An estimated 40,000 children avoided peanut allergy diagnoses after the guidelines for when to first expose kids to the food allergen changed, according to new research.

The dramatic drop in childhood peanut allergies arrives a decade after a watershed study found that feeding peanut products to babies reduced their chances of developing an allergy by over 80%.

For decades, parents were advised to avoid feeding common allergens, such as peanuts, to infants. In 2015, guidance shifted around peanuts for high-risk kids and expanded two years later.

A study published Monday in the journal Pediatrics found that peanut allergy rates in children under the age of 3 dropped by roughly 43% after the recommendations were expanded in 2017. Rates for all food allergies dropped by about 36%.

“What I was surprised by was the magnitude of the results,” said Dr. David Hill, an attending allergist at the Children’s Hospital of Philadelphia and senior author of the study. Even being able to say that allergy rates plateaued “would be huge news, but the fact that we actually saw a reduction in the onset of new food allergy in kids under 3 is incredible.”

Hill and his colleagues analyzed electronic health records from nearly 50 pediatric practices to track diagnoses of food allergies in about 120,000 children between 0 and 3. Fifteen months marks the peak onset of peanut allergy, according to the study.

Kids were deemed newly allergic if they received a diagnosis code by a provider for a food allergy and were prescribed an EpiPen, Hill said.

The reductions in diagnoses were found even though only about 29% of pediatricians and 65% of allergists reported following the expanded guidance issued in 2017, surveys found.

Confusion and uncertainty about the best way to introduce peanuts early in life led to the lag, according to a commentary that accompanied the study. Early on, medical experts and parents alike questioned whether the practice could be adopted outside of tightly controlled clinical settings.

Sung Poblete, chief executive of the nonprofit Food Allergy Research and Education, who was not involved in the study, hailed it for its focus on real-world data.

While it’s clear that the practice of “eat early, eat often” for foods that can provoke allergic reactions works in a clinical setting, “it’s really important to know that in the real world, this can also decrease incidence and prevalence for the infant population,” she said.

When a person has a peanut allergy, their body responds to the proteins in peanuts as though they’re dangerous. The immune system tries to fight them off, triggering symptoms ranging from hives and diarrhea to anaphylaxis, a life-threatening condition.

Poblete said that the findings highlight the need to change policy to further drive down life-threatening allergy diagnoses. That could include the U.S. Department of Agriculture including peanut products in their infant food packages, she said.

Food allergy prevalence has been on the rise, with 2.2% of U.S. children having a peanut allergy, according to commentary on the new study. Besides delayed introduction of allergenic foods, other risk factors include cesarean section births and antibiotic exposure, according to Hill.

“This is just a call to double down our efforts to understand why it is that children develop food allergies and how we can better treat and ultimately cure these diseases,” Hill said.

Since the period the researchers studied, children’s food guidelines have expanded further. In 2021, the Academy of Pediatrics recommended introducing major food allergens, including peanuts and egg, to all infants between four to six months of age.

The Associated Press contributed to this report.

Designing the system that would bring a slice of Mars back to Earth at NASA’s Jet Propulsion Laboratory — the Southern California lab that pioneered American rocketry and the scientific exploration of our solar system — was her dream job.

As she worked toward degrees in mechanical engineering, she watched JPL launches and became enamored with the photos the lab took on Mars. She attended a JPL open house, which she said felt like “Disneyland.” She applied to work at JPL more than 60 times. When she finally got the job working on the Mars Sample Return Mission, she hoped to spend the rest of her career there.

But on Tuesday, she was one of the 550 employees the lab laid off — representing more than 10% of the workforce.

It was the fourth round of layoffs in two years at the lab, which has struggled since Congress pulled funding for its flagship Mars Sample Return mission because of a ballooning budget and timeline.

Morale has tanked amid reports of management problems. Staffers say they’re following budget discussions in the national news while hearing little from the lab’s leaders.

“There’s been this creeping dread in anticipation,” said the mechanical engineer, who spoke on condition of anonymity to share her views candidly. “The boot was once again raised to stomp on us, but we didn’t know when it was going to drop.”

As a result, an institution with an illustrious record of solving the hardest problems in space now faces a daunting task here on Earth: reclaiming its place at the vanguard of exploration and innovation.

“People forget how much JPL is known internationally,” said Fraser MacDonald, senior lecturer in historical geography at the University of Edinburgh in Scotland and author of the book “Escape From Earth,” about JPL’s founders. To MacDonald, the lab is “a major scientific and technological anchor in Southern California.”

JPL — which is operated by Caltech in La Cañada Flintridge and funded primarily through NASA — was born in the 1940s, after experiments by Caltech rocket scientists caught the eye of the U.S. military.

Many of the tales of their early endeavors — including a 1936 test that ended with an oxygen line catching fire, creating, essentially, a flailing flame thrower — are now told in hyperbole, MacDonald noted. Regardless, they formed a “quintessentially Californian story,” he said, which helped fuel worldwide admiration.

After World War II, JPL was largely sidelined from the military’s rocketry endeavors, as the U.S. instead focused on a secret mission to bring Nazi scientists into the country to advance rocket development. But when the Cold War propelled the U.S. to seek technological dominance on Earth and beyond, it was JPL that developed the U.S.’ first successful satellite, Explorer 1, designed to study cosmic rays.

The same year, 1958, the U.S. government created NASA, and JPL found a new home.

Contracts for ambitious, high-profile NASA missions have become JPL’s lifeblood. But in recent years, there have been fewer of these to go around.

The White House and Congress — under both Presidents Biden and Trump — have increasingly focused on human spaceflight to the moon and Mars. Meanwhile, mission costs have risen because of economic factors ranging from supply chain expenses to employee cost of living, said Casey Dreier, chief of space policy at the Planetary Society, a space science advocacy organization led by Bill Nye.

At the same time, a series of well-documented recent management stumbles have not helped JPL’s cause.

After NASA’s Psyche mission to a metal-rich asteroid failed to meet its 2022 launch date, the agency commissioned an independent review, which found that internal reorganizations and personnel changes created distracted and uninformed managers and burned-out, stretched-thin staffers.

And, in 2023, another sobering independent review determined there was “near zero probability” of Mars Sample Return making its proposed 2028 launch date, and “no credible” way to fulfill the mission within its budget.

NASA sharply cut its spending on Mars Sample Return in anticipation of budget cuts from Congress — which, by extension, meant steep funding cuts to JPL. The agency eventually began seeking alternative plans from other NASA centers and the private sector, placing JPL in the humbling position of having to compete for its own project.

JPL had beefed up staffing from roughly 5,000 people in the early 2010s to roughly 6,500 to support its flagship missions including Europa Clipper, which is set to explore one of Jupiter’s moons, and Mars Sample Return. But with both Clipper and Psyche now in space and Mars Sample Return shelved, the lab couldn’t find roles for some of the projects’ workers.

“I struggled with balancing the passion that I had for the work with the knowledge that I could be moved off of projects anytime,” said the mechanical engineer, who said that JPLers don’t join the lab for the paycheck. “Why should I pour my heart and soul into it? … A lot of the stuff that we’re doing might never go anywhere. We’re just going to pack it up in boxes and put it on shelves.”

Then came the layoffs for which many had already braced.

In January 2024, the lab let go of 100 on-site contractors. A month later, 530 employees and 40 contractors. When it became clear NASA’s funding for JPL would not substantively change in 2025, the lab laid off an additional 325 employees.

JPL’s 2026 budget is still uncertain, with the government in its third week of a shutdown. But, regardless of which version of the budget Congress passes, the lab probably won’t see any significant new streams of cash.

That could explain why JPL — which says its latest layoffs are not due to the shutdown itself — chose October to send out the layoff notices.

Throughout the two years of steady layoffs — which, all in all, eliminated roughly a quarter of all staff — employees would pepper lab leaders with the same questions at town halls: When were layoffs happening and who was going to be let go? They received few answers.

The JPL Reddit forum, which had historically been a place for aspiring engineers and scientists to ask employees about getting hired and about life at the lab, turned sour. Employees vented their frustrations and posted layoff information that leaders wouldn’t share.

“The morale at JPL is horrid right now,” the mechanical engineer said. “There is a lot of distrust and dissatisfaction that’s been built up against the people who are at the top of decision making on lab.”

Yet, she still sees hope for Southern California’s premiere planetary science lab: “I do genuinely believe that JPL can weather the storm.”

This is not the first time JPL has faced a funding crisis.

In 1981, President Reagan’s administration proposed slashing NASA’s planetary science funding.

NASA’s administrator at the time responded that the cuts would make JPL “surplus to our needs.” JPL seriously considered returning to its origins by pivoting to Department of Defense work, but politically connected Caltech leaders managed to convince Congress and the White House to keep funding Galileo, JPL’s flagship mission at the time to explore Jupiter’s atmosphere.

Few have hope that Mars Sample Return will spur recovery as Galileo did. Dreier, for example, sees a different set of options for the lab in 2025: increasingly rely on defense and national security projects, and use its robotics and Mars expertise to support NASA’s new goal of landing humans on the moon and Mars.

“Who else has landed on Mars as many times as JPL has?” Dreier said. (Answer: No one. JPL has done it successfully nine times since 1976. In fact, a successful landing without JPL didn’t happen until China pulled it off in 2021.)

Saving JPL’s signature planetary science missions like the Mars rovers and Jupiter orbiters is more challenging. Unlike in 1981, the current proposals to cut government spending on science reach far beyond NASA.

And while human spaceflight to our nearby celestial neighbors is certainly a reasonable endeavor, Dreier said, “the cosmos is a lot bigger than just the moon and Mars.”

If you’re faced with a serious disease, you better hope it’s not a rare one.

After an often tortuous path to diagnosis, people with rare diseases are likely to find that good treatment options don’t exist and none is on the horizon. Many of these conditions are poorly understood, and conducting studies in tiny patient populations can be practically impossible. Most drugs won’t pan out, and those that do will have little demand and financial payoff, no matter how beneficial they are. Drug companies usually direct their attention elsewhere.

Recognizing these challenges, policymakers have worked since the 1980s to encourage rare-disease drug development. They’ve earmarked federal research funding, established dedicated programs at the Food and Drug Administration, extended protection against competition to help secure profits, and awarded lucrative incentives for rare pediatric disease drugs. Still, 95% of rare diseases, which affect an estimated 30 million Americans, lack any approved therapy.

Some blame the FDA. They say the agency is too rigid, imposing impossible requirements and demanding unreasonable proof of effectiveness and safety for rare disease drugs. Some suggest patients would be better off if the FDA just got out of the way — not only for rare disease treatments, but also more broadly.

The current administration now seems determined to do just that, at least for products that fit the “Make America Healthy Again” agenda, like stem cells and psychedelics (never mind recent intrusions on COVID vaccines and abortion drugs).

Large swaths of FDA experts have been DOGE’d or otherwise forced out of the agency. Commissioner Martin Makary has proposed approving drugs based just on their scientific plausibility, while the agency’s chief medical and scientific officer pledged to “take action at the first sign of promise for rare diseases” — potentially making treatments available far sooner, even though many drugs that look promising at the start turn out not to work.

Just last month, the FDA announced approval of a drug the commissioner claimed would help “hundreds of thousands of kids” with autism, not based on a clinical trial but on published case reports of 40 patients with a potentially related condition — alarmingly and unprecedentedly accepting anecdotes as evidence of efficacy.

The call to disarm the FDA is coming from inside the house.

Criticism of the agency’s gatekeeping is certainly not new, but critics are especially vocal now. Banking on expectations that the Trump administration would break through perceived red tape, they are calling on the White House and new FDA leadership to approve rare disease drugs with far less attention to safety and effectiveness than to keeping companies financially interested in developing rare disease treatments and bringing them to market.

It can be reasonable to assess rare disease drugs differently as usual market-driven incentives often fail to yield treatment options. That’s why the FDA has been remarkably flexible about these approvals for decades. Sure, the FDA sometimes says no — but what if the drugs it rejects just weren’t any good?

Ideally, you want to see a drug’s effectiveness replicated in at least two studies to be confident in the results. For drugs approved to treat common diseases (outside of cancer), that replication is typical. But only 13% of approved rare disease drugs (again outside of cancer) relied on more than one robust clinical trial to show they work. Recent FDA policy has made clear that this single-study approach will be the rule for rare disease drugs and potentially for other conditions going forward.

Rare disease drugs are also increasingly granted “accelerated approval,” a pathway that allows drugs for serious diseases to be approved based on predicted rather than proven benefit. Companies must complete required studies after approval, but the FDA has allowed drugs to stay on the market even if these studies fail. This happened for a recent gene therapy for Duchenne muscular dystrophy, a therapy that was later linked to patient deaths.

Even outside accelerated approval, the agency sometimes approves drugs that miss the targets chosen in advance to prove the drug works. A recent study found this happens in 1 of 10 approvals, about half of which were for rare disease.

Despite this flexibility, criticism continues. Rather than taking FDA’s refusal to approve a drug as a critical warning sign, these decisions are often met with the assumption that the FDA must be wrong.

Take the agency’s recent action on elamipretide for Barth syndrome, an ultra-rare, life-threatening genetic disorder characterized by heart, muscle and immune system abnormalities that affects about 300 patients globally.

Given the tiny patient population, Stealth BioTherapeutics, the company developing elamipretide, conducted a trial in just 12 people, which failed to show the drug worked. Some patients continued in an extended version of the trial and appeared to perform better on tests of walking distance and fatigue. However, the FDA reasonably worried this might be because of patients’ awareness that they were receiving the drug, leading to a placebo effect.

After reviewing the evidence, the FDA issued a letter in May 2025 refusing to approve elamipretide. Recognizing the need for flexibility, however, the agency left the door open to approval based on a new, unverified measure of patient benefit: improved muscle strength in the knee. In September, following substantial public criticism from patient advocates and members of Congress, the FDA granted the drug accelerated approval. Stealth will now have to complete another study to see whether the treatment really helps patients — but even if that study fails, the agency may not withdraw approval.

Even if elamipretide fails to pan out, one might wonder what harm lies in just approving it. Maybe it can provide some hope to patients who have nothing else, while encouraging companies not to abandon rare diseases.

The problem is this: Those who criticize the FDA for setting the bar too high shrug off trial data that fail to show benefit, arguing that it doesn’t mean the drug is ineffective. But it is very hard to prove that a drug doesn’t work. If that were the standard, FDA reviewers should just close up shop, leaving a universal “approved” stamp for any drug that appears not to kill patients.

When it’s working well, FDA approval signals to patients (and their doctors and insurers) “this drug has been shown to work” — or at least “this drug has been shown very likely to work.” If FDA approval means anything less, such as “this drug has not been shown not to work,” it fails to serve patients, leaving them no better off than if they were browsing unproven dietary supplements on Amazon. They might even be worse off, if duped into relying on FDA approval as a meaningful indicator of benefit.

Rare disease patients, like all patients, should have drugs that work. The burden must be on companies to prove that their drugs do. Shifting or altering that burden by changing FDA approval standards won’t help, but other changes might. For example, policymakers could improve existing legal approaches that allow patients with serious diseases to try investigational drugs that aren’t yet approved. The federal government could also increase support for the research needed to understand, diagnose and treat rare disease, helping companies focus on the most promising targets and minimizing failures. Unfortunately, the Trump administration’s ongoing decimation of federal health agencies and research instead sets back rare disease science.

Public trust in authorities like the FDA is already depleted. Demanding that the agency greenlight more rare disease drugs, evidence be damned, will make this problem worse — and likely won’t leave rare disease patients better off. Rather than blaming the FDA, the policymakers, companies and patient advocates should be doing all they can to get better drugs in front of the agency’s reviewers.

Holly Fernandez Lynch is a senior fellow at the Leonard Davis Institute of Health Economics at the University of Pennsylvania, where she is also an associate professor of medical ethics and law. Reshma Ramachandran is a family medicine physician and assistant professor at Yale School of Medicine, where she co-directs the Yale Collaboration for Regulatory Rigor, Integrity and Transparency.