• New York’s first measles death in more than 30 years.

• How fast do a mother’s measles antibodies fade?

• Do grandparents need a measles booster?

• How early should you get vaccinated before the due date?

• Shingles, pneumococcal, and RSV vaccines for grandparents.

New York’s first measles death in more than 30 years.

A child in Lewis County, New York, has died in the state’s measles outbreak. State and county officials announced the death on November 2nd and called it New York’s first known measles-associated death in more than 30 years. Yesterday I wrote about why some kids still don’t get vaccinated in New York despite the state eliminating non-medical childhood vaccination exemptions in 2019.

Thanksgiving is three weeks away, and many grandparents will be holding a new baby for the first time.

How fast do a mother’s measles antibodies fade?

In the United States, babies don’t get their routine measles, mumps, and rubella (MMR) vaccine until 12 months, although they can get an extra early dose as early as 6 months during an outbreak or before international travel. Until then, babies rely on antibodies from their mothers, which fade fast. In a 2019 study of 196 infants in Ontario, 92% of babies tested at 3 months had already lost protective levels.

So for months, the people around a baby are its main defense against measles.

Do grandparents need a measles booster?

Not if they have 2 documented MMR doses, which the Centers for Disease Control and Prevention (CDC) counts as proof of immunity. The CDC doesn’t recommend antibody testing after that, or a routine third dose for measles, even during an outbreak. People born before 1957 are assumed to be immune because nearly everyone caught measles as a child back then.

If you were born in 1957 or later and can’t find documentation of MMR vaccination, get at least 1 dose now. Some people who got a measles shot from 1963 through 1967 got an inactivated vaccine that didn’t work. If that’s you, or you don’t know which kind you got, get another dose.

An extra MMR dose does no harm if you’re already immune. A quick shot at the pharmacy is a lot easier and faster than having blood drawn to test for antibodies. MMR’s weakened vaccine virus doesn’t spread from you to the baby.

In April 2025, Texas told adults with 1 documented dose in its outbreak counties to get a second.

How early should you get vaccinated before the due date?

In a CDC study, Tdap given to mothers in the third trimester of pregnancy prevented about 78% of whooping cough cases in babies younger than 2 months. It prevented about 90% of the cases that put a baby in the hospital. Whooping cough is most dangerous before 2 months, when babies get their first shot.

Anyone around a baby should get Tdap if they haven’t had it as an adult, plus a flu and COVID shot every season.

Get these shots at least 2 weeks before you meet the baby, since your body needs about that long to make antibodies. But because about 1 in 10 U.S. babies are born before 37 weeks, I’d get Tdap, flu, COVID, and any MMR dose 4 to 6 weeks before the due date. Two MMR doses must be at least 4 weeks apart, so start 8 weeks out if you need both.

Shingles, pneumococcal, and RSV vaccines for grandparents.

Shingles, pneumococcal, and RSV vaccines mainly protect you, and a grandparent sick in bed can’t hang out with a newborn.

In trials, the shingles vaccine Shingrix, which the CDC recommends at 50 and older, prevented 96% to 97% of cases in people in their 50s and 60s, and 91% in people 70 and older. Its side effects can keep you from normal activities for 2 to 3 days. Don’t get it the week you’re planning to visit.

The CDC recommends the pneumococcal vaccine starting at 50, and most people need 1 shot. It recommends 1 RSV shot at 75, or from 50 to 74 for people with conditions such as heart or lung disease.

You can get most of these on the same day, in different arms. Medicare Part D has covered recommended adult vaccines with no cost-sharing since January 2023. Part B covers flu and pneumococcal shots.

Finally, stay away when you’re sick and wash your hands before holding the baby.

• New York’s first measles death in 30 years.

• What changed after New York ended non-medical exemptions?

• Schools that don’t check, and children who aren’t in school.

• Did federal vaccine changes cause this outbreak?

• Who can reach unvaccinated families?

New York’s first measles death in 30 years.

A child in Lewis County, New York, has died in the state’s measles outbreak. State and county health officials announced the death on November 2nd and called it New York’s first known measles-associated death in more than 30 years. They haven’t released the child’s age, vaccination status, or cause of death. They’ve asked for the family’s privacy.

Governor Kathy Hochul called the death “a devastating reminder of what’s at stake when Washington deliberately erodes decades of progress.” New York’s record matters more to this story, though, because the state ended non-medical vaccine exemptions in 2019, years before Washington started rewriting the childhood schedule.

What changed after New York ended non-medical exemptions?

In June 2019, during the Brooklyn outbreak, New York ended religious exemptions from school vaccine requirements. The law covers children in public, private, and religious schools alike. In the four states that fully repealed non-medical exemptions, kindergarten MMR coverage rose 4.0 percentage points more than in states that kept them. Medical exemptions rose by only 0.4 points.

Outside New York City, coverage rose 5.5 points in private and religious schools, against 0.9 points in public schools. Before the law, the average private or religious school had 84.1% of its students fully vaccinated, while public schools averaged 97.6%. By the 2021-22 school year, private and religious schools had reached 89.9%, still 8.6 points behind public schools.

Schools that don’t check, and children who aren’t in school.

New York’s law depends on schools checking records, and three Amish schools in New York don’t require proof of vaccination. The state fined them a combined $118,000. Amish parents and the schools sued, arguing the law violates their religious freedom. The U.S. Court of Appeals for the Second Circuit upheld the law on June 30, 2026, after the Supreme Court sent the case back for another look.

Homeschooled children never appear in school vaccination records. The state’s data also miss some private and religious schools. They were nearly four times as likely as public schools to be dropped from the data for incomplete reporting, so the gap may be wider than the averages show.

New York has reported at least 218 measles cases this year. Most came after August, in about 10 upstate counties. Amish communities, where vaccination rates tend to be lower, have been hit hard. The New York clusters appear linked to larger Amish outbreaks in Ohio and Pennsylvania.

In Lewis County, all 15 cases reported through October 14th were in unvaccinated people. The first 11 cases were all in the town of Lowville, and none of the children attended public school. Officials haven’t said which community these families belong to. The 2019 law covers private and religious schools, but it doesn’t cover homeschooling.

Did federal vaccine changes cause this outbreak?

In August, President Trump signed an executive order that recommends splitting MMR into three separate shots. The FDA hasn’t licensed those single-disease vaccines.

No one has reported that a Lewis County family delayed or declined a shot because of anything federal officials said. The national slide in vaccination rates also started before the 2025 federal changes. Among U.S. children with regular medical care, the share who got their first MMR dose on time fell from 79.9% in 2021 to 76.9% in 2024. Johns Hopkins has counted 5,053 cases in 45 states and D.C. this year through October 30th. It’s the most measles cases since 1991.

Who can reach unvaccinated families?

During the 2018-19 outbreak in Williamsburg, New York City combined community outreach, easy access to shots, and enforcement. First-dose MMR coverage among children aged 1 to 4 rose from 79.5% to 91.1%. It cost the city health department $8.4 million.

Governor Hochul’s Executive Order 65, signed October 5th, lets pharmacists give MMR to children 2 and older and adds paramedics and midwives as vaccinators. It expires on November 4th, and Hochul hasn’t yet announced an extension.

• Who funds DenseBreast-info?

• Does an MRI help women with dense breasts?

• Does cost keep women from coming back?

• Is my follow-up test free now?

• Who pays for an MRI after a normal mammogram?

• What if you’re on Medicare?

• What can Congress, regulators, employers, and you do?

On October 27th, the Department of Health and Human Services (HHS) announced a plan to help women get answers faster after an abnormal mammogram. Part of the plan is a new agreement with DenseBreast-info.org, a nonprofit that teaches women about dense breast tissue. Companies that make mammogram machines and run imaging centers help fund the group.

Who funds DenseBreast-info?

DenseBreast-info is a nonprofit funded by grants and donations. Its September 2025 press release lists 10 supporters. They include Hologic, which makes mammogram machines; RadNet, which runs imaging centers; Bracco Diagnostics, which makes contrast dye for scans; and Myriad Genetics, which sells genetic tests. The group also works with Congress to pass the Find It Early Act. That bill would require health plans to cover extra breast imaging, at no cost, for women with dense breasts or higher risk.

Under the agreement, HHS’s Office on Women’s Health will work with the group to expand education about breast density. Dr. Tammy Huizenga is an adviser at the Agency for Healthcare Research and Quality (AHRQ). In the HHS press release, she said women deserve information about “screening options beyond mammography.”

Dense breasts have more gland and fiber tissue than fat, and on a mammogram that tissue can hide a tumor. Since September 10, 2024, the Food and Drug Administration (FDA) has required every mammogram report to tell a woman whether her breasts are dense.

Does an MRI help women with dense breasts?

One large trial took place in the Netherlands. Researchers at University Medical Center Utrecht studied 40,373 women ages 50 to 75 who had extremely dense breasts and normal mammograms. Some were also invited to get an MRI. In that group, 2.5 cancers per 1,000 screenings showed up between scheduled mammograms over 2 years. The mammogram-only group had 5.0. The MRI also brought false alarms: about 80 for every 1,000 MRI screenings. The researchers didn’t track whether fewer women died.

In 2024, the U.S. Preventive Services Task Force, a panel of independent experts, reviewed the evidence. It looked at extra ultrasound and MRI for women with dense breasts and a normal mammogram. It found too little evidence to say whether those scans do more good than harm. The task force still recommends a mammogram every other year from age 40 through 74.

Does cost keep women from coming back?

A callback is different from extra screening. When a routine mammogram shows a spot that needs a closer look, the clinic calls the woman back for more tests. Those can include a second mammogram, an ultrasound, an MRI, or a biopsy.

Cost keeps some women from coming back. In a 2022 survey presented to the Radiological Society of North America, more than 1 in 5 women said a deductible would keep them from returning for more tests.

Who pays now depends on three things: your plan, your state, and whether you’re on Medicare.

Is my follow-up test free now?

Since January, most private plans have had to cover the extra imaging and biopsy results a woman needs “to complete the screening process” at no cost when you use in-network doctors. Regulations also require private plans to cover patient navigators, staff members who help with rides, scheduling, and paperwork. The Health Resources and Services Administration (HRSA) made that change in December 2024. HRSA approved the change under the previous administration, and it took effect under this one.

HHS didn’t mention this rule in its press release about the new plan. Before it took effect, benefits advisers warned employers that many plans billed follow-up imaging as “diagnostic,” which meant patients paid part of the bill.

Who pays for an MRI after a normal mammogram?

It matters most for extra scans after a normal mammogram. HRSA didn’t write a separate rule for dense breasts, and its guidelines don’t say whether an MRI for dense tissue after a normal mammogram counts. So the plan and the state decide. Since January 2022, Pennsylvania has required plans it regulates to cover extra screening for women with extremely dense breasts. DenseBreast-info’s release says state laws are inconsistent and that many kinds of plans are exempt from them.

What if you’re on Medicare?

HRSA’s rule covers private plans. Under traditional Medicare, a routine mammogram is free, but a follow-up mammogram costs the yearly Part B deductible plus 20% of Medicare’s approved price. are still in the task force’s screening range.

In January 2023, Medicare made a colonoscopy after a positive stool test part of its free screening benefit. It hasn’t done the same for follow-up breast imaging. That change is up to the Centers for Medicare & Medicaid Services (CMS), Congress, or both.

What can Congress, regulators, employers, and you do?

The Departments of Labor, HHS, and the Treasury enforce coverage rules for private plans. In January 2022, they told plans they couldn’t charge for a colonoscopy after a positive stool test. They could issue the same kind of guidance on HRSA’s rule. Congress would move on the Find It Early Act. Employers can ask how their health plans bill a follow-up after a routine mammogram. A woman who gets a callback can ask her plan, in writing, to cover the follow-up as preventive care under HRSA’s guidelines. She can ask for a navigator, too.

• What the new peptide policy says.

• What testing found in peptides sold online.

• What did the FDA’s scientists say?

• How the policy answers the FDA’s safety worries.

• Who’s watching after the vials ship?

• Hims & Hers plans to sell BPC-157.

• Who still has a say.

What the new peptide policy says.

On October 28th, the Department of Health and Human Services (HHS) released a notice, signed by Health Secretary Robert F. Kennedy Jr., saying it doesn’t intend to act against pharmacies, federal facilities, or doctors that make six unapproved peptides. They are BPC-157, TB-500, KPV, MOTS-c, Semax, and Epitalon. Makers must meet the policy’s conditions, and each product’s label must include the sentence “This is a compounded drug product. Not for resale.”

HHS told me the policy “does not grant FDA approval or establish clinical safety and effectiveness.”

What testing found in peptides sold online.

Thousands of Americans take BPC-157, and almost all of them buy it from online vendors that sell it “for research use only.” CBS News’ Adam Yamaguchi bought a $90 vial labeled as a different experimental peptide, retatrutide, at a Brooklyn bodega. University of Iowa scientists found filler and 13 bacterial colonies inside and no peptides. A preprint analysis covered about 6,400 samples that buyers and sellers sent to a testing lab. More than a third held a dose more than 10% higher or lower than labeled.

If people are going to inject these drugs anyway, yes, a licensed pharmacy beats a corner store. A pharmacy has to show where its ingredient came from. HHS told me its work “includes confronting the risks of black-market peptides.”

What did the FDA’s scientists say?

Federal law limits which bulk ingredients a pharmacy can use. The ingredient must be part of an approved drug, meet an official quality standard, or be listed by the Food and Drug Administration (FDA) as serving patients whose medical needs can’t be met by an FDA-approved drug. That list lets pharmacies use ingredients that no drug company ever took through approval, often old treatments that doctors have relied on for years. Before the FDA adds a substance to that list, it weighs what it is, whether it’s safe, whether it works, and how pharmacies have used it. That bar is lower than drug approval, which requires clinical trials.

The FDA’s drug reviewers recommended against listing all six. For BPC-157, they found “insufficient clinical safety information”. They also found no evidence that it treats ulcerative colitis, the specific use they reviewed. A 2026 systematic review of BPC-157 for injuries and tissue repair found 3 human studies with 30 people in all, and it rated the evidence very low.

In July, the FDA’s Pharmacy Compounding Advisory Committee voted the other way, 8 to 6 on BPC-157. The vote came weeks after the agency named eight new members, six of them with ties to the wellness industry. According to the Associated Press, citing six people who know how the policy came together, HHS changed course after FDA staff wouldn’t back a rule deeming the peptides safe.

A listing would have required a final FDA rule finding that the four criteria favored each peptide. This policy relies instead on a finding by Secretary Kennedy that the new policy protects patient safety. In May, the FDA took a similar approach to unauthorized vapes and nicotine pouches. It wrote that its restraint “in no way has a bearing” on whether those products would ever win authorization.

How the policy answers the FDA’s safety worries.

For TB-500, the policy says the reviewers’ safety concerns “are inferential rather than observed.” For Semax, they’re “mechanistic rather than observed.” For KPV, it says their conclusion “reflects an absence of information rather than an observed signal.”

Harm is “observed” only after someone collects the data, and the FDA’s reviewers counted about 80 people who had received BPC-157 across all human studies. A side effect that hits 1 user in 100 could go unseen in a group that small.

Who’s watching after the vials ship?

Because the policy discounts harms nobody has observed, much of its safety system depends on what happens after the vials ship. HHS may revisit the policy if, among other things, “a safety signal or product quality issue” emerges. Pharmacies must report serious side effects and product-quality problems to the FDA’s MedWatch program and to states. The policy sets no deadline for those reports, and it asks only for serious ones.

The policy’s main added quality check is a certificate of analysis for the raw ingredient, which tests the powder before a pharmacy turns it into a shot. It adds no test of the finished vial. That’s the step where the New England Compounding Center in Massachusetts failed in 2012. Its tainted steroid shots caused an outbreak of fungal meningitis and other infections that sickened 753 people and killed 64.

Last year, two women had to be put on ventilators to help them breathe after shots that included unnamed peptides at a Las Vegas anti-aging festival. Nevada’s pharmacy board couldn’t figure out why they got sick. Investigators tried to get the product to test it and couldn’t.

Hims & Hers plans to sell BPC-157.

The policy remains in effect until the FDA finalizes a rule that adds each peptide to the list or keeps it off. The FDA hasn’t proposed one, and the notice sets no end date. Meanwhile, Hims & Hers has said it plans to sell BPC-157 by prescription once regulators clear the way. Its CEO told shareholders he’s “extremely excited about this category.” A phase 2 placebo-controlled trial of BPC-157 for hamstring strains in about 120 study participants is estimated to finish collecting its main data in February 2027.

Who still has a say.

Because the policy is still a draft, the public gets a say. Anyone can comment at regulations.gov, docket HHS-OASH-2026-0496, starting tomorrow, October 30th through November 13th. A final version could require tests of each finished batch for strength, sterility, and endotoxin. It could set a deadline for side-effect reports, and an end date that forces the FDA to decide.

States license these pharmacies, and the doctors who prescribe from them, and every compounded vial still needs a prescription. In May, Alabama’s medical board warned doctors against prescribing “research-grade” peptides that the FDA hasn’t approved.

• The safety announcement that isn’t

• The reassurance misses the danger

• Why botulism testing isn’t required

• What the government did instead

• Fewer inspectors, same hazard

The safety announcement that isn’t

On October 29th, in Des Moines, HHS Secretary Robert F. Kennedy, Jr. announced the biggest update to infant formula standards in nearly 30 years. The FDA proposed a rule to modernize formula nutrients. It is the first such change since 1998. The agency also released new data on chemical contaminants, and it promised a searchable dashboard. The agency framed it as reassurance. Secretary Kennedy said the FDA would “put babies’ health first.” Most samples, the FDA said, had low or non-detectable contaminants.

Because the standards are nearly three decades old, updating the nutrients is worth doing. So is testing for heavy metals. But neither part of the announcement is about botulism. The proposed rule covers nutrients. The contaminant data covers metals like lead and arsenic. The FDA’s announcement has nothing to do with the outbreaks that sent babies to the hospital.

The reassurance misses the danger

Last year, 48 infants across 17 states were hospitalized with botulism after drinking powdered formula from ByHeart. Botulism paralyzes a baby from the head down. In the worst cases, it reaches the muscles used to breathe. The sickest infants went on ventilators, and some were still recovering months later. This spring, four more infants were hospitalized after drinking formula from Nara Organics. Both brands drew their milk powder from one shared supplier. The FDA found Clostridium botulinum in ByHeart’s milk powder, and its July letter reported a genetic match from that powder to a sick infant. California officials found the same bacteria in an open can of Nara’s. Investigators still cannot say how the powder was contaminated.

A nutrient rule doesn’t do anything about bacterial contamination. Chemical-contaminant testing looks for lead, arsenic, and PFAS. So the reassurance the FDA offered measures a different danger than the one these outbreaks exposed.

Why botulism testing isn’t required

Companies still do not have to test formula or milk powder for Clostridium botulinum. The FDA makes them test finished formula for two other bacteria, Cronobacter and Salmonella. Botulinum is not on the list. In a July letter to the industry, the FDA asked companies to watch their suppliers more closely, but it did not require testing. It sent the question of whether to test to two international science panels for review.

The FDA could label Clostridium botulinum a hazard that requires a control, the same way its rules already treat other germs. Congress could pass H.R. 7867, the March bill that would require this testing, but it hasn't moved since. And through WIC, the federal nutrition program which buys more than half of the formula sold in this country, the federal government could set formula safety requirements.

What the government did instead

The announcement introduced a new rule updating iodine and vitamin D, the first of three planned nutrient rules, and asked whether to require more of a fatty acid called ARA, which aids infant brain development. Last winter, a contaminated batch of ARA oil from a supplier in Wuhan compromised infant formula sold in 10 countries and sickened nearly 150 babies. That oil got into formula through a pathway called GRAS, short for “generally recognized as safe,” a pathway that lets companies add ingredients without clearing them with the FDA first. On GRAS, the new announcement offers only a promise to “consider” tighter rules.

Former FDA employees told my colleagues at KFF Health News that “Operation Stork Speed,” the initiative this falls under, was built by listing the infant formula work already underway, then branding it as a new push. HHS called that account false. The one action with real money behind it came from a courtroom. Last month, Abbott agreed to pay $385 million to settle Justice Department and state claims that it sold formula from an unsanitary plant and billed federal programs for it. Abbott admitted no fault.

Fewer inspectors, same hazard

The announcement landed days before the midterm elections, as some of the administration’s supporters have lost patience over slow progress on food safety.

And the people who could catch it are fewer. Roughly 40% of the inspector jobs for critical-foods plants, formula plants included, sit empty. The administration also shut down a food-safety committee that was studying which babies were most at risk from Cronobacter found at a formula plant.

A parent can check a lot number, but they can’t label a bacteria a hazard, staff an inspection force, move a bill, or set the terms on half the infant formula in the country. The public may submit comments until February 26th at regulations.gov under docket FDA-2026-56789.

Watch here (jump to 13:50): https://www.idweek.org/joseph-e.-smadel-lecture

Transcript:

The letter that fired the director of the National Institute of Allergy and Infectious Diseases is two paragraphs long.

It’s dated September 26, 2025. The reason? The HHS secretary’s constitutional authority.

Twenty-two days earlier, she had filed a whistleblower complaint with the Office of Special Counsel, the channel federal law gives employees who report wrongdoing. She described grants and clinical trials canceled without peer review, vaccine research pushed aside, and HIV and TB studies abroad cut off. She says she was fired for it. Her lawsuit is still pending.

Many of you know her. Some of you trained with her, and more of you were mentored by her.

Today she runs the society that hosts this meeting.

She’s Jeanne Marrazzo. Who in this field can afford to speak up? Who has already paid for it? And what do the rest of us owe them?

This lecture is named for Joseph Smadel. IDSA describes his work as “a bridge between the basic laboratory and the physician caring for infected patients.” In 1948, he took chloramphenicol into the field in Malaya and showed it worked against scrub typhus and typhoid. He worked at the Rockefeller Institute, Walter Reed, and NIH: academic, military, and government science in one career. He crossed lines this field often keeps separate. He crossed them by choice. Today, defending the work means crossing into politics. Few of us trained for that. Some who crossed lost their jobs.

Thank you to Ron Nahass, for that introduction, to Ron and Ruth Lynfield for the invitation, and Giulia DeMello and the IDWeek program team. And thank you, Jeanne, for letting me share your story.

I’m here as a physician and fellow of the IDSA. I am here speaking for myself, not representing any other organization. I have no disclosures.

Most of us signed the same deal. You train for a decade or more. You make a lot less money than your classmates. In return, you get to choose what you study. The public paid for the work, and peer review and the civil service were supposed to keep politics out of it. And we owe the people who paid for that work an explanation of what we did with their money.

The sociologists Gil Eyal and Ilana Löwy describe basic science as “a social arrangement, dating from the end of World War II, which gives scientists relative autonomy to fund esoteric research that does not have immediate applications.” That arrangement was conditional.

Between February and August of 2025, NIH grant terminations disrupted 383 clinical trials with more than 74,000 enrolled patients. This year, NIH stopped funding grants by score cutoff. NIAID’s policy now says it “will no longer use a score cutoff as the basis for funding.” A proposed rule would have a political appointee screen grant applications. In June, about 8,000 federal science and program jobs were reclassified as at-will, with no independent appeal.

Economists surveyed over 900 young biomedical scientists for the National Bureau of Economic Research. The share planning to stay in academic research fell from 66% to 44%. The share planning to stay in the United States fell from 93% to 72%. Other countries are now recruiting the scientists American taxpayers paid to train.

The old system wasn’t perfect, and we didn’t get everything right during the pandemic. Some of our messaging claimed more certainty than the evidence had. People noticed, and they remember. None of that justifies canceling trials while patients are still enrolled.

Many of us built our careers on federal money and Washington’s goodwill. In 2025, Washington started taking both away.

Under Commissioner Tom Farley, I served at the New York City Health Department. After the 2008-2009 recession, public health suffered big budget cuts. I realized I could only do so much from within when the public didn’t understand what public health was. And that’s how I ended up on the path I’m on now. I added another job. Bearing witness. And it’s come at a cost.

After my husband, Grant Wahl, died in December 2022, strangers told me I’d killed him with COVID vaccines. A local TV news van parked outside my apartment. People tried to break into my building. My protection was my neighbors. That’s not a policy. I reported some of the calls to the FBI.

Every path in this field comes with a trade.

• Clinical medicine gives you stability and patients who need you, but you take the work home, and your patients need you at all hours.

• Academic medicine lets you choose your questions, as long as the grants keep coming.

• Public health and government let you reach whole populations, and in the past two years, some of our best people lost their jobs there.

• Global health does the most good per dollar, but when USAID was eliminated, much of the funding evaporated. Meanwhile, NIH stopped allowing foreign subawards.

None of those paths is safe anymore.

One skill I’ve taken from job to job is explaining science to people who aren’t scientists. Taxpayers fund this work. That makes them our bosses, and most of us have never reported to them. We all need that skill, now more than ever. The next time there’s a measles case in your hospital, a reporter might call you. I’ve spent years on the other end of that call, so here are a couple of things I’d want you to know.

Most of you don’t call back, for seemingly good reasons. You don’t have time, or your institution sends every call through its press office. You’re afraid of being misquoted or attacked online. In a 2021 survey of over 460 physicians on Twitter, nearly one in four said they’d been personally attacked on social media. And talking to reporters doesn’t count toward promotion.

Talk to them anyway, because the public still trusts you. The public still trusts you. In KFF’s polling this summer, 86% of adults said they trust their own doctor or healthcare provider for reliable health information, and about half said the same about the CDC. In a Pew survey released in January, 77% said they have confidence in scientists to act in the public’s best interests. When I send questions to CDC now, two HHS press secretaries answer them. I can’t get a CDC scientist on the record. The scientists reporters can still reach… work outside government, and many of them are in this room.

Over the past two years, many people in our field have asked me to cover one story or another. But a story for the general public needs someone the audience can care about, someone willing to be named and say what happened to them. When nobody will be that person, it’s hard for me to tell the story.

Some of you may have fears about engaging with the media. Here’s some advice on how to navigate those conversations.

Settle the terms before you say anything. On the record, your name goes next to your words. On background, the reporter can use what you say but credits it to something like “an infectious disease physician at a large academic medical center.” Off the record, they can’t use it at all.

All three are legitimate, as long as you agree on them at the start, out loud. And if your institution limits what you can say, you can still correct the record on background.

Many reporters call to confirm a story they’ve mostly already written. They’re on deadline, so the time to shape the story is before the interview. Get to know the health reporters in your city before you need them, or they need you. Call back even when the story is small. When something big happens, reporters call the people who explained things clearly and picked up the phone last time.

The rest, you can’t fix alone, but you can ask for help. Ask your chair for protected time, and get the answer in writing. Put your public work in your promotion packet, so the committee has to decide whether it counts. Ask your institution what it will do if you’re harassed or someone tries to get you fired: who pays for a lawyer, who gets your home address offline, who calls security. Ask before you need the answer, because people who’ve spoken up… have lost their jobs.

Jeanne isn’t the only one who’s paid.

Susan Monarez. She was director of CDC for about a month. They asked her to commit in advance to approving every recommendation from the vaccine advisory committee, whatever the evidence, and to fire career staff without cause. She refused, and they fired her. She told the Senate, “I could have stayed silent, agreed to demands, and no one would have known… But I would have lost the one thing that cannot be replaced: my integrity.”

Fiona Havers. She led CDC’s network for tracking respiratory virus hospitalizations. When the secretary removed every member of the vaccine committee in June 2025, she resigned. She wrote that she no longer had confidence the data would “be used objectively.”

Tony Fauci. He ran NIAID for 38 years, and Jeanne succeeded him. Four days after his January 2025 pardon, President Trump canceled Tony’s federal security detail. The president told reporters, “They can hire their own security, too.” President Biden had renewed that detail year after year because of threats on Tony’s life.

Several of the people who lost their jobs have applied for positions for which they were well-qualified, but they didn’t get them, Jeanne told me. Some, she was told, were “too hot to handle.” A chair won’t take on that association, not when federal grants fund their department’s or division’s salaries.

Deb Houry, Demetre Daskalakis, and Dan Jernigan resigned from CDC the same week Susan was fired.

Kathleen Neuzil, who ran the Fogarty International Center, filed the whistleblower complaint alongside Jeanne, then resigned.

Paul Offit was dropped from the FDA’s vaccine advisory committee by email, with no reason given.

Peter Hotez has lived with threats for years and keeps talking anyway.

Most of the people who paid were leaders and career scientists in government, where job protections are weakest, and the June reclassification weakened them further. Of the people he left behind, Demetre said: “Not all people at CDC have the luxury to do so. They have families and financial circumstances to consider.”

Trainees are the ones organizing, writing, and showing up. Senior faculty have tenure and standing. Asking the most exposed people to take the most risk gets it backward.

But senior people in this field have too often been the quietest. And, yes, they have the most to protect: the department’s federal grants, the dean’s relationships, the next study section.

IDSA hears from members who keep their heads down because of their patients, too. They run clinics for people who can’t afford attention. That’s one reason I can’t argue with.

During the AIDS crisis, infectious disease doctors and scientists learned to fight in public, alongside activists.

Right now, a group of HIV researchers and clinicians is organizing again. They call themselves SPEAK UP and include Judy Auerbach, Ken Mayer, Tom Coates, Judith Feinberg, Gregg Gonsalves, and Steve Shoptaw. Judy told me they formed to speak up “as elders where other, younger colleagues cannot.” Several of them came up through the AIDS activist years, so they’ve done this before. But they’re not sure how they can have impact this time around.

Former FDA Commissioner Rob Califf is executive chair of a new group called Coethia. Health professions students run campus chapters that answer health misinformation online. Thirteen chapters are up, and several dozen schools have asked to join.

The people best protected to speak up are sitting in this room. The public still trusts you.

This summer I reported on an AI system that does discovery work. Anthropic said that the system, running hundreds of agents at once, sorted through more than 200,000 proteins in about a day and flagged a new CRISPR-like enzyme system. The company’s lab then made the protein and characterized it. But no one has shown it works as an editing tool in human cells, and finding out… could take years.

A team at Stanford and the Arc Institute went further. They trained a model on millions of genomes, asked it to write new bacteriophages, viruses that infect bacteria, from scratch, and built the designs. Sixteen of the genomes worked. A cocktail of them beat E. coli that had evolved resistance to a natural phage. That’s a new tool against resistant bacteria. It’s also the first time a model has written a working genome.

Private companies are building the capacity to do discovery science, fast, at the same moment Washington is cutting the public pipeline that trains scientists. They’re also making new organisms, and this field is who has to answer for them. Our assays have to detect a sequence nobody has seen before. Somebody has to say whether an existing antiviral or vaccine platform covers it, and run the trial that proves it. Health departments need a case definition, and the surveillance to know whether one case is an outbreak. Companies that print DNA to order screen those against a list of dangerous germs, but that screening is voluntary. The Biosecurity Modernization and Innovation Act, a bipartisan bill, would make it mandatory, but it has sat in the Senate since February and leaves the list of covered organisms to the Commerce Department. This field knows which organisms belong on that list. People in this field belong on the committees that decide what gets screened, approved, and paid for. Your institution’s biosafety committee. The FDA advisory panels. Your state lab’s advisory board. You can file a comment when Commerce publishes the screening rule. Those seats go to whoever shows up.

FDA approved an mRNA flu vaccine this summer. In August, Merck and Moderna reported that their personalized mRNA melanoma vaccine met its main goal in a phase 3 trial. Meanwhile, HHS canceled about half a billion dollars in mRNA projects, including pandemic flu work. Pandemic flu vaccines have no market until the day we need them. Much of infectious disease work is something the market won’t pay for: antibiotics, outbreak response, surveillance, vaccines for the next pandemic. That’s why the public has to.

Do both. Go where the science is moving, and fight for the public pipeline at the same time. And ask the companies that draw on that pipeline to help pay for it.

The companies whose products depend on publicly trained scientists and publicly funded discovery have a stake in this pipeline. NIH funding contributed to the research behind every one of the 210 new drugs FDA approved from 2010 through 2016. You can ask those companies to endow fellowships and bridge positions, held by universities or societies rather than by the companies themselves. If you do, create guardrails to protect against influence over who’s chosen, what the fellows study, or what they publish. But that money can never replace public funding. The moment “industry will cover it” becomes the excuse for the next cut, we’ve traded one conditional patron for another.

The rest is up to the institutions in this room. The public still trusts you.

At the University of Minnesota, CIDRAP’s Vaccine Integrity Project did the evidence review the federal government stopped doing. JAMA published it in September, and the IDSA used it for this season’s recommendations for immunocompromised patients.

At Yale, Anne Zink and Dan Weinberger built PopHIVE, a free public health data platform. It pulls hospital records, immunization surveys, and search data into one place. Anne was Alaska’s chief medical officer through COVID. CDC’s dashboards stopped updating during last year’s shutdown. PopHIVE was one of the few places left to watch RSV.

Jono Mermin ran CDC’s HIV and STI center until April 2025. He’s now dean at Columbia Mailman School of Public Health. He’s building a health policy initiative: research, issue briefs, and people working alongside the officials who write the laws. He says it’s to have the country ready for big changes ahead.

The University of Michigan recently launched a Center for American Dialogue to address polarization and declining public trust. One initiative brings public health faculty and students together with communities across Michigan, in partnership with local health departments. Rather than arriving with a predetermined agenda, these town halls will focus on each community’s health concerns and priorities, creating opportunities for genuine dialogue between residents and public health experts.

Craig Spencer at Brown has a proposal called In Health We Trust. He would hold three sessions in each of sixteen communities: residents, then local leaders, then the health department. Each group would hear recordings from the previous group. Health officials would go on the record first, sharing their understanding of what people need. The idea is to measure the distance between their answer and what residents say.

Universities are building places to do this work. But few reward it or protect the people who do it.

Credit. Most medical schools’ promotion guidelines now mention social media somewhere, but very few state that public work counts as scholarship. Nobody has measured whether promotion committees actually count it. Some of you sit on those committees, so you know the answer better than I do.

Protection. Most universities protect faculty with a list of phone numbers. When people tried to get into my building, it was my neighbors who stopped them.

Here’s a cautionary tale: Stanford’s Internet Observatory studied online misinformation. Lawsuits and congressional subpoenas followed. By 2023, legal fees approached seven figures. Its election project didn’t run in 2024. This July, two years after the project shut down, a judge dismissed the last lawsuit against its researchers.

Some of you may know the virologist Angela Rasmussen. She worked at a vaccine research institute whose leadership kept pressuring her to stop talking about vaccines. She’s received a continuous stream of death threats after an anti-vaccine podcaster pointed his audience at her. Her university offered her wellness resources. Then last November, her university’s administration called her in. They said it wasn’t Angela’s fault, but asked whether she wanted to say something to the colleagues who had gotten hate mail… because of her. One said they were worried about her. They asked, as her friend, whether it wasn’t time to stop talking.

A university’s job is to manage the university’s exposure. The cheapest way to cut exposure is to cut the speech. Your academic medical center will do the same math. They will tell you to stop talking.

In 2011, the American Tradition Institute demanded climate scientist Michael Mann’s emails from the University of Virginia. The university fought it alongside him and won in the Virginia Supreme Court in 2014. His colleagues still raised more than $100,000 for his legal bills. Scott Mandia and Joshua Wolfe turned that fundraising into the Climate Science Legal Defense Fund. The fund is free to scientists, and it has assisted hundreds since. It runs on about half a million dollars a year, which pays for legal advice and lawyer referrals. Journalists have the Reporters Committee for Freedom of the Press, which runs a free legal hotline. Meanwhile, we have no fund. Forty dollars from each of our over 13,000 members would pay for one.

Build the fund. Ask IDSA to advocate for it and to publish model institutional policies, work the society says it already does. But keep the money and the lawyers to defend you outside the universities.

Meanwhile, the people who decide whether this work counts are sitting in this room.

• If you chair a department or sit on a promotions committee: count public and policy work as scholarship. Put it in writing, with criteria. The public pays for this work, so explaining it is part of the job, not a favor your faculty do on their own time.

• If you train fellows: most of them won’t stay in academic medicine. Don’t treat that as failure. Treat it as sending your people out into the world as ambassadors for your institution. Help them network, and encourage them to consider different career paths.

• If you control a budget: fund protected time for that work, and bridge fellowships for the trainees whose grants were cut.

• If you lead an institution or a society: put in writing that talking to the public is part of the job, and that you defend the people who do it. Tell them who to call. Don’t tell them to stop talking.

• If you’re senior and secure: write the letters, take the reporters’ calls, and put your name down where a younger colleague can’t yet put theirs.

Count it. Fund it. Protect it. Because the public trusts you.

Smadel connected the laboratory to the bedside. This generation needs people who connect the lab to the newsroom, the clinic to the statehouse, and the university to the communities that pay for the science. Those of us with tenure and standing can do that work now. Our job is to make it safe for the people coming up behind us.

The secretary fired Jeanne 22 days after she filed a whistleblower complaint.

This room has tenure, standing, promotion committees, budgets, and a society of over 13,000 members. We decide what counts in this field and who we protect.

The people best protected to speak up are sitting in this room.

• Covered only after the disease it can help prevent.

• How many employers are dropping coverage of GLP-1s for weight loss?

• Are the health benefits worth paying for?

• Will Medicare save money on GLP-1 drugs?

• What happens to people who lose GLP-1 coverage?

• What should you ask during open enrollment?

Covered only after the disease it can help prevent.

Donna Morris, Walmart’s chief people officer, sent employees a letter on October 13th, ahead of open enrollment. Starting in 2027, the company will cover certain GLP-1 drugs for type 2 diabetes at no cost. The drugs on the list, which require prior authorization, are Ozempic and Mounjaro. Walmart already covered them for diabetes, but now employees won’t have to pay anything at all. According to SHRM, the plan still won’t pay for the drugs when they’re prescribed solely for weight loss.

Ozempic and Wegovy contain the same drug, semaglutide. Mounjaro and Zepbound both contain tirzepatide. In a three-year trial of adults with obesity and prediabetes, 1.3% of people on tirzepatide developed type 2 diabetes, compared with 13.3% on placebo. But the plan won’t make the drug free until someone has the disease the drug can help prevent.

How many employers are dropping coverage of GLP-1s for weight loss?

In a Business Group on Health survey of big employers, coverage of these drugs for obesity fell from 72% last year to 60% this year. Starbucks ended coverage for weight loss in October. Cigna, a health insurer, dropped it for its employees on July 1st. In Massachusetts, 112,000 fewer people with commercial insurance used a GLP-1 for weight loss in early 2026 than in 2025, a drop of 52%.

The employers have the cost data on their side. Economists at Indiana University, the University of Wisconsin, and the University of Southern California followed 288,907 people with job-based insurance for five years after they started a GLP-1. Their other medical spending increased by about 22 cents for every dollar spent on the drug before rebates. Daniel Sacks, one of the authors, told me that counting rebates makes that number larger. Rebates shrink the drug’s cost to patients while leaving the extra spending untouched. Most of those patients started older GLP-1s between 2017 and 2019. For semaglutide, the drug in Ozempic and Wegovy, other medical spending didn’t clearly rise or fall over two years. Sacks says it will take a few more years to learn how tirzepatide, the drug in Mounjaro and Zepbound, compares.

Three other studies found no savings either. Duke’s Jonathan Zhang, who co-wrote one of those studies, says the only studies showing savings are simulations projecting a decade or more into the future.

Are the health benefits worth paying for?

If the drugs don’t save money, the case for covering them rests on health benefits alone, and the health benefits are real. In a trial of 17,604 adults with cardiovascular disease and excess weight but no diabetes, semaglutide cut the combined rate of heart attacks, strokes, and cardiovascular deaths by 20%.

Most of those benefits fade when people stop. In one trial, people who stopped semaglutide regained about two-thirds of the weight they’d lost within a year. Their blood pressure and blood sugar drifted back up, too.

Will Medicare save money on GLP-1 drugs?

Medicare went the other way on GLP-1 drug coverage this summer. Since July, eligible enrollees can get Wegovy, Zepbound, or Lilly’s new pill, Foundayo, for $50 a month through a program called the GLP-1 Bridge. Under the program’s rules, drugmakers supply the drugs to Medicare at $245 a month. Chris Klomp, chief counselor at the Department of Health and Human Services, said this month that about 1.1 million people have signed up. Medicare’s open enrollment runs through December 7th, and the Bridge requires Medicare drug coverage.

CMS Administrator Dr. Mehmet Oz explained the logic on Face the Nation in August. At $50 a month, he said, “all the downstream costs that would have been incurred no longer happen.” When the Congressional Budget Office studied permanent Medicare coverage in 2024, it projected $35 billion in added federal spending over nine years, at the prices it expected then. The study of veterans, whose average age was about 65, found no savings.

So Medicare is also paying for health that doesn’t pay for itself, just at a lower price than employers get. Lilly sells Zepbound directly to employers for $449 a month. Jenny Goins, leads the National Alliance of Healthcare Purchaser Coalitions, said that price “doesn’t appear to be much lower” than what many employers were already paying after rebates. Medicare’s price came from a White House deal with the drugmakers, while employers rely on pharmacy benefit managers whose rebate contracts are confidential.

What happens to people who lose GLP-1 coverage?

A spokesperson for the Massachusetts Health Policy Commission told me it believes that some of the 112,000 fewer people with commercial insurance coverage for GLP-1s now pay cash, at roughly $4,200 to $5,400 a year. Dr. Laura Davisson, who runs the weight management program at West Virginia University, estimates that more than half her patients initially stopped when their coverage ended. Some restarted as cash prices fell, while lower-income patients often struggled most to stay on.

And people who keep paying cash may not get credit for it. Cigna told its employees that cash purchases won’t count toward their deductible, according to a document Reuters reviewed. I asked Cigna whether that’s still true. Its statement, which cited “managing affordability,” didn’t say.

Cigna also owns Express Scripts, a pharmacy benefit manager. In a settlement with the Federal Trade Commission, Express Scripts agreed to count TrumpRx purchases toward deductibles once, in the FTC’s words, “necessary legal and regulatory groundwork is laid.”

What should you ask during open enrollment?

Does my 2027 plan cover GLP-1s for weight loss, or only for diabetes? Look for Wegovy and Zepbound by name on the plan’s drug list, called a formulary. In SHRM’s 2026 benefits survey, 47% of employers covered GLP-1s for type 2 diabetes, and 15% covered them for weight management.

If my plan covers them, what does it take to qualify? Ask about prior authorization, weight cutoffs, and any required coaching program. Goins told me 68% of employers in her group’s 2026 survey limited access to certain patients or were considering it.

If I pay cash, will it count toward my deductible? Get the answer in writing, because plans differ. Stacie Dusetzina, a health policy professor at Vanderbilt, puts cash prices for the injections at $350 to $450 a month.

Which pharmacy benefit manager runs my plan? That company manages the drug list and negotiates confidential rebates with drugmakers. Express Scripts has agreed to credit TrumpRx purchases once the legal groundwork is laid.

What happens if my coverage ends? Your doctor can help you plan before your last covered fill. Dr. Davisson’s clinic tried behavioral programs, other weight-loss pills, and less frequent doses to slow the regain.

I’m on Medicare. How long will the $50 price last? The Bridge requires Medicare drug coverage, and Medicare’s open enrollment runs through December 7th. The Bridge ends December 31, 2027, and CMS hasn’t said what comes after it.

• A new study about choline supplementation in pregnancy

• What did the trial find?

• Why are so many nutrition studies misleading?

• Why does the study dropout pattern matter for IQ?

• Did choline prevent autism?

• What do we know about choline?

• Which prenatal supplements are worth taking?

• Is food better than a pill?

A new study about choline supplementation in pregnancy

A new study in the American Journal of Obstetrics and Gynecology suggests that pregnant women take 6,300 milligrams (mg) of phosphatidylcholine a day, starting as early in pregnancy as possible. The authors call all pregnant women potential candidates. Phosphatidylcholine is the form of the nutrient choline found in eggs, meat, and fish.

The trial itself gave women 9,000 mg a day, starting at 16 weeks. The authors’ recommendation of 6,300 mg is based on their own research as well as earlier research.

What did the trial find?

Researchers at the University of Colorado randomly assigned 206 pregnant women to take phosphatidylcholine capsules or placebo from 16 weeks of pregnancy until birth. They followed the children to age 4.

The team named two main goals in advance. The first was fewer attention problems. The second was less social withdrawal, including a child pulling away from others, avoiding eye contact, or showing little interest in people. Parents rated both on a standard checklist.

The two groups had similar rates of attention problems. On withdrawal, 17 of 71 children in the placebo group scored two or more, which can mean one problem that happens all the time. In the choline group, 3 of 71 did. At age four, children in the choline group scored about 8 points higher on an IQ test, 106 versus 98.

The result most likely to make headlines came from an analysis the team hadn’t planned. In the placebo group, 10 of 71 children had checklist scores in the range seen in autism. In the choline group, 1 of 71 did.

Why are so many nutrition studies misleading?

Ice cream sales and drownings rise and fall together. Ice cream doesn’t cause drowning. Summer drives both, since hot weather sends people to the beach and to the freezer aisle. Scientists call a hidden third factor like summer a confounder.

Nutrition studies run into confounding all the time. Women who eat more eggs in pregnancy may also earn more, sleep more, and see a doctor more often. If their children do better, nobody can tell which of those things made the difference.

A randomized trial is the best fix. When a coin flip decides who gets the real pill, the two groups start out alike in income, education, and health, including traits nobody thought to measure. This trial randomized its participants.

Why does the study dropout pattern matter for IQ?

A coin flip only works when people stay in the study. About a third of families dropped out before the behavior checks, and half dropped out before the IQ test.

Dropping out was associated with education. Among mothers with a graduate degree, 81% stayed through year four. Among mothers who hadn’t finished high school, 43% did. A mother’s education is one of the strongest predictors of her child’s IQ score, and the main IQ comparison didn’t adjust for mothers’ education. If more well-educated mothers stayed in one group, that group’s scores would rise for reasons that have nothing to do with choline.

The groups also differed in timing. In the choline group, 37 of 104 women were pregnant during the COVID pandemic, compared with 21 of 107 in the placebo group. A split that uneven rarely happens by chance. The placebo group likely had more children who were toddlers during 2020 social distancing, the same ages at which parents rated their social behavior. The paper doesn’t report the children’s birth dates by group. It also doesn’t report each group’s mix of boys and girls, though 10 of the 11 children with autism-range scores were boys.

During the pandemic, the researchers gave 20 women, 10 in each group, the real supplement on top of their study pills. When the team left those women out, the IQ difference was no longer clear.

Researchers told women that choline could help their baby. Parents later rated their own children’s behavior. If some parents guessed which pill they’d taken, their hopes could also skew the ratings.

Did choline prevent autism?

The autism finding is based on an unplanned analysis of the parent screening checklist. That checklist was developed by testing a large reference group of young children. The trial flagged a child as autistic if they scored higher than 90% for that group and also had language problems common in autism. The checklist cutoff alone flags about 1 in 10 children by design.

The Centers for Disease Control and Prevention (CDC) estimates that about 1 in 31 American 8-year-olds, or about 3%, have an autism diagnosis. When a checklist flags 10% of children and only about 3% are autistic, most of the children the checklist flags aren’t autistic.

Only 11 children had autism-range scores. One more flagged child in the choline group would cut the 10-fold difference in half. The difference in autism scores could be explained by different rates of loss-to-follow-up among more highly educated versus less highly educated women in the study.

Many of the checklist questions behind the autism score are the same ones that measured social withdrawal, such as avoiding eye contact and showing little interest in people. So the autism result largely repeats the withdrawal data rather than adding new evidence.

Meanwhile, a team at the University of North Carolina gave pregnant women 750 mg of phosphatidylcholine on top of dietary sources, and their babies showed no gains in memory, language, or development at 10 and 12 months.

In a 2012 trial, a team at the University of North Carolina gave pregnant women 750 mg of choline a day, as phosphatidylcholine, on top of their diets. Their babies showed no gains in memory, language, or development at 10 and 12 months.

What do we know about choline?

Choline helps build cell walls and brain cells. The National Institutes of Health (NIH) sets the target at 450 mg a day in pregnancy and 550 mg while breastfeeding. A 2025 review of 12 studies estimated that only about 11% of pregnant women reach that target.

Prenatal vitamins rarely close the gap. When researchers tested 47 prenatal vitamins in 2025, only 12 listed choline on the label. Only 5 of those 12 prenatal vitamins contained an amount of choline within 20% of what the label said. Choline is bulky, which makes a full dose hard to fit into one pill.

The best trial evidence is small. In a Cornell feeding study, women consumed 930 mg or 480 mg of choline a day in their last trimester. When researchers tested twenty of their children at age 7, the higher-dose group did better on an attention task. Studies that track women’s diets link higher choline levels to fewer pregnancy complications, though those studies face the same confounding problem as the ice cream scenario.

In a small trial of 69 heavy drinkers in Cape Town, South Africa, babies whose mothers took 2,000 mg of choline a day did better on learning and memory tests and grew more in their first year than babies whose mothers took a placebo, though no amount of alcohol in pregnancy is known to be safe.

In 2025, an international panel of 35 experts agreed that supplements of folic acid, iodine, iron, vitamin D, and the omega-3 fat DHA matter in low-risk pregnancies. The panel couldn’t agree on choline.

For now, the safest goal is the standard target of 450 mg of choline a day. One large egg has about 147 mg. Women who eat few eggs, little meat, poultry, or fish, or who’ve had weight-loss surgery may need a supplement to reach that target. More research is needed to determine whether higher doses are warranted.

Which prenatal supplements are worth taking?

Folic acid

The U.S. Preventive Services Task Force gives it an A grade. Anyone who could become pregnant should take 400 to 800 micrograms (mcg) a day. The neural tube, which becomes the spine, closes in the first month, often before a woman knows she’s pregnant. Women who’ve had a pregnancy affected by spina bifida, or who take certain seizure medicines, are typically prescribed higher doses of folic acid.

Iodine

The fetal brain needs thyroid hormone, which needs iodine. The American Thyroid Association recommends 150 mcg a day during pregnancy and breastfeeding. Check the label, since about half the prenatal vitamins in a 2025 testing study didn’t list iodine.

Omega-3 fats

A 2018 Cochrane review of 70 trials found that omega-3 supplements lowered the share of babies born prematurely, before 34 weeks, from about 4.6% to 2.7%. A 2019 Australian trial of about 5,500 women found no drop in those early births. Evidence that omega-3 pills produce a smarter baby is weak.

Iron

Most prenatal vitamins contain it, and doctors check for anemia during pregnancy. The U.S. Preventive Services Task Force says the evidence is too thin to weigh the benefits and harms of routine extra iron for women without symptoms. The American College of Obstetricians and Gynecologists (ACOG) recommends low-dose iron starting in the first trimester.

Vitamin D

The target is 600 international units a day, and most prenatal vitamins have 400 IU. The American College of Obstetricians and Gynecologists (ACOG) recommends testing only women at higher risk, such as those with darker skin or little sun exposure.

Calcium

The World Health Organization recommends 1.5 to 2 grams a day to lower the risk of preeclampsia in countries where diets run low in calcium. Women in the U.S. who eat little dairy should ask about it.

Vitamin B12

Vegans need a supplement. Babies breastfed by mothers with untreated B12 deficiency can suffer lasting brain damage.

What to avoid

More isn’t better for vitamin A. Studies have linked birth defects to supplements with more than 10,000 units a day of the preformed kind, called retinol.

Is food better than a pill?

Whole food also contains protein, fiber, and other nutrients a pill doesn’t. Eggs, salmon, and dairy also supply choline, omega-3 fats, and calcium. Still, most pregnant women in the U.S. are at risk of falling short on key nutrients from food alone, a 2023 study of 2,450 pregnant women found. Folic acid and iodine are harder to get from food than from a pill. The body absorbs at least 85% of the folic acid in a pill or fortified food, but only about half of the folate in food. Most salt in the U.S. diet comes from processed food, and manufacturers almost always use salt without iodine. The task force recommends a daily folic acid pill for anyone who might become pregnant, regardless of her diet. It also says that women don’t eat enough fortified food to get the full benefit, and only 20% to 40% of women who were recently pregnant or trying to be say they took it around conception. A supplement can fill other specific gaps: B12 for vegans and for women who take metformin or have had weight-loss surgery; extra iron for women with iron-deficiency anemia; vitamin D for women with low levels; choline for women who eat few eggs and little meat, poultry, or fish; omega-3s for women who don’t eat fish; and calcium for women who don’t eat dairy.

The team behind that 2023 study checked the labels of 20,547 U.S. supplements, including 421 prenatal vitamins, for the doses they calculated pregnant women need of vitamins A and D, folic acid, calcium, iron, and omega-3 fats. Only one product had all six, and it wasn’t a prenatal vitamin. It cost about $200 a month and meant taking seven pills a day. Under the Affordable Care Act, most private plans must cover folic acid at no cost for women who could become pregnant. In a 2021 KFF survey, every state Medicaid program that answered covered prenatal vitamins. A 2020 study in Health Affairs linked Medicaid expansion to more low-income women taking folic acid before pregnancy. In 2024, the U.S. Department of Agriculture added canned fish to WIC, its food program for low-income pregnant women and young children.

Supplement makers decide what goes in a prenatal vitamin, and the Food and Drug Administration (FDA) doesn’t approve supplements before they go on sale. In the 2025 testing study, 20 of the 25 products that listed iodine held more than 20% less than their labels claimed. Congress could add requirements for prenatal vitamins to the Dietary Supplement Health and Education Act, the 1994 law that governs supplements.

• Is it cancer?

• Small lung cancers have good odds.

• Why don’t more Americans get screened for lung cancer?

• What about people who never smoked?

Is it cancer?

Pope Leo XIV said on October 11th that doctors found a nodule in his lung “at a very early stage.” Surgeons in Rome will remove it this week.

Doctors find a lung nodule in about 1.6 million Americans a year, and more than 95% aren’t cancer. Most small nodules get a repeat scan or nothing. Surgeons operate when cancer looks likely. The Vatican hasn’t said whether he had a biopsy, so the pathology report after surgery will show whether it’s cancer.

Small lung cancers have good odds.

Two large trials studied early lung cancers 2 centimeters or smaller. Removing part of a lobe worked as well as removing all of it. Between 79% and 94% of patients were alive 5 years later.

Why don’t more Americans get screened for lung cancer?

Doctors diagnose only 24% of U.S. lung cancers while still confined to the lung. They diagnose about half after it has spread to distant organs, when 5-year survival is about 10%.

A yearly low-dose CT scan finds them sooner. In the National Lung Screening Trial, CT prevented 1 lung cancer death per 320 heavy smokers screened, compared with chest X-rays. The U.S. Preventive Services Task Force recommends the scan for adults 50 to 80 with at least 20 pack-years who smoke or quit within 15 years. Because the task force gave it a B grade, most private plans must cover it with no copay.

The scan is free, and only 18.2% of eligible Americans got one in 2022.

The health secretary appoints and can remove the panel that set that grade. The Supreme Court upheld that power in June 2025. Robert F. Kennedy Jr. named 8 new members, half the panel, in September.

What about people who never smoked?

We don’t know the Pope’s smoking history. At Northwestern University, a quarter of 997 lung cancer patients had never smoked. Only 35% met the screening criteria for a chest CT. Never-smokers have no screening recommendation, so insurers won’t cover a scan. The Environmental Protection Agency calls radon the leading cause of lung cancer in never-smokers. A home test kit measures it.

I’m part of an NYU Langone study that scans Asian women who’ve never smoked. On the first scan, 1.3% of the 1,000 women had lung cancer. The National Lung Screening Trial found 1.1% in heavy smokers. The NYU study doesn’t measure whether finding these cancers early saves lives.

• The strain that was supposed to be harmless.

• What happened in that Chicago lab?

• What the tidy version leaves out.

• Sixty-five contacts, thirty coworkers, no further cases.

• Pneumonia of undetermined etiology

The strain that was supposed to be harmless.

Professor Malcolm Casadaban worked with a strain of plague the federal government had decided was safe. The strain, KIM D27, is missing the stretch of DNA that carries the genes Yersinia pestis needs to pull iron out of a host. No iron means no growth, so researchers treated it as unable to infect a person. On that reasoning, the CDC left it off the select agent list, the registry that says who may handle dangerous germs. The exclusion kept his work outside those rules. In September 2009, it killed him.

Professor Scott Minnich, a microbiologist at the University of Idaho, wrote to me about Casadaban on Friday, in the middle of a week of questions about Irkutsk. “I say this only because lab accidents do rarely occur,” he wrote. “Malcolm’s death hit the news but nothing like this present incident.”

What happened in that Chicago lab?

At the University of Chicago, Casadaban was an associate professor working with Dr. Olaf Schneewind on a plague vaccine. He did his PhD in Professor Jonathan Beckwith’s lab at Harvard, where he worked out how to fuse a gene to a marker so you could watch it switch on. Thousands of people have used the trick since without knowing whose it was. “I used it routinely,” Minnich wrote, “and it remains a powerful technique and a stellar contribution to molecular biology.”

He got sick around September 7, and saw a clinician three days later. He reached a Chicago emergency department on September 13th, by which time he’d been sick for six days. About 12 hours after he arrived, doctors intubated him, and he died an hour later. He was 60.

“Olaf Schneewind and I were having a beer at Cold Spring Harbor Laboratories after a long day of meetings,” Minnich wrote, “when he received the phone call delivering the news of Malcolm’s death and the unexpected plague diagnosis.”

What the tidy version leaves out.

Casadaban had hemochromatosis, which nobody knew, so his blood carried too much free iron and the crippled bacterium fed on it.

The CDC’s report calls iron overload “one possible explanation.” The hemochromatosis diagnosis came after he died, from tissue, labs and a genetic test. He also had insulin-dependent diabetes, which that version drops. Nobody ever established how the bacteria got into him. The pathology pointed away from his lungs and toward his skin or a mucous membrane, and the report notes that gloves weren’t worn consistently in his lab.

“Though the exact route of exposure remains uncertain,” Minnich wrote, “the incident sent deep concern throughout the Yersinia research community.”

In 2012, Casadaban’s colleagues tested whether the iron explained it. Lauriane Quenee, with Dr. Schneewind as senior author, gave strains like KIM D27 to mice bred to model hereditary hemochromatosis, and those mice developed lethal plague while normal mice did not. They recommended a different product: people with hereditary hemochromatosis “may be protected with subunit vaccines but should not be exposed to live-attenuated plague vaccines.”

Sixty-five contacts, thirty coworkers, no further cases.

The response never rested on the classification being right. Investigators found 1 household contact and 64 other close contacts, and 61 of them took antibiotics. All 30 of his coworkers got a prescription too. Nobody else got sick, although Science reports that the case still pushed the US to tighten its rules for weakened strains.

Pneumonia of undetermined etiology

Darya Shipilova was hospitalized on September 29th and died on October 2nd, at 28. She authored two research publications: an abstract on which antibiotics are effective against Yersinia pestis, and a 17-author monograph her institute released in 2024 on improving live weakened vaccines for plague, tularemia, and anthrax. Russia told WHO she had been vaccinated against the high-threat pathogens she worked with.

Pneumonic plague needs close, sustained contact. Across eight documented outbreaks between 1907 and 1997, one case infected an average of 1.3 others before public health measures began. Isolating the sick and watching their contacts stopped every one of those outbreaks but Mukden in 1946. For 7 days Russia watched her contacts, with daily testing, and reported no symptoms in any of them.

The plague vaccine Russia uses is a live weakened strain. Nobody has said whether Russia tests the people it vaccinates for hemochromatosis, or whether anyone tested Shipilova.

Nobody outside Russia can check any of it. WHO asked for verification on October 3rd, 5th, 6th and 7th under the International Health Regulations, and again about a rumored second case. Russia answered that it had run more than 5,500 laboratory tests, that the panel covered influenza, SARS-CoV-2, plague, and streptococcus, and that among the contacts it found two cases of COVID, three of rhinovirus, one of Haemophilus, and other opportunistic infections. It still calls her cause of death “pneumonia of undetermined etiology.”

On October 10th, WHO said that “a laboratory-associated exposure therefore cannot be fully excluded at this stage,” and that the agency “is not in a position to independently assess whether the duration of the reported monitoring period for contacts was sufficient to exclude all plausible causes of the fatal illness.” It asked Russia to continue testing her clinical specimens, with reference laboratory support if needed, and offered support on high-threat pathogens and laboratory biosafety.

• What was already gone when WHO asked.

• How do you fact-check a rumor?

• What happened to the first reports?

• Why does a second case matter so much?

• Sverdlovsk took 13 years.

• What cuts against this?

What was already gone when WHO asked.

Darya Shipilova, a 28-year-old scientist at the Irkutsk Antiplague Research Institute, who had worked in its plague microbiology unit, died on October 2nd in a hospital in Shelekhov. Five days later, Dr. Tedros Adhanom Ghebreyesus, Director-General of the World Health Organization, said at a media briefing: “We have also sought verification of media reports of a second employee with pneumonia of undetermined cause.”

Several Russian outlets had already deleted their reporting on plague in Irkutsk.

How do you fact-check a rumor?

You trace it back to its source. Russian officials have been asking everyone to do exactly that. Rospotrebnadzor, the agency that runs Russia's anti-plague institutes and its disease surveillance, has called the second-illness reports fake on October 8th, the day after Dr. Tedros asked. Dmitry Peskov, Vladimir Putin’s press secretary, said that “the vast majority of what the media is reporting regarding this situation is false information.” He told reporters to rely on Rospotrebnadzor instead.

What happened to the first reports?

REN TV is a federal Russian channel. It ran five posts about plague in the Irkutsk region, then deleted all five. TGStat, an outside service that tracks Russian Telegram channels, had already indexed two of them. One said Shipilova had worked with plague strains. The other cited an unnamed source and said a suspected plague patient had died at a district hospital.

Lyudi Baikala, an Irkutsk newsroom, documented those deletions. Russia blocked its website in 2022, and its reporters do not use their names.

On October 2nd, Alexey Tsydenov, the head of the Republic of Buryatia, said a woman in the Irkutsk region had died of plague. He later edited the line to add “possibly.”

The Insider reported that on October 3rd, the Baikalsk city government warned residents to limit visits to Shelekhov. It cited “unofficial information about a confirmed case of plague.” The post stayed up about two hours before it came down.

On October 4th, Axios reported that the Trump administration was monitoring the situation, and Kommersant, a national daily, repeated it. Kommersant’s Telegram post and its website item came down 25 minutes later.

Why does a second case matter so much?

A second case is how you tell one exposure from a larger incident or outbreak.

Scott Minnich, a Distinguished Professor of Microbiology at the University of Idaho, assessed the captured trailers in Iraq for the Iraq Survey Group. I asked him what he would want if he were sent to Irkutsk. He’d start with the patient: whether she dropped a vial, whether she cut herself cleaning up a spill, whether she reported the exposure, whether anyone put her on antibiotics in time.

“The second priority is contact tracing and containment verification,” he wrote. “Have other personnel been exposed, and are any currently symptomatic? Are we looking at a common-source outbreak? Face-to-face interviews with laboratory members are essential.”

The tabloids got there first, but Minnich is asking the same question, as am I.

Rospotrebnadzor says nearly 5,000 lab tests found no dangerous germs. It hasn't said who those samples came from or what the tests looked for.

Sverdlovsk took 13 years.

“Ultimately, you are heavily reliant on human intelligence and testimonies. Obtaining truthful, uncoerced testimonies within Russia may be difficult given the potential institutional and political pressures personnel face,” Minnich said.

An anthrax release near Sverdlovsk killed dozens of people in 1979. The Soviet government blamed tainted meat. Nobody could inspect the site, so the meat story stuck. Thirteen years passed before Boris Yeltsin, then Russia’s president, acknowledged that the anthrax had come from a military facility and that the cause had been covered up.

The CIA reached the opposite conclusion about whether Iraq based on similarly shaky evidence. It leaned on secondhand material from an Iraqi defector code-named Curveball. Minnich wrote: “U.S. intelligence accepted this German intelligence at face value without ever vetting the information through their own direct interrogations.” He added, “Once an official determination has been made by authorities and endorsed by a select group of scientists, human nature makes it incredibly difficult to walk that conclusion back.”

The United States sent a classified démarche to Moscow on October 7th. It reminded Russia of its obligations under the International Health Regulations. WHO has offered technical support on lab biosafety, and says it still cannot finish a risk assessment.

What cuts against this?

Dr. Nahid Bhadelia is the founding director of Boston University's Center for Emerging Infectious Diseases (CEID). “I can’t say that we have any one scenario that has more evidence than elsewhere,” she said. “Putting my geopolitical hat on, to me, it seems it’s a greater cost to Russia to say it’s a ‘pneumonia of unknown etiology’ than it is to say it was a careless lab worker.” She doesn’t anticipate that Russia will provide more information about secondary cases.

Dr. Bhadelia is right that this doesn’t add up. She deferred to people who know Russian politics better than she does. I’ll say what she wouldn’t. If Russia is concealing something it regards as worse than a careless lab worker — research it does not want made public, for instance — then admitting the lab accident was the easier way out, and Russia has not taken it. I have no evidence for what that something is.

A note on the Russian sources. Lyudi Baikala is an Irkutsk outlet. Russia blocked its website in 2022 and detained its founders, who have since left the country, and its reporters inside Russia still work anonymously.

• What pathogens did Darya Shipilova work on?

• What’s a plague focus?

• What circulates in Gorno-Altai and Tuva?

• Who collects the strains?

• The 2024 monograph, and what is in it.

• What vaccination does and doesn’t establish.

• How laboratory exposures happen.

• What counts as an especially dangerous infection?

• The questions Rospotrebnadzor hasn’t answered.

What pathogens did Darya Shipilova work on?

Darya Shipilova died on October 2nd, at 28, in a hospital in Shelekhov, outside Irkutsk. Rospotrebnadzor, the agency that runs Russia's anti-plague institutes and its disease surveillance, says she died of pneumonia of unknown cause, that extended testing found no microorganisms connected to her professional activity, and that she was vaccinated and immune to the dangerous infections she worked with.

At least some of her work is public.

She’s a co-author, third of seven, on a conference abstract published in the proceedings of the XVIII Pokrovsky Congress, Russia’s annual national infectious disease meeting, which took place in Moscow this spring. The abstract was not peer-reviewed. The authors tested different antibiotics on Yersinia pestis bacteria to see what would kill them. The strains were sensitive to all of the antibiotics they tried: streptomycin, amikacin, gentamicin, doxycycline, kanamycin, chloramphenicol, cefotaxime, ciprofloxacin, and azithromycin.

The Irkutsk Institute’s director, Sergey Balakhonov, is one of 20 authors on the 2026 national plague forecast. It reports that in 2025 plague turned up in animals in three of Russia’s eleven natural foci, and her abstract names two of the three. The Altai animal outbreak was in Kosh-Agach district, the Tuva one in Mongun-Taiga. Russian teams isolated thirteen strains of plague bacteria that year, and twelve belong to a lineage called 4.ANT. Balakhonov also co-authored a 2025 paper that describes the 4.ANT strains in the Tuva focus as highly virulent and capable of causing human disease.

What’s a plague focus?

A natural plague focus is a place, not a pathogen. Yersinia pestis keeps circulating there among wild rodents and their fleas on its own, with no human cases required, sometimes for centuries. Russia has eleven. A focus is where Yersinia pestis persists and could spill into people. It says nothing about antibiotic resistance, vaccine protection, or how sick the organism would make you.

What circulates in Gorno-Altai and Tuva.

The Tuva mountain focus was found in 1964 and hasn’t gone quiet since. Between 1964 and 2009, Russian teams isolated 1,497 strains there. 95% of them came from one animal, the long-tailed ground squirrel, and its fleas.

From 1961 to 2011, the only organism recorded in the Gorno-Altai focus was Y. pestis subspecies altaica, which kills laboratory mice and doesn't kill guinea pigs. Most of the cultures came from Mongolian pikas and their fleas, and in fifty years not one person there got the plague. Russian scientists rated its epidemic potential as low. But when the main subspecies spread through grey marmot colonies, people started getting infected. One case in each of 2014, 2015, and 2016, all in Kosh-Agach district, all in people who had been butchering poached marmots. All three were bubonic. All three survived.

Russia has recorded no human plague since 2017, none since those three cases. Both subspecies keep circulating, and the 2026 forecast expects outbreaks among the animals in Gorno-Altai from both.

Who collects the strains?

Russia inherited the Soviet anti-plague system, a standing network of regional stations and research institutes whose job is to watch the foci: zoologists, entomologists, field epidemiologists, and laboratory staff, organized to catch a change in the animals before it reaches people. One station is assigned to each focus.

Gorno-Altai belongs to the Altai Antiplague Station in Gorno-Altaisk. Tuva belongs to the Tuva Antiplague Station in Kyzyl. Both sit roughly 1,000 kilometers from Irkutsk, in other republics. On the field papers they appear as co-author institutions alongside the Irkutsk Institute.

In a 2019 survey of the Saylyugem focus along the Mongolian border, a joint Russian and Mongolian team worked out of a laboratory mounted on a truck. They walked 2,668 square kilometers, tested 282 mammals and 261 ectoparasites, and pulled 47 strains from grey marmots and their fleas. They found plague in 63% of the fresh carcasses and raptor prey remains they examined.

The person who traps the marmot, the laboratory that identifies the isolate, and the scientists who later sequence its genome are usually three distinct teams. The Irkutsk Institute types and stores the strains that the stations in Gorno-Altaisk and Kyzyl collect.

Nothing on the record puts her in the field. University of Idaho professor Scott Minnich, who has peer-reviewed papers from Irkutsk, told me the Irkutsk Institute publishes environmental studies of natural plague reservoirs, and tracking of how local strains vary.

Shipilova finished a residency in clinical laboratory diagnostics in 2024. She worked in the plague microbiology unit at least through January 2025, Lyudi Baikala reported in an account Meduza republished. Clinical laboratory diagnostics is a bench specialty.

She could have been exposed at the Irkutsk Institute, on a field expedition, or in the community. WHO sent its verification request under the International Health Regulations following media reports of an accidental lab exposure. Russia’s written response reports a community-acquired infection.

She was also out of Irkutsk five weeks before she fell ill. According to the Irkutsk Institute’s news page, Shipilova took part in international rapid-response exercises at Russia’s national anti-plague Institute, “Microbe,” in Saratov with two other young specialists from the Irkutsk Institute. It doesn’t say what the exercises involved.

The 2024 monograph, and what is in it.

Her only other publication is a 2024 monograph from the Irkutsk Institute, edited by Balakhonov and written with sixteen colleagues. It’s about making the vaccines for plague, tularemia, and anthrax work better. These are all live weakened strains, including Yersinia pestis EV, the one Russian lab workers get. The researchers immunized mice, waited 21 days, then infected them with a virulent strain. At a reduced dose, the plague vaccine alone protected 60% of those mice.

What vaccination does and doesn’t establish.

Rospotrebnadzor says she was vaccinated and immune to the dangerous infections she worked with. A former classmate who trained at the institute says it required the plague vaccine even of students on placement.

Her monograph says the weakened germ in vaccines like this one can still cause disease. The shot causes side effects. It can also leave the person who gets it primed to overreact to the next dose.

Javier Pizarro-Cerdà, who heads the Yersinia unit at the Institut Pasteur in Paris and directs the WHO’s collaborating centre for Yersinia, told Science that former Soviet countries give this vaccine roughly every year, because protection fades fast. He also said there’s less evidence it works against pneumonic plague than against bubonic. Western countries stopped using plague vaccines decades ago over side effects and questions about whether they worked.

Maria Van Kerkhove, WHO’s epidemic and pandemic preparedness lead, said Shipilova “would have had training and vaccinations against some of these pathogens,” and that WHO is seeking more information on that.

How laboratory exposures happen.

In March, as the congress opened, the Microbe Institute accepted for publication a review of biosafety in work with infected laboratory animals. The review reports that working with infected animals is the biggest risk in handling dangerous pathogens. The usual way in is a break in the skin: a bite, a scratch, a needle stick, or a scalpel cut during an autopsy. Shipilova’s monograph describes experiments on infected mice, although it doesn’t say who handled them.

Nobody has established how she was exposed. Reports of a broken test tube trace back to a single Telegram post by one Russian journalist, which Science reports has not been confirmed. Shipilova’s mother, who works in medicine herself, told Lyudi Baikala that the tubes can’t be broken or spilled, even hypothetically, and that the lab is that well protected. A classmate told Sibir.Realii said she couldn’t imagine what would have to happen for infection to be possible.

What counts as an especially dangerous infection?

V.A. Vishnyakov is the seventh author on Shipilova’s abstract and head of the isolation unit at the Irkutsk Institute. He was also the sole author of another conference abstract arguing that Russia has no objective scientific definition of “especially dangerous infection.” The existing criteria, he wrote, can be arbitrarily interpreted and have no objective assessment. He proposed a ten-point scale to replace them.

He wrote that in March, and in October, Rospotrebnadzor classified a death at his institute as pneumonia of unknown cause.

The questions Rospotrebnadzor hasn’t answered.

We still don’t know what killed Shipilova.

We know she worked on Yersinia pestis from two foci where the organism is circulating in the animals.

Rospotrebnadzor says nearly 5,000 laboratory tests found no dangerous pathogens. It hasn’t said whose samples those were, or what the tests were looking for.

Which experiments were hers. The monograph has seventeen authors and attributes no section to any specific person.

Whether she ever handled live cultures of anything but plague. The monograph establishes what the laboratory was working on, but not what she touched.

What the Saratov exercises involved. Which agents the scenario involved, whether she did laboratory work or field sampling, and whether live cultures were in the building at all.

Which department she worked in after January 2025. Lyudi Baikala puts her in plague microbiology at least until then.

Which vaccines she received, and when. Russia’s vaccination guidelines recommend the plague, tularemia and brucellosis vaccines for those working with live cultures of those particular pathogens. This would tell us what she was working on.

What did the nearly 5,000 tests look for, and in whom? What did the extended studies on her own samples test for? Was there an exposure, and if so, when did she report it, and what was she given, and when?

A note on the Russian sources. Lyudi Baikala is an Irkutsk outlet. Russia blocked its website in 2022 and detained its founders, who have since left the country, and its reporters inside Russia still work anonymously. Meduza publishes from Latvia. Sibir.Realii is a project of Radio Free Europe/Radio Liberty. Moscow has designated both Meduza and RFE/RL "undesirable organizations," a label that makes it a crime in Russia to cooperate with them or pass them information. In April 2024, a Russian court sentenced a man to penal labor for giving Radio Liberty a street interview. That's why the Institute staff quoted aren't named.

• Flu season started early this year.

• Last season’s mismatched shot still cut hospitalizations.

• Has the virus changed again?

• The first mRNA flu shot in the US.

• How mFlusiva compares with a standard flu shot.

• More side effects, a higher price, and no high-dose comparison.

• Who should consider the mRNA flu shot?

• Will your insurance pay for the mRNA flu shot?

Flu season started early this year.

Flu season started early this year. In the week ending September 26th, flu accounted for 0.44% of emergency department visits nationwide. That’s nearly triple the 0.16% at the same point last year, and the share has risen every week since August.

The CDC still calls activity “low nationally, but … increasing,” and every state was still at its lowest activity level in late September. The early rise is concentrated in a few places. Hawaii leads at 3.25% of ED visits, followed by Alaska and Washington. San Francisco reported its first flu death of the season on October 1st.

Almost all of it is H1N1. Of the 55 samples that public health labs typed in late September, 96.4% were that subtype. Taiwan, where flu season overlaps ours, reported 219 severe cases and 44 deaths in the week ending October 5th, more than in any week of its past season, and H1N1 accounted for about 91% of its flu A cases.

An early start doesn’t predict a bad winter. It does mean many people will meet this year’s virus soon, so how well the shot matches it matters now.

Last season’s mismatched shot still cut hospitalizations.

Not as well as it usually does. Last season was driven by an H3N2 strain called subclade K, which made up 93% of the H3N2 viruses the CDC characterized. Subclade K emerged after scientists had already picked the strains for the 2025-26 vaccine, so the shot was built for a different H3N2.

The shot still helped. The CDC’s interim estimate, from about 142,000 patients across three networks, put vaccine effectiveness at roughly 30% to 40% against outpatient visits and hospitalization. Protection against flu B, which matched better, ran about 63% to 66% in adults.

In a study published in September, researchers at the VA in St. Louis followed veterans who did and didn’t get a flu shot at a primary care visit. Vaccinated veterans had 22% fewer flu-related ED visits and hospitalizations combined, and 32% fewer flu hospitalizations. In absolute terms, that’s about five fewer ED visits or hospital stays for every 10,000 people vaccinated.

When the virus drifts away from the vaccine, the shot still lowers your risk, but by less.

Has the virus changed again?

H1N1 hasn’t stood still, though. Jesse Bloom’s lab at the Fred Hutchinson Cancer Center tested 355 blood samples against 148 recent flu strains this summer. It found newer H1N1 strains carrying a mutation called G155E that antibodies block less well. The drop was biggest in people younger than about 25. The study is a preprint and hasn’t been peer-reviewed.

Bloom included blood from Australians who got the 2026 Southern Hemisphere vaccine, which carries the same H1N1 strain as the US shot. Vaccination still raised their antibodies against the G155E strains, though for some people the rise was smaller. In an email, he called the vaccine “partially mismatched” but “close enough to somewhat boost neutralizing antibody titers” against those strains. He described the G155E effect as “a quantitative decrease in titers rather than complete escape” from the vaccine.

Antibody levels in a lab aren’t the same as protection in people. The first real estimate of this season’s effectiveness usually comes in February. Until then, last season’s data are the best guide, and they show a partially matched shot still cut flu hospitalizations by roughly a third.

The first mRNA flu shot in the US.

It’s the first mRNA flu vaccine licensed in the US. The FDA approved Moderna’s shot in August for adults 50 and older, with full approval for ages 50 to 64 and accelerated approval for 65 and older. Accelerated approval means Moderna still has to confirm the benefit in that older group. Moderna told me the shot is available at more than 13,000 locations.

Most flu shots are made from inactivated virus grown in eggs. mFlusiva instead delivers genetic instructions that tell your cells to make the flu’s surface protein, so your immune system learns to recognize it. Because it isn’t grown in eggs, it can’t pick up the mutations that sometimes creep into egg-grown H3N2 strains. Cell-based and recombinant flu shots share that advantage.

How mFlusiva compares with a standard flu shot.

In one large trial, yes. Moderna’s study enrolled about 40,800 adults 50 and older and compared mFlusiva with Fluarix, a standard-dose shot. Over the season, 2.0% of people who got mFlusiva caught flu, compared with 2.8% who got Fluarix. That’s a 26.6% relative reduction. Among people 65 and older, the numbers were 1.9% and 2.6%.

For every 1,000 people who get mFlusiva instead of a standard shot, about 8 fewer catch flu in a season.

More side effects, a higher price, and no high-dose comparison.

It hurts more. In a subset of about 6,000 trial participants, about 66% had a sore arm after mFlusiva, compared with 30% after the standard shot. About 45% reported fatigue, compared with 20%. Severe reactions, meaning symptoms bad enough to stop normal activities, hit 6.4% of people after mFlusiva and 1.0% after the standard shot. Most of those were whole-body symptoms such as fatigue and aches.

Serious side effects look rare. In the big trial, serious adverse events that investigators judged vaccine-related occurred in 3 of about 20,350 people on mFlusiva and 2 of about 20,350 on the standard shot. One large trial can’t rule out rare problems, which is what post-marketing surveillance is for.

Nobody has tested it against the high-dose shots older adults already get. For people 65 and older, the CDC prefers high-dose, adjuvanted, or recombinant shots, because they work better than standard-dose ones. Moderna compared mFlusiva only with a standard-dose shot. A Moderna-funded modeling study estimates mFlusiva might beat high-dose by about 12%. The FDA’s approval letter makes that comparison a condition of the 65-and-older approval. It requires a randomized trial of mFlusiva against high-dose in that age group, scheduled to start August 31, 2026, finish in July 2029, and report by May 2030. A separate study comparing antibody responses with high-dose is due by the end of 2028. Until then, older adults are choosing without a head-to-head answer.

It costs more. Moderna’s list price is $149.56 a dose. Medicare sets its own payment rates, which are $171.21 a dose for mFlusiva, $122.52 for high-dose, adjuvanted or recombinant shots, and as little as $23.89 for standard ones.

Who should consider the mRNA flu shot?

Adults 65 and older.

mFlusiva is one of four enhanced options. The medical societies’ joint guidance lists high-dose, adjuvanted, recombinant “or mRNA” for this group and ranks none above the others. The American Academy of Family Physicians took the lead on that recommendation. Its chief medical officer, Dr. Margot Savoy, told me the AAFP didn’t compare the enhanced shots with one another because there wasn’t enough data. She said the evidence review “alone is insufficient to elevate the vaccine to preferred for 65 years and older,” so the academy “also considered access issues” in listing it. In practice, a pharmacy with mFlusiva and no high-dose shot can still give an older adult something better than a standard dose.

Adults 50 to 64.

AAFP calls mFlusiva an “acceptable option” that “may provide enhanced protection compared to a standard flu vaccine” for this age group. This is the group where the trial’s comparison actually applies, because most people this age get a standard shot. It’s also the group whose insurance coverage is least certain.

Flu shots during pregnancy.

No mRNA flu vaccine is licensed or recommended in pregnancy, according to the American College of Obstetricians and Gynecologists. A standard or recombinant flu shot is recommended in any trimester.

Adults under 50.

mFlusiva isn’t approved for you.

Whatever your age, the CDC advises getting vaccinated by the end of October, and that applies no matter which shot you choose.

Will your insurance pay for the mRNA flu shot?

Ordinarily, the CDC’s Advisory Committee on Immunization Practices would weigh exactly these tradeoffs in public, and its vote would decide who pays. Under the Affordable Care Act, private plans must cover, with no copay, any shot the committee recommends and the CDC director adopts. Medicare is the exception, because Congress wrote flu vaccine into the statute.

That committee can’t vote. On June 9, 2025, Health Secretary Robert F. Kennedy Jr. removed all 17 members and named his own. On March 16, 2026, Judge Brian Murphy of the federal district court in Massachusetts stayed 13 of those appointments in AAP v. Kennedy, along with the panel’s votes back to June 2025, because he found the panel was likely seated in violation of the Federal Advisory Committee Act. HHS appealed on April 29. The committee hasn’t met since December 2025, and the CDC’s flu page doesn’t mention mRNA.

So each insurer decides on its own. Over the past two weeks, I asked six of the biggest whether they’ll cover mFlusiva. Elevance said it covers the shot for adults 50 and older. CVS Health said Aetna covers it, at no cost for Medicare members, and that its CVS Caremark program generally covers it at $0 unless a client opts out. UnitedHealthcare said it is “currently evaluating coverage.” Centene, Cigna, and Humana didn’t respond by my deadline. Moderna told me it is “seeing broad coverage for mFLUSIVA across commercial plans,” but that for people 50 to 64, “coverage varies by individual health plan.”

The insurance trade group AHIP has promised its members will cover “all ACIP-recommended immunizations with no cost-sharing” through the end of 2027. But that doesn’t apply to self-funded employer plans, which cover 67% of workers with employer-based insurance.

If you’re on Medicare, you’re covered by law. If you’re 50 to 64 with private insurance or Medicaid and want mFlusiva, call your plan before you make an appointment. Standard flu shots are still covered either way.

Restoring a lawful ACIP is “entirely within the government’s control,” Dr. Andrew Racine, president of the American Academy of Pediatrics, which brought the suit, told me. Meanwhile, Judge Murphy’s ruling left the committee “unable to carry out its core responsibilities,” Secretary Kennedy says.

• What killed Daria Shipilova?

• What do the International Health Regulations require?

• Who checked the Irkutsk Institute?

• Why hospitalize 189 contacts?

• What three plague and bioweapons experts had to say.

• How an official finding got overturned from inside.

• Nobody will say what the United States can still see.

What killed Daria Shipilova?

We still don’t know what caused the pneumonia that killed Daria Shipilova, the young lab assistant who died in Siberia earlier this month. She worked at the Irkutsk Anti-Plague Institute, which studies Yersinia pestis plague and other deadly infectious diseases. Rospotrebnadzor, the agency that runs Russia's anti-plague institutes and its disease surveillance, calls her death pneumonia of unknown etiology.. Russia separately told WHO that no case of Yersinia pestis plague has recently been registered in Irkutsk oblast.

What do the International Health Regulations require?

WHO saw the first news reports of Shipilova’s death on October 2nd and asked Russia for more information the next day, under the International Health Regulations. Article 10 gives a country 24 hours to acknowledge such a request. It gives 24 more hours to hand over the public health facts it has.

Russia replied on October 6th, three days after WHO asked. Shipilova’s illness at first looked like an acute respiratory viral infection, the catch-all that covers everything from a common cold to a severe case of COVID or influenza, Russia told WHO, according to WHO spokesman Christian Lindmeier. WHO also asked which germ caused the pneumonia. Lindmeier says Russia did not answer.

Who checked the Irkutsk Institute?

The Irkutsk Institute has its own Commission on Biosafety Compliance. It looked into Shipilova’s death and found no mishap in handling pathogenic microorganisms. Dr. Anna Popova, the head of Rospotrebnadzor, sent a panel of experts, which backed the Commission’s findings.

Extensive testing of Shipilova’s samples “did not detect any microorganisms associated with her professional duties.” That wording leaves two possibilities open: a pathogen the Institute studies but Shipilova was not assigned to, or a pathogen the Institute does not study at all. It also says nothing about what the tests did find. Russia has not told WHO what testing was done, or what the tests found.

Rospotrebnadzor employed Shipilova, runs the Irkutsk Institute, ran the tests, and sent the panel that cleared Rospotrebnadzor.

Professor Sonia Ben Ouagrham-Gormley studies the old Soviet anti-plague network at George Mason University. “We can’t really trust that they will be truthful,” she said of the agency. “The test will be whether they involve the WHO in the investigation.” WHO has offered technical support and is still waiting for a response.

Why hospitalize 189 contacts?

Rospotrebnadzor ruled out any laboratory accident, but also said that “given the nature of the employee’s work, a full range of anti-epidemic measures was immediately implemented.” Those measures put 189 of about 197 contacts in the hospital for daily testing. Russia has since told WHO that medical observation of all identified contacts is complete and that it found no high-threat pathogens among them.

WHO said on October 6th that it has asked Russia to clarify three things: the cause of the severe pneumonia, the pathogen that prompted the public health measures, and media reports of a second employee with pneumonia of undetermined cause. Rospotrebnadzor said the same day that it had recorded no infectious disease among the institute’s staff in 10 days.

“Timely, complete and transparent information sharing under the International Health Regulations is essential to clarify conflicting reports,” posted WHO Director-General Tedros Adhanom Ghebreyesus.

What three plague and bioweapons experts had to say.

Professor Henry Heine spent his career at the United States Army Medical Research Institute of Infectious Diseases (USAMRIID) testing drugs against Yersinia pestis plague. Asked what could kill a Yersinia pestis plague patient on treatment, Professor Heine said, “Of course resistance is a possibility. More likely is that therapy was initiated too late.”

The CDC tells clinicians to start antibiotic treatment based on suspicion and not wait for test results. Doctors initially took her illness for severe COVID. They sent her to the hospital’s COVID center, the Irkutsk outlet IRK.ru reported. There they put her on a ventilator. “Regardless of what the infection was caused by, that indicates to me that it was too late,” said Professor Heine.

“An agent alone, even if virulent, cannot serve as a weapon,” Professor Ben Ouagrham-Gormley wrote in Barriers to Bioweapons. She does not rule out a bioweapons-related incident, although she thinks it unlikely. The old Soviet program worked on drug resistance, and the know-how is still in Russia.

How an official finding got overturned from inside.

Professor Scott Minnich studies microbes at the University of Idaho. From October 2003 to May 2004, he served in Iraq as a subject matter expert for the Defense Intelligence Agency. He was attached to the Iraq Survey Group, the team hunting Iraq’s weapons. The group sent him to inspect two truck trailers that American troops had captured after the invasion. The administration offered them as proof of the mobile germ labs Secretary of State Colin Powell described to the UN Security Council in February 2003. A CIA and DIA assessment that May called one a mobile production plant “despite the lack of confirmatory samples.”

Each trailer held one reaction vessel of about 600 liters, with no sampling ports, no way to pump in oxygen, and no way to harvest what was inside. A fermenter is the vessel you need to grow bacteria in quantity. These were not fermenters, he said.

A British team had said the same for months without being heard, Minnich said. They went back out together and found lye and aluminum filings in one vessel. That pairing makes hydrogen. The Iraq Survey Group’s final report called the trailers “almost certainly designed and built for the generation of hydrogen,” filling weather balloons for artillery wind readings.

Professor Minnich was contradicting the CIA and DIA assessment while working for one of the two agencies that wrote it.

In Irkutsk, the Commission sits inside the Institute. The panel came from Rospotrebnadzor, which oversees the Institute. Neither has published its findings, and nobody outside Russia has seen the lab, the samples or the records.

Nobody will say what the United States can still see.

The United States left WHO on January 22, 2026. The CDC says the country now “coordinates with WHO solely in a limited fashion to effectuate our withdrawal”.

The International Health Regulations are a separate treaty from WHO membership. The State Department and HHS rejected the 2024 amendments to the Regulations in July 2025. Those amendments, they wrote, “were set to become binding on the United States regardless of our withdrawal from the WHO.”

The Regulations give every country a National IHR Focal Point, an office that must be reachable at any hour to send and receive outbreak notifications. HHS still has that role, and we do not know how the U.S. withdrawal from WHO has affected U.S. participation in the IHR.

Those outbreak notifications arrive on the Event Information Site, a secure page focal points can read before the public gets anything. It is unclear whether the United States still has access to it.

I asked the State Department for comment again today:

And their response:

I also followed up with HHS:

And their response:

HHS answered none of my questions and referred me to the State Department. The State Department partially answered two of my six questions. Neither agency would say whether Russia has answered, and the State Department would not say whether Russia's disclosures satisfy the treaty.

• No cause of death has been released.

• What does “plague” actually mean here?

• Is bioweapons research done in Russia or the United States?

• Is Russia’s response out of proportion to the risk?

• The questions, and the non-answers.

• What is Yersinia pestis plague, and how dangerous is it?

• Why was the Black Death so deadly?

• How worried should Americans be?

No cause of death has been released.

Daria Shipilova, a 28-year-old laboratory assistant at Russia’s Irkutsk Anti-Plague Institute, fell ill in late September, was hospitalized with severe pneumonia, and, according to a WHO spokesperson, died on October 2nd. Russian officials have not released her cause of death. HHS Press Secretary Emily Hilliard called this "reports of a suspected case of pneumonic plague," but this hasn't been confirmed. Nobody has reported which tests were run, on which samples, at which lab, or on what date. No additional cases have been reported. A national reference laboratory holding almost two thousand strains of Yersinia pestis should be able to sequence an isolate in days, but Russia has released nothing.

It's been reported that on September 25th, Shipilova allegedly broke a test tube containing Yersinia pestis, the bacterium that causes pneumonic plague. Plague takes a higher exposure to infect than tularemia does. We have no further details about Shipilova's specific duties or which organisms she handled. Russia’s health agency says tests on her samples found nothing tied to her work. We don’t know if she was vaccinated against plague. WHO recommends Yersinia pestis vaccination for high-risk groups including laboratory personnel. Nobody has said what antibiotics Shipilova was given, or which ones failed. That answer separates a late presentation from the wrong first drug from a resistant strain.

What does “plague” actually mean here?

“Plague” can mean two different things, and the institute’s name carries the looser meaning.

Frank Snowden, an emeritus professor of history at Yale, writes in Epidemics and Society that the word plague has become shorthand for catastrophe itself. That is the old, wider meaning, the one in “a plague of locusts.” The narrow meaning is infection with Yersinia pestis bacteria, which can cause bubonic plague, pneumonic plague, or septicemic plague, and was the cause of the Black Death across Europe in the Middle Ages.

Maj Gen (ret) Paul Friedrichs was a senior director for biodefense on the National Security Council. He describes the Irkutsk Anti-Plague Institute as “a well-known public health lab which has been around since the 1930s doing research into endemic” disease found in Siberia, Central Asia, and the Far East. That work includes tularemia, brucellosis, anthrax, cholera, tick-borne encephalitis, and hantavirus, as well as Yersinia pestis plague. Much of this work involves surveillance of rodent populations and the development of diagnostics and vaccines for infectious diseases circulating among rodents. The Irkutsk Anti-Plague Institute has, for example, developed a live tularemia vaccine.

Across Siberia and Central Asia, Yersinia pestis circulates among wild rodents. The last case Russia reported before this one was a 10-year-old boy with bubonic plague in the Altai region in 2016.

Because tularemia infects at a much lower dose, lab-acquired tularemia is historically far more common than lab-acquired Yersinia pestis. A USAMRIID team reviewing the worker-illness records of the United States’ offensive biological warfare program from 1943 to 1969 found that pathogens requiring higher doses to infect, Yersinia pestis among them, were largely prevented by protective equipment and training alone. But protective equipment was not enough for pathogens like tularemia, which infect at lower doses. Tularemia infections continued at an average of 15 a year until a vaccine became available.

In Fort Collins, Colorado, the CDC's Bacterial Diseases Branch studies many of the same infectious diseases as the Irkutsk Anti-Plague Institute, including Yersinia pestis, tularemia, bartonella, Lyme disease, and tick-borne infections.

During the Los Angeles plague epidemic of 1924, Yersinia pestis spilled over from urban rats into rural rodents, becoming endemic among rodents in the American West. The CDC reports an average of 7 human cases of Yersinia pestis plague a year, most of them in northern New Mexico and Arizona, with more in California, southern Oregon, and western Nevada.

Is bioweapons research done in Russia or the United States?

Japan, the Soviet Union, and the United States have all studied weaponizing Yersinia pestis. Japan’s program ended in 1945, and America’s in 1969, but the Soviet anti-plague network likely supplied virulent strains of Yersinia pestis to the Soviet bioweapons program. The State Department’s 2025 arms control compliance report says Russia and North Korea “maintain their offensive biological weapons programs in direct violation of their treaty obligations.” We do not know if the Irkutsk Anti-Plague Institute has such an active program.

The United States’ ongoing research on Yersinia pestis is defensive: surveillance, diagnostics, treatment, and vaccines. Yersinia pestis is considered a Tier 1 select agent, meaning any laboratory holding it must be registered and inspected under the Federal Select Agent Program. Russia has published no equivalent registration, inspection, or incident record for Irkutsk.

Is Russia’s response out of proportion to the risk?

Dr. Anna Popova, Chief State Sanitary Doctor of the Russian Federation since 2013, flew to Irkutsk on October 2nd, the day Shipilova died, to convene an emergency commission. Six medical facilities are under quarantine.

Health workers are testing about 200 contacts daily. So far, they’ve picked up two cases of COVID and two cases of rhinovirus, which causes the common cold. 189 of those contacts have been hospitalized for observation and preventive antibiotics.

“Quarantining people who were exposed is appropriate,” Dr. Friedrichs said, “but timeline of exposure and death are unclear, what antibiotics were unsuccessful is unclear, status of close contacts is unclear.”

“Pneumonic plague can spread person-to-person through respiratory droplets during close contact, which is why rapid identification and treatment of infected individuals, and monitoring of potentially exposed individuals, are important public health measures,” HHS Press Secretary Hilliard emailed me.

However, the Russian response is out of proportion to a case of Yersinia pestis plague pneumonia acquired in the lab. It’s an infection that’s easily contained with personal protective equipment, contact tracing, isolation, testing, and preventive antibiotics. This could indicate an abundance of caution in a world still traumatized by the COVID pandemic, an infection that cannot be controlled with preventive antibiotics, or an infection other than Yersinia pestis plague.

Shipilova’s contacts will require seven to ten days of quarantine, during which time they should complete the seven-day course of preventive antibiotics. Igor Kobzev, governor of the Irkutsk region, said on October 5th that close contacts could go home next week.

The questions, and the non-answers.

Dr. Friedrichs called this a “missed opportunity for Russia to be truthful and transparent.”

WHO is verifying this event under Articles 9 and 10 of the International Health Regulations. A spokesperson says WHO has asked Russian authorities for more details including “the laboratory tests underway, and their results,” and the health status of the contacts. WHO has also had a formal tie to “Microbe,” Russia’s anti-plague reference institute, since November 2021, when it became a WHO Collaborating Centre.

The United States hasn’t revealed much either.

Here are the questions that I submitted to the State Department:

And their response:

From HHS:

Between them, they answered only one of my questions.

Neither said whether the U.S. has asked Russia for the laboratory results directly or only through WHO, whether Russia has replied, whether the U.S. offered confirmatory testing, or whether it knows which pathogen Shipilova handled. Neither said whether CDC’s own plague laboratory in Fort Collins has seen a strain, a sequence, or any laboratory data. Neither addressed whether this is being handled as a public health event or a Biological Weapons Convention matter, or whether the National Security Council has convened on it. I asked both which office is the lead. State said it is working with “the CDC and our other interagency partners.” HHS said it is “working across the U.S. government.”

What is Yersinia pestis plague, and how dangerous is it?

It’s a bacterial infection, and it causes three different diseases depending on where it lands.

Bubonic plague is the most common. A flea bites you, the bacteria travel to your lymph nodes, and those nodes swell into what doctors call “buboes.” You get sudden fever, chills, body aches, and vomiting. Septicemic plague is when the bacteria infect the blood, and pneumonic plague is when the bacteria infect the lungs; these are the most deadly forms of plague. Pneumonic plague causes shortness of breath and a cough, often with blood in the sputum. Symptoms usually start within 1 to 7 days, and sometimes in under a day.

It spreads less easily than its reputation suggests. Bubonic plague doesn’t pass between people at all, because you get it from a flea. Pneumonic plague can pass person to person in respiratory droplets. Dr. Jacob Kool, then a CDC infectious disease epidemiologist in Fort Collins, reported in 2005 that patients usually infect others only in late illness, when they cough copious bloody sputum, and only in close contact. Because spread needs that level of closeness, masks and distance could stop these outbreaks even before antibiotics existed.

WHO puts the case fatality ratio for untreated bubonic plague at 30% to 60%, and calls the pneumonic and septicemic forms “always fatal” when left untreated. Pneumonic plague can kill within 18 to 24 hours of onset.

Researchers at CDC’s Bacterial Diseases Branch in Fort Collins reviewed 762 treated cases of Yersinia pestis plague reported between 1937 and 2019. Overall mortality was 20%, 27% for pneumonic plague, but 13% in patients who got an aminoglycoside. Common antibiotics work, especially when started promptly. CDC's treatment guidelines recommend 10 to 14 days for all three form, but the drug depends on the form: doxycycline for an uncomplicated bubonic case, and an aminoglycoside (like streptomycin or gentamicin), a fluoroquinolone (like levofloxacin), or a tetracycline (like doxycycline) for pneumonic and septicemic plague.

Doctors diagnose plague from blood, sputum, or fluid drawn from a bubo. WHO has validated a rapid dipstick test now used in Africa and South America. Close contacts get a 7-day course of preventive antibiotics, usually doxycycline, ciprofloxacin, or levofloxacin.

Most human Yersinia pestis plague cases reported to WHO between 2019 and 2025 came from the Democratic Republic of the Congo and Madagascar.

Why was the Black Death so deadly?

Because nobody had modern medicine, specifically antibiotics, in the Middle Ages.

The Black Death ran from about 1347 to 1351 and killed tens of millions of Europeans. Researchers reconstructed the pathogen’s DNA in 2011 from skeletons at the East Smithfield mass burial ground in London; the deaths were caused by “a variant of Y. pestis that may no longer exist.” And that pandemic did not kill randomly. The elderly and people with underlying conditions were at highest risk.

How worried should Americans be?

Based on the information we have, not very.

A WHO spokesperson says that “based on unofficial information available, the public health risk to the general population appears to be low.”

Update, October 7th. Rospotrebnadzor said on October 6th that its Commission on Biosafety Compliance found no incident in the handling of pathogenic microorganisms at the Irkutsk institute, and that testing of Shipilova's samples "did not detect any microorganisms associated with her professional duties." The agency's denial covers the reported broken test tube described above. Russia also told WHO that medical observation of all identified contacts is now complete and that it found no high-threat pathogens among them. Russia has still not said what caused her pneumonia. WHO said on October 6th that it has asked Russia to name the cause, to name the pathogen that prompted the public health measures, and to address media reports of a second employee with pneumonia of undetermined cause.

• What did the retatrutide trials find?

• Why obesity drugs now work like a subscription.

• Cost and side effects top the reasons people stop.

• How are people getting GLP-1 drugs?

• What additional research is needed?

What did the retatrutide trials find?

Eli Lilly’s experimental obesity shot, retatrutide, did well in two late-stage trials. In one, published Tuesday in the New England Journal of Medicine, people without diabetes on the top dose lost about a quarter of their body weight over 80 weeks. In the second, in The Lancet, people with type 2 diabetes lost about 19%.

Lilly plans to ask the FDA for approval early next year. Meanwhile, vials marked “research use only” sell online for up to $600.

Why obesity drugs now work like a subscription.

For years, Lilly and Novo Nordisk told insurers, employers and Congress that obesity is a chronic disease, like diabetes. Chronic diseases need lifelong treatment, the companies argued, and insurance should pay for it.

People taken off Zepbound in a 2024 trial regained 14% of their body weight within a year. Dr. Fatima Cody Stanford, an obesity medicine physician-scientist at Massachusetts General Hospital and Harvard Medical School, tells her patients they are “potentially long-term treatment.” (Lilly and Novo Nordisk have both paid her consulting fees. So have Boehringer Ingelheim and Amgen, which are testing rival obesity drugs.)

Insurers and states have been less willing to pay. Four state Medicaid programs have dropped obesity-drug coverage since last fall. Several Massachusetts insurers dropped it too, and GLP-1 use in the state fell sharply. CMS has delayed the Part D portion of its planned GLP-1 coverage model.

So the drugmakers started selling directly to patients for cash. Lilly lists Zepbound at $299 to $449 a month. Novo now calls Wegovy buyers “customers” and tells them to “Live lighter.”

In any other industry, a lifelong product with a monthly cash price is a subscription. Obesity has become a subscription disease.

Cost and side effects top the reasons people stop.

About 4 in 10 commercially insured people stop these drugs within a year. Older studies put it higher, at 50% to 65%. “That is consistent with the real-world challenges we see,” Dr. Stanford told me. Her patients stop because of “cost or loss of insurance coverage, prior-authorization barriers,” side effects and “injection fatigue.”

In one study of 125,474 patients, clinical notes most often cited side effects and cost as the reasons people stopped. Among patients with diabetes, those with estimated incomes over $80,000 quit at a lower rate than those under $30,000.

Across the GLP-1 drug class, more weight loss has come with more side effects. In the TRIUMPH-1 trial, about 4 in 10 people on the top dose had nausea. About 1 in 8 felt burning or tingling skin. “A roughly 1-in-8 rate at the highest dose is not trivial,” Dr. Stanford said. About 1 in 11 had low blood pressure, including people taking no blood pressure pills. “We should not assume it is only a medication-interaction issue,” she said.

How are people getting GLP-1 drugs?

While direct sales may reach people who aren’t actually overweight or obese, they miss medically eligible patients who can’t afford $449 a month. “Ability to pay is increasingly influencing who can start and, especially, who can stay on treatment,” Dr. Stanford said. Direct sales “may help some people,” she added, but “they can create a two-tiered system in which treatment continuity depends on disposable income.”

Retatrutide has no approved route to patients except clinical trials and a narrow Lilly early-access program. Before that, Lilly gave the drug to one 79-year-old man through compassionate use. STAT News asked the White House several times whether he was President Trump, also 79 at the time. Afterward, White House spokesperson Kush Desai said on X the request wasn’t for the president.

Online sellers offer vials on a gray market. So do some doctors and nurse practitioners. In June, my CBS News colleagues found more than 120 websites selling or promoting retatrutide, including more than 50 clinics with licensed staff. Two Los Angeles clinics listed it at $330 and $450 a month, about what Lilly charges for Zepbound. Since 2024, the FDA has sent 14 warning letters to sellers. At least six kept offering it. Poison centers logged about 95 retatrutide exposures a month in early 2026, up 265% from late 2025.

What additional research is needed?

Dr. Stanford won’t offer retatrutide outside a trial until the FDA approves it. She wants “larger and longer studies,” starting with “cardiovascular safety, including heart rate and rhythm effects.”

For the burning skin, she wants to know “where the symptoms occur, how long they last, whether they recur with dose escalation.” She asks what rapid weight loss does to “lean mass, nutrition, hydration, and gallbladder disease,” especially in older adults. She also wants data on what happens after people stop, and on patients with several chronic conditions.

A late-stage trial is now testing retatrutide against tirzepatide, the drug in Zepbound. Until it reports, “it would be premature to conclude that retatrutide is clinically superior,” Dr. Stanford said. “My concern is that enthusiasm can outrun the evidence.”

Lilly plans to file for FDA approval early next year.

• RFK Jr. pitched AI against the experts on launch day.

• What is America.gov?

• What does it say about vaccines and autism?

• How does it decide which page wins?

• The chatbot never mentioned the court order.

• What else did it get wrong?

• Accurate answers on raw milk and smoking.

• Should people use America.gov?

RFK Jr. pitched AI against the experts on launch day.

On Tuesday at the Make America Healthy Again Summit, HHS Secretary Robert F. Kennedy Jr. said that artificial intelligence can give patients “a second opinion that is much better informed than any doctor in the country.” He said it has the power to “free us from medical tyranny.” He also said public officials who tell Americans to defer to experts will “never ever again” be able to dominate them. “If somebody tells you that a vaccine will prevent transmission and infection, or you need to take it to protect your grandmother, A.I. may say it actually doesn’t do that.”

The same day, the White House launched America.gov. It promises “simple answers only from official government sources.” Its health answers cite agencies Kennedy runs: the Centers for Disease Control and Prevention (CDC), the Food and Drug Administration, the National Institutes of Health, and the Centers for Medicare & Medicaid Services. So I asked America.gov for a second opinion. I sent it 37 questions. The answers were often complicated, because the official sources no longer agree with each other.

What is America.gov?

An executive order signed September 29th makes the site “a single digital point of entry” for federal services. The White House’s National Design Studio built it, the General Services Administration (GSA) runs it, and it’s powered by Gemini and xAI’s Grok, drawing on about 29,000 government websites. But when I asked which AI it runs on, it said, “I do not share details about the underlying software.”

What does it say about vaccines and autism?

I started with autism and vaccines. In November 2025, Kennedy ordered a change to a CDC page. The page now says, “The claim ‘vaccines do not cause autism’ is not an evidence-based claim.” Yet other CDC pages still say studies find no link.

The chatbot didn’t pick a side. Since both pages are official, it opened with “CDC pages currently do not all say the same thing.” It quoted both versions and pointed to health officials in New York and Colorado, who call the new wording inconsistent with the research. Then it said, “I cannot verify a single federal conclusion that covers every vaccine.”

How does it decide which page wins?

I asked. It said it prefers “the agency that runs the program for a final determination.” After that, it prefers “the most recent official source.” The rule sounds sensible, because agencies usually update a page when the science or the policy changes. It fails when a court has paused the newest page, as happened with the hepatitis B shot.

Last December, CDC dropped its hepatitis B shot recommendation for every newborn. But in March, a federal judge paused that change and restored the recommendation for every newborn. The government has appealed. I asked about the newborn hepatitis B shot in three different ways. Because the December page is the newest, the chatbot went with that, but that’s wrong. It is still recommended that a newborn get a hepatitis B shot.

The chatbot never mentioned the court order.

In June 2025, Kennedy fired all 17 members of the CDC's vaccine advisory committee, calling the panel "little more than a rubber stamp for any vaccine." He named his own replacements. That December, his appointees voted 8 to 3 to end the recommendation of a hepatitis B shot at birth for every newborn.

On March 16th, Judge Brian E. Murphy of the federal court in Massachusetts paused that vote. He found that Kennedy’s new committee likely broke a federal law requiring balanced advisory panels. So his order also paused appointments for the new members and a January overhaul of the childhood schedule.

The American Academy of Pediatrics and other groups had sued. His order also paused the new committee members’ appointments and a January overhaul of the childhood schedule. The earlier schedule, including the birth dose, went back into effect.

I asked America.gov about the newborn shot in three different ways. Each time, the chatbot told me to prefer the December 2025 CDC and HHS fact sheets, never mentioning the court order.

The chatbot said CDC had announced a new “three-category” schedule in January. The court has also paused the January schedule.

What else did it get wrong?

On insulin, it said, “most Part D plans cap insulin at $35 for a month’s supply (participating plans; check your formulary).” “Participating plans” describes a voluntary Medicare pilot that started in 2021. About 38% of Part D plans joined it, according to KFF. In the 2022 Inflation Reduction Act, Congress extended the $35 cap to every Part D plan. The chatbot described the pilot and left out the law that replaced it.

Accurate answers on raw milk and smoking.

When I typed chest pain and a numb left arm, its first words were “Call 9-1-1 now.” For a toddler who swallowed blood pressure pills, it put Poison Control’s number first. It walked me through Medicare Part D enrollment windows and Medicaid appeals. Its Spanish answers matched its English ones. It said raw milk isn’t healthier than pasteurized milk and that smoking causes lung cancer.

Should people use America.gov?

Use it to find forms and deadlines. Take your health questions to your healthcare provider.

Republished on CBSNews.com.

• Teen suicide risk hit a 10-year low in 2025.

• Guns, sexual violence, and self-harm.

• Where are the 2025 data on transgender teens?

• Missing data changes public health policy.

Teen suicide risk hit a 10-year low in 2025.

CDC asked more than 10,000 high school students over 100 questions in spring 2025. Yesterday, it published a report covering a fraction of them. The report focsed on five areas: diet, physical activity, sleep, digital behaviors, and mental health, reflecting the priorities of the Make America Healthy Again movement.

There was good news. Fewer teens seriously considered suicide in 2025 than in any survey year since 2015: 14%, down from 20% in 2023. Sadness fell.

Sports participation, which dipped to 49% in 2021, is back to its 2015 level of 58%. Daily PE fell from 27% to 20% in two years, while fewer students eat breakfast daily than a decade ago.

Guns, sexual violence, and self-harm.

The questionnaire is public, and it goes well beyond those five areas. Students answered questions about sexual violence, self-harm, and whether they got help when they felt sad or anxious. They said how quickly they could get a loaded gun. They reported whether food ran out at home and whether they felt close to people at school. There were questions about sexual identity, including whether they were transgender.

The report did not publish any of the answers students gave to those questions.

CDC posts the raw survey data online, so anyone can see what the report left out.

The report doesn’t cover the health of sexual minority students. Among students who identified as gay, lesbian, bisexual, questioning, or another non-straight identity, persistent sadness fell from 66% to 57% between 2023 and 2025. Serious thoughts of suicide fell from 41% to 30%. Those drops were bigger than the ones among straight students, although the gap remains wide: 28% of straight students reported persistent sadness, about half the rate of LGBQ+ students.

About 1 in 7 girls reported sexual violence in the past year. One in 6 students hurt themselves on purpose. Among students who needed help for sadness or anxiety, half rarely or never got it. And among students who had seriously considered suicide, 43% said they could get a loaded gun without an adult’s permission.

One protective factor rose sharply. In 2023, 55% of students said they felt close to people at their school; in 2025, 63% did. CDC’s earlier research ties that kind of connection to better mental health.

Where are the 2025 data on transgender teens?

In 2023, CDC's national survey asked high school students, for the first time, whether they were transgender. Their answers are the only nationally representative data on transgender teens, and their mental health is by far the worst of any demographic. That year, 71% of transgender students reported persistent sadness, 54% seriously considered suicide, and 26% attempted it, compared with 38%, 18%, and 8% of cisgender students. But we don’t have 2025 data.

The timing matters because students took the survey in spring 2025, as federal policy toward transgender youth shifted. On January 28th, an executive order directed agencies to end federal support for gender-affirming care for people under 19, and several hospitals paused that care within days. In February, the CDC said it would no longer process transgender identity data. Later in 2025, the Supreme Court upheld Tennessee’s ban on gender-affirming care for minors, and HHS ended the 988 Lifeline’s dedicated line for LGBTQ+ youth.

Researchers disagree about how much state policy shapes trans teens’ mental health. A 2024 Trevor Project study estimated that anti-trans laws increased suicide attempts among trans and nonbinary youth. Critics dispute its methods and its estimates. The 2025 survey could have added national evidence to that fight. CDC collected the data but hasn’t publicly released it.

Missing data changes public health policy.

Surveillance works as a feedback loop. It can show officials whether their policies help or hurt. The CDC didn’t provide data on the students most exposed to transgender policy shifts.

The report kept measures that frame health as personal choice, like fruit, breakfast, and sports. It left out measures that track policy, like gun access, mental healthcare, food at home, and school climate, all of which respond to what legislators and agencies decide.

HHS did not respond to questions that I sent on September 21st.