OBJECTIVE. The goal was to investigate the epidemiologic features of antipsychotic prescribing to children and adolescents in general practice in the United Kingdom.
METHODS. A total of 384 participating general practices from the United Kingdom General Practice Research Database were used to identify patients 0 to 18 years of age who were prescribed ≥1 antipsychotic medication between January 1, 1992, and December 31, 2005. Annual age-specific prevalences and incidences of antipsychotic prescribing were calculated.
RESULTS. The overall prevalence of use of all antipsychotics increased from 1992 (0.39 users per 1000 patient-years) to 2005 (0.77 users per 1000 patient-years). The prescribing prevalence for patients 7 to 12 years of age almost tripled between 1992 (0.23 users per 1000 patient-years) and 2005 (0.61 users per 1000 patient-years). Atypical antipsychotic prescribing increased 60-fold from 1994 (0.01 users per 1000 patient-years) to 2005 (0.61 users per 1000 patient-years). However, typical antipsychotic prescribing decreased significantly from 2000 (0.44 users per 1000 patient-years) to 2005 (0.18 users per 1000 patient-years). The incidences for typical and atypical antipsychotics showed trends similar to those of the respective prevalences. However, the overall incidence (number of new starters) for all antipsychotics was relatively stable between 1992 and 2005, which suggests that patients remain on treatment longer.
CONCLUSIONS. The overall prevalence of antipsychotics almost doubled between 1992 and 2005; however, the rate of increase was much lower than the reported figures in the United States. The prescribing of atypical antipsychotic drugs has increased despite the lack of conclusive evidence showing their superiority over older conventional antipsychotics. Additional investigation is required to evaluate their efficacy and safety in children and adolescents.
General practitioners in the United Kingdom play an important role in prescribing and monitoring antipsychotic drugs for children with emotional and behavioral problems in the community. Antipsychotic prescriptions for children are usually initiated by specialists in secondary care, such as child psychiatrists and behavioral pediatricians; however, general practitioners continue to monitor and to prescribe these drugs.1 Antipsychotic drugs not only are used for the treatment of psychotic disorders, such as schizophrenia, but also have been shown to be effective in the treatment of pervasive developmental disorders, disruptive behavioral disorders, aggression, and tic disorders.2–5 So-called atypical antipsychotic drugs were introduced in the late 1980s with the licensing of clozapine, followed in the 1990s by a variety of other medications, including risperidone, olanzapine, and quetiapine, which shared some of the advantages of clozapine, particularly the reduced incidence of movement disorders.2,6,7
However, a review by Cheng-Shannon et al7 found that the use of atypical antipsychotics by children and adolescents was associated with many of the same adverse effects as seen with the older, typical, antipsychotic drugs. Concerns regarding the safety of atypical antipsychotics (for example, the occurrence of neuroleptic malignant syndrome, weight gain, and cardiac and metabolic effects such as QTc prolongation and hyperprolactinemia) also have been raised.7–10 More recently, in a meta-analytic review of trials of antipsychotics in pediatric patients, Armenteros and Davies11 concluded that atypical antipsychotics had no efficacious advantage, produced more-severe sedation and weight gain, and had similar rates of extrapyramidal symptom occurrence, compared with typical antipsychotics.
Studies in the United States and the Netherlands showed that antipsychotic use in children is increasing. In the Netherlands, there was a 1.5-fold increase in new users of antipsychotic drugs between 1995 and 1999 (from 1.1 to 1.7 users per 1000 children); the prevalence of antipsychotic use increased twofold in the same period (from 1.6 to 3.4 users per 1000 children).12 A study in the United States showed that the number of new users (2-18 years of age) of antipsychotics almost doubled between 1996 and 2001 (from 23 to 45 users per 10000 children).13 In the United States, a recent study estimated a sixfold increase in the number of “office-based visits by youth that included antipsychotic treatment” from 1993 to 2002.14 This rapid increase in antipsychotic prescribing is of concern because there is little robust evidence of a corresponding increased prevalence of psychotic illness in young people.15 A previous study suggested that the number of psychotropic drug prescriptions for children in the United Kingdom increased by 68% between 2000 and 2002, which is much greater than the rate of increase in the United States and other countries.16 Therefore, the aim of this study was to investigate the trends in antipsychotic prescribing to children and adolescents in United Kingdom general practice.
Data for this study were obtained from the General Practice Research Database (GPRD). The history of the GPRD and validation of its data are described elsewhere.17,18 Data from the GPRD are collected from anonymized patient records from participating general practitioners in the United Kingdom. Currently, the GPRD contains >35 million patient-years of data from ∼400 United Kingdom general practices, with 789467 registered patients ≤19 years of age.19
Data available from the GPRD include patient demographic data, patient registration details, practice details, therapy records (including medicines prescribed for patients), consultation details (with a unique identifier for each consultation), clinical records, laboratory test results, and referrals. However, the GPRD does not provide information regarding the filling of prescriptions.
The study population encompassed all patients in the GPRD who were 0 to 18 years of age (ie, the denominator). All prescriptions for this study population were screened for antipsychotics. Patients who received ≥1 prescription for an antipsychotic during the defined study period (between January 1, 1992, and December 31, 2005) were classified as case subjects in the study (ie, the numerator). Ethics approval was granted by the GPRD Scientific and Ethical Advisory Group and the Independent Scientific Advisory Committee for Medicines and Healthcare Products Regulatory Agency database research.
Patients had ≥1 year of research standard data available, a known gender, and an acceptable patient registration status. Subjects who were registered temporarily with a practice were excluded.
Antipsychotic drugs were categorized as typical or atypical according to the British National Formulary (Table 1). 20 Data were extracted by using BusinessObjects (Business Objects, Paris, France), which is the standard tool supplied by the GPRD for querying the database.
Data analysis was conducted by using Stata SE 9.1 (Stata Corp, College Station, TX). Once all patients in the study population were identified, prescribing data for the first year for each patient were screened for types of prescriptions. If patients received their first antipsychotic prescription after this screening period, then they were classified as a new starter and counted as an incident case. Incidence was defined as the frequency of new starters of an antipsychotic in the general pediatric population. Incidence was calculated as incidence = (NI × 1000)/K, where NI is the number of patients with a new prescription for an antipsychotic in a particular year and K is the total number of patient years in the GPRD population ≤18 years of age in a particular year.
Prevalence was defined as the number of patients who were prescribed antipsychotic treatment in the general pediatric population. Prevalence was calculated as prevalence = (Np × 1000)/K, where Np is the number of patients with a prescription for an antipsychotic in a particular year and K is the total number of patient years in the GPRD population ≤18 years of age in a particular year.
The prevalence values were stratified according to 6-year age bands (0–6, 7–12, and 13–18 years of age). A χ2 test (Cochran-Armitage test for trend) was used to examine the yearly trend of antipsychotic drug prescribing.
The duration of each prescription for an antipsychotic was calculated from the prescribed quantity and daily dosage. The prescription duration was calculated as prescription duration = Q/D, where Q is the quantity of drug prescribed and D is the daily dosage of the prescribed drug. For example, a patient who was prescribed 56 tablets with a daily dose of 2 tablets per day would have a prescription duration of 28 days.
For psychiatric conditions, general practitioners are usually informed of a diagnosis requiring specialist medications from secondary or tertiary care. Unfortunately, there is no direct link between medical diagnoses and prescriptions in the GPRD. Therefore, all clinical records for patients in the study cohort were screened for possible diagnoses associated with antipsychotic prescriptions. The clinical codes were presented as Read and Oxford Medical Information System codes, which consist of syndrome, symptom, and clinical event descriptors. The codes were mapped by Mr Rani into 5 different syndromal categories derived from the International Classification of Diseases, 10th Revision. The classification of codes was checked by Dr Byrne. The 5 categories were (1) psychosis, schizophrenia, hallucinations, delusions, and paranoia; (2) behavior, conduct, and personality disorders; (3) autism and Asperger syndrome; (4) depression and mood and affective disorders; and (5) tics and Tourette syndrome. The numbers of patients with diagnoses in the different categories within 6 months before or after their first prescription were then investigated. Because there is no direct link between diagnoses and prescriptions, the first prescription for each subject may be associated with >1 diagnosis.
During the study period, 16497 patients received a total of 38286 prescriptions for antipsychotic drugs. Prescriptions for prochlorperazine were subsequently excluded from the analysis, however, because prochlorperazine is often prescribed for the treatment of nausea and vertigo and is not usually administered for treatment of psychotic disorders in United Kingdom primary care. After exclusion of prescriptions for prochlorperazine, a total of 2767 patients received 21089 prescriptions for antipsychotic drugs. There were 1871 patients who received a total of 10135 prescriptions for typical antipsychotics, with a median of 2 prescriptions per patient (interquartile range: 1–4 prescriptions), and 1103 patients who received 10954 prescriptions for atypical antipsychotics, with a median of 5 prescriptions per patient (interquartile range: 2–12 prescriptions). There were 207 patients who were prescribed both atypical and typical antipsychotic drugs (Table 1). The median prescription duration for all antipsychotic drugs was 28 days per prescription (interquartile range: 23–30 days for all antipsychotics, 20–33 days for typical antipsychotics, and 28–30 days for atypical antipsychotics).
The most commonly prescribed atypical antipsychotic was risperidone (73% of all prescriptions for atypical antipsychotics), followed by olanzapine (16%) and quetiapine (7.3%). The most commonly prescribed typical antipsychotic was thioridazine (27% of all prescriptions for typical antipsychotics), followed by haloperidol (26%) and chlorpromazine (19%). The overall proportions of antipsychotic prescriptions according to type of drug and number of patients are given in Table 1.
Figure 1 shows that the overall prevalence of use of all antipsychotics (either atypical or typical) almost doubled from 1992 to 2005 (from 0.39 to 0.77 users per 1000 patient-years; P < .01). Typical antipsychotic prescribing rates were stable from 1992 to 2000 and then decreased significantly from 2000 (0.44 users per 1000 patient-years; 95% confidence interval [CI]: 0.38–0.49 users per 1000 patient-years) to 2005 (0.18 users per 1000 patient-years; 95% CI: 0.15–0.21 users per 1000 patient-years; P < .01). In contrast, Fig 1 shows that the prevalence of atypical antipsychotic prescriptions increased steadily, from 0.006 users per 1000 patient-years (95% CI: 0.002–0.02 users per 1000 patient-years) in 1994 (when the first atypical antipsychotic prescription was recorded) to 0.61 users per 1000 patient-years (95% CI: 0.55–0.67 users per 1000 patient-years) in 2005 (P < .01).
Figure 2 shows the incidence (new starters) of both typical and atypical antipsychotics, mirroring the prevalence shown in Fig 1, with a steady increase in new starters of atypical antipsychotics and a decrease in new starters of typical antipsychotics. However, unlike the prevalence of use of all antipsychotics shown in Fig 1, the overall incidence of use of all antipsychotics was stable, without a significant increase between 1992 (0.30 users per 1000 patient-years; 95% CI: 0.24–0.37 users per 1000 patient-years) and 2005 (0.33 users per 1000 patient-years; 95% CI: 0.29–0.37 users per 1000 patient-years; P > .10).
Figure 3 shows the prevalence of use of all antipsychotics according to 6-year age bands (0–6, 7–12, and 13–18 years of age). The prevalence of use of antipsychotics for teenagers (13–18 years) increased significantly from 1992 (1.09 users per 1000 patient-years; 95% CI: 0.87–1.31 users per 1000 patient-years) to 2005 (1.57 users per 1000 patient-years; 95% CI: 1.41–1.73 users per 1000 patient-years; P < .001). However, the prescribing of antipsychotics to patients in middle childhood (7–12 years of age) showed the most significant increase throughout the study period, from 1992 (0.23 users per 1000 patient-years; 95% CI: 0.15–0.34 users per 1000 patient-years) to 2005 (0.61 users per 1000 patient-years; 95% CI: 0.51–0.71 users per 1000 patient-years; P < .0001).
A total of 1557 patients (56% of all patients) had diagnoses in ≥1 of the 5 categories within 6 months of their first prescription. The distribution of patients with diagnoses according to different age bands is presented in Fig 4.
There were greater proportions of patients with diagnoses from categories 2 (behavior, conduct, and personality disorders) and 3 (autism and Asperger syndrome) in early childhood (2–6 years of age) and middle childhood (7–12 years of age), compared with adolescent patients (13–18 years of age). In contrast, adolescent patients had greater proportions of diagnoses from categories 1 (psychosis, schizophrenia, hallucinations, delusions, and paranoia) and 4 (depression and mood and affective disorders), compared with patients in the other age groups. Seventeen percent of patients in the 13- to 18-year age band had a diagnosis related to category 1, compared with 2% of patients in middle childhood and 0.8% of patients in early childhood.
Comparison With Utilization Studies in the United States
To our knowledge, this is the first large pediatric cohort study to examine the epidemiologic features of antipsychotic prescribing in the United Kingdom. Our results raise many important issues related to the prescribing of antipsychotic medications in the pediatric population.
Our results showed a very different picture from the study by Olfson et al14 in the United States, which showed a sixfold increase in antipsychotic treatment between 1993 and 2002. During the same period of our study, there was only a 36% increase in the prescribing of antipsychotics. The difference in prevalence, compared with the study in the United States, may be attributable to a number of factors. First, different methods were used. Although the study by Olfson et al14 used outpatient practices, similar to our study, the authors reported on office-based visits, where there could be duplication of patients and hence overestimation of antipsychotic prescribing prevalence. Their study population included patients ≤20 years of age, whereas the study population in our study was ≤18 years of age. Another possible reason for the difference is the difference in prescribing practices between physicians in the United States and physicians in the United Kingdom. Indirect evidence comes from the prescribing of psychostimulants; the prevalence of stimulant prescribing for children in the United States was estimated as 1% to 6%, compared with 0.03% for the pediatric population in the United Kingdom.21
Another important finding was that the overall prevalence of antipsychotic prescribing almost doubled between 1992 and 2005; however, the overall incidence was relatively stable during the same period. This suggests that patients are being prescribed antipsychotics for longer durations, rather than that more new patients are starting treatment. This is in sharp contrast to the United States data published by Cooper et al,13 who reported that the number of new users (2–18 years of age) of antipsychotics almost doubled (from 23 to 45 users per 10000 children) from 1996 to 2001.
Effect of Restricted Indications for Thioridazine
The most commonly prescribed typical antipsychotic in this study was thioridazine, which represented 27% of all typical antipsychotic prescriptions. In December 2000, the Committee on Safety of Medicines restricted the indications of thioridazine to “the second line treatment of schizophrenia in adults” and also declared that the “balance of risk and benefits is unfavourable for its previous indications (anxiety, agitation and restlessness in the elderly, moderate to severe psychomotor agitation, violent and dangerously impulsive behavior, mania or hypomania, and behavioral disorders and epilepsy in children)” because of cardiotoxicity associated with its use.22 The effect of this warning is evident, in that thioridazine prescribing decreased significantly after 2000, which slowed the increasing trend of overall antipsychotic prescribing.
Antipsychotics in Early and Middle Childhood
In the United Kingdom, atypical antipsychotic prescribing to patients ≤18 years of age is “off-label” or use outside the product license, except for amisulpride, risperidone, and clozapine. Amisulpride and risperidone may be prescribed for patients ≥15 years of age with schizophrenia with positive and/or negative symptoms, and clozapine may be given to patients ≥16 years of age with treatment-resistant schizophrenia that is unresponsive to other antipsychotics. However, there has been increasing interest in the use of atypical antipsychotics for the treatment of other psychiatric conditions, such as bipolar disorder, obsessive-compulsive disorder, autism, tic disorder, and Tourette syndrome, in children and adolescents.8,23 Analysis according to age bands showed that the increase in prescribing was greatest in middle childhood (Fig 3). Data on the associated diagnoses showed that adolescent patients in the 13- to 18-year age band had greater proportions of diagnoses related to schizophrenia and psychosis, compared with patients in early and middle childhood (Fig 4). This demonstrates that schizophrenia spectrum disorders are less prominent in early and middle childhood, compared with adolescence, and therefore are unlikely to account for this upward trend in antipsychotic prescribing. This suggests that these drugs are used for indications other than psychotic disorders. Furthermore, our results showed that patients in early and middle childhood had relatively greater proportions of diagnoses related to behavior disorders and autism spectrum disorders, compared with patients in the 13- to 18-year age band. This accentuates the need for more-rigorous investigations concerning the use of antipsychotic medications for such indications in youths.
A recent study suggested an increase in the prevalence of autism spectrum disorders.24 Although the reason for this increase is unclear, that is, whether the increase is attributable to better ascertainment, broadening diagnostic criteria, or increased incidence,24 such trends may fuel a demand for medications to modify behaviors associated with developmental disorders.
Increasing Trend of Antipsychotic Prescribing
Results of the study also showed an increase in atypical antipsychotic prescribing and a decrease in typical antipsychotic prescribing during the study period (Fig 1); this implies that the use of typical antipsychotics in the pediatric population has been largely replaced by the use of atypical antipsychotics. Interestingly, a similar trend in antidepressant prescribing to young people was observed between 1992 and 2001, when selective serotonin reuptake inhibitors gained popularity over tricyclic antidepressants despite a paucity of research on their efficacy and safety.25 Subsequently, the United Kingdom regulator, the Medicines and Healthcare Products Regulatory Agency, contraindicated the use of selective serotonin reuptake inhibitors other than fluoxetine as new treatment for patients <18 years of age with depressive illness after concerns were raised about their safety and the increased risk of suicidal behavior associated with their use.26 Do we have sufficient evidence to ensure that history will not be repeated in antipsychotic prescribing? Although scores of clinical trials have been devoted to adults with schizophrenia, relatively few trials of reasonable quality have focused on children with this condition in the past 20 years.11 Many trials involving a younger population were conducted with small sample sizes and short durations of study. Experience from the antidepressant saga raises the possibility that publication bias may exist for studies investigating the effects of antipsychotics.27 There have been numerous reports of adverse drug reactions associated with atypical antipsychotic use in children and adolescents, including neuroleptic malignant syndrome, weight gain, metabolic abnormalities, diabetic ketoacidosis, hepatotoxicity, and hyperprolactinemia.7,8,28 This highlights the need for long-term safety investigations and ongoing clinical monitoring of atypical antipsychotic use in children and adolescents, particularly if the prescribing rate of these medicines continues to increase.
It was recently reported that most physicians have financial relationships with the pharmaceutical industry, including receiving drug samples, payments, or reimbursements for costs associated with professional meetings.29 However, the Medicines Act in the United Kingdom prohibits the pharmaceutical industry from promoting unlicensed and off-label uses of medicines. Therefore, it is unlikely that the trends shown by our results are influenced by such relationships, because the majority of antipsychotic medications are not licensed for our study population.
This study also raises the issue of the availability and appropriateness of properly established antipsychotic therapy guidelines for the young population. International readers may be surprised to learn that the current National Institute for Health and Clinical Excellence guidance on schizophrenia does not cover patients <18 years of age.30 Our finding of a significant increase in the trend of antipsychotic prescribing for pediatric patients underlines the need for improved clinical guidelines in this area. This would be helpful for general practitioners, because they are likely to be the first point of contact for the majority of patients and are the ones who continue to prescribe and to monitor antipsychotic treatments.
The study does not provide any direct evidence on the pathway into care (including secondary and tertiary care) that children follow that results in prescription of an antipsychotic or on the quality of diagnostic decisions. The study calculated accurate prevalence and incidence rates of prescribing in primary care; however, we were unable to capture patients who were cared for solely in secondary or tertiary care or by prison services. Because the United Kingdom National Health Service is a primary care-driven health care system, general practitioners in primary care are the gatekeepers of referral to both secondary and tertiary care. Patients cannot proceed to secondary or tertiary care without primary care referrals.1 Furthermore, many children, including those with severe forms of mental disorders, are not referred for assessment to specialist services and remain in primary care.31 In addition, pediatric psychotropic prescription analyses have shown that psychotropic drug prescribing is becoming increasingly common in general practice.16 Therefore, the lack of data from secondary and tertiary care is unlikely to have a significant effect on the overall national trend shown by our results. Finally, not all general practitioners recorded the diagnosis in clinical records; this limits the analysis of diagnostic indications. Nevertheless, this is the largest United Kingdom pediatric cohort study to investigate the trends of antipsychotic drugs prescribing, and the results are representative of the trends occurring in United Kingdom general practice.
We have identified 3 major trends in the prescribing of antipsychotic medications for children and adolescents in the United Kingdom. First, the prevalence of antipsychotic prescribing in general practice almost doubled between 1992 and 2005. Second, the greatest increase is affecting children 7 to 12 years of age. Third, there is a strong trend for the replacement of typical antipsychotics with atypical antipsychotics. This increase in prescribing is predominantly for off-label and unlicensed indications. Although this trend is nowhere near the magnitude of such changes in the United States, where psychotropic prescriptions for young children are increasingly common, there may be grounds for caution. Additional studies are required to investigate durations of treatment, long-term outcomes and adverse effects of treatments, factors influencing discontinuation, and the information available to children and families regarding safe prescribing.
Professor Wong was supported by a Department of Health Public Health Career Scientist Award. The license for the GPRD was funded by the European Commission via the Taskforce European Drug Development for the Young Network of Excellence European Commission Framework 6 Programme (2005–2010).
- Accepted September 17, 2007.
- Address correspondence to Ian C.K. Wong, PhD, Centre for Pediatric Pharmacy Research, The School of Pharmacy, University of London, 29-39 Brunswick Square, London WC1N 1AX, United Kingdom. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject
There has been an increasing trend in antipsychotic prescribing in studies conducted in the United States and the Netherlands. There is little information regarding the prescribing trend for antipsychotics in the pediatric population in the United Kingdom.
What This Study Adds
This study provides descriptive analyses of the prescribing trends for antipsychotic medications to children and adolescents in United Kingdom primary care including the prescribing trends for typical and atypical antipsychotics as well as associated diagnoses according to age groups.
- ↵Frangou S, Byrne P. How to manage the first episode of schizophrenia. BMJ.2000;321 (7260):522– 523
- ↵Schirm E, Tobi H, Zito JM, de Jong-van den Berg LTW. Psychotropic medication in children: a study from the Netherlands. Pediatrics.2001;108 (2). Available at: www.pediatrics.org/cgi/content/full/108/2/e25
- ↵Sutcliffe AG, Wong ICK. Rational prescribing for children. BMJ.2006;332 (7556):1464– 1465
- ↵Wong ICK, Murray ML, Camilleri-Novak D, Stephens P. Increased prescribing trends of paediatric psychotropic medications. Arch Dis Child.2004;89 (12):1131– 1132
- ↵Medicines and Healthcare Products Regulatory Agency. The General Practice Research Database. Available at: www.gprd.com. Accessed June 27, 2006
- ↵Joint Formulary Committee. British National Formulary. 54th ed. London: British Medical Association and Royal Pharmaceutical Society of Great Britain; 2007
- ↵Rey JM, Sawyer MG. Are psychostimulant drugs being used appropriately to treat child and adolescent disorders? Br J Psychiatry.2003;182 (4):284– 286
- ↵Committee on Safety of Medicines. Thioridazine: restricted indications and new warnings on cardiotoxicity. Available at: www.info.doh.gov.uk/doh/embroadcast.nsf/f011981a95f31f4180256c07003d34a0/fc5d1f63a16da86b80256dad00480816?OpenDocument. Accessed June 13, 2006
- ↵Coghill D. Current issues in child and adolescent psychopharmacology, part 2: anxiety and obsessive-compulsive disorders, autism, Tourette's and schizophrenia. Adv Psychiatr Treat.2003;9 :289– 299
- ↵Murray ML, de Vries CS, Wong ICK. A drug utilisation study of antidepressants in children and adolescents using the General Practice Research Database. Arch Dis Child.2004;89 (12):1098– 1102
- ↵Committee on Safety of Medicines. Selective serotonin reuptake inhibitors: use in children and adolescents with major depressive disorder. Available at: www.info.doh.gov.uk/doh/embroadcast.nsf/0/183a970c6d74afad80256df800330c99/$FILE/CEM2003-20.doc. Accessed June 22, 2006
- ↵National Institute for Clinical Excellence. Schizophrenia: core interventions in the treatment and management of schizophrenia in primary and secondary care: clinical guideline 1, December 2002. Available at: guidance.nice.org.uk/CG1/niceguidance/pdf/English. Accessed April 17, 2007
- Copyright © 2008 by the American Academy of Pediatrics