ADHD screening before embryo transfer: what the genetics actually show

If you or your partner has ADHD, or it runs in your family, chances are you've thought about whether your child might have it too. Maybe you've been told it's about parenting, or diet, or screen time, or that kids grow out of it. Most of the conventional advice for parents in your situation focuses on environmental factors, because those were the only things anyone could change.

But that advice is missing the forest for the trees. The biggest determinant of whether or not a child develops ADHD is not parenting. It's not diet. It's not screen time.

It's genetics.

ADHD's heritability is 74-88%. That's higher than breast cancer. Higher than type 2 diabetes. Higher than most conditions families already consider during embryo screening. And for families going through IVF, the tools to read that genetic signal at the embryo stage now exist. Five years ago they didn't.

The genetic signal was always strong. We just couldn't read it.

Most people still think of ADHD as a behavioral condition. It's not. A meta-analysis of 37 twin studies put the mean heritability at 74%. A large Swedish twin study of nearly 60,000 twins found 88% heritability for clinical ADHD. Genetics accounts for the vast majority of why some people develop ADHD and others don't.

And it doesn't go away with age. That same Swedish data found 72% heritability in adults. This matters more than most people realize. The adults who got diagnosed at 35 after years of "underperforming" at work, after decades of feeling like something was wrong but not knowing what, didn't develop a new condition. They had it all along. The genes were there from the beginning. The idea that kids grow out of ADHD is one of those persistent myths that the data just doesn't support.

Compare that to breast cancer, where heritability sits around 30-40%, or type 2 diabetes at roughly 50-65%. Families who'd never hesitate to screen for those conditions often don't realize that ADHD has a stronger genetic signal than either one. We think of ADHD as behavioral and diabetes as biological, but the heritability data says ADHD is more genetically driven.

So why hasn't ADHD been on the screening list? Not because the biology was missing. Because the tools to read a polygenic signal this complex weren't good enough until very recently.

For years, the knock on ADHD polygenic scores was that they explained too little variance to be useful. With older methods, a single polygenic score captured about 3.6-4% of the variation between people. Not nothing, but not enough to act on in the clinic. That criticism was fair. But it was a limitation of method, not biology.

Then the methodology changed.

ADHD shares genetic architecture with depression, educational outcomes, risk-taking behavior, sleep problems, and metabolic conditions. Rather than building a score from ADHD cases alone, researchers began combining information from hundreds of genetically correlated traits. A 2023 Nature Communications study showed that combining 937 related polygenic scores increased prediction accuracy by up to ninefold. The genetic signal for ADHD hadn't gotten stronger. The way we read it did.

And the picture keeps getting clearer. A 2025 Nature paper mapped thousands of common DNA variants that account for the vast majority of ADHD's measurable genetic signal, with 27 genome regions reaching significance in large case-control studies. For some families, the story also includes rare variants with large individual effects. If ADHD appears in every generation of your family at very high rates, a rare high-impact variant could be part of the explanation, and monogenic testing may be relevant alongside polygenic screening.

Do these scores actually work between siblings?

Here's the question that actually matters for embryo screening: do these scores predict differences between siblings? Embryos from the same IVF cycle share parents, ancestry, and roughly half their DNA. If a polygenic score only works because it's picking up on population-level patterns rather than actual causal genetics, it'll fall apart when you compare siblings. And that would make it useless for the one context families actually need it.

We tested this directly. Nature has already run the experiment for us: every time two siblings are created, the genes they inherit from each parent get scrambled and randomly assigned. In Moore et al. 2025, we evaluated 17 disease polygenic scores on sibling pairs, and 16 out of 17 showed no decrease in predictive performance compared to population-level analysis. The scores are picking up real causal genetic variants, not statistical artifacts. The predictive accuracy holds up under the hardest relevant test.

If you're curious about what ADHD screening looks like for your embryos specifically, our team can walk you through it.

What this actually means if you're going through IVF

For parents with ADHD in their family who are already doing IVF, here's what the process looks like. A small biopsy is taken from each embryo on Day 5 or 6. This is the same biopsy used for standard chromosomal screening, so there's no additional procedure. That genetic sample gets sequenced, and polygenic scores are computed for ADHD and whatever other conditions you're interested in. Your family history gets factored into the baseline risk, and you get a risk estimate for each embryo broken out by ancestry group so the numbers reflect your background accurately.

We calibrate across 8+ ancestry groups because most ADHD genetics research has been done in European-ancestry populations and genetic testing should work for everyone. And we incorporate family history directly because if you or your partner has ADHD, that changes the baseline risk in ways a raw polygenic score alone won't capture. You can then weigh that information alongside morphological grading and chromosomal results when deciding which embryo to transfer.

There's a piece here that most people miss. Polygenic scores for psychiatric traits like ADHD explain less total variance than scores for conditions like type 2 diabetes or heart disease. Environment still matters, and severity isn't predictable from genetics alone. But ADHD exists on a spectrum, and what screening does is quantify genetic predisposition toward the clinical end of that spectrum, where the condition impairs learning, relationships, and the ability to follow through on things that matter to the person living with it.

And these conditions aren't all-or-nothing. Someone with a high ADHD polygenic score may never get a formal diagnosis but still spend their life fighting through brain fog and missed deadlines, the constant feeling that they should be doing better than they are. The line between "has ADHD" and "doesn't" is blurrier than a risk percentage suggests, and one of our counselors who understands this can explain what the scores actually mean for quality of life, not just for a diagnostic label.

And the genetic variants involved don't just affect attention. Recent research shows that ADHD polygenic load also contributes to metabolic and cardiovascular health outcomes, conditions you might not associate with attention deficits at all. So the information you get from screening for ADHD connects to a broader picture of your embryo's health profile that you wouldn't see if you only screened for the conditions you'd already thought of.

If you're interested in learning more about how Herasight's screening applies to ADHD and your family's specific situation, please reach out to us.

‍