Article Text
Abstract
Background Attention-deficit hyperactivity disorder (ADHD) in childhood can have adverse effects on mental health, learning, and employment opportunities. There is evidence of a relationship between socioeconomic disadvantage and likelihood of ADHD in childhood; however, most studies use area-level measures which may underestimate inequality compared to individual/family-based measures. This study aimed assess whether area-level and individual-level measures of social disadvantage were predictive of child ADHD via dispensed prescriptions.
Methods We used birth data for all children born in Scotland 2010–2012 (n=195,419) linked to Prescription Information System up to March 2018. Two measures of socioeconomic circumstances (SECs) at the child’s birth were used: Scottish Index of Multiple Deprivation (SIMD) (area-based), and four class measure of the NS-SEC of the mother (individual-based). Prescription use was defined as a record of any dispensed prescription for ADHD up until March 2018 (median age 6). We used binary logistic regression to estimate risk ratios (RRs) for prescription uptake by each SEC measure before and after adjusting for covariates/confounders (sex, age of child as of March 2018, number of births in pregnancy, mother’s age at first live birth, mother’s birth country, relationship status of parents).
Results Prescription use varied by area deprivation (0.58% of children born in the most deprived SIMD decile compared to 0.14% in the least deprived) and social class (0.62% for children born to mothers in the unemployed/other social class compared to 0.16% in the managerial/professional social class). The strength of association narrowed slightly after adjustment for confounders/covariates; the fully adjusted RR for prescription use was 2.14 (95% C.I.: 1.33–3.44) in the most compared to least deprived SIMD decile and 2.32 (95% C.I.: 1.78–3.04) for children born to mothers who were unemployed/other compared to managerial/professional. After mutual adjustment for both SEC measures, the effect of SIMD was reduced whilst that of mother’s social class remained consistent.
Conclusion Both area-level and family-level deprivation at birth are associated with increased prescription use among young children in Scotland, suggesting disadvantaged SEC are associated with higher prescription use. Inequalities in ADHD prescriptions across childhood may not be fully captured since prevalence increases with age (and our data only follow children up to the maximum age of 8). Prescription data may underestimate prevalence of ADHD as not all children with ADHD symptoms will be diagnosed and/or prescribed (and this may vary by SEC). Future analyses will explore this using data from child health checks.