What can genetically informative designs add to the understanding of ADHD?

Abstract

Attention-deficit/hyperactivity disorder (ADHD) is a highly prevalent and clinically heterogeneous neurodevelopmental disorder affecting approximately 3.4–7.3% children and adolescents and 2.5–3.4% adults worldwide, leading to adverse consequences for affected individuals, their families, and society at large. Despite the growing body of research, the etiology of ADHD remains obscure. Epidemiologic research is crucial for detecting disease risk factors. However, prior studies have rarely moved beyond identifying associations between proposed risk factors and later outcomes to rigorously testing the causality of these associations. The stagnation is partly due to the intrinsic limitations of observational studies, such as insufficient adjustment for genetic confounding. Large register-based data enable researchers to incorporate classical epidemiologic designs into genetically informative samples. Such genetically informative study designs are useful tools for unraveling the role of genetic and environmental factors in the development of diseases. In this thesis, the four studies used genetically informative designs to enhance the current understanding of the etiology of ADHD and the mechanisms underlying the associations between ADHD and its adverse health outcomes, including high body mass index (BMI) and suicidal behavior.

In Study I, we attempted to quantify the familial aggregation of ADHD in a large representative Swedish family sample, consisting of siblings and cousins of varying degrees of genetic relatedness. The results demonstrated that the familial aggregation of ADHD generally increased with increasing genetic relatedness between family members. Persistence of ADHD into adulthood in family members was associated with higher degree of the familial aggregation compared to remission of ADHD prior to adulthood.

In Study II, we assessed the co-aggregation of ADHD and overweight/obesity within families and further tested the hypothesis that the familial co-aggregation of the two conditions is at least in part due to common familial causes. The findings suggested an etiological overlap between ADHD and overweight/obesity.

In Study III, we investigated the association between high maternal pre-pregnancy BMI and offspring ADHD both at the population level and within siblings. The findings indicated the contribution of shared familial factors to the association, which supported the conclusion from Study II. A direct causal effect of high maternal pre-pregnancy BMI on offspring ADHD was, however, not evident.

In Study IV we examined the association between the use of ADHD medication and concomitant suicidal behavior. When the same individual was compared with him of herself for the rate of suicide-related events during on- and off-treatment periods, the results did not support that the use of ADHD medication increased the risk of concomitant suicidal behavior.

In conclusion, observational studies using genetically informative designs may provide valuable insights into the etiology of ADHD. The familial aggregation of ADHD increases with increasing genetic relatedness. ADHD and overweight/obesity share etiological factors. In family settings, such shared etiological factors lead to the familial co-aggregation of the two conditions. Moreover, the shared etiological factors manifest as unmeasured familial confounding, which may partly account for the population-level association between high maternal pre-pregnancy BMI and offspring ADHD. The use of ADHD medication does not appear to increase the risk of concomitant suicidal behavior. Genetically informative designs should be increasingly used in future epidemiologic research to efficiently control for confounding and strengthen causal inferences.