National record-linkage study of hospital admissions for schizophrenia in childhood and adolescence in England

Background There is a lack of information on changes in hospital admission rates for childhood-onset schizophrenia (COS), or on patient characteristics, to inform clinical research and health service provision. Aims To report age- and sex-specific incidence rates of hospital admissions and day patient care for schizophrenia (ICD-10 F20) and non-affective psychosis (ICD-10 F20-29), by year of occurrence and age, in childhood and adolescence. Methods Population-based study using person-linked data for England (available 2001-2016); time-periods in single years and 4-year groups. Results Hospitalised incidence for schizophrenia increased with increasing age, from 0.03 (95% confidence interval (CI) 0.02-0.05) and 0.01 (0-0.01) per 100,000 in, respectively, males and females aged 5-12 years, to 3.67 (3.44-3.91) in males and 1.58 (1.43-1.75) in females aged 13-17 years. There was no gender difference in hospitalised incidence rates in children aged 5-12, but in 13-17 years old, there was a male excess. Rates for schizophrenia were stable over time in 5-12 years old. In ages 13-17, rates for schizophrenia decreased between 2001-2004 and 2013-2016 in males, from 6.65 (6.04-7.31) down to 1.40 (1.13-1.73), and in females from 2.42 (2.05-2.83) to 1.18 (0.92-1.48). The hospitalisation rates for schizophrenia and non-affective psychosis, combined, in 13-17 years old decreased in males from 14.20 (13.30-15.14) in 2001-2004 to 10.77 (9.97-11.60) in 2013-2016, but increased in females from 7.49 (6.83-8.20) to 10.16 (9.38-11.00). Conclusions The study confirms that childhood-onset schizophrenia is extremely rare, with only 32 cases identified over a 15-year period in the whole of England. The incidence of schizophrenia and non-affective psychosis increased substantially in adolescence; however, the marked reduction in the proportion of those diagnosed with schizophrenia in this age group suggests a possible change in diagnostic practice.