Fewer False Alarms: How Two Countries Made a Newborn Hormone Test Far More Accurate

What families should know about 2026 research from Sweden and from Ontario, Canada, on improving newborn screening for congenital adrenal hyperplasia

A test every baby gets, and the problem it has had

Every newborn has a few drops of blood taken from the heel in the first days of life, and one of the conditions that blood is checked for is congenital adrenal hyperplasia — a treatable hormone disorder that can be fatal if missed. For decades this particular test has raised far more false alarms than real cases, especially in babies born early. Two studies published in 2026, one from Sweden and one from Ontario, Canada, show how to fix that.

What the condition is, in plain terms

The adrenal glands sit on top of the kidneys and make several hormones the body cannot do without. One is cortisol, which helps the body handle stress and keep blood sugar steady. Another is aldosterone, which tells the kidneys how much salt to keep. In congenital adrenal hyperplasia — usually shortened to CAH — a baby inherits a gene change that leaves out one of the enzymes needed to make these hormones. The most common version involves an enzyme called 21-hydroxylase [1].

When that enzyme is missing or barely working, two things happen at once. The body cannot make enough cortisol and, in the more severe form, enough aldosterone. And the building blocks that would have been turned into those hormones pile up behind the blockage and get converted into male-type hormones instead [2].

The severe form is called salt-wasting CAH. Without treatment, a baby with it loses too much salt in the urine. In the second or third week of life the baby may start vomiting, stop feeding well, lose weight, become floppy, and then collapse. This is called an adrenal crisis and it can be fatal. Girls with this form are usually born with genitals that look partly masculine, which prompts doctors to investigate immediately. Boys look completely typical at birth, which is exactly why they were historically the ones who died — nothing about them raised suspicion until the crisis arrived. Once the condition is known, treatment is straightforward: hormone replacement, given every day, and extra doses at times of illness [3].

What families and doctors faced before newborn screening

Before screening existed, families found out the way you would expect — after something went badly wrong. Screening for CAH began in the 1970s, using a substance called 17-hydroxyprogesterone as the signal, because it is one of the building blocks that piles up behind the missing enzyme. Within a decade the approach had spread widely enough for researchers to survey its results across countries, and it is now part of the newborn blood-spot panel in more than forty of them [4]. Sweden started in 1986, and Swedish researchers were later able to show what it bought: fewer babies collapsing from salt loss, particularly boys, and much shorter waits for families of newborn girls whose appearance at birth had raised questions [5].

But the test came with a persistent flaw. The substance it measures is naturally high in babies born prematurely, and it rises further in any baby who is unwell. So the test flags a great many babies who do not have CAH. A long Swedish review covering 26 years of screening found that among all the babies the test flagged, only about 13% actually had the condition. Among babies born early, it was 1.4% — roughly one true case for every seventy alarms [6]. Researchers confirmed that the hormone levels of healthy premature babies genuinely overlap with those of affected babies, so no single cut-off line can separate them cleanly [7]. Programmes around the world have struggled with the same thing, with wide variation in how well the test performs from place to place [8].

For families, a false alarm is not harmless. It means a phone call, often a frightening one, a repeat blood test, sometimes a hospital visit, and days of worry — all for a condition the baby does not have. Nobody wanted to solve this by making the test less sensitive, because the whole point is not to miss the baby whose life depends on being found.

The idea both research teams tested

There is a second substance the body makes when the 21-hydroxylase enzyme is missing. It is called 21-deoxycortisol. What makes it useful is that in a healthy baby it is barely present at all — the body has no reason to make it. In a baby with CAH, it climbs. Specialists have argued for years that it is a much more telling signal than the one screening programmes have traditionally relied on [9]. Researchers in the Netherlands had already shown, first in a pilot and then over two years of everyday use, that adding this substance to the testing process could cut out false alarms and spare babies a second heel prick [10], [11].

The question the two 2026 studies set out to answer was whether a whole screening programme could be rebuilt around this idea without losing a single baby with the dangerous form.

Importantly, this does not mean an extra needle. Both programmes run the additional test on the same dried blood spot the baby already gave — a second, more precise laboratory analysis of blood that has already been collected.

What the two studies did and what they found

The Swedish study was carried out by the country's single national screening laboratory at Karolinska University Hospital in Stockholm, which handles blood spots for all of Sweden's roughly 100,000 births a year. The team first went back to stored blood spots — 49 from babies known to have CAH, 155 from babies who had triggered a false alarm, and 46 from healthy babies — to work out where the cut-off lines should sit. They then ran the new test alongside the old one for six months before switching over, and adjusted the thresholds twice as they learned more, using different settings for babies born at different gestational ages [12].

After three years, false alarms had fallen by 88%. Among full-term babies, the chance that a flagged baby actually had CAH rose from 14% to 84% — meaning that instead of roughly one in seven alarms being real, more than four in five now are. Among premature babies, false alarms fell by 65%. No cases were missed during those three years, although the researchers are careful to say that three years is not long enough to be certain of that.

The Canadian study came from Newborn Screening Ontario, the provincial programme that screens about 145,000 newborns a year and reaches more than 99% of them. Rather than changing the programme first, this team analysed 1,710 blood spots that had already tested positive and used computer simulation to test eight different possible rules, running each one ten thousand times to find the best combination of thresholds [13].

Their chosen rule required three things to be true at once, and the results were striking. Every case of the severe salt-wasting form was still found. But the proportion of flagged babies who truly had it rose from 3% to 70%. Among premature babies, the number of unnecessary callbacks fell from 174 to 3. Overall, the number of families who would have been called back for further testing dropped by 96% — from 259 down to 11.

Two independent programmes, on two continents, using different methods, arrived at the same answer. That is the kind of agreement that makes a finding trustworthy.

What this means for families

If your baby's screening programme uses one of these newer methods, two things follow.

First, you are much less likely to receive a worrying phone call that turns out to be nothing. That is a real benefit, and it is the main point of this work.

Second — and this matters — if you do receive that call, it should be taken seriously. Under these newer methods a positive screen in a full-term baby is far more likely to be a real finding than it used to be. It is still not a diagnosis: confirmatory blood tests, and often genetic testing, are needed. But the old reassurance that "these are usually false alarms" no longer applies in the same way.

There is one further point families should understand. Both programmes deliberately tuned their tests to catch the dangerous salt-wasting form as reliably as possible, and that came with a trade-off: the milder forms of CAH are now less likely to be picked up by the newborn screen [14]. Milder forms do not cause a life-threatening crisis in the newborn period, but they can cause symptoms later in childhood — early signs of puberty, unusually fast growth, and in girls, irregular periods in the teenage years. A normal newborn screen does not rule these out. If your child develops symptoms like these later, it is worth mentioning CAH to your doctor even if the newborn screen was normal.

What researchers are working on next

The false alarms that remain are concentrated in the smallest, earliest babies, where the levels sit right at the edge of what the laboratory equipment can detect. Both teams want to work out whether separate cut-off lines for different gestational ages would clear up those last few cases — neither study had enough babies to answer that yet.

Both also want longer follow-up. Three years of clean results is encouraging but not conclusive; researchers generally say it takes more than a decade to be confident that a screening programme is not missing anyone. The Swedish team is also looking at a closely related substance, 21-deoxycortisone, which early evidence suggests may be an even clearer signal.

For now, the practical takeaway is a hopeful one. A test that has frightened many thousands of families unnecessarily has been made substantially more precise, without loosening its grip on the babies who genuinely need to be found — and it has been done using blood that is already being collected, in programmes that are already running.

References

  1. Auer MK, Nordenström A, Lajic S, Reisch N. Congenital adrenal hyperplasia. Lancet. 2023;401(10372):227–244. doi:10.1016/S0140-6736(22)01330-701330-7) ↩
  2. Claahsen-van der Grinten HL, Speiser PW, Ahmed SF, et al. Congenital Adrenal Hyperplasia—Current Insights in Pathophysiology, Diagnostics, and Management. Endocr Rev. 2022;43(1):91–159. doi:10.1210/endrev/bnab016 ↩
  3. Speiser PW, Arlt W, Auchus RJ, et al. Congenital Adrenal Hyperplasia Due to Steroid 21-Hydroxylase Deficiency: An Endocrine Society Clinical Practice Guideline. J Clin Endocrinol Metab. 2018;103(11):4043–4088. doi:10.1210/jc.2018-01865 ↩
  4. Pang SY, Wallace MA, Hofman L, et al. Worldwide experience in newborn screening for classical congenital adrenal hyperplasia due to 21-hydroxylase deficiency. Pediatrics. 1988;81(6):866–874. doi:10.1542/peds.81.6.866 ↩
  5. Thilén A, Nordenström A, Hagenfeldt L, von Döbeln U, Guthenberg C, Larsson A. Benefits of neonatal screening for congenital adrenal hyperplasia (21-hydroxylase deficiency) in Sweden. Pediatrics. 1998;101(4):E11. doi:10.1542/peds.101.4.e11 ↩
  6. Gidlöf S, Wedell A, Guthenberg C, von Döbeln U, Nordenström A. Nationwide neonatal screening for congenital adrenal hyperplasia in Sweden: a 26-year longitudinal prospective population-based study. JAMA Pediatr. 2014;168(6):567–574. doi:10.1001/jamapediatrics.2013.5321 ↩
  7. Nordenström A, Wedell A, Hagenfeldt L, Marcus C, Larsson A. Neonatal screening for congenital adrenal hyperplasia: 17-hydroxyprogesterone levels and CYP21 genotypes in preterm infants. Pediatrics. 2001;108(4):E68. doi:10.1542/peds.108.4.e68 ↩
  8. Speiser PW, Chawla R, Chen M, et al. Newborn Screening Protocols and Positive Predictive Value for Congenital Adrenal Hyperplasia Vary across the United States. Int J Neonatal Screen. 2020;6(2):37. doi:10.3390/ijns6020037 ↩
  9. Miller WL. Congenital Adrenal Hyperplasia: Time to Replace 17OHP with 21-Deoxycortisol. Horm Res Paediatr. 2019;91(6):416–420. doi:10.1159/000501396 ↩
  10. Stroek K, Boelen A, Bouva MJ, et al. Second-tier Testing for 21-Hydroxylase Deficiency in the Netherlands: A Newborn Screening Pilot Study. J Clin Endocrinol Metab. 2021;106(11):e4487–e4496. doi:10.1210/clinem/dgab464 ↩
  11. Olthof A, Bouva MJ, Claahsen-van der Grinten HL, et al. 21-deoxycortisol as a second-tier test in congenital adrenal hyperplasia newborn screening in The Netherlands: two-year evaluation. Arch Dis Child. Published online 23 September 2025. doi:10.1136/archdischild-2025-328929 ↩
  12. Engström K, Zetterström RH, Wedell A, Nordenström A. Neonatal Screening for CAH in Sweden—Results of Implementing Second-Tier Testing. Int J Neonatal Screen. 2026;12(2):29. doi:10.3390/ijns12020029 ↩
  13. Lawrence SE, Marcadier J, Auger S, et al. Improved Performance of Newborn Screening for Congenital Adrenal Hyperplasia Using 21-deoxycortisol Measurement. J Endocr Soc. 2026;10(1):bvaf157. doi:10.1210/jendso/bvaf157 ↩
  14. Nordenström A, Falhammar H. Management of Endocrine Disease: Diagnosis and management of the patient with non-classic CAH due to 21-hydroxylase deficiency. Eur J Endocrinol. 2019;180(3):R127–R145. doi:10.1530/EJE-18-0712 ↩