The short version
A polygenic score (PGS) adds up thousands of small DNA differences into one number for a person. In research it is useful for studying groups. It is not a diagnostic test for ADHD, and a low score does not rule ADHD out. (Tier B)
The usual mix-up
"There is a DNA score for ADHD, so a DNA test could prove or disprove it." The score is a risk tendency measured against a reference population. Two people with the same score can differ completely in whether they have ADHD.
The plain-English layer
- A GWAS measures how strongly each of thousands of common variants is linked to ADHD.
- A PGS takes one person's DNA and adds those links up, weighted by their strength.
- The result is a position on a spread, high, middle or low, relative to other people.
What a recent study showed. In three separate cohorts, PGSs built from the combined symptoms-plus-diagnosis meta-analysis predicted ADHD traits better than PGSs built from symptoms alone or diagnosis alone [1]. That is progress in the science. It is not evidence that a PGS is ready to be a clinical test.
The deeper layer: why a score cannot diagnose
- It is a probability shift, not a yes/no. Each person's risk also depends on rare variants, environment and chance (GEN-04).
- Overlap. People with and without ADHD have overlapping score distributions, so a given score can occur in both groups. (Standard in polygenic prediction. Tier B; specific accuracy figures not yet verified.)
- Ancestry. Scores built mainly from one ancestry group usually perform less well in others. (Tier B, general.)
- Diagnosis is clinical. The NICE ADHD guideline (NG87) bases diagnosis on a full clinical and developmental assessment by a suitably qualified specialist. No genetic or blood test is part of that diagnosis. (Standard guideline position; Tier A/B. Wording to be checked against the current text before publishing.)
- Direct-to-consumer "ADHD scores". Commercial reports are not clinical-grade for this purpose. Treat them as entertainment, not evidence.
What would have to be true before a PGS became a clinical tool
- It would need to add measurable accuracy beyond standard clinical assessment.
- It would need validation in the people it will be used on, across ancestries.
- It would need to change a decision (for example treatment choice or access) for the better.
- It would need safeguards against misuse, for instance being used to deny care to someone with clear symptoms.
Point 4 is relevant to this site. A service that says "your score does not show ADHD" is applying a tool outside its evidence. Clear clinical symptoms and a clinical diagnosis outrank any score.
Limits
- This page does not report accuracy figures for PGS in ADHD because none has been verified for this site.
- It does not say PGSs will never become useful.
- It makes no statement about any individual's score or risk.
Sources
Bibliographic details checked against PubMed.
- 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
- NICE. Attention deficit hyperactivity disorder: diagnosis and management (NG87). Check the current version at nice.org.uk.