Using a genetics-based statistical approach, researchers from the United Kingdom found attention-deficit/hyperactivity disorder (ADHD) to have a causal effect on childhood obesity and coronary artery disease. Below, researcher Evie Stergiakouli, PhD, describes the study approach, details of the findings, the significance of the results and possible future research.
Read the transcript:
Hello, my name is Dr. Evie Stergiakouli. I'm a Senior Lecturer in Genetic Epidemiology based at the MRC Integrative Epidemiology Unit at the University of Bristol in the UK. My research focuses on childhood mental health and using genetics to identifying causes and consequences of mental health problems in children.
Today, we'll discuss a study that we performed on the effects of ADHD on physical health. The main message of our study is that ADHD can cause childhood obesity and coronary artery disease.
Attention deficit hyperactivity disorder, or ADHD, is a childhood neurodevelopmental disorder. It affects about 5 percent of school‑age children, can persist into adulthood, and can sometimes lead to educational, social, and occupational difficulties.
The economic impact of ADHD ranges from about $143 billion to $266 billion in the United States, so it is a substantial cost. Most of these costs are incurred in adulthood.
Some of these costs are because having ADHD as a child is linked to a broad range of physical health problems in adulthood, including obesity, type 2 diabetes, hypertension, asthma, autoimmune conditions, and migraine.
We wanted to find out whether ADHD actually causes these health problems directly or whether they're just correlated with ADHD in order to inform strategies to prevent poor health in people with ADHD.
Current Research (1:45)
Most of the studies reporting on the link between ADHD and physical health are observational. They're called observational studies because the investigators observe individuals without any manipulation or intervention. This is unlike a randomized controlled trial where investigators intervene and look at the effects of their intervention.
Observational studies are useful where randomized controlled trials would not be ethical or practical to be performed. However, observational studies use large data sets to see if there are patterns of correlations between factors, but correlation and causation is not the same thing.
The conclusions from observational studies cannot provide causal evidence because they face 3 major problems: selection bias, confounding, and reverse causation.
Firstly, selection bias can arise because the populations recruited within such large studies are often not representative of the wider population.
Secondly, confounding is where a third factor, which is associated with both the exposure and the outcome, could be explaining the link between them. For example, education could be confounding the association between ADHD, which is the exposure in our case, and obesity, which is the outcome, since it is associated with both of them.
Thirdly, in reverse causation, the outcome could be actually causing the exposure. For example, maternal obesity could be causing ADHD in a child. This is called reverse causation and it cannot be assessed in an observational study.
Study Method (3:26)
What did we do? Because of these problems with observational research, our study used a different method, which is called Mendelian randomization. Mendelian randomization is a statistical approach that uses genetics to provide information about the relationship between an exposure, in our case ADHD, and an outcome.
The combination of genetic variance a person receives from their parents is randomly assigned at conception. This makes Mendelian randomization a natural experiment, which is potentially less likely than observational studies to be subject to confounding and reverse causation. It is also much quicker and easier to perform than randomized controlled trials.
In this study, we used Mendelian randomization to test if ADHD, as indexed by genetic instruments, is causing metabolic, cardiovascular, autoimmune, allergic, or neurological conditions.
We tested whether ADHD has a causal effect on adult BMI, childhood obesity, coronary artery disease, heart attack, hypertension, type 2 diabetes, migraine, epilepsy, rheumatoid arthritis, inflammatory bowel disease, and allergic diseases.
We used data from large‑scale genetic studies, which are called genome‑wide association studies. Large genetic studies have been carried out on all the diseases that I mentioned earlier, as well as ADHD. We used their findings to perform Mendelian randomization.
Study Results (5:00)
What did we find? We found evidence that ADHD has a causal effect on childhood obesity and coronary artery disease. We did not find evidence that ADHD causes heart attacks, hypertension, type 2 diabetes, neurological, or autoimmune diseases.
We checked our analysis using various tests and found it was robust. We also checked if the genetic variance that we used as instruments in Mendelian randomizations have multiple functions, which can sometimes stop Mendelian randomization from working well, and we did not find any evidence. This makes us more sure of the results that we are reporting.
Our next step was to find the causal pathway that relates ADHD to childhood obesity and coronary artery disease. Our results show that the effect of ADHD on coronary artery disease was reduced when we took into account childhood obesity, suggesting that childhood obesity may be a potential mediator. In other words, ADHD causes childhood obesity, which in turn causes coronary artery disease in adulthood.
Some of the effect of ADHD on coronary artery disease is, however, direct, and not related to childhood obesity. Other factors, such as smoking, which is more common in people with ADHD, could be involved.
Significance of the Results (6:22)
What do our results mean? It seems that some of the impact of ADHD on coronary artery disease occurs early in life. This means that any intervention that aimed to lower the risk of coronary artery disease in people with ADHD would need to begin early.
We did not find evidence of a casual effect of ADHD on adult BMI, which could indicate that the effects might be more pronounced during childhood, when ADHD symptoms are at their peak.
Since motor hyperactivity is a hallmark of ADHD, it may appear counterintuitive that patients with ADHD will have a higher risk for obesity. However, patients with ADHD have been reported to spend more time watching television, so lower levels of physical activity and increased dysregulation of eating behavior. All of this could be contributing.
Our findings strengthen the argument for effective treatment of children with ADHD. Reducing ADHD symptoms in children could result in reduced childhood obesity, less impact on physical health in adulthood, and perhaps reduce the economic impact of ADHD.
Future studies should include more factors linked to physical health in ADHD, such as education, and disentangle the individual contribution of each factor on physical health.
We also suggest that clinicians who support people with ADHD are aware of the risk of childhood obesity to reduce future risk of coronary artery disease, and that they are also prepared to discuss these risks with patients and their families.
Leppert B, Riglin L, Wootton RE, et al. The effect of ADHD on physical health outcomes-a two-sample Mendelian randomization study. American Journal of Epidemiology, 2020;kwaa273.
Evie Stergiakouli, PhD, is a Senior Lecturer in Genetic Epidemiology and Statistical Genetics based at the MRC Integrative Epidemiology Unit (IEU) at the University of Bristol in the United Kingdom. Her research focuses on childhood mental health and using genetics to identify causes and consequences of mental health problems including neurodevelopmental disorders and cleft lip with or without palate.