Attention-deficit/hyperactivity disorder (ADHD) is substantially influenced by genetic variation. Family studies show that ADHD occurs more often among biological relatives of affected people; twin and adoption studies support a large contribution from inherited factors; and genome-wide research now identifies numerous associated regions of DNA alongside an emerging contribution from rare, higher-impact variants. The best-supported model is a complex, polygenic developmental liability that interacts with the conditions in which a person grows and lives. Genetic influences shape probabilities across populations, while clinical ADHD is established through developmental history, persistent symptoms, cross-setting impairment, and qualified assessment. The evidence supports a strong answer to the common question “Is ADHD genetic?” while explaining why no ordinary genetic test can presently diagnose ADHD or reliably predict an individual's clinical course. Faraone and Larsson, 2019; Demontis et al., 2023; NICE, NG87.
What Genetic Research Means When It Says ADHD Is Heritable
Heritability is a population-level estimate of how much observed variation in a characteristic is statistically attributable to genetic differences among people in a specified population and environment. For ADHD, researchers can study variation in clinically diagnosed disorder, symptom ratings, or an underlying liability inferred from diagnoses. These are related measurements with distinct ascertainment rules. An estimated heritability of roughly 70–80% means that inherited differences explain a large proportion of variation in the studied population under the model used. It says nothing like “80% of this particular person's ADHD was caused by genes,” and it does not indicate that 80% of children born to someone with ADHD will have ADHD. Faraone and Larsson, 2019.
Heritability also depends on the population, age range, method of assessment, environmental variation, and statistical assumptions. If social or environmental conditions change, a heritability estimate may change even when the underlying biology remains stable. Heritability is compatible with powerful educational, psychological, medical, and environmental interventions. It also coexists with uncertainty about the mechanisms connecting DNA variation to attention regulation, activity levels, inhibition, motivation, and functional difficulties. Faraone et al., 2021.
Familial aggregation, heritability, and molecular association
Familial aggregation is the observation that relatives resemble one another more often than unrelated people do. It can arise from genetic similarities, shared environments, assortative partnering, correlated exposure, and shared patterns of recognition or diagnosis. Heritability is a model-based estimate that attempts to partition sources of variation. Molecular association examines measured DNA variants and their statistical relationship with the condition. These lines of evidence answer different questions and become more compelling when they converge. Faraone and Larsson, 2019.
The term genetic risk refers to a probabilistic association with ADHD liability. Risk variants have a frequency and an effect size that depend on the studied population; neither a single associated allele nor a collection of alleles automatically confers a diagnosis. A causal variant, a statistical marker that tags a causal variant, and a gene whose expression may be influenced by a locus are different entities. Genome-wide association studies identify associated genomic regions first; the biological explanation often requires additional functional research. Demontis et al., 2023.
Family Studies: What Happens Among Biological Relatives?
ADHD is more common among first-degree biological relatives of people with ADHD than in the general population. A historical review reported increased risk in siblings, and later population studies and family-based designs have reinforced familial clustering. The risk for any given relative varies with the definition of ADHD, study population, age of assessment, ascertainment, and whether the relative has been formally evaluated. Family history raises clinical suspicion and makes a careful developmental interview informative; it is not itself a diagnostic criterion. Faraone and Larsson, 2019; NICE, NG87.
A parent may have unrecognized ADHD when a child is evaluated, and a child's diagnosis may prompt an adult to recognize longstanding difficulties with attention, organization, impulsivity, emotional regulation, or time management. Conversely, siblings with related genetic backgrounds may show different symptom profiles and functional outcomes. Family resemblance follows distributions rather than a fixed family script, because children inherit different combinations of variants and experience different developmental contexts.
Relatives share genes and environments
Family studies alone cannot assign a particular proportion of resemblance to DNA. Parents and children share both inherited variation and practical conditions such as household organization, neighborhood resources, sleep routines, access to care, and stress exposure. They may also share characteristics that influence when someone seeks an assessment. Genetically informative designs compare relatives with different degrees of genetic relatedness or with separated biological and rearing relationships, giving a stronger basis for inference. Faraone and Larsson, 2019.
Family history remains clinically useful even without precise recurrence prediction. A careful history asks which relatives were diagnosed, who showed convincing childhood-onset traits, which conditions might instead explain those traits, and whether the information comes from firsthand assessment or retrospective family impressions. Different generations have had unequal opportunities for recognition, especially women and older adults. NICE, NG87.
Twin Studies and the Meaning of the Heritability Estimate
Twin studies compare monozygotic twins, who ordinarily share nearly all their inherited DNA sequence, with dizygotic twins, who on average share about half their segregating variants. If identical twins resemble one another more strongly than fraternal twins in ADHD measures, that pattern supports genetic influence, subject to assumptions about relevant environments and measurement. Across 37 twin studies summarized by Faraone and Larsson, the mean estimate was approximately 74%. Earlier syntheses commonly cited a value near 76%. Different studies produce different estimates because instruments, participant ages, statistical specifications, and clinical thresholds vary. Faraone and Larsson, 2019; Faraone et al., 2005.
Twin models often divide observed variation into additive genetic effects, common or shared environmental effects, and nonshared environmental effects plus measurement error. Such components are statistical abstractions; they do not list the individual experiences that matter. An estimate of a small shared-environment component in a classical twin model should not be interpreted as proof that parenting, schooling, poverty, trauma, or environmental exposures are irrelevant. Model assumptions, correlated experiences, rater effects, gene–environment correlation, and the measured range of exposure all matter.
Why identical twins can differ
Monozygotic twins can differ in symptom severity, clinical outcome, and support needs. They may encounter distinct illnesses, friendships, school demands, injuries, prenatal conditions, and social responses. Stochastic developmental processes, somatic genetic differences, measurement error, and potentially epigenetic differences can also contribute. Discordance is consistent with substantial genetic influence because heredity concerns relative probabilities and distributions. Faraone and Larsson, 2019.
Adoption and extended-family evidence
Adoption studies compare the occurrence of ADHD among biologically related and adoptively related family members. The broad pattern reported in ADHD research supports a substantial inherited contribution, while reminding researchers that adoption is selective and adoptive families differ from the general population in several ways. Studies involving siblings, half-siblings, cousins, and parent–offspring relationships further test whether similarity changes with genetic relatedness. The strongest conclusion comes from triangulating several approaches rather than treating one model as definitive. Faraone and Larsson, 2019.
The Polygenic Architecture of ADHD
Most ADHD liability involves many genetic variants whose individual contributions are small. Polygenic means that numerous positions across the genome contribute to variation in susceptibility. A person may carry common alleles associated with higher risk, alleles associated with lower risk, and rare variants with larger effects; their combined contribution is embedded in a developmental history. The genome is not divided into a single set of “ADHD genes” possessed by all diagnosed people and a separate set possessed by everyone else. Demontis et al., 2023.
In the early candidate-gene era, research focused on plausible neurotransmitter-system genes, including dopamine-related candidates. Some initial associations proved inconsistent across samples or weaker than anticipated. Genome-wide association studies (GWAS), which examine variants across the genome while controlling for multiple testing, changed the field by allowing discovery that was less dependent on prior hypotheses. Modern findings support highly distributed neurodevelopmental pathways, rather than a simple dopamine-gene explanation. Faraone and Larsson, 2019; Demontis et al., 2023.
How a genome-wide association study works
A typical ADHD GWAS compares variant frequencies among a large group with ADHD and a comparison group, often incorporating multiple biobanks, clinical cohorts, and national registers. Each single-nucleotide polymorphism, or SNP, is tested statistically; researchers account for ancestry-related population structure, relatedness, and many simultaneous tests. A genomic locus reaches genome-wide significance only under a stringent threshold. Researchers then use fine-mapping, expression data, chromatin information, and functional studies to identify plausible genes or regulatory mechanisms. An associated SNP can be a marker for a nearby causal variant rather than the mechanism itself.
A larger study tends to detect more small effects. Counts of significant loci are therefore a statement about the data and statistical power available at the time, not a fixed biological inventory. A locus can contain several genes, and the same gene can be implicated by multiple analytic methods. Association also changes with how ADHD is defined in the sample, including psychiatric interviews, diagnostic codes, symptom measures, and developmental age. van der Laan et al., 2025.
The landmark 2023 ADHD GWAS
Demontis and colleagues analyzed 38,691 individuals with ADHD and 186,843 controls and reported 27 genome-wide significant loci. They highlighted 76 potential risk genes and found enrichment in biological signals relevant to early brain development and particular neuronal cell types. Their work also explored genetic links between ADHD liability, cognition, and other psychiatric or behavioral traits. These findings showed that population genetics could move beyond generic familial risk toward reproducible locations in the genome. The 76 genes were candidates identified through mapping and integration, not 76 individually validated clinical biomarkers. Demontis et al., 2023.
This study drew extensively on available cohorts and used statistical approaches designed for very large datasets. Important questions remained about representation across ancestries, variation in case definitions, and the gap between association signals and mechanisms in developing neural systems. Every new result must be interpreted alongside these design features.
The 2025 analysis combining childhood symptoms and diagnoses
A 2025 Nature Genetics study led by van der Laan combined information from childhood ADHD symptom measures with clinically ascertained ADHD. The analyses included 290,134 symptom assessments from 70,953 distinct individuals and identified 39 independent loci in the combined ADHD analysis, including 17 reported as new. The authors proposed 22 potential effector genes using a fine-mapping-informed method. Their result supports the idea that clinically recognized ADHD occurs toward the impairing end of a broader distribution of attention and activity-related liability. It also shows that pooling diagnostic and symptom data requires careful treatment of rater, instrument, and age differences. van der Laan et al., 2025.
Neither the 2023 nor the 2025 locus count should be interpreted as a list of deterministic causes, and the two figures are not directly comparable as though the same discovery design had simply been repeated. Sample composition, outcomes, meta-analysis strategy, and gene-mapping rules differ. Together they establish robust molecular evidence for polygenic architecture.
Rare Variants and Larger Individual Effects
Common SNPs are only one part of the genetic picture. Rare coding variants, protein-truncating variants, and copy-number changes can disrupt molecular processes more substantially, although only a minority of ADHD cases currently receive an explanation from a recognized high-effect alteration. Rarity alone does not imply pathogenicity; many rare variants have little known functional importance. Molecular evidence increasingly suggests that some uncommon variants converge on pathways already implicated by common variants. Demontis et al., 2023.
A 2024 sequencing study by Olfson and colleagues of affected children and their biological parents reported enrichment of rare de novo damaging variants and implicated KDM5B through combined family and case–control analysis. The modest number of trios and challenges of interpreting rare variants limit its use for routine prediction. These results chiefly offer a discovery route into neurodevelopmental mechanisms, not a clinically generalizable test for ADHD. Olfson et al., 2024.
Evidence published in 2025 and appearing in the 2026 Nature issue
Demontis and colleagues studied exome-sequencing data from 8,895 individuals with ADHD and 53,780 controls. They identified rare coding-variant associations implicating MAP1A, ANO8, and ANK2, with reported odds ratios roughly 5.55 to 15.13 for the associated variant-burden contrasts. Their pathway findings implicated neuronal architecture, synaptic processes, and RNA-related mechanisms. The study was published online on November 12, 2025, and appeared in a January 2026 journal issue. The same paper estimated a modest aggregate burden-heritability contribution for the studied classes of rare variants. Demontis et al., 2026.
An odds ratio in a rare-variant study compares group odds under a particular design; it does not give the probability that a carrier will develop ADHD. Low population frequency, incomplete penetrance, variable expressivity, ascertainment, and co-occurring developmental disorders complicate any individual interpretation. Identifying a biologically informative rare variant is an achievement in mechanism research, while clinical deployment requires separate validation.
Copy-number variants and syndromic contexts
Copy-number variants are deletions or duplications of larger DNA segments. Some variants contribute to multiple developmental outcomes, such as intellectual disability, autism, congenital features, and attention-related problems. In a child with ADHD symptoms plus developmental delay, unusual physical findings, seizures, or a strong syndromic pattern, clinicians may consider genetics referral according to the broader phenotype. The referral addresses the overall clinical presentation and any suspected genetic syndrome. A routine ADHD diagnosis alone does not create a standard indication for broad genomic testing.
Why Twin Heritability and SNP Heritability Differ
Twin studies infer the contribution of inherited differences from patterns of resemblance among relatives. SNP heritability estimates the proportion of liability variation tagged by a measured set of common genetic markers under a particular statistical method and reference population. These estimates capture different portions of genetic architecture and depend on different assumptions. The common-variant component captured by current arrays is lower than many classical twin-study estimates. Some of the gap may reflect rare or poorly tagged variants, imperfect phenotype measurement, nonadditive effects, and differences in statistical design. It should not be filled with an invented single missing mechanism. Faraone and Larsson, 2019; Demontis et al., 2023.
Researchers distinguish total heritability, narrow-sense additive heritability, SNP-based estimates, and variant burden estimates. Each is conditional on design choices and prevalence assumptions when calculated on a latent liability scale. Comparing them requires reporting the measure and method, not merely quoting a percentage in isolation.
Polygenic Scores: What They Measure and What They Cannot Predict
A polygenic score combines many measured variants, weighting them by effect estimates obtained from GWAS. Scores can help study risk distributions, developmental persistence, population-level associations, and correlations with educational or mental health outcomes. People with relatively high scores have, on average, higher ADHD liability than those with relatively low scores in a matched study population, but the individual distributions overlap substantially. A low score does not exclude ADHD, and a high score does not establish it. Demontis et al., 2023.
A genetic score trained in a cohort with a particular ancestry mix can perform differently in a population with another ancestry profile. Differences in linkage disequilibrium, allele frequencies, discovery-sample representation, measurement, and environmental structure reduce transportability. Performance also varies across ages, diagnostic definitions, and healthcare systems. A research association, even one repeatedly replicated, does not automatically satisfy standards for clinical calibration, individual discrimination, fairness, or demonstrated benefit in decision-making. Martin et al., 2019.
Clinical use remains a separate evidence question
Genetic testing is not part of the routine diagnostic criteria for ADHD. The current clinical pathway uses developmental history, reports from relevant settings, psychiatric and psychosocial assessment, impairment, and evaluation of alternative or co-occurring conditions. A genetic score neither substitutes for that work nor indicates which medication will work best for an individual patient. Promising research into pharmacogenetics and treatment response requires independent, clinically meaningful replication before entering routine ADHD care. NICE, NG87.
Commercial claims about an “ADHD DNA test” should be evaluated by asking whether there is a clinically validated diagnostic or predictive use, not whether the test reads real variants. Analytical accuracy, association, clinical validity, and clinical utility are different standards. Someone can receive technically accurate genotype results that provide little actionable information about their symptoms, support needs, or treatment options.
Genetic Overlap With Other Conditions and Traits
Genetic correlations estimate the degree to which common-variant effects across the genome are statistically shared between two studied traits. ADHD has reported genetic correlations with autism, major depression, some substance-use disorders, insomnia, and numerous behavioral or cognitive measures. The direction and strength vary by cohort and phenotype. Correlation means partially shared inherited liability at a population level; it does not establish that one clinical condition caused another or that everyone with ADHD has the correlated trait. Demontis et al., 2023.
The ADHD–autism overlap is especially relevant because the conditions can co-occur and both involve developmental differences in attention, behavior, and adaptive functioning. Cross-disorder genetic overlap supports investigations of common mechanisms while leaving autism-specific diagnosis and combined clinical presentations to their respective scientific questions. Shared variants can also have distinct effects depending on developmental pathways, context, and phenotype definitions.
Depression, sleep, and substance-use relationships
Genome-wide findings indicate that ADHD liability shares variants with depression, insomnia, and substance-use phenotypes. Clinical co-occurrence involves multiple processes, including environmental exposures, reciprocal functional effects, and differential identification. A correlation with insomnia does not imply that sleep deprivation is a genetic form of ADHD; rather, sleep conditions can coexist with ADHD, magnify attention complaints, or produce an alternative presentation that warrants assessment. The research on sleep-related ADHD symptoms is developed in ADHD and Sleep: Insomnia, Circadian Rhythms, and Daytime Symptoms. Demontis et al., 2023.
Genetic correlations with substance-use disorder may coexist with causal pathways involving impulsivity, stressful circumstances, access to substances, and other psychiatric conditions. Clinicians assess the person and their context, including current intoxication or withdrawal, before attributing attention complaints to lifelong ADHD. Genetic correlation alone cannot decide diagnosis or treatment.
Genetic Continuity From Childhood Into Adult Life
ADHD is a neurodevelopmental condition whose symptoms and impairment may persist or change over the lifespan. Genetic liability can be associated with childhood ADHD, continued adult impairment, and late recognition of symptoms. The relative contribution of particular common and rare variants may differ among samples defined by persistent childhood diagnosis versus first diagnosis in adult services. A study comparing childhood, persistent, and late-diagnosed groups found higher ADHD polygenic scores in the persistent group and differences in genetic overlap with other psychiatric conditions. The findings are important for hypothesis generation but also depend on how each category was defined and followed over time. Rajagopal et al., 2022.
Adult ADHD evaluation still requires attention to developmental onset, longitudinal symptoms, alternative diagnoses, and actual functioning. Later diagnosis can represent missed childhood symptoms; it can also be complicated by depression, substance use, sleep disturbance, trauma, acquired brain injury, and other conditions. Family history may support a developmental account, yet the history cannot be replaced by polygenic scores. See ADHD in Older Adults: Diagnosis, Aging, and Differential Assessment for age-specific assessment questions. NICE, NG87.
Sex, gender, and ascertainment
Clinical recognition differs across boys, girls, men, women, and gender-diverse populations. Differences in referral, symptom descriptions, masking or compensation, and cultural expectations can shape who enters a genetic study as a diagnosed case. The genetic architecture of ADHD may be partially shared across sex-defined cohorts, while the interpretation of effect differences depends on sample size and ascertainment. Genetic data do not remove the need to evaluate inattentive symptoms and functional impairment in underrecognized groups. NICE specifically flags under-recognition among girls and women. NICE, NG87.
Genes, Brain Development, and Executive Functions
Associated loci converge statistically on biological processes active during nervous-system development. Brain development involves gene expression in changing cell populations, synaptic formation, neuronal communication, and circuit maturation. GWAS gene-mapping methods connect inherited variation to candidate developmental systems, but an enrichment signal at a cell type or tissue is not proof that an anatomical structure is uniquely responsible for ADHD. Brain development is distributed across interconnected systems. Demontis et al., 2023; Demontis et al., 2026.
Executive functions include mechanisms such as working memory, inhibition, flexible control, and the regulation of goal-directed activity. These functions vary widely among people with ADHD, and a person can meet clinical diagnostic criteria without performing poorly on every executive function test. Genetic correlations between ADHD and cognition do not establish a one-to-one mapping from a gene to a psychological ability. The research question is how multiple pathways shape developmental probability and heterogeneous functioning over time.
Dopamine, reward, and timing models
Dopamine and other catecholamine systems are important to the regulation of attention and action, and medications that affect these systems can reduce ADHD symptoms in many patients. Molecular genetics supports a broader network of processes than a single transmitter deficit. A treatment's mechanism also does not identify a unique genetic cause of the condition it treats. The distinct behavioral and computational literature on reward and delayed outcomes is treated in ADHD and Motivation: Reward Processing and Delay Aversion.
Gene–Environment Interplay and Developmental Causation
Genetic factors operate through development in environments. Gene–environment correlation occurs when genetic differences are statistically associated with experienced environments, for example because family characteristics influence routines or because a person's traits influence social responses. Gene–environment interaction describes a situation in which the statistical effect of a genetic factor varies across environmental exposures, or vice versa. These concepts require careful causal study designs; neither can be established by showing that a stressful environment and high genetic score occur together.
Prenatal exposures, prematurity, lead exposure, severe early adversity, family resources, sleep, and educational demands have been investigated in relation to ADHD traits and diagnoses. Many observational associations are vulnerable to familial confounding, correlated exposures, selection, and measurement differences. A more stringent causal interpretation may draw on sibling comparisons, negative controls, genetically informed cohorts, or quasi-experimental designs. This article establishes how inherited risk is studied; the independent question of environmental and developmental causation is a distinct scientific investigation.
Epigenetics and popular explanations
Epigenetic regulation concerns changes in gene expression and chromatin regulation that do not require a change in the underlying DNA sequence. ADHD-related epigenetic associations are an active research area, but blood or saliva methylation findings may reflect differences in cell composition, medication exposure, stress, smoking, or correlated environmental variables. Most findings do not support individual diagnostic testing or a claim that a specific event directly switched on an “ADHD gene.” Claims that families transmit ADHD through a particular parenting style or that daily habits permanently rewrite ADHD genetics go beyond the established evidence.
What Genetic Findings Mean for Diagnosis and Family Care
The primary clinical implication of genetic research is that ADHD deserves the same careful, developmentally informed assessment as other conditions with substantial familial risk. Clinicians ask about symptoms across settings, the age at which problems began, the degree of impairment, strengths, supports, and possible co-occurring conditions. Rating scales and screening instruments can organize information, but diagnosis requires trained clinical judgment. Neuropsychological tests may be indicated for particular differential questions; neither these tests nor a genetic panel independently establishes ADHD. NICE, NG87.
For a family in which ADHD has been diagnosed, a biological relative with meaningful difficulties can be offered an appropriate assessment based on symptoms and impairment. The decision to seek help is not predicated on a DNA result. Families benefit from evidence-based support, accessible education, reasonable adaptations, and interventions appropriate to age and goals. The genetics literature does not supply a deterministic forecast about a child's future school, work, or relationships.
Genetic counseling and unusual presentations
Genetic counseling can be appropriate when the wider medical picture includes developmental delay, intellectual disability, seizures, congenital anomalies, or an established genetic syndrome. The counselor or medical geneticist addresses the whole phenotype, family history, and possible testing indications. Some families have questions about recurrence risk, interpretation of a rare-variant report, or privacy. Those questions require individualized clinical discussion rather than an online risk calculator.
Medication response and preventive claims
Present genetic studies have not established a generally usable genotype-based rule for selecting stimulants, nonstimulants, psychotherapy, or school support for ADHD. Treatment is guided by clinical indications, patient preference, age, concurrent conditions, benefits, adverse effects, and monitoring. Research on medication effects and safety has its own evidence standards; see Stimulant Treatment for ADHD: Methylphenidate and Amphetamines. NICE, NG87.
No credible evidence demonstrates that a supplement, restrictive diet, commercial “detox,” or genetic interpretation service can erase inherited ADHD susceptibility. Nutritional or medical interventions may be relevant to specific, independently identified problems, but genetic predisposition should never be treated as an invitation to unsupported medical claims.
Research Methodology and the Limits of Current Evidence
A robust ADHD genetics paper reports its phenotype definition, ages, recruitment, ancestry composition, genotyping or sequencing process, quality control, sample overlap, statistical thresholds, replication, and treatment of comorbidity. Studies based on health records may be very large but capture clinically recognized or treated ADHD unevenly. Studies using symptom scales can detect dimensional variation but may compare raters and instruments with different reliability. These design tradeoffs can shift estimates and implicated loci.
Ascertainment, ancestry, and portability
Many influential GWAS have drawn heavily from participants of European genetic ancestry, and participation in genomic studies is socially patterned. Results can transfer imperfectly to groups whose ancestry or healthcare access differs from discovery cohorts. The limitation is scientific and ethical: unequal discovery data can create unequal predictive accuracy. A central research priority is representative recruitment, community trust, sound cross-population validation, and fair interpretation of findings. Martin et al., 2019.
Statistical power, effect sizes, and multiple testing
Large genetic datasets can reveal statistically significant associations with extremely small effects. Such findings are scientifically useful for pathway discovery while offering little discriminatory power for a single person's diagnosis. An effect discovered through thousands of simultaneous tests needs appropriate correction, replication, and sometimes independent functional experiments. Rare-variant studies face different difficulties: sparse carrier counts, uncertainty about pathogenicity, and challenges in estimating penetrance.
Causal inference and biological interpretation
A genome-wide association demonstrates correlation between genetic variation and a measured outcome under the study's design. It does not by itself reveal a causal molecular mechanism, especially when a marker tags several nearby variants. Functional genomics, cell models, animal models, developmental datasets, and fine-mapping can strengthen hypotheses, but translational relevance must be evaluated carefully. Genetic correlation across conditions similarly does not establish that one condition is the cause of another.
Ethical interpretation, stigma, and privacy
Genetic explanations can reduce misconceptions that ADHD arises from weak character or simple unwillingness to try. Their ethical value depends on accurate communication. Probabilistic findings must not be converted into assumptions about intelligence, competence, behavior, parental quality, or a person's future. Genetic information introduces privacy considerations and may affect relatives who never consented to a particular interpretation. Research consent, ancestry equity, and responsible communication are integral to this scientific field.
Frequently Asked Questions
Is ADHD genetic or caused by the environment?
ADHD risk is strongly influenced by inherited variation, and development also involves environmental circumstances. Family, twin, and adoption studies indicate substantial genetic contribution; GWAS and sequencing studies identify part of the underlying molecular architecture. Neither type of evidence assigns a simple percentage of causation to an individual. Faraone and Larsson, 2019.
What percentage of ADHD is hereditary?
Classical twin research commonly estimates heritability in the range of roughly 70–80%, with a frequently cited pooled value around 74%. The figure describes population variation in the specific studies. It does not measure how much of one person's ADHD was inherited and cannot be directly converted into the probability that a child will inherit a diagnosis. Faraone and Larsson, 2019.
Can ADHD skip a generation?
Families may recognize symptoms in one generation and not another because inheritance is polygenic, individual combinations of variants differ, impairments vary, and opportunities for diagnosis have changed. Apparent generation-skipping also occurs when an older relative was never assessed. The pattern does not imply a single recessive ADHD gene.
If a parent has ADHD, will their child have ADHD?
Having a biological parent with ADHD raises risk compared with the population average, but outcomes remain variable. Many children of parents with ADHD do not meet diagnostic criteria, and some children have ADHD despite no recognized family history. A meaningful assessment depends on the child's developmental and functional profile, not family history alone.
Can a DNA test confirm or exclude ADHD?
Routine ADHD diagnosis is clinical. The tested person's developmental history, symptoms, impairment, and differential assessment matter. Existing polygenic scores lack the individual clinical accuracy and established utility needed to confirm or exclude ordinary ADHD. Rare-variant testing may be considered for a wider syndromic medical presentation, under separate clinical indications. NICE, NG87.
Is there one gene for ADHD?
The central genetic architecture is polygenic. Studies have identified many associated loci, and rare variants in some genes may carry substantially larger risk in subsets of cases. Researchers study their biological effects across neural development. No universal single-gene explanation fits the full clinical spectrum. Demontis et al., 2023; Demontis et al., 2026.
Can inherited risk be changed through treatment or support?
A person's inherited DNA sequence generally remains stable, while symptoms, skills, functional outcomes, and environmental demands can change. Evidence-based treatment and supports target functioning and wellbeing. A high heritability estimate is fully compatible with clinical improvement. Faraone et al., 2021.
Scientific Synthesis
Independent genetic methods converge on a coherent account of ADHD as a strongly heritable, heterogeneous neurodevelopmental condition. Twin and adoption studies establish large inherited contributions to population differences; family studies show clustering; common-variant GWAS supports a widely distributed polygenic architecture; and modern exome analyses identify uncommon, potentially higher-impact contributions. The strongest molecular findings advance biological research but currently explain only part of individual variation. Genetics informs a developmental model of risk and strengthens family-aware clinical understanding, while diagnosis and treatment remain grounded in careful assessment and observed needs. Faraone and Larsson, 2019; Demontis et al., 2023; van der Laan et al., 2025; Demontis et al., 2026.
Related Articles
Cluster ADHD and Executive Function — The published research chapter integrating diagnostic, developmental, executive-function, and treatment evidence across ADHD.
ADHD and Motivation: Reward Processing and Delay Aversion — Theories and empirical findings on reward valuation, delay aversion, and motivation.
ADHD and Sleep: Insomnia, Circadian Rhythms, and Daytime Symptoms — Comorbidity, overlapping symptoms, and sleep-related differential assessment.
ADHD and Organization: Planning, Sequencing, and Routines — Functional executive challenges and environmental support.
ADHD in Older Adults: Diagnosis, Aging, and Differential Assessment — Age-specific recognition and differential diagnosis.
References
Demontis, D., Duan, J., Hsu, Y.-H. H., et al. (2026). Rare genetic variants confer a high risk of ADHD and implicate neuronal biology. Nature, 649, 909–917. https://doi.org/10.1038/s41586-025-09702-8.
Demontis, D., Walters, G. B., Athanasiadis, G., et al. (2023). Genome-wide analyses of ADHD identify 27 risk loci, refine the genetic architecture and implicate several cognitive domains. Nature Genetics, 55, 198–208. https://doi.org/10.1038/s41588-022-01285-8.
Faraone, S. V., Banaschewski, T., Coghill, D., et al. (2021). The World Federation of ADHD International Consensus Statement: 208 evidence-based conclusions about the disorder. Neuroscience and Biobehavioral Reviews, 128, 789–818. https://doi.org/10.1016/j.neubiorev.2021.01.022.
Faraone, S. V., and Larsson, H. (2019). Genetics of attention deficit hyperactivity disorder. Molecular Psychiatry, 24, 562–575. https://doi.org/10.1038/s41380-018-0070-0.
Faraone, S. V., Perlis, R. H., Doyle, A. E., et al. (2005). Molecular genetics of attention-deficit/hyperactivity disorder. Biological Psychiatry, 57, 1313–1323. https://doi.org/10.1016/j.biopsych.2004.11.024.
Martin, A. R., Kanai, M., Kamatani, Y., et al. (2019). Clinical use of current polygenic risk scores may exacerbate health disparities. Nature Genetics, 51, 584–591. https://doi.org/10.1038/s41588-019-0379-x.
National Institute for Health and Care Excellence. (2018, current guideline). Attention deficit hyperactivity disorder: diagnosis and management (NG87). https://www.nice.org.uk/guidance/ng87/chapter/recommendations.
Olfson, E., Farhat, L. C., Liu, W., et al. (2024). Rare de novo damaging DNA variants are enriched in attention-deficit/hyperactivity disorder and implicate risk genes. Nature Communications, 15, 5870. https://doi.org/10.1038/s41467-024-50247-7.
Rajagopal, V. M., Duan, J., Vilar-Ribó, L., et al. (2022). Differences in the genetic architecture of common and rare variants in childhood, persistent and late-diagnosed attention deficit hyperactivity disorder. Nature Genetics, 54, 1117–1124. https://doi.org/10.1038/s41588-022-01143-7.
van der Laan, C. M., Ip, H. F., Schipper, M., et al. (2025). Genome-wide association meta-analysis of childhood ADHD symptoms and diagnosis identifies new loci and potential effector genes. Nature Genetics, 57, 2427–2435. https://doi.org/10.1038/s41588-025-02295-y.