Newborn Genetic Screening

Where We Started and Where We're Headed

My interest in newborn screening began in fifth grade because my mother, Dr. Mary Louise Efron, was the physician in charge of Massachusetts’s pilot screening program for phenylketonuria, or PKU.

PKU was the first condition to be widely screened for in American newborns through a population-based laboratory testing program. In 1963, Massachusetts became the first state to require PKU screening for newborns. By the end of that decade, screening had become routine in most states.

Growing up, I heard discussions of the program at home. They gave me an early sense of what it means when medicine gets something profoundly right. Later, in medical school in the 1980s, I watched newborn screening continue to expand.

The ethical case for PKU screening was quite clear. Infants with PKU cannot properly metabolize phenylalanine, an amino acid found in protein-containing foods. Without treatment, the condition can cause severe and irreversible neurocognitive harm. When affected babies are identified early and promptly started on a carefully maintained low-phenylalanine diet, most can avoid that devastating outcome.

The diet is demanding and restrictive—a real cost to children and families. But the benefit of early identification is so substantial that the rationale for screening is compelling. Early newborn-screening programs were built around conditions for which timely detection could clearly change a child’s life.

What is unfolding now is more complicated.

Several research programs in the United States and elsewhere are evaluating whole-genome sequencing for newborns. Rather than testing for a defined group of treatable conditions, sequencing can produce information relevant to hundreds or thousands of genetic conditions. Some findings may be actionable in infancy. Others may be uncertain, relevant only later in life, or impossible to interpret with confidence.

The ethical questions are substantial, and so are the practical ones. Families may need counseling about results that range from immediately useful to ambiguous to deeply unsettling. Health systems already stretched for genetics expertise will have to decide which findings should be sought, returned, stored, and acted upon.

Because this technology is advancing rapidly, while protocols and policies are still being shaped, I want to explore some pressing concerns about how we proceed. I will emphasize the human consequences that require careful—not hasty—thought and wide public debate.

There are many excellent sources on these dilemmas (1–3). Here, I want to begin with the lived experience behind the policy questions. I hope this essay will encourage readers to take the matter seriously while there is still time to influence how genomic testing of our children develops in the United States and around the world.

The story of one mother, one child, and one chromosomal finding is instructive.

A mother noticed that her daughter, born in 1993, was not developing normally. When the child was six months old, her mother sought genetic testing. The girl—whom I will call Anne—was found to have a rare chromosomal abnormality involving one of her X chromosomes.

The condition was so uncommon that the geneticists had very little information about its likely course. Anne’s mother was told that her daughter would probably require special education and might never walk or talk.

The prognosis was frightening—and, as it turned out, far too bleak. Anne will always need supervision and someone with whom to live. But at age 33, she is communicative, dances, rides horses, and is a joyous person.

Her mother reflects that, if newborn screening had identified Anne’s condition at birth, it might have spared her six months of being told she was “hyperattentive” when she knew something was wrong. Anne’s diagnosis also helped her family obtain insurance coverage for badly needed developmental services, including physical and occupational therapy.

On the other hand, the scanty information conveyed after Anne’s genetic testing at six months—combined with the grim and inaccurate prognosis her mother was given—terrified her unnecessarily.

This family’s story illustrates both the potential value and the potential burden of earlier genetic knowledge. A result may provide critical information that improves a child’s health; in some cases, interventions must begin at birth or soon thereafter to prevent harm. An early diagnosis can also open the door to much-needed medical care, services, or surveillance.

But, when a rare or never before identified genetic abnormality occurs, the consequences of that abnormality may be quite uncertain. As in Anne’s case, this may lead to predictions being offered by well-meaning medical providers that turn out to be wrong, and wreak uncalled-for havoc with a family’s peace of mind.

These are some of the questions genomic sequencing has placed before us — and before every family whose child is born into this new landscape. The challenges are already arriving: conditions uncovered that are neither preventable nor treatable, and that may not manifest until adulthood. Results of uncertain significance will multiply as these technologies spread. In all such cases, at the least families need qualified clinicians who are available when it matters — professionals with the training and the time to explain what is known, what remains uncertain, and what, if anything, should be done next.

Should children and their parents bear the burden of foreknowledge of conditions that cannot be clearly defined, or that may be neither preventable nor treatable? How might receiving such information affect family relationships and children's sense of themselves as they grow? These are not easy questions, but they are essential ones.

Part 2: Where newborn genome sequencing stands today

For more than sixty years, newborn screening has generally involved a heel prick and a panel of targeted tests. The goal is to identify a defined list of conditions for which early treatment or intervention can improve outcomes.

In the United States, the federal Recommended Uniform Screening Panel currently includes 38 core conditions and 26 secondary conditions, although the conditions screened for differ from state to state. Conventional newborn screening is highly effective at what it was designed to do: detect specific disorders for which early detection can alter childhood care.

Whole-genome sequencing is different. It aims to analyze nearly all of a baby’s DNA—roughly three billion base pairs—and produces far more information than a targeted screening panel. But more information does not necessarily mean more useful information.

A genome may contain variants associated with disease, variants whose significance is uncertain, and variants that are common and harmless. It may also contain findings related to conditions with variable severity, incomplete penetrance, or no currently available treatment. Sequencing therefore does not simply answer the question, “What disease does this baby have?” It raises a series of decisions: What should we look for? Which findings are sufficiently reliable and actionable to return? Who will explain them? And how should the information be stored and protected over a lifetime?

Several major research programs are trying to answer these questions.

The BabySeq Project, a collaboration involving Boston Children’s Hospital and Brigham and Women’s Hospital, was among the earliest randomized trials to sequence newborns and return selected findings to families. It has generated important information about clinical utility, parental response, and the considerable work health-care systems must do to integrate genomic information responsibly into pediatric care.

NC NEXUS, based in North Carolina, examined the use of exome sequencing in newborn screening and explored the difficult choices families face about what kinds of results they want to receive. It focused not only on laboratory findings, but also on the decision-making framework around them.

GUARDIAN, led by Columbia University, is among the largest U.S. genomic newborn-screening studies. The study evaluates genes associated with more than 450 conditions, with the aim of identifying childhood-onset disorders for which earlier recognition may lead to meaningful medical action.

But documented concerns have emerged from this new landscape. Families have sometimes been given genetic findings when the implications were not clearly understood, when prognosis was uncertain, or when no treatment could change the course of the condition. Such information may be useful in certain circumstances, but it can also create anxiety, alter how parents see their children, lead to difficult and sometimes costly follow-up testing, and place families in a medical limbo without clear guidance about what to do next.

These efforts are not confined to the United States. In England, the NHS Generation Study is evaluating whole-genome sequencing alongside traditional newborn screening, with plans to enroll up to 100,000 babies. Other countries, including Australia, are conducting their own pilots, each within different health systems, consent practices, and assumptions about the role of genomic information in public health.

The science is advancing rapidly. But the ethical and clinical infrastructure needed to absorb what sequencing can uncover is not developing at the same pace. Meanwhile, pediatric-care systems in many places are already stretched to the breaking point. In the United States, for example, some families who lose Medicaid coverage have no realistic means of paying for medical care out of pocket. Introducing complex genomic results into an already fragile system raises urgent questions about whether children and families will receive the follow-up care they need.

We still need robust standards for interpretation, appropriate consent processes, sufficient counseling capacity, responsible data governance, and long-term systems for helping families understand results as scientific knowledge changes. Medicine’s oldest principle—first, do no harm—counsels against moving faster than our medical and social infrastructure can responsibly support.

Part 3: These are some of the questions that must be addressed before newborn genomic sequencing is broadly implemented.

CONSENT AND AUTONOMY

1. As newborn genomic sequencing scales to entire populations, will parents receive the counseling they need to make truly informed decisions? For adults considering predictive testing for conditions such as Huntington disease, genetics programs have historically used careful counseling protocols to give people time to consider what they wish to know—and what they may prefer not to know. If research studies are providing that level of discussion now—and there is reason to question whether this is always happening even in research settings—what will happen when genomic screening is offered at population scale?

2. For conditions that cannot be prevented or treated and may not manifest until adulthood, is it ethical to ask parents to consent on behalf of infants who cannot decide for themselves? Or should clinicians withhold such information until individuals are old enough to provide informed consent?

3. At what age, and through what process, should the person whose genome was sequenced gain meaningful control over how that information is used, stored, or shared? Genomic information generated in infancy may outlast childhood by decades. The policies governing it must be designed accordingly.

4. Can consent be fully informed when the universe of possible findings is itself unknown? Traditional testing may yield upsetting results, but patients can generally be told what they might learn. Genomic sequencing does not always offer even that assurance.

CLINICAL INFRASTRUCTURE & WORKFORCE — ARE WE READY?

5. Are health systems prepared to deliver genomic findings responsibly? There are not enough genetic counselors in the United States to meet current demand — including for cancer patients who already require genetic evaluations to define optimal treatments. Even at existing, limited levels of genomic testing, access to genetic services is profoundly uneven. Primary-care clinicians may not have the time, training, or specialist support needed to interpret complicated genomic findings and guide families through their consequences. What would responsible delivery require — and do we have those systems in place?

6. How should programs handle uncertain findings? Not every genetic variant found has a known meaning. Responsible screening programs may choose not to return variants of uncertain significance, but these decisions require clear standards. As sequencing expands, who determines which findings are sufficiently reliable to report? How are those judgments updated? And who is accountable if counseling falls short?

EQUITY AND ACCESS

7. Will genomic newborn screening widen existing health disparities? Genetic reference databases historically have underrepresented many populations, including people of non-European ancestry. That underrepresentation can make some variants more difficult to interpret. Will genomic screening benefit all children equally—or create more uncertainty and potential harm for families who are already underserved?Will genomic sequencing become another advantage concentrated among families with better insurance, good access to health systems, and proximity to academic medical centers?

Even if sequencing itself is covered, will families have equitable access to genetic counseling, confirmatory testing, specialty care, and long-term follow-up?

WHO OWNS THIS DATA, WHO WILL PROTECT IT — AND FOR HOW LONG?

8. Who owns a newborn’s genomic data, and for how long? Genomic data is uniquely identifying and permanent. It has implications not only for the child but also for biological relatives. Who controls it? Who should be allowed to access it? What happens if the institution or company holding it is acquired, hacked, subpoenaed, or closed?

In the United States, we have not done an adequate job of protecting highly sensitive personal information. The repeated exposure of Social Security numbers through data breaches should make us cautious about assuming that genomic information—which is permanent and has implications for biological relatives—will automatically be kept safe.

9. What are the broader social consequences of collecting genomic information at population scale? The United States has legal protections against some forms of genetic discrimination in health insurance and employment, but important gaps remain, including in life, disability, and long-term-care insurance. As genetic sequencing moves into broader use, what guardrails exist—and are they sufficient?

I will be examining many of these questions more deeply in future posts. I hope this initial overview will help spark thoughtful discussions and debate on this important matter.

References:

• NPR: The Secrets in Your Baby’s Genes

• ACLU: Widespread Newborn DNA Sequencing Will Worsen Risks to Genetic Privacy

• Front Line Genomics, "The State of Newborn Sequencing" (2024)