A Nature News Feature published on 19 August 2026 synthesized the global wave of genomic newborn screening programs and opened with a concrete result. In the GUARDIAN study in New York, the published analysis of its first 4,000 newborns found that 147 infants, about 3.7%, received a positive screen result, most for conditions not covered by traditional newborn screening, with some findings prompting life-saving interventions such as bone-marrow transplants. The study reported 120 true-positive diagnoses against only 10 identified by standard newborn screening.Other programs report findings in the same range. BabyScreen+ in Victoria, Australia confirmed findings in 1.6% of 1,000 newborns, and BabyDetect in Belgium confirmed genetic conditions in 1.8% of nearly 4,000 infants, including 0.8% whose conditions would have been missed by conventional screening. These are enrolled cohorts with confirmed diagnoses, and they shift the question the field needs to ask next.
Conventional newborn screening relies on a dried blood spot and mostly chemical analysis of proteins and metabolites. US guidelines recommend testing for 66 conditions, France for 16, and the UK for 10. Of roughly 3.6 million US births each year, about 98% are screened, and an estimated 6,600, near 1 in 600, test positive. Genomic newborn screening reads DNA from the same blood spots and can evaluate hundreds of genes, with some pilots screening for more than 700 disorders.
Dozens of initiatives are now running worldwide, and the accumulated data answers the feasibility question directly: population-scale sequencing of newborns can be done, and it finds treatable conditions that current panels miss. The earliest of these efforts, BabySeq, has run since 2013; across its two trials about 11% of sequenced infants carried disease-associated variants, and roughly a third of those were already showing early signs of disease. What remains costly and difficult is scaling this reliably, and cost is only part of the constraint.
These programs return results, not diagnoses. Every screen-positive finding requires confirmatory testing, and a meaningful share is ruled out. In GUARDIAN, 64 of 475 initially flagged infants showed no signs of disease at confirmation, and North Carolina's Early Check ruled out 22 of 50, while BabyScreen+ reported none.
Gene-list design compounds the difficulty. There is no consensus panel: BabyScreen+ analyzes 605 genes, BabyDetect 405, GUARDIAN began near 250 and expanded to 450, and Early Check evaluates 169. Even well-characterized genes do not always predict disease, and variant databases disagree on pathogenicity, which leaves interpretation and counselling carrying real weight.
The Nature feature makes the stakes of that downstream work concrete through two families. Safi Ford, screened through the UK Generation Study, started growth-hormone therapy at 6 months for a deficiency her mother was not treated for until 17, after the critical growth window had closed. A GUARDIAN family received a Smith-Magenis-associated result with little support, told to look it up online; later testing showed the variant was unlikely to be disease-causing, but the experience left them shaken. What separated those two outcomes was the work after detection: confirmatory testing, informed family communication, structured natural-history capture, and a route to the right trial or therapy.
That downstream chain is where Sano works. Sano combines genetic testing, recruitment, and long-term engagement in a single platform, which maps onto the exact workflow a screen-positive result triggers. Confirmatory testing and counselling run through Sano, including at-home saliva kits and kits-at-site options spanning whole exome, whole genome, panels, and genotyping through GxP and CLIA-CAP-certified labs, paired with genetic counselling and personalized reporting.
Finding dispersed, genetically defined families is a separate problem. In the US launch of a genetic hearing-loss program, Sano reached over 70% pre-screener completion and nearly 90% of participants meeting eligibility criteria through digital campaigns, advocacy partnerships, and a streamlined referral process. Cohort building depends on staying connected over time: Sano supports recontact for future research and has expanded testing including polygenic risk scores and repeat-expansion sequencing for decentralized studies, the kind of infrastructure natural-history capture requires.
Three problems will determine whether genomic newborn screening delivers net benefit at scale. The first is gene-list harmonization. Divergent panels mean an infant's findings depend heavily on where they are born, and the field needs shared criteria for which conditions clear the actionability threshold. The second is variant interpretation at scale. As volumes grow, disagreement between databases turns into false positives and avoidable family distress, so consistent, well-supported interpretation becomes central. The third is follow-through infrastructure to ensure there is a reliable path to confirmation, counselling, longitudinal data, and trial access.
The pilots have shown the science works. Our recap of an episode of The Genetics Podcast with Zornitza Stark, an investigator who worked on BabyScreen+, and our analysis of newborn genome sequencing cover the clinical case in more depth. The remaining question is operational, and it is the harder one to answer. Sponsors and screening programs that build the identification-to-recruitment chain now will be the ones ready to convert a screen-positive result into a treated child.