The comorbidity between attention-deficit/hyperactivity disorder and other neurodevelopmental disorders : aetiology, treatment and outcomes

Abstract

Attention-deficit/hyperactivity disorder (ADHD) is a neurodevelopmental disorder (ND) defined by the presence of impairing levels of inattentiveness, hyperactivity, and impulsivity. These symptoms are often accompanied by impairment in several functional domains, and by the presence of symptoms or diagnoses of other disorders, especially other NDs, including Autism Spectrum Disorder (ASD). In addition, risky behaviours and health issues are more common among individuals with ADHD, as compared to their peers who do not display ADHD symptoms. The overall aim of this thesis is to investigate the comorbidity between ADHD, ASD and other NDs, in order to clarify shared aetiology, treatment effectiveness and adverse health outcomes.

Study 1 examined the association between clinically diagnosed ASD and ADHD across different types of relatives and explored potential differences between low- and highfunctioning ASD (that is, with or without intellectual disability) in the link with ADHD. Data for the study came from a linkage of national Swedish registers in order to identify different types of relatives, from twins to cousins, and clinical diagnoses of ASD and ADHD. Logistic regression was used to estimate the association between ASD and ADHD within the same person (within-individual association) and within relative pairs (within-family association). Results demonstrated that individuals diagnosed with ASD and their relatives had an increased risk of ADHD. The association in twins and siblings was higher than the association in cousins. The magnitude of the association was larger in high-functioning ASD. Study 2 focused on the phenotypic and aetiological overlap between traits related to ADHD and ASD in young adult twins from the general population. Data for the study came from a web-based survey within the Swedish Twin Registry. Four different trait dimensions were considered: inattention (IA), hyperactivity/impulsivity (HI), repetitive and restricted behaviours (RRB), social interaction and communication difficulties (SIC). Structural equation modelling was used to decompose the covariance across these trait dimensions into genetic and non-genetic influences. Results showed that at the phenotypic level, the correlation between IA and RRB was similar to the one between IA and SIC, while the correlation between HI and RRB was stronger than the one between HI and SIC. Genetic and non-genetic contributions accounted for a similar amount of the covariation across all trait dimensions under study. The largest genetic correlation between traits related to ADHD and traits related to ASD was between HI and RRB. Taken together, results from Study 1 and 2 suggest that comorbidity between ADHD and ASD may reflect shared aetiological factors, which are in part of genetic origin and which may be specific to certain symptom domains.

Study 3 tested the association between use of ADHD medication and risk of unintentional injuries in children and adolescents with ADHD, including those with co-occurring NDs. Data for the study came from a linkage of national Swedish registers. All residents in Sweden with at least one diagnosis of ADHD and one diagnosis of unintentional injury were included and followed during the study period. Follow-up time was divided into consecutive periods, which may be on-treatment or off-treatment, and the rate of injuries during periods on-treatment was compared to the rate of injuries during periods off-treatment within the same individual, using stratified Cox regression. Results indicated that ADHD medication use was associated with a lower rate of all unintentional injuries, among children and adolescents, among males and females, and among individuals with NDs, as well as among the subgroup with ASD. Study 4 investigated the association between different NDs and the risk of violent victimization in adolescents and young adults, considering the role of familial and mediating factors. Similarly to Study 1 and 3, a linkage of national Swedish registers was used to identify diagnoses of different NDs and inpatient or outpatient visits or deaths due to assault in the study population. The association between the NDs and violent victimization was explored using Cox regression. Results revealed that being diagnosed with any ND was associated with an increased risk of later violent victimization in males and females. After adjustment for familial factors and mediators, all the associations were attenuated and only ADHD was associated with an increased risk of violent victimization among males and females. Taken together, results from Study 3 and 4 suggest that comorbidity between ADHD and other NDs does not seem to affect treatment effectiveness with regard to ADHD medication and injuries. On the other hand, risk of violent victimization, which seemed to be related to NDs as a group, may be specifically linked to ADHD.

In conclusion, the work presented in this thesis supports the notion that NDs are a group of disorders characterised by both general and specific aspects in terms of aetiology, treatment effectiveness and negative outcomes.