The short version
- ADHD runs in families. Family, twin and adoption studies put its heritability at about 74% (Tier B) [1].
- No single gene causes ADHD. Large studies find many regions of DNA that each shift risk slightly. The 2023 consortium study reported 27 genome-wide significant regions [2]. A 2025 study that added ADHD symptom scores to diagnosis data reported 39, of which 17 were new [3] (Tier B).
- Rarer, larger-effect variants exist too. A 2024 sequencing study identified KDM5B as a high-confidence risk gene and estimated that about 1,057 genes contribute to ADHD risk [4] (Tier B, early).
- Genes describe risk in groups of people. They do not diagnose one person and they do not pick a medication.
The usual mix-up
"74% heritable" does not mean "74% of your ADHD comes from genes." Heritability is a statement about why people in a population differ, not a share of any one person's condition. It also says nothing about how changeable the outcome is. Height is highly heritable, yet average height has risen across generations as nutrition improved. Heritable does not mean fixed, and it does not mean untreatable.
The plain-English layer
How we know it is genetic
Researchers compare relatives. Identical twins share virtually all their DNA and non-identical twins share about half. If ADHD traits line up more closely in identical twins than in non-identical twins, genes are doing some of the work. Studies of this kind are where the 74% figure comes from [1].
What the DNA studies found
| Layer | What it is | What we know |
|---|---|---|
| Common variants | Small DNA differences that are common in the population | Thousands each add a tiny bit of risk. Together they account for roughly a third of the heritability [1] |
| Rare variants | Uncommon changes, sometimes new in the child ("de novo") | Larger effect per variant. KDM5B is the best-supported gene so far [4] |
| Copy-number variants | Missing or extra stretches of DNA | Account for part of the heritability [1] |
What this says about ADHD as a trait
The 2025 study found that combining a clinical diagnosis with symptom scores from tens of thousands of children gave more statistical power than diagnosis alone. The authors interpret their findings as supporting the view that clinical ADHD sits at the extreme end of a continuous liability [3]. In plain terms: the genetics looks like a spectrum of risk, not an on/off switch.
The deeper layer
- Why "loci", not "genes". A GWAS finds a region of DNA where variants differ between people with and without ADHD. The region is not automatically a gene, and the nearest gene is not automatically the one that matters. The 2025 paper used a gene-mapping method to propose 22 potential effector genes at its loci [3]. Treat these as candidates.
- Why the old "dopamine gene" headlines faded. Before large GWAS, studies picked genes that seemed plausible, such as DAT1, DRD4 and DRD5. A 2009 meta-analysis found some of these associated with childhood ADHD, but with significant heterogeneity between studies [5]. The large modern studies have a different design. See GEN-02 for why that matters.
- Dopamine is still central to treatment. Stimulants act on dopamine and noradrenaline signalling, and the main non-stimulant options act on noradrenaline signalling (Tier A). What is not yet settled is how the genetic findings connect to those systems (Tier D). Saying so plainly is stronger than forcing a link.
Limits
- Heritability and loci counts describe populations, not individuals.
- Most large GWAS to date have mainly included people of European ancestry. Results may transfer imperfectly to other groups. (Tier B, general; exact shares not stated here.)
- The 27-loci and 39-loci counts come from different studies with different designs. They are not the same experiment updated, so do not present them as a trend line.
- KDM5B and the 1,057-gene estimate are from one early sequencing study. Replication will refine them.
- This article makes no claim about any single person's cause.
Sources
Bibliographic details checked against PubMed.
- Faraone SV, Larsson H. Genetics of attention deficit hyperactivity disorder. Mol Psychiatry 2019;24(4):562-575. DOI · PMID 29892054
- Demontis D et al. Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains. Nat Genet 2023. DOI · PMID 36702997
- van der Laan CM et al. Genome-wide association meta-analysis of childhood ADHD symptoms and diagnosis identifies new loci and potential effector genes. Nat Genet 2025;57(10):2427-2435. DOI · PMID 40962958
- Olfson E et al. Rare de novo damaging DNA variants are enriched in attention-deficit/hyperactivity disorder and implicate risk genes. Nat Commun 2024;15:5870. DOI · PMID 38997333
- Gizer IR, Ficks C, Waldman ID. Candidate gene studies of ADHD: a meta-analytic review. Hum Genet 2009;126(1):51-90. DOI · PMID 19506906