A paralysing conflict inside the nation's product safety regulator has prevented the agency from taking action against a popular baby product that studies have linked to at least 48 infant deaths over 27 years and that public health officials say should be banned, according to a Washington Post investigation.
The cause of the breakdown is a small team inside the U.S. Consumer Product Safety Commission that does not believe the product - padded cot bumpers - played a role in most, if any, of the infants' deaths, derailing the agency's attempts to regulate or ban cot bumpers. The position of the CPSC's health sciences directorate, charged with assessing product risks, is opposed by pediatricians and safety advocates and contradicted by medical studies. It also has led senior officials at the CPSC to question whether their own scientists are producing reliable research about the risks of infant sleep products.
"I'm troubled by what appears to be a wide gulf between the views of our highly skilled technical staff and experts in the public health community," said Robert Adler, a longtime commissioner who recently took over as the CPSC's acting chairman. Adler plans to invite experts from outside the agency to discuss cot bumpers at a public hearing early next year.
The CPSC's health sciences team reached its opinion on cot bumpers by rejecting rulings from medical examiners who found that bumpers were associated with at least 35 deaths from across the country.
That is what happened after the 2012 death of Dylan Micjan, a 3-month-old found by his parents with his face pressed against a padded cot bumper at their home in Suffolk, Virginia. Police investigated. And an autopsy found a distinct white line along the baby's cheek and nose.
"This means that the child's nose was compressed by something after he was dead," pathologist Wendy Gunther, an assistant chief medical examiner, later testified in a deposition.
She ruled the baby's death was "likely due to accidental suffocation in bumper pad."
But CPSC staff, reviewing the case four years later, said they believed the bumper wasn't to blame.
The CPSC's stalled response to evidence of the dangers of cot bumpers is a story of an agency struggling to fulfill its basic mission to protect the public from unreasonably dangerous products. Even as the CPSC's leadership has sought alternative views within and outside the agency, it appears almost powerless to stop the sale of products that a consensus of public health authorities say should be avoided, according to agency documents reviewed by The Post and interviews with 24 medical researchers, agency employees and doctors.
The current dispute over the agency's failure to act on cot bumpers also is playing out just months after the CPSC was widely criticized for its handling of inclined sleepers - another infant-sleep device - when the agency was accused of being slow to react to a series of infant deaths in Fisher-Price's Rock 'n Play.
"It absolutely makes no sense at all. We should be able to come up with a better system," said Rachel Moon, a pediatrician who studies infant sleep deaths and helped write the American Academy of Pediatrics' safe-sleep guidelines. Those guidelines recommend against using both cot bumpers and inclined sleepers.
The CPSC health sciences directorate's stance on cot bumpers, which wrap inside a cot and are marketed as products that protect babies, is unusual. The AAP, National Institutes of Health and Centers for Disease Control and Prevention all have warned the public against using bumpers for years. Three states and two municipalities have banned the sale of padded bumpers because of safety concerns. The latest was New York state in August.
Walmart and Target stopped selling padded bumpers. Safety advocates - frustrated by the CPSC's lack of action - recently persuaded federal lawmakers to introduce legislation for a nationwide padded bumper ban.
But bumpers remain popular with parents. An estimated 1.2 million sets were sold in the United States last year, according to an industry official. They can be found at retailers including Pottery Barn Kids, Amazon and Buy Buy Baby. They can be seen in Instagram photos and magazine ads showcasing the perfect nursery. There are traditional padded bumpers made of fabric, which provoke the greatest concern from experts, along with newer mesh cot liners, which claim to pose less of a suffocation risk.
But there is little evidence that padded bumpers protect babies from injuries, according to the pediatrics academy. And while deaths from bumpers are relatively rare - ranging from just 27 deaths over two decades to as many as 62 deaths in 11 years, reflecting different data sets and the challenge of tracking cases - the deaths are entirely preventable, experts say.
"The burden of proof should be in proving safety - especially with something as devastating as preventing sleep-related deaths," said Benjamin Hoffman, a pediatrician in Portland, Oregon, who chairs the pediatric academy's Committee on Injury, Violence and Poison Prevention.
The lonely position of CPSC staff on cot bumpers was spelled out in agency memos in 2013 and in 2016, when medical examiner's findings were repeatedly rejected. The 2013 memo noted that the staff "does not believe the available records suggest that there is a risk" of death from cot bumpers. That analysis proved influential in curtailing the CPSC's actions.
The author of the staff memos was Suad Wanna-Nakamura, a longtime researcher and physiologist in the federal agency's health sciences directorate.
She has worked at the CPSC for nearly three decades and holds a University of Illinois doctorate in physiology, a branch of biology that studies how organisms, including humans, function. She has investigated hundreds of products at the CPSC. She was among the first to warn almost 20 years ago about the dangers of babies being smothered in their parents' bed - now considered mainstream advice. And it is credited with saving lives.
But her recent work on infant-sleep products hase been widely criticized, according to interviews with doctors, past and present agency employees and consumer advocates who have discussed the issue with the scientist over the years. Their accounts were supported by agency documents.
She also worked on the agency's investigation into the risks of inclined sleepers, which failed to identify why babies were dying in the devices, contributing to the agency's failure to push for a recall even as deaths mounted and criticism from public health officials grew. At least 35 babies died in the Rock 'n Play over 10 years.
The CPSC ended up asking a biomechanics expert at the University of Arkansas for Medical Sciences for answers. That study, released in October, found that the design of inclined sleepers was inherently dangerous. The agency now is considering rules that would effectively ban the entire class of products.
On cot bumpers, the view of CPSC staff, led by Wanna-Nakamura, is also an outlier.
"I have had multiple conversations with her about this and I still don't understand what her rationale or what her opposition is," said Moon, who works at the University of Virginia Medical Center.
"We were absolutely and totally opposed to her analysis," said Dean Woodard, a former CPSC deputy director for hazard identification, who co-wrote two studies on cot bumper deaths.
But Carol Pollack-Nelson, a safety consultant and former CPSC employee, said she thinks criticism of Wanna-Nakamura is misguided.
"I have no reason to doubt CPSC staff," Pollack-Nelson said. "They're experts and they don't refrain from saying something is good or bad."
Wanna-Nakamura did not respond to multiple requests for an interview, eventually declining to comment through a CPSC spokesman.
CPSC spokesman Joe Martyak said the agency's decisions on cot bumpers reflect the work of the entire health science staff, not the opinion of any single staffer.
But, he said, the agency "recognizes further data analysis is needed to identify the risks associated with cot bumpers."
One plan, according to an agency official who spoke on the condition of anonymity to discuss internal deliberations, is to ask the same outside biomechanical expert who looked at inclined sleepers to conduct a safety review of cot bumpers.
Figuring out why babies die during sleep is hard. There rarely are witnesses. Multiple risk factors are common. And doctors still don't fully understand why some babies just stop breathing.
An estimated 3,500 babies die unexpectedly during sleep each year in the United States. That number is half what it was in the mid-1990s, when public health authorities started urging parents to put babies to sleep on their backs, not their stomachs. Yet there are still only theories to explain why that advice works.
Today, about a quarter of unexpected sleep deaths are blamed on accidental suffocation.
And products often play a role, experts say.
But by law the CPSC - which regulates 15,000 consumer products - doesn't assess the risks of products before they're sold to the public.
The CPSC might only react to problems, Moon said, "but it's still unclear to me why they haven't acted on cot bumpers."
At the CPSC, Wanna-Nakamura's skepticism about the dangers of cot bumpers was well known.
In 2004, she wrote a CPSC letter, in response to a baby-products industry inquiry, saying the health sciences team found "insufficient detail" to conclude that cot bumpers caused deaths. Three years later, she led a staff review that found "there was no conclusive evidence" that bumpers were the primary cause of fatalities. And she wrote a 2010 CPSC research paper that discounted reports of problems with bumper pads.
Yet outside the CPSC, doubts soared.
Chicago and Maryland moved to ban padded bumpers by 2013. The lack of federal oversight was noticed.
"So if Maryland does this, it actually could push the CPSC to do this nationally," said Peter Beilenson, then a county health official involved in the state's effort, according to a hearing transcript.
Alarm bells had been going off about cot bumpers since the publication of a 2007 Journal of Pediatrics study that found 27 deaths involving the products over two decades. The study, led by St. Louis neonatologist and sleep-death researcher Bradley Thach, used CPSC data to find the deaths. And it quoted from medical examiner case summaries, such as, "face obstructed by cot bumper pad - positional asphyxia."
"Because bumpers can cause deaths," the study read, "we conclude that they should not be used."
It was considered by pediatric researchers to be a landmark finding.
But the study was disputed by Wanna-Nakamura, along with CPSC's health sciences staff, and attacked by the Juvenile Products Manufacturers Association, a trade group that represents the mostly small companies that make cot bumpers.
The JPMA hired a consulting firm called Innovative Science Solutions to produce a report in 2009 that found no "causal relationship" between cot bumpers and infant deaths. The JPMA said the report would be published "in a peer-reviewed medical journal" - important validation of the findings. That never occurred.
Another company hired by the JPMA, Exponent Failure Analysis Associates, then wrote a report claiming Thach's study suffered from methodology problems.
"We weren't looking at debunking anything," Rick Locker, the JPMA's general counsel, said in an interview. "We were looking at where the data was."
And the data, Locker said, has failed to show that bumper pads cause suffocations. Locker also pointed to an industry-approved voluntary safety standard for cot bumpers.
Bumpers are needed, he said, because they prevent babies from getting their arms and legs caught between cot slats.
"That problem is the No. 1 issue we're seeing right now," he said.
Keeping babies from getting their limbs trapped is a main selling point for the products, along with softening blows from an infant tumbling against a cot's side.
But the evidence that bumpers work is weak. No study has shown that bumpers prevent entrapments - or that it is something that needs guarding against.
"There is no doubt they are unsafe and unnecessary," said James Kemp, a pediatrician and sleep medicine researcher at St. Louis Children's Hospital.
Young babies aren't strong enough to hurt themselves by hitting a cot's side, said Hoffman, the pediatrician in Portland. The result of a baby's leg being trapped between the slats is a nuisance, not an injury, he said. And by the time a baby is older and stronger, the bumper needs to be removed to avoid the risk it will be used to climb out of the cot.
The trade-off is not worth it, Hoffman said.
"We can fix arms and legs," he said. "We can't fix suffocation."
In late 2011, the American Academy of Pediatrics updated its safe-sleep guidelines, including a warning against using all cot bumpers.
"We felt the evidence was compelling," said Moon, who helped draft the guidelines.
The JPMA decided it had seen enough.
A few months later, it turned to the CPSC. The trade group filed a petition in 2012 asking the agency to draw up a national safety standard to distinguish between what it saw as "hazardous pillowlike" bumpers and "non-hazardous traditional" bumpers.
A federal standard would create a single set of rules nationwide.
It also would cancel out state and local bans.
But first Wanna-Nakamura and her team needed to investigate the risks of cot bumpers.
The prospect of the health sciences staff investigating cot bumpers worried some officials at the product safety agency.
Inez Tenenbaum, then the CPSC's chairwoman, wanted a second opinion.
In a highly unusual move, she asked an agency employee outside the health sciences directorate to write a separate risk assessment on the same topic.
"There were so many incidents I just wanted a fresh set of eyes to look at it one way or another," Tenenbaum said in an interview. She said she trusted CPSC career staff but felt they were missing something.
Tenenbaum tapped NJ Scheers, a statistician who, in the early 1990s, led the agency's Infant Suffocation Project, which investigated the dangers of adult soft bedding. Wanna-Nakamura had worked on the same project with Scheers. The project's report was instrumental in declaring adult pillows and comforters as infant suffocation hazards.
But the two women held very different views on cot bumpers.
In 2013, they wrote separate memos that painted contrasting pictures of whether cot bumpers were safe.
Wanna-Nakamura's report looked at records over a 22-year period and found "no conclusive or persuasive evidence implicating cot bumper pads as the primary cause of death."
Her report ruled out many deaths linked to bumper pads, despite medical examiners' findings, because the information was considered contradictory or unclear or lacked detail. Other cases were excluded because "the bumper pad was not physically touching the infant's face."
In contrast, Scheers's report found 42 bumper-related deaths over 27 years, plus another six cases likely related to bumpers.
Scheers assumed that medical examiners and pathologists provided "the best evidence for how the deaths occurred" because of their training and immediate access to the cases, according to her report.
Scheers's report suggested the commission issue tough standards or even ban padded bumpers.
"I thought, perhaps when they see this, they will change their minds," Scheers said.
The commission instead decided in 2013 the issue required additional study.
The battle over cot bumpers blew up with a second study in the Journal of Pediatrics in 2016.
The title left no doubt: "cot Bumpers Continue to Cause Infant Deaths: A Need for a New Preventive Approach."
Thach was not the lead author this time. Scheers was. She had retired from the CPSC, but her interest in cot bumpers continued. The study said public warnings and voluntary rules were not working. Babies were still dying in cot bumpers - 48 fatalities from 1985 to 2012.
And the problem seemed to be getting worse. Three times as many deaths - 23 - had occurred in the most recent seven-year period than during any prior seven-year period.
Wanna-Nakamura and her staff responded months later with another memo to CPSC commissioners as they prepared to take a second crack at writing a federal safety standard.
The memo methodically laid out how health sciences staff whittled down the death toll likely attributable to cot bumpers. CPSC staff ruled out two-thirds of 107 possible deaths for a variety of reasons, mostly because they said the bumper was not found near the baby's face. Another 24 cases were classified as "unknown," meaning they couldn't tell whether the bumper played a role or they had conflicting information.
Wanna-Nakamura's team found only nine deaths that seemed to be caused by the bumper.
But even those cases were not blamed on bumpers alone, with the memo noting that "the majority included extenuating or confounding circumstances."
Her memo also took pains to put the number of cot-bumper deaths in perspective, one that seemed designed to minimize their importance. The memo said staff's findings suggested that, considering the product's popularity, the annual death rate from cot bumpers was about 0.087 per million products in use, meaning they "account for only a small proportion of cot-related deaths."
This kind of death-rate calculation was dismissed by American Academy of Pediatrics officials. The deaths might be rare, Hoffman said, "but for any family who loses an infant to suffocation in a cot bumper, it's 100 percent of their babies."
The 2016 report from Wanna-Nakamura's health sciences team - which, in effect, argued that cot bumpers were not dangerous enough to require regulation - meant the CPSC was blocked.
Agency leaders felt they needed the staff's scientific backing to justify any action, according to interviews with four commissioners.
So Adler, then a commissioner, attempted to counter the CPSC staff's own findings: He hosted a hearing in late 2016 on cot bumpers with public health experts.
Scheers testified. So did Thach. They were followed by the director of the National Center for Fatality Review and Prevention, Teri Covington, who said the data her group collected from state teams investigating child deaths found 62 cases tied to cot bumpers in just 11 years. The number was so much higher than the CPSC's data that it was clear the agency was missing deaths, Covington said.
Moon testified. So did Joshua Sharfstein, a doctor who pushed Maryland's padded-bumper ban into law when he was the state's health secretary.
Sharfstein criticized the idea that if more than one hazard existed in a cot, you can't blame any of them.
"You don't just say, well, since there was more than one thing, it couldn't be anything," he said during the hearing.
The next month, in November 2016, the CPSC commissioners voted 4-1 to issue a joint statement asking parents to not use padded cot bumpers. The commissioners noted "we disagree with the approach and conclusions" of the health sciences staff and think there is a "clear risk of injury or death associated with padded cot bumpers."
The commissioners asked the agency's staff to go back and draw up ways to make bumpers safer.
In September, the staff came back with its third briefing package on cot bumpers, laying out possible fixes to make the products safer in hopes of winning over the CPSC commissioners.
Wanna-Nakamura was not the author of the health sciences directorate's memo this time. And health sciences staff avoided delving again into the number of deaths tied to cot bumpers. They focused instead on cot-bumper design changes that could help avoid hazardous situations.
Safety advocates and public health officials said the memo did not go far enough. Adler, the CPSC chairman, said he hoped the public hearing next year with outside experts would help.
But the new health sciences memo did note that seven more babies had died in bumper-associated incidents in the three years since the last CPSC memo.
Scheers and Thach never looked at Dylan Micjan's case. It wasn't included in their cot-bumper studies because of the lag time before records appeared in the CPSC system.
Wanna-Nakamura and CPSC staff did.
It is record No. 83 in the health science team's 2016 memo on the hazards of cot bumpers - the same memo that led a majority of the agency's commissioners to conclude that the CPSC's staff had underplayed the threat.
Dylan's case is only briefly summarized. The memo provides a date: 2012. And a victim: a 3-month-old male. The brief narrative says, "The infant was found face-down against a bumper pad that was attached to his cot" and the father attempted CPR. The official cause of death is listed as "probable asphyxia."
The memo also provides the CPSC staff's interpretation of what happened: "The victim was found prone in the corner of his cot with his face 'down' (per the data record sheet) and against the installed bumper."
And it then gives the staff's opinion on whether removing the cot bumper would have changed the outcome: "Unlikely."
The CPSC staff did not think the cot bumper played a key role in Dylan's death.
Case No. 83 is listed among other cases that staff also ruled out, writing: "because the face was not into the bumper in these cases, and the child was not wedged or entrapped, removing the bumper is unlikely to have prevented these fatalities."
The CPSC staff had decided that Dylan's nose and mouth were not pressed into the bumper pad.
But that is not what the medical examiner in Virginia found.
The baby was discovered "dead with his faced pressed against the bumper pad," the medical examiner report reads. The police report said the same thing. So did a report by a medicolegal death investigator.
And Gunther, the assistant chief medical examiner, noted that she took all of these reports, along with her autopsy findings, into consideration to reach her official pathological diagnosis of "Sudden unexpected infant death, likely due to accidental suffocation in bumper pad," according to court records. She declined to comment further when contacted by The Post.
Dylan's death, like almost all infant-sleep deaths, involves some uncertainty, Gunther said, according to court records.
An autopsy can't show suffocation because it rarely leaves behind physical signs. But Gunther had conducted thousands of death investigations in her career by this point, including hundreds involving children who died in unsafe sleep conditions. She explained in court records how she noticed the unusual "linear blanching" running across the baby's cheek and nose - the white line - and that it appeared to be an imprint from the top of the cot bumper.
It wasn't definitive proof, she said. That didn't exist.
"It's just my best medical judgment," she said.
The records also don't tell the entire story of what happened after Dylan died.
They don't describe the panic his father, Travis Micjan, felt as he pulled his son from his cot or the shock his mother, Stefanie Micjan, described feeling as she watched it all unfold. The records don't mention that Dylan was cremated. Or how his mother only learned about the cot bumper's role from the autopsy report, which devastated her. She'd bought the bumper because she thought it would make sleep safer for her baby.
"You put a little bit of trust in that," she said in an interview. "And it fails you."
Dylan's parents sued Walmart, the store where they bought the cot bumper, and the company that made it, Triboro Quilt Manufacturing, claiming they should have known the product was dangerous. The case ended in 2017 with a confidential settlement.
Triboro declined to comment on the case. Walmart, which did not respond to multiple requests for comment, no longer sells padded cot bumpers.
Pottery Barn Kids and Amazon did not respond to a request for comment about why they still do. (Amazon CEO Jeff Bezos owns The Washington Post.) Buy Buy Baby spokeswoman Jessica Joyce said the retailer is "aware of the current debate" and monitoring developments.
Last year, Stefanie Micjan gave birth to a boy. They named him Conner. They bought him a cot. No bumper pads this time. No pillows. No quilt. He sleeps with just a light sheet.
And Conner still sleeps in that cot, next to his parents' bed, where even today his mother wakes up during the night, again and again, just to check on him, to make sure he's all right.