Is ADHD more than a brain disorder?

Transcript: Jan Buitelaar – Is ADHD a systemic disease?

(Automatically generated with Whisper; minor recognition errors are possible.)

[00:00:15] Hello, my name is Jan Buitelaar. I am a child and adolescent psychiatrist, researcher and working at the Radboud University Medical Center and the Donders Institute for Brain, Cognition and Behaviour in Nijmegen and the Netherlands. I would like to thank, I am very grateful for the invitation to this meeting about ADHD and ADHD as a systemic disease as a systemic disease. The first show about ADHD and ADHD as a systemic disease is characterized by persistent patterns of inattention, hyperactivity, impulsivity that is associated with significant impairment in functioning. It's very rare that ADHD is presented in pure form, so many times there is comorbidity with mental disorders in particular. ADHD is not just ADHD

[00:01:12] associated with increased mortality rates over the lifetime. This is a presentation and a result of a study of some years ago, which shows you that is adjusted increased mortality rates in children with ADHD below 86, between six and 70 years, and strikingly increased significantly increased mortality rates in those over the age 18. And this cumulative mortality is shown even more strikingly in this prospective cohort study from the United States and the data show you that year by year over about nine years, eight to nine years follow up, mortality increases in those with ADHD, compatible with ADHD. This mortality is of course multifactorial, it could be that it's just death by let's let's say suicide or increased mortality due to impulsive acts or substance use over impulsive dangerous behavior but it could also be mortality due to increased morbidity due to physical diseases.

[00:02:32] So ADHD is a very serious public health concern, public health problem that's due to its high preference is associated impairment in functioning which address many important areas of life like family life, school functioning, educational performance, failure to deliver in the workplace, the limited effectiveness of treatment on the long term, and the chronic persistent cause of ADHD over the lifetime just from adolescence to adulthood. Comorbidity significantly increases to this mental health and public health concern. And my talk will address the importance of recognizing and managing somatic comorbidities, so comorbidities with physical diseases. And the question I will approach and address is ADHD, just a disorder of the brain or of the body, or is it the diseases of both, of the brain and the body? Let me first give you some background on the genetics of ADHD, so more than 30 twin studies over the last 20 to 30 years have indicated that ADHD has a higher retability, and the quantitative behavioral retability indicates to which extent differences between ADHD ADHD symptoms within the population is due to genetic factors and so there's that difference due to genetic factors that variance why the one person is more hyperactive and inclusive and independent of the other is about 70% to 80% is a role for environmental risk factors but is mostly not shared by siblings and twin studies also suggest that ADHD can be represented conceptually as the extreme tale of one or more irritable quantitative traits.

[00:04:41] And there is genetic overlap between ADHD and other mental disorders like list of those here, cognitive impairment, intellectual disability, autism, schizophrenia, bipolar disorder, anxiety, and major depression. The genetic causes of diseases can be separated This is a classic separation in two models of jacuzzi, very rarely, like, let's say, there is one dominant autosomal dominant disease like Huntington's disease and multifactorial diseases where the combination of maybe some hundreds of thousands of genes of very small effect sizes increased the probability of a disorder of a significant rate. There is even a third possibility, and that in 5 to 10 percent of those with ADHD, there is a major rare variant with a strong effect, which plays a role, which is called oligogenic diseases. The way to study multifactorial genetics is to run GWAS, the genome-wide association study. And this is the latest, GWAS, which is from 2000 to 2023, where, across more than 20,000 features of the GWAS and close to 200,000 controls, 27 genome-wide significant load sievers were identified, and those were indicating potential leads to an expression in the brain.

[00:06:25] What is important is that the SNP heritability is only explaining 40% and 14% of the heritability of ADHD, but that around 80 to 90% of those ADHD relevant variants, SNPs, are also shared and overlap with the genetics of all the psychiatric disorders. So a strong genetic overlap between ADHD and other psychiatrists.

[00:07:00] The one way to do research, particularly, is to calculate polygenic scores, and that's a way to calculate the genetic liability for a trait by calculating computing the sum of all risk alleles in terms of weighted by their effect sizes, So there is a significant difference between the presence in the ADHD and the control population and computing the effect sizes overall in one summary account. And so a high polygenic scores indicates a high polygenic genetic liability for ADHD in our case. So let's now turn to somatic comorbidities or multimorbidity. Let me explain that concept. So co-morbidity is when we have a disease, a primary or main disease, and another disease that's co-occurring at the same time in the same person. But multimorbidity means that a person has two or more long-term health conditions at the same time, where you can hardly differentiate between the primary and secondary and they both impact together interactively on morbidity and also morbidity for that individual.

[00:08:40] And so the multi-morbidity can be long-term chronic conditions, can be both physical and mental conditions. This is a readable, well very comprehensive review paper on non-mental diseases associated with ADHD across the lifespan, and this figure displays the main finding. So ADHD is associated with increased risk for not surprisingly, but also let's say neurological All the diseases except mental, psychiatric conditions, all the diseases from the central nervous system like increased risk for migraine, epilepsy, respiratory, and vasculitis, needs like 1 and 2, obesity, elimination disorders, immunological disorders, autoimmune diseases, So reassess inflammation and atopic disorders and allergies.

[00:10:00] And so over the lifetime, growing older from chill ADHD and those without ADHD, so the difference between the ADHD and the old age, multimodality accumulates and cascades displayed visualized in the figure, and lower functioning increased impairment, physical mental health dangers, also. Let me focus and back in particular, the link between ADHD and obesity, and later cardiovascular diseases. So this is the result of a meat analysis already a few years ago. A meat analysis of about 42 studies, plus a 50,000 ADHD subject, increased, more than 600,000 controls, which established increased odds ratio, 50% increased obesity in children with ADHD and more than 70% increased risk for obesity in adults with ADHD. Of course, we have to be careful on claiming this association because the common ability could be true that third disorder, so the opposition to find disorder, oppression, mood disorder, personality disorder, that could be other psychiatric disorders that confound this association. And so we have also to carefully adjust for other potentially confounding factors like education level, neighborhood, income, and so on and so forth. So if we look for the adjusted odds ratio, at least then there is an increased risk of 27% for obesity in those with ADHD from my independent other media, that is also from about 10 years ago with rise also least 30%. So this is a solid question of course is if ADHD and obesity associated are occurring together more significantly than to be expected on the basis of chance by coincidence, then is there any causality in this association? And here I list four possibilities.

[00:12:54] One possibility is ADHD proceeds and contributes to obesity by, for example, regulating the energy balance. For example, simply by eating too much. So ADHD precedes and contributes. The second option is it's just the reverse causality. Obesity leads to ADHD symptoms. The third option is bi-directional links between the two. So it could be the combination of option 1 and 2, and then option 4 is disorders share obesity and ADHD, common pathophysiological mechanisms. Is there any evidence that ADHD proceeds and contributes to obesity and deteriorating the energy balance? Yes, I will just focus on that first study that I mentioned in this slide from Cortese, a 2013 Pediatrics paper and this shows you that childhood ADHD is associated with a higher BMI body mass index and higher rates of obesity in adolescence and adulthood many, many years later.

[00:14:14] Let's look at the data, it's not a huge study, it's a small study of more than 200 white Caucasian boys with ADHD were systematically assessed at ages 18, 25, and 41 years, so many, many years later. And then at the left you see that for those with childhood ADHD compared to an increased prevalence with 100% increased risk. That risk was for increased BMI and also for obesity. If you break down the ADHD, the problem of the study was that those who who had been identified with ADHD in childhood and we needed so they went out of had increased rates of later about option 2. Obesity leads to ADHD symptoms. There is also symptoms. For that I have to introduce you a little bit about the technique of Mendelian randomization and that's because it's it's very complicated to draw to draw causal inferences, to draw causal conclusions from observation, and in many cases these causal inferences are based on observational long term studies.

[00:16:14] So we are all familiar with the concept of randomized clinical trials where you randomize the population according to a target intervention or control interval condition, and then when When you randomly assign a power of that design, it is based on the fact that confounders like SES, IQ, background factors, neighborhood, cultural subgroups, and so forth, then these confounders are equally distributed between the groups and so if you compare the outcomes you may assume and you can also test whether these confounders are equally distributed that the difference in the outcome can be absolutely different than the difference in the medical intervention. Now, by the same principle, you can use them indeed in randomization, because if you have exposure to, let's say, ADHD and you are interested in the outcome, whether it's high BMI or obesity, then the ADHD exposure can be associated with confounding factors in observational study.

[00:17:42] But the genetic background of the participants with and without ADHD genetic background and the alleles are randomly assigned. So you can check, you can calculate an instrumental variable which indexes the exposure and then the same like in the randomized clinical trial, the co-founders are equally distributed across the different genetic groups and outcomes can then be compared in the same way. So here it is again visualized. So you construct a genetic instrumental variable that indexes the exposure. The assumptions are that the SNP is associated with the exposure. It is not associated with the confounders. and the SNP effect, the exposure, is only with the outcome and not by a third way. Now this technique is increasingly used to try to deal with these hidden confounders and to create a possibility to carefully draw cost conclusions from observational studies.

[00:19:07] and this was done in this study and here they tested whether ADHD was causally influencing BMI and using Mendelian randomization. This direct causality was not significantly supported in this study, but the reverse causality, the test by BMI influences severity and the incidence of ADHD look at the red quadrant of this left graph, you see that all these all significant effect sizes were crossing 0.0 lines of levels that BMI could influence ADHD, it's the severity incidence. Then what about option 3, are the bi-directional links between the two conditions? This is very likely, of course, so if ADHD precedes and supports option 3, but what about option 4? Are there any evidence that could be shared by the physiological mechanisms?

[00:20:24] Yes, I cannot fully unpack all these possibilities, so we'll particularly draw on whether there is shared genetics between the genetics of obesity and the genetics of ADHD that could also be shared by the physiological mechanisms in terms of that if you have weak executive functioning so you have difficulty inhibiting stronger pulses for example and you are inclined to be triggered by exposure to food or sorts of food, then you are too much, for example. And when this goes together with, for example, that you have also difficulties in terms of rewards, your reward system of the brain is differently sensitized to food, for example, and you like strong food and then that increases that impedes your executive functioning, you like a fantastic food, you eat too much, you have a positive energy balance that could be both underlying ADHD and obesity. The other option is for example a positive energy balance eating too much in terms of given your low physical activity you are not physically active, not not sporting, not jogging, not swimming, not biking, all by car and sitting all day and so on.

[00:21:58] It could also be that difficulties in terms of sleep and insufficient sleep or inflammation. They both have been suggested and also indicated by empirical studies to both underlie sleep and obesity. and obesity. This is some data to spot the idea that a shared familial so genetic factors may allow for a relationship between ADHD and family-based sex. Sibling spray increases the risk for ADHD compared with siblings of index males without index males with normal weight. Even if you adjust for the ADHD status of the index male

[00:22:59] If we look at family members of individuals with obesity, you have increased rates of ADHD. Family members of problems with ADHD have increased overweight, independent of their affected status. So whether they have ADHD or non-ADHD, they have parents and siblings, independent of having ADHD. They have more likely to be overweighted or obese, go back to family members of problems without ADHD. This is a recent Israeli study which shows you that a very large population of more than 1 million 17-year-olds, Those with severe ADHD were about 1.3 point more likely to be overweighted and 1.8 times more likely to be obese compared to their typically developing peers. That's confirmation, that's a replication, confirmation of what we already discussed.

[00:24:18] But the more important both for males and females, the upper graph is for males, the lower graph is for females. even for those who is mild ADHD, so just for subthreshold ADHD. Also those are increased risk for overweight and obesity. You may think about what is the effect of medication, what medication you are controlled for in so many studies. This is a also recent study that addressed that question. A meat analysis of eight studies, close to 20,000 individuals It was found that those with ADHD who were taking medication were just less likely, significantly less likely to be obese than those not taking medication. So this is important for our thinking about the clinical implications of this work. So, those without medication, not on medication, they had a significant increase for being on obese and those patients were less likely to be obese.

[00:25:41] This is fortunately for just a strong claim that medication is protected. This is a first data conclusion, data set. And, of course, this needs to be replicated in all the studies. So, let me summarize the potential mechanisms that could link ADHD and obesity. There are indications for familial and genetic factors, so shared genetic factors that pose drive increased liability for ADHD and obesity that might go together with shared environmental risk factors like stress, poor neighborhood, low SES, that could be shared weakness, poor performance on the psychological features and as I mentioned, there could be shared problems in terms of poor sleep hygiene, sleep disorders, poor sleep, and poor sleep also discharge because metabolism, your amount of balance, and this could both drive increased risk for ADHD and obesity, a lot of physical activity, and also overeating because of differences in dysregulations of the reward system. So there could be many, many causal mechanisms, many shared pathophysiology or mechanisms that are underlying this connection.

[00:27:20] Let's shift to another important physical illness, physical condition, cardiophascular disease. This is a systematic review and meta-analysis of a huge sample, more than 18 million individuals among more than 4,000 with ADHD and it's reported ADHD was assessed, that's good to know, because in many times this is not all based on clinical diagnosis, in some cases the ADHD was assessed by symptom-related questions on questionnaires, it's not based on careful clinical diagnostic like interview-structured interview with clinical diagnosis by high scores on symptoms scales, or electronic health records linked to DSM or ICD codes that could index ADHD diagnosis. But of course, it's still very noisy. So the power of these design is the huge sample sizes, but it's more noisy diagnostically. And the cardiovascular disease outcome was in most studies in any way hypertension or other sort of vascular diseases and cardiovascular diseases indexed by codes of SED diagnostic system.

[00:28:42] And then this study indicated that ADHD was associated with a significantly increased risk of cardiovascular disease, adjusted for all sorts of potentially confounding or biasing factors and that risk is almost 90% increased. It's a huge, huge increase. This is to indicate from the same paper that if we look at the ADHD risk, the risk is in here indicated at least 70% increased. If we compare that to the risk of other risk factors for So it's in about the same ballpark, so the effect of ADHD or cardiovascular disease controlling for all sorts of risk factors is about as powerful as obesity, sleep, disorder, sedentary behavior. But it's not quite specific. There is also other psychiatric illnesses that are associated with risk for cardiovascular disease, like schizophrenia, depression, anxiety disorders and stress-related disorders. and also stress. It is interesting to reflect on, if you think about the link between ADHD and cardiovascular disease, it's interesting to reflect on whether these other factors could be found, or could be mediators, or could be both. So it could be both that So ADHD is associated with this increased, let's say, and this was a kind of compilatine, this propensity, diabetes type 2, a healthy lifestyle.

[00:30:35] It could be that the ADHD effect is mediated by this artifact, this ADHD effect of cardiovascular disease. But it was also that these are just potentiators or modifiers that they accumulate that it's a mediated confounder, or even a modifier. The difficulty is you can never, let's say, adjust your design and your analysis for all sorts of potentially confounding factors. You can adjust for sex, race, race, yes, but it could also be a measure of potential confounders. One way to deal with this way of whether it's a confounder or a mediator or a modifier is to again use Mendelian randomization. So to use GWAS data to index the instrumental variable, and in this analysis, if you look at the association between the two together, according to the childhood obesity and the effect of stress than lifestyle smoking can be both a mediator, but ADHD has also a direct non-mediated by smoking and overweight, a direct effect on coronary artery disease.

[00:32:21] Let me broaden the scope and let's again look at this conceptual framework like somatic disease can lead to ADHD, ADHD can lead to somatic disease, how can we utilize all the sorts of very large GWAS data to perform latent causal variable analysis and check whether there are causal associations between complex traits and ADHD. This was based on a huge database again and I have indicated a summary of this and this This indicates that all those, maybe you can see my cursor, all those diseases and physical conditions and rates that are indicated in the left part of this graph, all these genetic variants associated with iron deficiency and e-mail, obesity, type 2 diabetes, substance use, neck, shoulder pain, genetic diabetes, all these physical conditions, these partially increase genetic risk in ADHD and so they contribute to genetic conservation of ADHD.

[00:33:46] So if you have high rates of let's say perfect heart rate disease, neck, shoulder pain, iron deficiency, anemia, sleep problems and obesity, there is care genetics and this is increasingly It causally increases your risk for ADHD, reversing the right side of this graph. There is also genetics variants with ADHD, genetic liability for ADHD. That has a genetic causality effect on your risk for, for example, COPD, chronic obstructive pulmonary disease. So it is degenetic, it's in this concept. The most interesting of this is that it is also genetic. negative, genetic relations. So if you do a lot of unpaid, voluntary rework, for example, you did never smoke and you are a school teacher, if you have high genetic lowly produce, those constraints that protects you, protects you, protects you genetically.

[00:35:05] This lowers your risk for ADHD. So this is negative genetic correlation, just genetic liability for these traits lowers your risk for ADHD. So what is the interesting thing of these protective effects? These are particularly so, if you are likely to participate in socially supportive and interactive activities. And on the other hand, genetic liability to a whole array of very complex physical traits, cardiometabolic phenotypes, physical pain contributes to the etiology for ADHD. That further impact the role of family genetics and so the role of the high polygenic score for ADHD. This shows you that for a lot of physical diseases, central nervous system diseases, respiratory diseases, musculoskeletal diseases, endocrine diseases, circulatory diseases, gastro-internet diseases, and so on, that the degree of genetic relatedness increases your risk for these diseases, so if you have ADHD, you have more risk for central nervous system diseases, respiratory diseases, and that is association with the individual, but also full siblings are at increased risk and even half siblings, but the degree generally at that risk.

[00:36:57] Here is another study that looks at whether polygenic scores for ADHD, independent of where that individual had clinically phenotypical ADHD, that hypoallergenic scores for ADHD increase a significant increase, as was, for example, diabetes type, rheumatoid arthritis, heart failure, cardiovascular disease, peripheral vascular disease. And many of these effects are also So mediated or moderated by BMI substance use and educational level, which is indicated in this graph, it's particularly BMI, which is significantly mediating and moderating those effects.

[00:37:58] This is a summary of all the separate analysis, which extend the polygenic score risk for all the somatic diseases is mediated by education in three out of the six somatic outcomes it's mediated by a educational level about half of the analysis the BMI is a mediator tobacco use in five out of sixteen analysis and alcohol use in eight out of sixteen analysis. Let me finish with a very recent paper, one or two months ago, and that's addressing the issue of differences between males and females. There's no option to fully discuss this paper, but it's of course important to look at sex differences. We look, we know that the clinical presentation of ADHD may differ between females and males. There is a male preponderance in childhood and adolescence, and then in adulthood there is more equal sex balance. But so the suggestions that I did, so slightly difference in symptoms, the development of the lifetime and also long term outcome for males and females.

[00:39:16] And this is a study of the Danish national registers and they were able to identify sex related factors and that shows you that particularly it could confirm what we already discussed that there is increased risk for somatic diseases and those with ADHD could also indicate that for example at younger ages the risk for physical diseases was significantly higher in females. So males had 20% increased risk and it's 40 to 23, but males had about 80% increased risk. And then these differences converge to about 50% increased risk at 23. So the take-home message of this study is that you are aware of sex differences and that sex differences in that risk, particularly with women, high risk for co-morbid physical diseases at younger ages. That's an important message for people's attention.

[00:40:35] So let me summarize what I have discussed so far. We had a question, is ADHD a disorder or the brain or the body of both? And I am tempted to say it's a disorder of both. We have seen that ADHD has a complex genetic architecture with a substantive role of playotropy. Playotropy didn't discuss that concept earlier, but it means that the phenotypic expression is very variable and could be in different organ systems. ADHD has shared genetics, shared genetics with many other mental traits and disorders. It has phenotypic and genetic overlap with multiple somatic diseases and there is a whole list of potential mechanisms that may underlie this association and that may require more study. The first of all is genetics, the role of inflammation. There could be a reverse causality, so medical diseases increasing risk for ADHD. We have to study carefully the role of medication, whether that's protective and under which circumstances.

[00:41:51] But I didn't discuss so far also, which may be important, that for example lifestyle factors, poor diet, poor sleep, lack of physical activity, but again think about ADHD might also negatively impact on the prognosis and the outcome of somatic diseases. So for example, patient delay, not seeking referral for doctor's assistance, and if you go to be treated for physical treatment and you have ADHD for physical diseases and you have also ADHD, this higher risk that you are showing non-adherence also might also be that you are at risk for your current physical disease, which is for multi-morbidity risk, but that you also your prognosis of the outcome of those physical diseases might patiently are non-adherence to medical treatment is worse, and it all so contributes to long-term morbidity, polymorbidity, and multimorbidity, and also more mortality in the long term. So what are the clinical implications of this? ADHD is a disease of the brain.

[00:43:07] So all individuals with ADHD, of any age, where the children and the lessons of adults should be screened for so many diseases. diseases. They should also be screened for a healthy lifestyle, and lifestyle advice, diet, physical exercise, check for sleep, stress management, non-stopping or non-starting smoking and use of substances. All this is important to be part of ADHD treatment. Physical care of individuals with ADHD should be sensitive.

[00:43:54] Importance is also that if you are a doctor, not working in a psychiatry, but in somatic health care, so if you are a doctor, not working in an ADHD in our patient unit, but you are in all the services in every service for mental and somatic health care, individuals with sometimes under diagnosed ADHD may be seen to be. ADHD multi-morbidity requires a multi-disciplinary treatment approach and an early recognition and prevention of multi-morbidity is key to improve functioning quality of life and to reduce impairment of functioning and improve the prognosis of ADHD across the lifetime. time. Thank you for your attention. And I'm stopping this presentation.

Jan Buitelaar
Radboud University
Nijmegen, The Netherlands