In this retrospective study, we analyzed the trends in antipsychotic and off-label usage in children and adolescents from 2016 to 2021. To the best of our knowledge, this was the first study related to trends in the use of antipsychotics and off-label prescriptions in children and adolescents in China. Our study assessed the real-world usage of antipsychotics, which provides insights into how these drugs are actually being used in clinical practice.

In our study, we noted that the safety and efficacy of certain antipsychotics in pediatric patients have not been established, and several antipsychotics were not authorized for use in the pediatric population. Moreover, information regarding drug dosage was incomplete. As a result, our study focused solely on analyzing off-label indications. In addition, we discovered that certain antipsychotics, such as sulpiride and tiapride, were not included in the FDA list. Furthermore, the indications approved by the NMPA for antipsychotics vary from those approved by the FDA. For instance, in China, aripiprazole is solely approved for treating schizophrenia, whereas in addition to this indication, it has also been approved for bipolar I disorder, irritability associated with autistic disorder, and Tourette’s disorder. Consequently, off-label prescriptions were assessed based on the NMPA, which aligns more closely with the circumstances in China. Furthermore, the results of our study may differ from those of studies conducted in other countries.

Our study demonstrated a general increase in both the prescription of antipsychotics and their off-label usage between 2016 and 2021. Notably, the number of patients receiving antipsychotics decreased in 2020, which may be related to the prevention and control policy for COVID-19 at that time [33]. In addition, we noticed a decline in the trends of typical antipsychotic prescriptions and off-label usage from 2016 to 2021, whereas the proportions of SGAs increased, thus mirroring the findings in other studies [31, 32, 34, 35]. For example, the prevalence of SGAs increased from 1997 to 2005 and from 2004 to 2014 in studies conducted in Taiwan [31] and Hong Kong [32], China.

Furthermore, we observed that 61.7% of the usage was off-label, which is consistent with findings from previous studies [12, 32, 36]. Due to the vague definition of a suitable population of children and adolescents approved by the NMPA or FDA, we focused solely on off-label indications in our analysis. As a result, the proportion of off-label usage may be lower than the actual situation. Moreover, prescriptions for treating bipolar disorders with medications such as olanzapine, quetiapine, and risperidone were classified as on-label usage. This contributed to the lower rate of off-label usage.

In our study, age, sex, region, years, department, antipsychotic type, drug expenses, polypharmacy frequency and number of diagnoses were found to be associated with off-label prescriptions, which was essentially consistent with the findings of other studies [37,38,39,40]. However, the risks for off-label prescriptions of female sex and SGAs were greater in our study, which was different from the results obtained in the Netherlands [40] and United States [39]. Additionally, we observed that the risk of off-label usage declined in 2018 and 2019 and then increased in 2020 and 2021, which may be related to the COVID-19. Studies have shown that antipsychotic prescriptions increased after the outbreak of COVID-19 [33, 41, 42]. However, there is no direct evidence showing that increasing off-label antipsychotic prescriptions are associated with COVID-19, and additional studies on this topic are needed.

The highest proportion of off-label prescriptions exceeded 90% for chlorpromazine. This was primarily due to its utilization in treating upper respiratory tract infections in children and adolescents. Chlorpromazine was approved for the treatment of schizophrenia, nausea and vomiting instead of upper respiratory tract infections. However, as one of the components of the lytic cocktail, chlorpromazine has been widely used for preoperative sedation since its introduction [43, 44]. In addition, chlorpromazine has been shown to cause cooling and respiratory inhibition [45, 46]. These findings may contribute to the widespread usage of chlorpromazine in children and adolescents.

For SGAs, the most frequent off-label antipsychotic was aripiprazole, followed by quetiapine and olanzapine. Aripiprazole has been approved by the NMPA solely for the treatment of schizophrenia. Correspondingly, FDA approval included additional indications, such as the treatment of bipolar disorder, irritability associated with autistic disorder, and Tourette’s disorder. Concerning olanzapine, the FDA has approved its use for treating resistant depression, whereas the NMPA has not granted approval for this indication. With regard to off-label use for treating depressive states, some RCTs have shown that aripiprazole and quetiapine could be used as adjunctive treatments for major depressive disorder [47,48,49,50,51,52]. These findings may be related to the off-label use of aripiprazole, quetiapine and olanzapine.

The increase in SGAs from 2016 to 2021 may be related to the increase in mood or affective disorders and neurotic, stress-related and somatoform disorders in children and adolescents. For these diseases, the rates of off-label prescriptions were greater than those for other diseases and increased over the six years. This may be responsible for the increasing proportions of off-label usage in SGAs. Except for schizophrenia, the most common diagnoses were depressive state and mental disorders, which were classified as off-label usage in SGAs, thus causing a higher rate of off-label prescriptions compared to FGAs.

One of the most common diagnoses in our study was behavioral and emotional disorders, with onset usually occurring in childhood and adolescence (F90-F98), which was consistent with the results of other studies [9, 21, 31, 53]. In accordance with a study conducted in Taiwan [31], the most common diagnosis in pediatric patients in our study was tic disorder (F95, 22.0%). However, studies conducted in the United States and Germany indicated that the most common diagnosis in children and adolescents who received antipsychotics was hyperkinetic syndrome of childhood (F90) [9, 12, 39, 53]. Additionally, our study revealed a high prevalence of mood or affective disorders among pediatric patients. Furthermore, we observed a significant increase in the prescription rates of antipsychotics, including those for off-label use, for treating mood or affective disorders in this population between 2016 and 2021. These findings underscore the need for prioritizing psychological issues in children and adolescents and advocating for evidence-based and standardized approaches to antipsychotic usage in pediatric patients.

In our study, we observed that antipsychotic monotherapy was of high frequency, constituting more than 50%. For the schemes containing two drugs, antipsychotics were often combined with antidepressants for the treatment of depressive states, thus causing the highest risk in polypharmacy with two drugs compared to other polypharmacy. Additionally, polypharmacy with two or more antipsychotics is common in children and adolescents. This may lead to greater harm due to the increased incidence of adverse reactions compared to antipsychotic monotherapy. According to guidelines [6, 54, 55], the primary choice is to choose monotherapy for treatment, and clozapine or polypharmacy with antipsychotics are recommended when monotherapy was ineffective. However, evidence-based polypharmacy is needed. Currently, there is insufficient evidence to support the superiority of antipsychotic polypharmacy over antipsychotic monotherapy in terms of effectiveness [56].

In our study, we observed that patients with common off-label usage and polypharmacy of antipsychotics were mostly diagnosed with a depressive state. Studies have suggested the efficacy of certain antipsychotics as adjunctive treatments for depressive disorder [48, 50]. Meanwhile, certain antipsychotics have received approval for treating depressive symptoms, such as sulpiride. These may lead to the widespread use of antipsychotics in depressive pediatrics. Furthermore, it is worth noting that depressive state was not the same as depressive disorder. It is widely recognized that during both the acute and remission phases of schizophrenia, individuals may experience depressive symptoms. Additionally, among individuals diagnosed with bipolar disorder, it was common for them to experience periods marked by depressive symptoms. Therefore, depressive state may represent depressive symptoms during episodes of schizophrenia and bipolar disorder. This may also be responsible for the use of antipsychotics in pediatric patients with depressive states.

In alignment with depressive symptoms, antidepressants emerged as the most frequently prescribed medication in association with antipsychotics. Among these medications, sertraline was identified as being the most frequent drug for children and adolescents. Similar to fluvoxamine and fluoxetine, sertraline has been approved for the treatment of depression. However, none of these three drugs were authorized for depression treatment in children and adolescents, except for fluoxetine, which obtained FDA approval for major depressive disorder in the pediatric population over the age of 8 years. Concerning another frequently prescribed non-antipsychotic drug in our study, lithium has received approval for the treatment of mania, bipolar disorder, schizoaffective disorder, and recurrent major depressive disorder in children and adolescents over the age of 12 years. It was also authorized for bipolar disorder in the pediatric population over 7 years of age according to FDA indications. Given the high prevalence of mood or affective disorders (F30-F39), these medications were commonly prescribed.

There were several limitations in our study. First, the prescriptions were extracted from outpatients instead of from hospitalized patients. As a result, the findings of our study may not reflect all the situation of the entire pediatric population. Nevertheless, given the larger number of outpatients compared to inpatients, which provides a more substantial reference for medical professionals and the public, we believe that the results from outpatients may better reflect the pediatric population as a whole than those from inpatients. Meanwhile, other studies related to antipsychotic medications also use outpatients as their research population [9, 21, 57]. Second, certain diagnostic information within the database lacked standardization. For instance, certain prescriptions associated with bipolar disorder diagnoses did not provide specific details regarding the type of bipolar disorder, thus posing challenges in determining whether the use was on-label or off-label. In our study, these prescriptions were identified as on-label prescriptions if bipolar disorder patients were treated with olanzapine, quetiapine, or risperidone. Finally, in terms of factors associated with off-label prescriptions, some unknown factors may not have been included in the multivariate logistic regression analysis. Therefore, the unknown confounding factors could not be adequately adjusted for in our analysis.