate indexes of functioning and ‘hard’ outcomes such as school attendance, academic achievement, suicide attempts, emergency room access, hospitalisation and health costs.

Treatment choice should be based on an analysis of the estimated benefits and harms of the available interventions and their alternatives. A number of psychosocial interventions have proven to be effective in child psychiatry,⁴⁴ and are the first-line approach in the management of many disorders. While a few clinical trials have directly compared pharmacotherapy with psychosocial treatments, it will be critical to expand this research in practice settings.

Beyond symptom improvement, towards disease-modifying interventions

Mental disorders tend to be chronic and to continue in adulthood. While current pharmacotherapy helps manage symptoms, it is not curative, and long-term treatment is often required. Progress has been made in assessing safety during chronic treatment, but data on distal benefits of pharmacotherapy have been more difficult to evaluate. In particular, whether early treatment leads to a better prognosis in adulthood seems to be a critically important issue to clarify. There is currently evidence that some psychosocial interventions, such as cognitive-behavioural therapy for anxiety disorders, can be curative, in addition to reducing symptom intensity.⁴⁵ For child psychopharmacology, the evidence for disease-changing effects is still limited. In schizophrenia, the association between shorter duration of untreated psychosis and better functional outcome suggests that this may be the case,⁸ but, for conditions such as ADHD, depression or bipolar disorders, distal benefits have been difficult to ascertain.

While the most convincing evidence that an intervention is disease-modifying would come from large randomised trials, it may not be feasible to retain patients in randomly assigned treatment arms for long periods of time. Alternative approaches can be considered. Analyses of large and integrated population databases have been informative about associations between stimulant medications and incidence of accidents, substance abuse, illegal behaviours, hospitalisation and mortality.⁴⁶ The increasing availability of ‘big data’ and the development of appropriate analytical methods are especially promising resources for investigating how early pharmacotherapy may influence the trajectory of psychopathology with respect to functional outcomes.^{47 48}

Beyond categorical nosology: targeting dimensions of psychopathology

The limitations of the traditional psychiatric nosology, which is based on clinical description of symptoms, have become more apparent. The current diagnostic categories do not seem to map the genetic and neurobiological underpinning of psychopathology.⁴⁹ Nosology is obviously highly relevant to psychopharmacology because medications are developed, tested and eventually approved for clinical use for specific diagnostic indications. For example, the concept that children could suffer from bipolar disorder is relatively recent, and this explains that absence of paediatric clinical trials in this indication in the period up to 1998 (table 1). The inability to find more innovative and effective medications for common conditions such as depression and other mood disorders may be in part due to the heterogeneity of the current diagnostic constructs.

Alternative approaches have been proposed that are based on linking neurobiology with dimensions of psychopathology across the existing diagnostic categories.⁵⁰ Child psychopharmacology research is in the ideal position to develop treatments focused on dimensional constructs of high clinical relevance. For example, mood dysregulation with its clinical manifestations of irritability can be a target for pharmacotherapy.⁵¹ Clinical trials to test risperidone and aripiprazole for irritability in the context of neurodevelopmental disorders such as autism have already led to the acceptance of this construct, which is not per se a diagnostic category, by drug regulatory authorities.⁵² Likewise, psychopharmacology aimed at evaluating treatments for mood dysregulated youths seems to be an innovative approach to address an important clinical need.

Beyond serendipity: towards a neuroscience-informed developmental psychopharmacology

The current psychotropic medications have been in large part the result of serendipitous discovery, and paediatric psychopharmacology research has mainly consisted in applying to children medications already used for adults. A better understanding of the mechanisms behind mental disorders should ultimately lead to a neuroscience-based drug development. This approach holds promise for an entirely new chapter of paediatric psychopharmacology.

Precision medicine is already applicable to several areas of medicine such as oncology and cardiology.⁵³ The situation with psychiatry is much more complex, with hundreds of genetic variants contributing to the pathogenesis of disorders such as autism or schizophrenia. Genetic polymorphisms of genes encoding for several cytochrome P enzymes, such as CYP2D6 AND CYP2D19, influence the metabolism of a number of commonly used psychotropics (eg, risperidone, fluoxetine, escitalopram and atomoxetine), with potential implications for both safety and efficacy.^{54 55} However, the current data do not support routine testing of children for these genetic polymorphisms when prescribing these medications, even though they can be obtained for individual patients if there are concerns about adverse effects or treatment resistance. Testing for CYP2D6 and CYP2C19 is desirable before use of tricyclics antidepressants, which are nowadays seldom used, due to the risk of cardiotoxicity that increases with higher plasma levels. The search continues for markers that can predict efficacy and/or toxicity with sufficient precision to guide treatment.⁵⁶

A glimpse on what could be the future scenario of a pathogenesis-driven drug development may come from recent research on the few disorders whose genetic and aetiopathogenetic bases have been elucidated. For example, in the case of Fragile X syndrome, an abnormal expansion of the CGG triplet repeats on the chromosome X gene for the FMR1 protein causes silencing of the gene and lack of the protein, thus ultimately resulting in dysregulation of the balance between the glutamatergic excitatory neurotransmission and the GABAergic inhibitory one.⁵⁷ Building on this knowledge, pharmacological approaches have recently been tested in an attempt to correct the neurotransmission imbalance and thus improve the associated behavioural symptoms.⁵⁸ Even though this line of research is in progress and has not produced effective treatments as yet, it signals a new approach to drug development in child psychopharmacology.

Conclusions

Major accomplishments have been made in evaluating benefits and risks of psychotropics in children. With respect to both data acquisition and research methodology, paediatric psychopharmacology is clearly in a much stronger position now than it was 20 years ago. The need remains for data on the long-term benefits and risks of pharmacotherapies and on how individual characteristics can inform personalised and more precise approaches to treatment. The future of the field depends on how neuroscience advances can be translated into innovative clinical applications. It also depends on applying a mix of methods, including randomised practical trials and large database analyses, to determine what is the best use of psychotropics in children for changing the trajectory of psychopathology and thus improving psychiatric outcomes throughout life.