A clinician (off screen) wearing blue gloves presses a diapered infant’s heel against a paper card to collect blood samples.
A clinician (off screen) wearing blue gloves presses a diapered infant’s heel against a paper card to collect blood samples.

Did Researchers Really Uncover the Cause of SIDS?

An interesting but preliminary biomarker study’s reception illustrates the challenges of conducting and communicating nuanced research in the era of social media.

black and white image of young man in sunglasses with trees in background
Dan Robitzski

Dan is a Staff Writer and Editor at The Scientist. He typically works on the news desk and joined the team in 2021. He has a background in neuroscience and earned his master's in science journalism at New York University.

View full profile.

Learn about our editorial policies.

May 18, 2022

ABOVE: A clinician conducting a heel spot blood test on a newborn © ISTOCK.COM, ISAYILDIZ

Earlier this month, scientists published a finding suggesting that reduced activity of a particular enzyme is associated with a greater risk of sudden infant death syndrome, a mysterious condition that killed about 1,250 infants under the age of one in the United States in 2019. The study, published May 6 in eBioMedicine (a Lancet journal), offers preliminary evidence suggesting that measuring this cholinergic enzyme in living infants could indicate a newborn’s risk of SIDS. But experts tell The Scientist that the study itself oversells its findings—and that subsequent news coverage and social media chatter have exaggerated them further, reaching the point of incredulity.

“Sydney researcher who lost son to SIDS makes breakthrough in preventing ‘every parent’s worst nightmare’,” a tweet from the Australian Broadcasting Corporation exclaimed on May 7. Other early coverage of the paper was even more zealous. “Researchers Pinpoint Reason Infants Die From SIDS,” read the headline of BioSpace’s heavily shared article about the paper on May 10. “As the cause is now known, researchers can turn their attention to a solution,” the article concluded at the time.

Parenting Twitter was elated. “THEY FOUND THE CAUSE OF SIDS,” exclaimed a tweet from @TheDebbieMia that, at the time this article was published, had received more than 279,000 likes and 66,000 retweets.

Lead study author and SIDS researcher Carmel Harrington tells The Scientist via email that she was surprised by the study’s exuberant reception. “I was hoping that this work would grab the attention of researchers in my field, because the more of us researching the SIDS phenomena, the quicker the results will come, but I did not expect this level of attention,” she says. “In science the message is nuanced and it is easy for non-scientific people to misinterpret scientific language.”

“By way of clarification,” Carmel, who works at the Children’s Hospital at Westmead in Sydney, adds, “a biomarker does not equal cause and what we have found is a potential biomarker and not a cause.”

A screenshot of a scientifically-inaccurate tweet that reads, in part, “They found the cause of SIDS”
A viral tweet that exaggerates the findings of the new SIDS study

But experts who spoke to The Scientist were deeply critical of the hype surrounding the study—including some of the comments from Harrington herself that suggest that the study could help prevent SIDS. For instance, a May 7 press release from The Sydney Children’s Hospitals Network (with which Harrington is affiliated)—which sparked much of the study’s initial coverage—includes quotes experts say go too far, such as: “This discovery changes the narrative around SIDS and is the start of a very exciting journey ahead. We are going to be able to work with babies while they are living and make sure they keep living.” Harrington has also claimed to the press that SIDS will “be a thing of the past” within three to five years—an assertion that’s experts tell The Scientist is unrealistically optimistic and risks giving parents false hope.

“I would say there’s enough here for us to look further, but the idea that it’s gone around the world as a cure for SIDS just shows that there are real problems with social media,” says Richard Goldstein, director of the Robert’s Program on Sudden Unexpected Death in Pediatrics at Boston Children’s Hospital and Harvard Medical School who was not involved in the research. He and others say they are concerned that the breathless coverage not only oversells the paper’s results, it could actually result in more children at risk if parents rely on prospective biomarkers in the place of tried-and-true prevention methods such as safe sleeping practices.

A potential, tentative biomarker for SIDS

Harrington tells The Scientist that the study arose from a popular hypothesis that many SIDS deaths stem from the infant’s inability to rouse themselves from sleep when they can’t breathe. Some of the existing literature ties the neurotransmitter acetylcholine (ACh) to arousal, and, subsequently, deficits in acetylcholine to SIDS. Because the enzymes butyrylcholinesterase (BChE) and acetylcholinesterase (AChE) modulate ACh activity, Harrington explains that she “hypothesized that a deficit in the cholinergic system may well be evidenced by an abnormal level of one or both of these enzymes.”

To probe this question, she and colleagues determined the levels of AChE and BChE in dried blood samples taken from infants who later died of SIDS and compared them with the levels of those enzymes in blood taken from living babies or babies who died of known causes. The SIDS samples contained lower levels of BChE than the living or non-SIDS death groups, which Harrington explains can be used as a proxy for ACh levels and, therefore, as a potential indicator of an infant’s risk of death from SIDS.

“This finding represents the possibility for the identification of infants at risk for SIDS prior to death and opens new avenues for future research into specific interventions,” Harrington tells The Scientist. “I stress it is not a cause, but a marker of vulnerability. We know that not all infants with low BChE will die, and we need to better understand the complexity of this system. Future research will do this.”

See “Serotonin Neurons Implicated in SIDS

However, the relationship between SIDS and BChE is less convincing than some of the paper’s press coverage might suggest, experts say. For one, the findings differ somewhat from those of prior biomarker studies, including a 2010 PLOS ONE study that found an overabundance of cholinergic enzyme expression in SIDS cases compared to healthy controls—and that levels of the enzyme increase with age, which is noteworthy given the living babies analyzed in the new study were older than the deceased infants considered.

Furthermore, as researcher K. Iceman notes in a Twitter thread, a figure from the paper suggests that there is actually very little difference between the SIDS and control groups, calling into question the paper’s key assertion that BChE could be a useful biomarker. In an email conversation with The Scientist, Iceman adds that they feel deeply skeptical of the way Harrington relied on public donations to fund the research and continues to ask for crowdfunding support.

Eugene Nattie, a Dartmouth emeritus professor whose research focused on the neuroscience of breathing, tells The Scientist that he had reservations about the study, and passed along a newsletter by Brown University economist Emily Oster arguing how the study, though not invalid, was vastly exaggerated to the point of being misleading. Noting that a particular problem was the oft-repeated refrain that screening for BChE levels would help prevent SIDS, Nattie adds that the newsletter “pretty much sums up my thoughts.”

Screenshot of a tweet that includes a figure from the SIDS biomarker paper and indicates that there’s notable overlap between the experimental and control groups
A tweet showing a figure from the study that indicates there’s a smaller difference between the SIDS group and controls than press coverage indicated.

And experts point to other weaknesses in the research. In addition to the overlap between groups, Goldstein notes that some infants’ data were excluded from the study, leaving only a handful of newborns in the SIDS group, and that little is known about their exposure to other, more established risk factors such as secondhand smoke or having their mother smoke during pregnancy. Still, he says he likes the premise of the study and that he’s overall “very positive” about it. It’s simply “extremely preliminary,” he says.

“I think their methods are fine, but I just don’t think you can claim very much other than ‘very interesting study, let’s move forward’,” Goldstein adds.

Harrington didn’t respond to specific critiques, instead telling The Scientist, “I think it is really important that this study engages scientific debate and critique.”

The dangers of oversold science

It’s perhaps unsurprising that so many social media users celebrated the study online. When an infant dies of SIDS, their family members are rarely given any explanation as to how their child died or whether it was preventable. The autopsy procedure typically conducted is woefully outdated, says Goldstein, as it lacks a genomic screen that might actually prove useful for SIDS research. Because of that, the medical examiner is only able to offer families in search of answers the equivalent of a shoulder shrug. What mourning parents are left with, however, is the knowledge that any future children and even future generations of their family will face a greater risk of SIDS—and that scientists cannot yet explain why. The ordeal, as Goldstein describes it, is not remotely comforting.

Harrington shares that she was driven to unravel the mysteries of SIDS after losing her son to sudden unexplained death in childhood (SUDC), which is similar to SIDS but for children older than one year. She was unable to accept that no one could tell her why it happened, she says. For those who have lost children to SIDS or SUDC, a biological explanation for their child’s death, no matter how tenuous, may come across as much-needed reassurance that they’re not to blame—or suggest that there’s hope for future infants.

Such sentiments were all over social media after the press release and early coverage of the paper: “This is momentous. Parents of babies lost to SIDS can now know it’s not their fault, despite being told it’s from blankets or belly sleeping or overheating. Relief finally for these already heartbroken families,” read one tweet that had more than 60,000 likes before it was deleted. “I’m hoping this can at least provide an answer for grieving parents & eventually a cure,” says another.

As many have noted, it’s hard to blame the tweeters, given the way the research was presented initially. Reporters and the public simply repeated hyperbolic statements made in the press release—like that the findings could render SIDS a thing of the past in a matter of years. And while one could argue those covering the story should have done their due diligence, one can hardly expect the average non-scientist on Twitter to pore over the intricacies of a figure deep in a scientific study.

A tweet arguing that the general public is not to blame for overhyping the SIDS research, given its uncritical reception in the news media
A tweet highlighting the uncritical nature of early coverage of the study

An interesting but perhaps over-celebrated study wouldn’t necessarily pose much of a problem, Goldstein says, except that some of the writeups of the study that were widely shared suggest that a biomarker test could replace current recommendations for preventing SIDS, which include safe sleep practices meant to prevent the asphyxiation that experts say remains important regardless of biological risk factors. “Previously, parents were told that SIDS could be prevented if they took proper precautions,” the BioSpace article initially said, for instance. “These families can now live with the knowledge that this was not their fault,” it goes on to quote Harrington as saying.

“The dangerous part is that some of those tweets say, ‘You see, it’s not sleep position,’” Goldstein says. “Actually, there’s no way that you can take that away from this study. If there is an arousal defect, that arousal defect needs to be stressed. But somehow, as always happens with the social media world, it just becomes twisted.”

“I’ve already seen a ton of tweets about parents relieved that they can let their babies sleep on their tummies,” reads a tweet from Iceman’s thread. “NO, you should not! ALWAYS on the back and on a safe surface!”

Some of the initial coverage of Harrington’s study has been edited post-publication to reiterate the importance of precautions and to remove the language suggesting those measures were outdated. BioSpace, for instance updated their headline to more accurately reflect the nature of the work and added a clarification that “despite this breakthrough, it is still abundantly important that anyone caring for a baby should follow safe sleeping practices.” And The Sydney Children’s Hospitals Network added, “While these findings do offer some hope for the future, parents are strongly encouraged to continue following safe sleeping practices including placing babies on their backs to sleep, keeping their head and face uncovered during sleep and ensuring a safe sleeping environment in both the day and night” to its original release.

But experts’ and clinicians’ concerns linger that parents may miss the corrections and may falsely believe safe sleeping precautions are now unnecessary. Even if that’s not the case, the mental whiplash from the abrupt change in messaging may undermine confidence in future SIDS studies. And exaggerations of findings—whether from the researchers directly, in press releases, or in social media posts—can undercut people’s trust in scientific research, notes Goldstein. “It kind of demeans science,” he says.

Finding the true “cause of SIDS” may be impossible

One thing the hype got right is that figuring out the mechanisms underlying SIDS would be a momentous breakthrough. Just last week, Goldstein wrote in a commentary piece in the New England Journal of Medicine that identifying the molecular and genetic factors that result in increased SIDS risk is crucial. But one major challenge to that goal is the fact that the condition is what Goldstein calls “a heterogeneous collection of the unknown” and a “diagnostic dustbin,” referring to the way SIDS is given as the cause once all other known causes of death are ruled out. His and other research has convinced him that there are likely multiple variations of SIDS out there, each with its own risk factors and demographics resulting in the same outcome.

See “Genetic Mutation Linked to Cot Death

Better genomic profiling and screening can help differentiate among subpopulations of vulnerable infants, he notes, allowing scientists to take the crucial first step of defining what the problem is. Indeed, as genetic profiles or biomarkers linked to SIDS have been identified over the years, they’ve helped rename certain conditions as distinct from SIDS—effectively whittling away at the remainder of SIDS that remains shrouded in mystery.

In that sense, research like Harrington’s can serve as important pieces of the puzzle. But celebrating incremental progress as a breakthrough may hold the field back, Goldstein says.

“What we need to do is to make the leap beyond just these very crude, superficial, circumstantial observations to things that are more rooted in the biology,” he says. “Ultimately, our ideal is not to explain factors contributing to what happened, but to prevent it from happening; to treat whatever it was or prevent the trigger.”

But much like the danger of positing lofty goals such as preventing all heart attacks or aneurisms, Goldstein cautions that the complete prevention of sudden infant deaths may be a pipe dream. “I don’t think we can do that with a thing that is called ‘SIDS.’”

Correction (May 19): This article has been updated to correctly reflect that the infants analyzed in the study included those who died of known causes.

Correction (May 20): This article has been updated to reflect that a figure from the paper shows an overlap in values instead of error bars. The Scientist regrets the error.