Journalism in the Public Interest

The $13 Test That Saved My Baby’s Life. Why Isn’t it Required For Every Newborn?

Dozens of babies die every year because hospitals do not perform a simple test that detects congenital heart defects. Seventeen states have yet to require the exam for newborns.

Dozens of babies die every year because hospitals do not perform a simple test that detects congenital heart defects. Seventeen states have yet to require the exam for newborns. (Dave Herholz, Flickr)

On July 10, my wife gave birth to a seemingly healthy baby boy with slate-blue eyes and peach-fuzz hair. The pregnancy was without complications. The delivery itself lasted all of 12 minutes. After a couple of days at Greenwich Hospital in Connecticut, we were packing up when a pediatric cardiologist came into the room.

We would not be going home, she told us. Our son had a narrowing of the aorta and would have to be transferred to the neonatal intensive care unit at NewYork-Presbyterian Hospital at Columbia, where he would need heart surgery.

It turned out that our son was among the first in Connecticut whose lives may have been saved by a new state law that requires all newborns to be screened for congenital heart defects.

It was just by chance that we were in Connecticut to begin with. We live in New York, where such tests will not be required until next year. But our doctors were affiliated with a hospital just over the border, where the law took effect Jan. 1.

More Resources: Which States Require This Screening?

American Academy of Pediatrics legislation tracker

Newborn Foundation screening progress map

As we later learned, congenital heart problems are the most common type of birth defect in the United States. The Centers for Disease Control and Prevention estimate that about one in 555 newborns have a critical congenital heart defect that usually requires surgery in the first year of life.

Many cases are caught in prenatal ultrasounds or routine newborn exams. But as many as 1,500 babies leave American hospitals each year with undetected critical congenital heart defects, the C.D.C. has estimated.

Typically, these babies turn blue and struggle to breathe within the first few weeks of life. They are taken to hospitals, often in poor condition, making it harder to operate on them. By then, they may have suffered significant damage to the heart or brain. Researchers estimate that dozens of babies die each year because of undiagnosed heart problems.

The new screening is recommended by the United States Department of Health and Human Services, the American Heart Association and the American Academy of Pediatrics. Yet more than a dozen states — including populous ones like Massachusetts, Pennsylvania, Florida, Georgia, Wisconsin and Washington — do not yet require it.

The patchy adoption of the heart screening, known as the pulse oximetry test, highlights larger questions about public health and why good ideas in medicine take so long to spread and when we should legislate clinical practice.

Newborns are already screened for hearing loss and dozens of disorders using blood drawn from the heel. The heart test is even less invasive: light sensors attached to the hand and foot measure oxygen levels in the baby’s blood. This can cost as little as 52 cents per child.

Our son’s heart defect was a coarctation of the aorta, a narrowing of the body’s largest artery. This made it difficult for blood to reach the lower part of his body, which meant that the left side of his heart had to pump harder.

In the hospital, though, he appeared completely healthy and normal because of an extra vessel that newborns have to help blood flow in utero. But that vessel closes shortly after birth, sometimes revealing hidden heart problems only after parents bring their babies home.

Depending on the heart defect, the onset of symptoms can be sudden.

This is what happened to Samantha Lyn Stone, who was born in Suffern, N.Y., in 2002. A photograph taken the day before she died shows a wide-eyed baby girl lying next to a stuffed giraffe. The next morning, her mother, Patti, told me, she was wiping Samantha’s face when she heard a gurgle from the baby’s chest.

Before her eyes, Samantha was turning blue. Blood began to spill from her mouth. Ms. Stone dialed 911, and minutes later, a doctor who heard the call over a radio was there performing CPR. Samantha went to one hospital and was flown to another.

But the damage was irreparable. Samantha had gone 45 minutes without oxygen: She lapsed into a coma and died six days later.

It wasn’t until several years later that Ms. Stone learned about the pulse oximetry test. “This could have saved my daughter,” she told me. “There is no parent that should ever have to go through what I went through.”

Pulse oximetry is not a costly, exotic procedure. Most hospitals already have oximeters and use them to monitor infants who suffer complications. You can buy one at Walmart for $29.88.

A recent study in New Jersey, the first state to implement the screening, estimated that the test cost $13.50 in equipment costs and nursing time. If hospitals use reusable sensors similar to those found on blood-pressure cuffs, the test could cost roughly fifty cents.

As medical technology advances, few screenings will be so cheap or simple. Recent years have seen controversy over prostate cancer and mammography screenings. Medical ethicists have to weigh the costs of each program and the agony caused by a false positive against the lives saved.

But with pulse oximetry, the false positive rate is less than 0.2 percent — lower than is seen for screenings newborns already get. The follow-up test is usually a noninvasive echocardiogram, or an ultrasound of the heart. A federal advisory committee came down in favor — three years ago.

“There’s really no question, scientifically, this is a good idea,” said Darshak Sanghavi, a pediatric cardiologist and a fellow at the Brookings Institution. “The issue is, how do we change culture?”

Opposition has taken two forms. One is from doctors who believe policy makers shouldn’t interfere with how medical professionals do their jobs. The other is from smaller hospitals, which worry about access to echocardiograms and the costs of unnecessary transfers.

These concerns can be addressed fairly easily. Nurses in New Jersey and elsewhere have been able to work the test into their normal routines. A rural hospital should already have a protocol to transfer a newborn in serious condition. If Alaska can do it, less remote states can, too.

But this is not simply a rural health care problem. Cardiologists and neonatologists I’ve spoken with said they knew of hospitals in New York City, Boston and metropolitan Atlanta that weren’t screening newborns for heart defects.

“It’s completely the luck of the draw of where you deliver,” said Annamarie Saarinen, who has pushed for the screening since her daughter narrowly avoided leaving the hospital with an undetected heart defect.

Fortunately, our son’s condition was also caught and corrected. The only lasting effects are a three-inch scar on his side and checkups with a cardiologist. He will live a normal life. He will be able to play sports and climb things he’s not supposed to.

Shouldn’t every baby have that chance?

Catherine Grant

Sep. 23, 2013, 12:14 p.m.

In 1990 my third child was born with coarctation of the aorta.
Unfortunately, it was not discovered at birth. Ten days later he suffered heart failure overnight. I awoke at 6 AM on a Sunday to find him blue, but thankfully still breathing. We rushed him to our community hospital, big mistake, where we were told there was nothing they could do and to prepare for his inevitable death. Fortunately, there happened to be a pediatric cardiologist in the hospital taht day for another child and he told the emergency room staff that the baby needed to be transferred immediately to a Pediatric Hospital for cardiac surgery. He was rushed to Children’s Hospital In Newark, NJ (which has since been folded into Beth Israel Medical Center).There we were told he would have the coarctation repaired as soon as they could assemble the surgical team, if he survived that long. In less than 24 hours he was successfully operated on. I, too, learned that heart defects affect 1 in 100 births. My son has lived these last 23 years the absolute light of his father’s, two sister’s and my lives! To come so close to losing him was a life altering experince for all of us. I would never want anyone to have that experience and if a $13 test can prevent that, all babies should have it.

watch out for that Gadolinium metal contrast trial to “see how blood flows” and MRI studies. clinical trials gov and gadolinium, embeds in bones, brain, eyes, so on…. a safety issue. check your baby’s delicate kidneys. read about this metal. Do IT!

Natasha Thoren

Sep. 23, 2013, 1:56 p.m.

In 2012, my daughter was born with coarctation of the aorta as well, but she wasn’t diagnosed until she was 2 months old. There were several signs, she failed to nurse adequately and gain weight, she was tired often and had cold feet, but our physician kept explaining those things away as “first” time mom worries. We ended up in the pediatric emergency room because of a fluke fever and from there they noticed her respiratory rate was high. It took several days and lots of tests before they diagnosed her heart defect. She was in pretty severe heart failure, but had surgery and is fine now. If all of that could have been prevented with a small $13 test, then I agree all babies should receive it. They say heart defects are the most common defect in babies, so why wouldn’t we want to find them before they become fatal.

E. R. Kelly

Sep. 23, 2013, 2 p.m.


In the interest of accuracy, Coarctation Of The Aorta is not by itself a cyanotic congenital condition i.e., it does not show up as a drop in oxygen sat on a Pulse Ox meter.  Some times it (Coarctation of the Aorta) is associated with other conditions that can make it cyanotic. 

Coarctation of the Aorta is picked up by other signs usually noted by a careful examination of the neonate by a general pediatrician specializing in new borns,.  I am glad the writers story has a happy ending but I am writing this specifically to correct some wrong ideas pushed along with the happy story.  We cannot throw technology willy nilly at a broken system and hope for good outcomes.

W. Larch MD, FACC

Yogini Rachel

Sep. 23, 2013, 2:20 p.m.

In 1952, my older sister was the youngest person to have open heart surgery. She was 6 weeks old. The surgery was to repair a “hole over her heart.” The hospital where she was born refused to do anything. They simply told my mother, “You have a very sick baby.” The doctors were just waiting for her to die. Fortunately, my 21-year-old mother was in nursing school and had the courage to remove her baby from that hospital. Mom took my sister to a different hospital in Detroit where the life saving surgery was performed. A picture of my smiling mother holding her healthy baby was published in the Detroit newspaper with a caption announcing that open-heart surgery had been performed on a 6-week-old baby. The yellowed newspaper clipping was a valued family keepsake. I’m amazed that after 60 years, hospitals still respond inconsistently to these kinds of cases. My sister is 60 now. She has 3 grown children and grandchildren.

Thank you for this informative article. To find out what conditions your state screens for check out This is the nation’s newborn screening educational resource center.

The operative word is CAREFUL. Instead of rushing newborns and their moms out the door, we should be doing serious examination of both and having good home follow up by qualified nurses. Minimum cost and maximum suffering averted.

This article hits close to home! I *wish* the hospital would have done a pulse-ox on my first son Connor when he was born. Perhaps his heart defect would have been caught sooner and so many of the little symptoms that he had when he was a newborn (excessive sleepiness, disinterest in nursing, weight loss leading to a near diagnosis of failure to thrive) would have been attributed to their actual cause as opposed to this new mom thinking that I just wasn’t cut out for this whole motherhood thing.

The pediatrician did a pulse-ox on him when he was sick as an infant and it was low, but we attributed the low reading to his fever and congestion. Luckily, when he was 17 months, we finally went to a pediatric cardiologist who diagnosed him with atrial septal deviation and he had open-heart surgery to correct the problem when he was 20 months old. We are really fortunate to have the healthy 8 year old boy we have today.

Julie Meier Wright

Sep. 23, 2013, 6:31 p.m.

If the pulse oximeter is $30 and available over the counter, if I were pregnant I’d buy one and ask my doctor to use it!  It is absolutely tragic that these babies die or are at risk of dying when such a simple test is available.  It should not take a law for the medical community to adopt it.

Barbara Brooks

Sep. 23, 2013, 7:24 p.m.

Since many mothers are only in the hospital for 24 hours, one test might not be enough since some heart defects don’t appear for several days.  The best plan would be to repeat the test at the same time the baby is tested for jaundice.

This test could be even cheaper if pulse oximeter companies would make a reusable pulse ox probe that could be wiped off with disinfectant.  It’s such a waste to use a probe for one test, then toss it.  As it is now, they only make reusable probes in adult sizes.

Arthur Strauss, MD

Sep. 23, 2013, 10:09 p.m.

Universal screening for critical congenital heart disease (CCHD) via pulse oximetry has been mandated via legislation > never submitted for efficacy or cost-effectiveness via a scientific study in as large and diverse a country as the United States. It is estimated that 200 infants/year with CCHD would be found by screening in the newborn nursery before discharge. Only 100 of these infants would have their life ultimately extended by surgery (multiple for some) > the rest would die in childhood. This is not funded > the cost of the screening test for an individual infant is low > for 4 million babies it is not. This doesn’t account for the cost of the heart ultrasound, the cardiologist’s charges in the hospital, transport costs if no cardiologist is available locally, etc. Parental anxiety about screening already being done for other diseases is not incidental. This is not way to change medical practice.

This seems to be the trend in the medical industry, which no legislation about insurance can ever touch:  A cheap test that can be administered at home for pocket change?  No, let’s not talk about that, and let’s certainly not make it an automatic part of care.  That could waste, like, ten bucks per baby, and (to simplify Dr. Strauss’s comment) would “only” save a hundred or so per year.

It’s nice to know that’s the balance, by the way, that the medical industry thinks about.  I can’t think of many people who wouldn’t give up ten bucks per birth for even a small chance to save a kid’s life, myself.  I’m sure I’ve wasted much more money on far less worthwhile things.  I’m sure the bill for giving birth in a hospital wastes far more money on far less worthwhile things, too.

It reminds me of the charity advertising where they talk about fighting some disease that we still have in the United States for something like “just a dollar a day.”  If we can take care of a kid overseas for a dollar a day, then we can presumably do it here, too.

Filipe Ribeiro

Sep. 24, 2013, 10:04 a.m.

One of the reasons that prostate and breast cancer tests are controversial is because of the high false positive rate that can lead to unecessary treatment, as the author mentions. He cites a false positive rate of 0.2% that is claimed to be low. But the real question is, if the test is positive what are the chances that the condition is life threatening? Assuming the numbers given are right, there is a chance in 555 of a baby having a life-threatening heart condition. That means that 554 babies will be just fine. The test has a false positive rate of 0.2% which means that, of those 554 babies that are just fine, about 1.18 of them (0.002 x 554) will have a positive test when there is nothing wrong with them. So, for every baby that gets a positive test and has a life-threatening condition (a true positive), there are 1.18 babies with a positive test that are healthy (a false positive). So, if a baby has a positive test there is about a 50/50 chance that he has a life-threatening condition. This is actually not that bad compared with prostate and breast cancer screening. Still, it means you need to treat twice as many babies as needed. I am writing this just as an illustration of how presumably low false-positive rates can be misleading. You need to know how frequent the condition is and take that into account. In this case, a thorough analysis with more precise numbers would have to be performed to get the accurate chance. I’m taking these numbers at face value. Maybe there is a typo and there is a missing zero somewhere. Therefore, I’m sure 50/50 is not really the right number. But it illustrates the point. One needs to be very careful when claiming that a false positive rate is low. Especially for rare conditions. See for more details.
My 0.02$.

Elizabeth Bremner

Sep. 24, 2013, 5:40 p.m.

Since heart defects are the most common birth defect, we should do one better—use fetal echocardiography on every pregnant woman instead of generic ultrasounds that often miss heart anomalies.  This would allow cardiac care teams to stand by during birth to aid babies, and allow women whose babies have anomalies to opt for cesarean deliveries in the event natural delivery would stress their babies.

THANK YOU for publicizing this important issue!!

In response to Elizabeth.  Blood circulating through the heart of a fetus takes some different pathways than the way it circulates in a human existing outside of the womb.  These differences can protect the fetal heart in utero, but immediately upon the cutting of the umbilical cord after birth - changes occur that will affect the path of circulation.  Those changes then affect the functioning of the heart, and added strain. The act of birthing through the vaginal canal - while the newborn is still attached to the umbilical cord would not unduly stress the heart; therefore a C-section is not warranted. Fetal echocardiograms are more costly and there are probably fewer practioners to perform them than an anatomical ultrasound.  The pulse oximetry is a much cheaper alternative and can be repeated if abnormal.

Filipe, there’s a small flaw in your reasoning, I think, which is that $13 x 555 is about seven thousand bucks per life saved.  I’m not sure there are many people out there who would agree that a baby’s life (or quality of life, if you’re worried about people who would survive until their teens with a mediocre heart) is worth less than that.

Matthew DeNardo

Sep. 25, 2013, 1:24 p.m.

With a little digging into additional sources, we can determine the effectiveness of this screening test.

1/555 have the condition = 0.18%, which means that
554/555 do not have the condition = 99.82%

Of those 0.18% who have the condition, 99.99% will test positive via pulse oximetry (source:

Of those 99.82% who do not have the condition, 0.2% will test positive anyway (false positive)

Doing the math (or using any of a number of available online Bayes calculators such as results in a 47.4% chance that a baby has this condition given that they received a positive test (i.e. roughly 50/50).

There are approximately 4 million babies born in the US each year (source: If we expect 1/555 to have this condition, that is about 7,200 babies born per year in the US with this condition.

According to this article, 1,500 of those babies are not treated prior to leaving the hospital. That means we are already catching approximately 80% of babies born with that condition using existing methods. So, we are really trying to catch something more along the lines of 1/2700 babies rather than 1/555. (Note: This is assuming the pulse oximetry screening does not replace some other method outright, but rather augments what is being done today.)

Using this test as a screener (at $13, per the headline), would cost $52 million annually. Given that a positive test is only actually helping half of these babies, we would be spending $52 million a year to help 750 babies, a cost of about $70,000 per life saved (assuming that every diagnosed baby can be treated, something that Dr. Strauss indicated above is unlikely). Even if you take the eventual $0.52/test over the long haul, that’s still nearly $3,000 per life saved.

I intended this comment to clarify the effectiveness and costs involved so that we are all talking about the same thing. I hope I achieved that.

Arthur Strauss, MD

Sep. 25, 2013, 8:55 p.m.

Laypersons are often confused about medical testing and screening - CCHD screening via pulse oximetry has limited utility as a screen for all congenital heart disease. Fetal echocrdiograms miss 50% of serious heart defects. The pulse oximetry test only picks up 70% of life-threatening heart conditions that affect newborn infants and is not applicable to other lesions (like VSD/ASD) that may require surgery during infancy. To pass a law that mandates screening of all newborns for rare lesions without a payment mechanism for hospitals that have to do the testing and without a timely reporting mechanism is absurd. California did this as an unfunded experiment > data will be collected for 3 years and re-evaluated to see what should have been done properly in the first place.

Dr. Gerard Martin

Sep. 26, 2013, 11:40 a.m.

In response to Dr. Strauss’ comment, I wanted to mention that in the past two months the CDC posted a Pediatrics study that looked specifically at the cost-effectiveness for critical congenital heart defects using pulse oximetry and found that the newborn screening appeared to be a good value for the money. The sooner we can identify an issue the sooner we can intervene and help these children, extending and improving their lives. It also means that over the course of that child’s lifetime we’re actually bringing down the cost of care by using these early interventions. A Lancet study from 2010 written by Tennant reported estimates of up to twenty years of survival for children born with congenital differences and the CDC reports another Pediatrics study that found 69% of infants with CCHD survived to 18 years of age. In regards to the funding of these programs, there are programs that have implemented screening programs without draining resources and money, specifically in New Jersey which was reported on by the Journal of Pediatrics in August 2013.

annamarie saarinen

Sep. 26, 2013, 3:07 p.m.

Thank you again for this important piece.  Grateful things turned out well for your child.  As the mother of a child with CCHD, I wanted to weigh in on some of the comments above.

E.R. Kelly – see for where the screening is implemented or has been legislatively or statutorily added to the routine screening panel.

Dr. Larch – while you are correct about COA, those of us looking at implementation data have been surprised by the asymptomatic cases being detected prior to discharge. You cannot discount the fact that those cases may have left the hospital undiagnosed and risked collapse (or worse) at home. Since my baby was nearly discharged in profound heart failure, I commend all pediatricians and GPs who are able to pick up “hidden” symptoms in clinical evaluation.  Perhaps you are one of them. But babies would never leave the nursery with undiagnosed critical heart lesions if EVERY nurse or physicians was able to “see” the signs. The fact is, at least hundreds – but closer to thousands of newborns DO risk leaving the hospital undiagnosed. I personally know 15 families that had this happen. 8 of them ended up having funerals for their babies. 5 others had much tougher course of surgical care and ongoing feeding or neurodevelopmental issues.

Julie -  the pulse oximeter being used in the hospitals is an FDA approved piece of medical technology, geared for neonates and able to capture a clinical number reading through low perfusion and motion. The over the counter ones are not suitable for use on babies – in a hospital – or at home. 

Barbara - the algorithm was the result of years of research and studies to determine the most effective thresholds/cutoffs for this particular screening.  24 hours of age is the best time to rule out transitional circulation, yet capture a clinical measure that allows time to intervene. Pulse ox companies DO make reusable probes for neonatal use – and actually encourage their use in the newborn nursery for this screening. You can download or view the reuse of probes in a CCHD screening training video ( done in collaboration with the University of Minnesota Amplatz Children’s Hospital and the Minnesota Department of Health.  Properly disinfecting the probe is simple, meets CDC guidelines and, as you suggest, can save money for hospitals and state newborn screening programs.

Dr. Strauss -  I don’t know what data you are or are not looking at, but there is a cost effectiveness study already available. Additionally, you can clearly see the issues associated with late detection of CCHD in the following manuscripts:  Dr. Ted Thompson is one of the hundreds of neonatologists around the country who have championed the screening. He was part of the first hospital system pilot done in collaboration with a state department of health on newborn heart screening using pulse ox. We are still talking about thousands of newborns each year (more than just the “7 core defects” defined as CCHD) that have asymptomatic, yet critical heart problems. And there are many other causes of hypoxemia in newborns (neonatal sepsis, pneumonia, and PPHN) almost all requiring some type of important intervention. These are not false positives, they are true positives for a secondary screening target. The Minnesota study suggests that doing the screening may actually REDUCE the number of echocardiograms (see recent MN manuscript in Pediatrics And parents WANT to know about low oxygen saturation even if it’s ruling out problems. We are talking about minutes or hours here, not days. This condition was nominated for inclusion in the federal panel and as such was subject to a rigorous process that included review of every piece of domestic and international literature, input from physicians in multiple specialties and public health. Two of the parents that testified with me at these federal meetings lost their children – at 7 and 8 weeks of age, to TAPVR (often asymptomatic, very treatable, and very sensitive to pulse ox screening). One of these mothers is a doctor, the other was a nurse. These are NOT conditions that are apparent, and while some cases of CCHD do still have difficult courses of care, the vast majority are very successfully treated with timely intervention. There is no reason not to check O2 in newborns. It is simply a vital sign that happens to have the ability to detect hidden hypoxemia in babies. As a point of note, any condition included in the Recommended Uniform Screening Panel (RUSP) will now be part of an essential benefits package and is a covered cost.

Annamarie Saarinen, Mother to Eve, CCHD survivor and co-founder 1in100/ Newborn Coalition,

We were one of these families that went home undiagnosed. My daughter wasn’t diagnosed until 4.5 months of age, when she was diagnosed with 3 critical heart conditions (pulmonary atresia, TOF, and MAPCAS) and at that time had fallen from the 50% to off the charts. I knew something was wrong or off, but everyone (2 doctors and 2 lactation consultants) told me I was crazy. Whenever I got the diagnosis I couldn’t believe that if she had had a pulse ox test, I would have not been told I was crazy for 3 months. I tell all my friends to have a pulse ox test, as it would have saved us a couple months of crazy. It’s the number one birth defect…. why wouldn’t we have a test for that. Just because a lot are caught on ultrasound doesn’t mean we shouldn’t have a test to try to catch it afterwards.

Arthur Strauss, MD

Sep. 26, 2013, 8:53 p.m.

The 2013 Pediatrics article estimates that the universal CCHD screening program would avert 20 infants deaths/year >  total cost of screening per infants = $20 per infant. At 4 M births/year = $80M. This equates to $4M per infant “saved”. Universal bilirubin screening for the low risk of kernicterus (1/100,000 infants) has been estimated at $10-30M per case averted. I don’t know who determines what is “cost-effective"in the American medical system - but none of this makes sense!

How is the test supposed to be done?
How long does the pulse oximetry need to remain on, and what are the parameters for positive vs negative results?

Hi Deann:  Baby’s First Test and Children’s National Medical Center created this video for parents that answers these great questions if you click on the link below.

Heart Smart:  CCHD Screening for Parents

Our hospital in Ca. Implemented CCHD testing several months ago.  We test all of our babies after 24 hours. The test takes virtually minutes, and if we can catch even 1 baby with CCHD, isn’t that worth the investment to parents, babies, and hospital staff?  With 24 hour discharges returning, this simple screening can help prevent a baby from turning into a statistic.

Jenny Brown, RNC, MS, WHNP

Oct. 12, 2013, 8:48 p.m.

My son was diagnosed with Ebstein anomaly in May after failing his pulse ox test the day he was supposed to be discharged. The hospital started this test only a few months before he was born. His condition was missed on fetal echo, and eventually he will need surgery to correct. I will always be grateful that this test was part of regular hospital discharge. I will educate my OB patients in the future about this test.

Get Updates

Our Hottest Stories