This is the first article using a large national sample that describes the target disorders and target symptoms of antipsychotic medication prescribed for children and adolescents attending mental health services in Ireland between July to December 2021. We illustrate that the prevalence of antipsychotic prescribing by specialist child and adolescent mental health services in Ireland is 12%, and that the main indication for prescribing an antipsychotic is to target symptom clusters and not a disorder. We also illustrate in a real time pragmatic way, the starting doses and maintenance doses that clinicians are using to target symptom clusters and psychotic disorders.

The data highlights a high amount of off-label prescribing of antipsychotic medications to young people attending specialist child and adolescent mental health services in Ireland. There is limited licensed indications for use of antipsychotics in children or adolescents in Ireland. As such, any indication based on target symptoms alone without a defined condition (e.g., conduct disorder or psychosis) is deemed off-label. There is a recognition that off-label prescribing does not imply improper or illegal use and is often necessary practice in children and adolescents in order to benefit the individual patient [13, 14]. Off-label prescribing may occur because the age of the child is lower than indicated on the license (often adults), due to indication, dose or route of administration [6, 7]. A recent scoping review by Meng et al. (2022) highlighted that antipsychotic agents were the most frequently studied medication to understand their off-label use [14, 15].

The antipsychotic most commonly prescribed to children and adolescents in our cohort was quetiapine (29%), followed closely by risperidone (28.6%) and aripiprazole (24.5%). In a study by Dinnissen and colleagues, who investigated adherence to antipsychotic prescribing guidelines among Dutch child and adolescent clinicians, the most prescribed antipsychotic was risperidone (68.3%), with aripiprazole (9.2%), olanzapine (6.7%) and quetiapine (1.6%) prescribed less frequently [15, 16]. A study by Radojčić and colleagues examined trends of antipsychotic prescribing to children and adolescents in England between 2000 and 2019, using primary care data [16, 17]. They reported the most frequently prescribed antipsychotics was risperidone (54.6%), followed by aripiprazole (17.6%), quetiapine (13.8%) and olanzapine (11.3%). The starting doses of antipsychotics prescribed in our dataset were aligned with the findings from the Dutch cohort [15, 16]. Radojčić et al., 2023 also reported that doses of antipsychotics in their English cohort were mostly within therapeutic ranges [16, 17].

The current study identified that antipsychotics were most commonly prescribed to target symptoms such as agitation and irritability rather than for a specific condition such as psychosis. The literature on indication for antipsychotic prescribing in children and adolescents is mixed. Some studies including those by Penfold et al. (2013) and Dinnissen et al. (2020) report findings similar to the current study [16, 18]. Radojčić et al. reported that indications associated with antipsychotics included ASD (12.7%), non-affective psychosis (8.6%), anxiety disorders (7.5%), ADHD (7.1%), depression (6.4%) and conduct disorders (6.1%) [16, 17]. 11.2% of prescriptions were associated with a non-specific mental health code and importantly, a limitation of the study identified by the authors was the inability to identify an indication in almost a third of prescriptions (Radojčić et al., 2023). A study by Rao and colleagues who surveyed child and adolescent psychiatrists in England on their prescribing patterns of atypical antipsychotics reported an indication of 8.6% in non-affective psychosis, 2.8% in affective psychosis, albeit 12.7% in autism spectrum disorder, , 7.1% in ADHD and 6.4% in depression [18, 19]. Similarly, in our study 12.4% of target conditions with an antipsychotic was for psychosis, albeit there was greater target symptom prescribing in this study which may highlight to a greater prevalence of prescribing based on target symptom clusters rather than defined disorder. Despite this, it is important to acknowledge that while off-label use cannot be labelled as poor clinical practice, the use of antipsychotic medication off-label may increase the likelihood of adverse events associated with antipsychotic use e.g., obesity, extrapyramidal side effects, hyperlipidaemia. Albeit there is mixed results of adverse events reported on off-label use in antipsychotic use in children or adolescents [20,21,22].

As expected, the main association of being prescribed an antipsychotic was by disorder in adjusted models. In our analysis, a higher number of co-morbid conditions were not associated with being prescribed an antipsychotic, which would be in contrast to similar adult studies ([22, 23]). Positively, in adjusted analysis, having a definitive diagnosis helped to illustrate a lower odds of being prescribed an antipsychotic (e.g., anxiety disorder). This in our view may highlight the importance of a clear diagnosis documented during the course of treatment to better understand the rationale of why a certain psychotropic was prescribed. In particular, as diagnostic changes or adjustments are frequent in a child and adolescent population and may provide greater clarity and transparency as to why a particular psychotropic was prescribed or changed longitudinally [23, 24].

The above discussion should be understood knowing the following limitations of this study. First, the available data does not capture if other specialists prescribed an antipsychotic (e.g., family doctors, neurologist, paediatricians), only child psychiatrists. The data was collected at source by paediatric psychiatry services and not trained data controllers, as such there may be input error. Conversely, this is a strength as services provided the information anonymously about service users and as such may have allowed greater transparency as to child psychiatrists prescribing for target symptoms and not definable disorders. The diagnosis classification used was not requested by services, albeit within the Irish context it is the International Classification of Disease–11 (ICD-11) or Diagnostic and statistical manual of mental disorders-5 (DSM-5). The data does not include other psychosocial interventions (i.e., non-pharmacological interventions); this is important to understand whether antipsychotics were prescribed in crisis (e.g., emotional dysregulation, aggression) or following unsuccessful non-pharmacological approaches. We did not have improved participant demographics (e.g., physical health conditions, family factors, socioeconomic factors and educational factors) which are important when considering child outcomes. With the available dataset, we were not able to determine whether these young people were drug naïve or if their prescriptions were repeated. This is deserving of further analysis using a different dataset. While the dose of medication was deemed to be maintenance daily dose, it is not possible with available data to determine if the dose was given pro necessitate rather than regular or daily. The prescribing standards were closed questions to determine if they had adhered to standard practice (e.g., monitoring), it did not provide explicit information (e.g., specific bloods or monitoring side-effects). Finally, we did not have available data on adverse reaction or if prescribing an antipsychotic alleviated the condition or symptoms.

In our study when clinicians in teams were asked to provide anonymously the rational/indication for antipsychotic use there was a high indication for target symptom cluster rather than a defined disorder or condition. This highlights the main implication which is transparency in indication documenting for clinicians and the prevalence of off-label prescribing. There is a need to understand if off-label prescribing has higher reported adverse reactions and adherence to monitoring standards for antipsychotic safety. In particular, it is surprising that quetiapine was the most prescribed antipsychotic in this study with known significant cardiometabolic adverse effects and little to no licensed indication in this age cohort. Similarly, a recent study highlighted that quetiapine was the most commonly prescribed antipsychotic by condition (e.g., psychosis) and risperidone by symptom (e.g., aggression) with a condition (e.g., autism spectrum disorder) [25]. These findings demonstrate a potential notable adherence gap between antipsychotic indication and guideline use. Furthermore, the need for a process to monitor psychotropic prescribing by dose and indication in Ireland is now necessary (e.g., national yearly audit or reporting system of prescriptions in CAMHS). Positively maintenance doses were low in this study which may be associated with a higher off-label use. This in turn may reflect a clinician’s hesitancy in increasing the dose and this is an area worthy of further research.