Risky Behaviours Amongst Catalan Children With ADHD: Does Pharmacological Treatment Improve Outcomes?

Abstract

Objectives

Attention-Deficit/Hyperactivity Disorder (ADHD) is the most common mental health disorder amongst young children. Prevalence rates around the world range from 3-7% of school-aged children. Notwithstanding, there is an ongoing debate about the existence of over-diagnosis across developed countries because ADHD is controversial due to the fact that stimulant medication is typically administered as part of the treatment regime. We test whether ADHD plus medication use decreases the likelihood of risky behaviour or injuries (sexual behaviour, alcohol, tobacco and drug consumption) amongst children aged 6-18. Our objective is to explore whether the diagnosis of ADHD and subsequent treatment could alter behavioural choices made by children and their parents which could arguably reduce prevalence of injuries or risky conduct. We explore both medication and diagnosis impact on risky behaviours. In doing so, we make use of a scale that contains alternative definitions of ADHD in which over-diagnosis is identified.

Methods

We use a large administrative dataset that includes information from several providers for the whole population of Catalan children that were born between 1998 and 2012 (1,225,406 individuals), including those diagnosed with ADHD and those not. This database contains information on: primary care, mental health and all other hospitalizations, emergency care, and community mental health care. We established several definitions of children diagnosed with ADHD: (i) a measure provided by the public administration, (ii) those diagnosed with the ICD-9 codes related to ADHD in any health care provider, (iii) children consuming drugs related to the disease, and; (iv) our own classification informed by clinical expertise. Specifically, with regards the latter, we categorized children into three groups: (i) true ADHD diagnosis; (ii) likely ADHD diagnosis, and; (iii) not very likely ADHD diagnosis.

We used a quasi-experimental approach consisting of using a Difference-in-difference (DiD) strategy where each children is his/her own control. This can be done using only the ‘compliers’, that is, those ever diagnosed with ADHD. This reduces the risk of confounding given that it compares individual risky behaviour before/after date of diagnosis or treatment. We used the same strategy used by Dalsgaard et al. (2015) but taking into account the appropriate time zero span (Hernán and Robins, 2016).

Results

Results suggest that, in line with previous literature (Dalsgaard et al., 2015 and Chorniy and Kitashima, 2016), ADHD and medication use reduce the prevalence of risky behaviours, whilst for those children with ADHD and not on medication, the risks are increased. Our study contributes to the scant evidence on the effects of ADHD treatment on children’s outcomes and suggests pharmacological treatment has substantial benefits.