Abstract
Many psychotropic drugs have not been properly tested in clinical trials for efficacy and safety in children and adolescents. The increasing use of psychotropic drugs in young people is better established in North America; however, it is unclear to what extent these studies are applicable to UK practice. Using data from the UK General Practice Research Database (GPRD), an overall prevalence of psychotropic drug prescribing trend was initially carried out involving children and adolescents, aged 0-18 years, between January 1992 and December 2000. Upward prevalence of prescribing was observed in stimulants (aged 3-18 years), antidepressants (aged 13-18 years) and anticonvulsants (aged 6-18 years). Following this finding, two drug utilisation studies on stimulants and anticonvulsants were carried out. The use of antidepressants was not included, as its prescribing pattern had already been published elsewhere. An increasing prescribing of stimulants was attributable to methylphenidate use. This was coincident with the change of methylphenidate licence status in 1995. The prevalence of methylphenidate prescribing significantly increased from 0.01 per 1000 patients (95%CI 0.004-0.02) in 1992 to 2.13 per 1000 patients (95%CI 1.92-2.36) in 2000. The age-specific prevalence was highest between the aged 10-12 years. The most commonly diagnosis associated with its prescription was behavioural disorders. As some patients were taking methylphenidate as needed and some individuals may have stopped taking it during school holiday, this was problematic to measure methylphenidate exposure using a computerised database. Consequently, it was not possible to conduct further safety studies for methylphenidate use. The second drug utilisation study on anticonvulsants has shown the incidence of prescribing remain stabled from 1.40 per 1000 person-years (95%CI 1.41-1.57) to 1.05 per 1000 person-years (95%CI 0.90-1.22) throughout study period. The rate was highest in younger children aged 0-2 years. The use of conventional anticonvulsants appeared to decline whilst the use of newer anticonvulsants constantly increased. The majority of subjects prescribed anticonvulsant had a diagnosis of epilepsy. A small proportion of anticonvulsant-treated subjects died during the study period. As sudden unexpected death in epilepsy (SUDEP) is a phenomenon which is not fully understood at this time, hence, a descriptive case series study was conducted to assess the causes of death. Unfortunately, this was not a successful approach due to the incomplete information recorded in the database. In addition, we did not utilise GPRD verification service to obtain additional information. As a result, this study did not provide any compelling evidence. Due to the concern of an increasing use of lamotrigine (newer anticonvulsant) and the limited evidence of risk cutaneous reactions associated with anticonvulsants in paediatric population, a case-control study was subsequently conducted to investigate the association of severe cutaneous reactions (Stevens-Johnson syndrome, toxic epidermal necrolysis, erythema multiforme, exfoliative dermatitis) and anticonvulsants use. There was no evidence to show an increased risk of severe cutaneous reaction associated with anticonvulsant use (odd ratio: 0.93; 95%CI 0.48- 2.05), several risks factors were found significantly associated with these conditions. It is suggested to use case-crossover study design to reassure our findings.This thesis has illustrated strengths and limitations of using a computerised primary care database in the field of paediatric psychiatric research. It has provided an in- depth understanding of psychotropic drug prescribing patterns in young people in UK practice. Comparing it with published studies from North America, there is variation of prescribing trends between countries. It is suggested that GP questionnaire should be utilised in order to obtain additional information to confirm diagnosis and treatment. To collaborate with other countries in order to document psychotropic drug use in paediatric population at the multinational level is needed. Considerable research is still required to ensure that psychotropic drugs can be used appropriately in clinical practice.