When the people urging you to eat three times the recommended amount of something also profit from it, it’s worth looking carefully at the evidence before following along. Chicken and beef, bars and powders, tofu and crickets. So I researched what I could on this airplane, and here is my take. References include one of my own posts about the sketchily inverted food pyramid, a post by Dr. Eric Topol, and a plethora more upon request from OpenEvidence (otherwise formatting hyperlinks on an iPad is too painful!).

What “high protein” even means

The longstanding recommendation of 0.8 grams per kilogram of body weight per day was not meant to be optimal. It’s the floor, the amount that prevents deficiency in a healthy adult, derived decades ago from nitrogen-balance studies. Few Americans fall below it. By USDA figures, more than half of men and a third of women already exceed it, and roughly a quarter of us eat double.

The influencer target is something else. Last I checked many including Attia recommend 2.2 g/kg/day, close to three times the usual number. What are the possible benefits, and more importantly, are there risks to consuming this much protein? Beef at the top left of the food pyramid?! What are the best sources of protein? Is there a convergent conspiracy between sincere ideologies and notable conflicts of interest?

Maybe there is room for compromise and healthy debate, even as everything in America careens towards conflict and extremism? The controversial 2025 Dietary Guidelines raised the recommended target to 1.2–1.6 g/kg/day. That is a meaningful increase, and it rests on some decent justifications. So maybe the disagreement that actually matters, then, is not between 0.8 and 2.2. It is between a reasonably well-supported move toward 1.2–1.6 and a fashionable leap to twice that.

Let’s dive in…

Where more protein genuinely helps

There are situations where eating more protein is clearly wise, and we shouldn’t lose them in the skepticism.

The first is weight loss. In a calorie deficit, higher protein helps preserve muscle while we lose fat, and it helps us feel full. The effect on the scale itself is modest (meta-analyses put it somewhere between 1-5 pounds of advantage) but the benefit to body composition is the real point. We want to lose weight, but not become weaker. This matters more than ever now that so many patients are taking GLP-1 medications, where muscle loss is a real concern. For them, adequate protein paired with resistance training can be crucial rather than optional.

The second is aging. After about sixty-five, muscle becomes somewhat less responsive to protein (a phenomenon called anabolic resistance). The slow erosion of muscle can lead to sarcopenia (poor body muscle), one of the under-recognized threats of later life. Here the evidence supports a modestly higher intake, around 1.0 to 1.5 g/kg/day. But protein on its own does very little for aging muscle. The benefit appears only when it is paired with resistance training. We can’t just gobble down bars, powders, chicken breasts. The protein is just the raw material but the movement is the signal to build with it. Without some exercise, much of the extra is simply spun into generic calories or even excreted.

Here’s a summary of the ranges recommended by different sources and in different situations:

A few points of nuance and clarification are in order. The traditional RDA of 0.8 g/kg/day at the top represents the minimum to prevent deficiency, not optimal intake, and has been criticized as insufficient for muscle preservation, especially in older adults. The updated 2025–2030 Dietary Guidelines now recommend 1.2–1.6 g/kg/day based on evidence for muscle, satiety, and metabolic benefits. But these were also heavily influenced by MAHA and RFK adjacent ideologies, so read on for more considerations.

For individuals with obesity, protein targets should be based on adjusted or ideal body weight rather than total body weight, since adipose tissue does not require the same protein support as lean mass.

Higher intakes up to 2.0 g/kg/day may benefit athletes, older adults with sarcopenia risk, and post-surgical patients, though evidence for benefit beyond 1.6 g/kg/day in the general population remains limited.

I don’t know about you, but I suck at conceptualizing protein based on the metric system. So here are some translations into typical American pounds of body weight, and grams of protein per day. I’m including 300 pounds to demonstrate the folly in just following grams of protein per unit of body weight… otherwise people at these higher weights would be chasing over 200g a day!

And now, here is a sampling of grams of protein from some common foods.

In this table you’ll see a “protein quality score.” This estimates how well a food provides all nine essential amino acids in digestible form, scored using a system where 1.0 is the highest rating. Animal proteins (meat, dairy, eggs) consistently score ~1.0, while most plant proteins score lower (0.4–0.7) due to limiting amino acids, particularly lysine in grains and methionine in legumes — meaning larger portions or complementary combinations are needed to achieve equivalent metabolic benefit. Vegans especially have to know this nutritional science to get all their essential amino acids.

Where the evidence runs out, and where we might have concerns

Is there too much of a good thing? Usually the answer to this generic question is yes.

Several lines of research now implicate leucine (the amino acid most abundant in animal protein, and the one some athletes supplement deliberately) in activating a cellular pathway called mTOR. When chronically overstimulated, mTOR appears to accelerate atherosclerosis in animal models, with similar signals in humans. Large observational studies link habitually high intake, particularly of animal protein, with more cardiovascular disease and higher mortality, and there is a cancer signal in middle age tied to elevated IGF-1 that appears to reverse after sixty-five.

To get a general sense of where the studies suggest potential problems with eating too much animal protein, but not necessarily plant protein, check out this table below:

We should consider these findings loosely, because most of the work is observational or done in mice. But it usually comes back to all things in moderation, not eat as much as you like because too much protein can’t hurt you. At the extremes, like pretty much everything, it plausibly can.

Instead we see that the clearer and more useful lesson is about source. Across most outcomes like mortality, heart disease, kidney health, the kind of protein matters more than the amount. Plant protein is consistently associated with lower all-cause and cardiovascular mortality, while the concerning signals attributed to “animal protein” come mainly from red and processed meat, not fish, poultry, or dairy. Replacing even a small amount of processed-meat protein with plant protein tracks with meaningfully lower mortality.

I had to look up TMAO (trimethylamine N-oxide). It’s a metabolite produced when gut bacteria convert nutrients found mainly in red meat into trimethylamine, which the liver then oxidizes to TMAO, a compound linked to increased cardiovascular risk through promotion of atherosclerosis, platelet reactivity, and inflammation.

We also need to consider the massive environmental and energy costs of producing red meat as our planet is really screwed, and therefore so are every one of us. Cows are sweeties, too.

Politics and ideology

Briefly.

The 2025–2030 Dietary Guidelines for Americans, heavily shaped by RFK Jr. and the MAHA initiative, markedly elevate protein targets and explicitly promote animal‑source protein (red meat, full‑fat dairy, eggs) as “real food.” They downplay concerns about saturated fat compared with prior mainstream guidance, while sharply elevating concern about seed oils and ultra‑processed foods. Concerns like the ones I mentioned above about red meat are usually dismissed. More tallow. Gross. At the same time, news sources have documented substantial financial ties between key new nutrition advisors and meat/dairy industry groups, raising conflict‑of‑interest concerns, even though formal findings of corruption have not been made. A careful characterization is that their final dietary guidelines come from a mostly sincere, MAHA‑aligned preference for animal protein and animal fats, though ignoring much inconvenient evidence about risks, and then accepting convenient reinforcement by industry‑aligned conflicts of interest.

What I will tell patients

If you are reading this then rest assured; you are probably getting enough protein already. No worries. The daily gram-counting and gaming burns mental calories on a problem you most likely don’t have. Bars and powders you probably don’t need.

The real exceptions deserve attention: adults past their mid-sixties, who lose muscle more readily, and anyone losing weight (especially on a GLP-1 medication) who needs to protect lean mass. In both, protein works only alongside resistance training. Without that, the extra mostly passes through.

Consider the newly increased 1.2–1.6 g/kg/day guidance loosely, too. It was hatched through a politicized process with industry ties, despite ironic protestations to the contrary, not the usual evidence-based one. And the nuance of animal versus plant protein is just not there at best, upside down and ideological at worst.

Source matters more than amount. Shifting from red and processed meat toward beans, nuts, fish, poultry, and Greek yogurt tracks with lower mortality and a lighter burden on the planet.

Protein is a building material, not a cure.

I like to enjoy mine when I eat it, rather than thinking of literally everything as a longevity and wellness tool of optimization to be commodified.

Take good care!

P.S. I’ll respond to any comments in a couple days 🌞

I’ve grown older with people now. I have a stake. It matters to me that our country is turning 250 years old. I’ve been alive for 20% of that time, and a lot of it has been invested trying to help people. The health of our country is the grand sum of hundreds of millions of unique lives and circumstances. I help take care of only a few thousand of those… but I can think bigger, step back farther, and write about what I think the United States herself would say about the lowest lows and the highest highs in American health history, if we were to ask.

Five things we got catastrophically wrong

1. The Tuskegee Syphilis Study (1932–1972). The U.S. Public Health Service (adjacent to my profession) enrolled about 600 poor Black men in Alabama, roughly 400 with syphilis, told them they were being treated for “bad blood,” and withheld penicillin even after it became the standard cure, so they could watch the disease run to autopsy. Forty years. It ended only when a whistleblower went to the press in 1972. The debacle helped give us informed consent. It also caused a mistrust we are still trying to earn our way back from.

2. Eugenics and forced sterilization (Buck v. Bell, 1927). Another terrible Supreme Court ruled 8–1 that the state could sterilize a young woman against her will, Justice Holmes writing the line that should hang in every ethics class: “Three generations of imbeciles are enough.” Roughly 60,000 Americans were sterilized across some 32 states. Nazi lawyers cited the case at Nuremberg. We didn’t import this idea; we exported it.

3. Leaded gasoline (1923 onward). We started adding a known neurotoxin to fuel to stop engines knocking. Refinery workers hallucinated and died, and a plant earned the nickname “the loony gas building.” Industry demanded a half-century burden of proof, so we fouled the air every child breathed and shaved IQ points off a generation. Lead in gasoline has been tied to over 150 million excess cases of mental health disorders. We didn’t finish banning leaded road fuel until 1996.

Historically, leaded gasoline was the dominant source of widespread lead exposure in the U.S., even though most identified childhood lead poisoning cases these days are tied to lead-based paint, contaminated dust, soil, and water in older housing and urban environments. It has been estimated that lead is responsible for the loss of 824,097,690 American IQ points as of 2015. Among last year’s federal cuts was a CDC division that helps communities deal with lead in their water. Flouride out, lead back in?

4. The opioid epidemic (1990’s to present). Purdue Pharma launched OxyContin in the mid‑1990s and trained its sales force to tell clinicians that the addiction risk was “less than one percent,” a claim that misrepresented the true risk in chronic pain populations and was transmitted to patients via trusted prescriber relationships. The Sackler family earned billions from OxyContin sales while the U.S. opioid epidemic unfolded. CDC estimates that from 1999 to 2023, approximately 806,000 people died from opioid overdoses, including those involving prescription opioids, heroin, and illicit fentanyl. Of course this is a multi-faceted problem, but producers (and prescribers) share part of the blame.

5. The deliberate dismantling of American Public Health by a radical regime (2025–PRESENT). This is happening now, on purpose, and the body count and human suffering is not hypothetical. Here is just a partial accounting, with every figure sourced:

• Coverage. The reconciliation law signed July 4, 2025 enacted the largest Medicaid cut in history. The CBO projects about 10 million more uninsured over the next 8 years from this alone — and roughly 16 million more once the lapsed ACA subsidies are counted. Researchers at Penn and Yale estimate the coverage and nursing-home provisions will kill about 51,000 Americans a year. A Harvard/CUNY analysis in Annals of Internal Medicine put the Medicaid-cut death toll alone at 8,000 to 25,000 deaths a year.

• The childhood vaccine schedule. In January 2026 “the CDC” stripped universal recommendations from six childhood vaccines, including the hepatitis B birth dose, in place since 1991 and responsible for a 99% drop in pediatric infection, and the arbitrarily halved HPV dosing. One published model finds that delaying the hep B birth dose for a single year’s birth cohort adds 482 deaths and 304 liver cancers. A federal court has since paused the changes, but the president is trying his hardest to force this through all by himself.

• The vaccine advisers. Secretary Kennedy fired all 17 members of the CDC’s immunization committee and replaced them with skeptics after personally promising Republican Senator and onetime proud medical doctor Bill Cassidy that he wouldn’t.

• Health and Human Services shed roughly 20,000 jobs and collapsed from 28 divisions to 15, gutting the labs, the disease detectives, and the staff without whom the scientists can’t function. The list of offices eliminated entirely or reduced to the point of non-functionality is what you might expect if an archenemy were trying to do a disservice to our health and humans.

• NIH funding and science research, gutted by billions, with political strings attached to funding and a massive brain drain away from America. This will hurt our economy, competitiveness, and future treatments and cures that will not be researched and discovered. Click to read more of the note I wrote about this travesty last week, and leave a public comment before it’s too late. More background from YLE published this morning, too.

• Preventive care. “You’re fired” to the leaders of the U.S. Preventive Services Task Force, the body that guarantees no-cost cancer screenings, HIV prevention, and depression screening for roughly 100 million insured Americans. Put a copay back on a mammogram and fewer women get one. Cancers simply show up later, larger, and less survivable. For those of us in primary care, the impact is very concrete: more patients delaying all sorts of preventive care, or simply not qualifying for it, more confusion about what’s covered, and a steady erosion of the progress we’ve made in preventing cancer, heart disease, and other chronic illnesses.

• Planned Parenthood. They severely cut Medicaid reimbursement to a health network where abortion is actually less than 10% of services. About 1.1 million patients risk losing care and some 200 clinics face closure, 60% of them in rural or medically underserved areas, red and blue. The early data: breast-exam visits down 25%, IUD insertions down nearly 40% already.

• Bodily autonomy. The abortion bans this radical Supreme Court made possible keep killing. Maternal mortality rose 56% in Texas in the ban’s first full year, while it fell in states that protected access. Researchers attribute roughly 478 excess infant deaths across states with bans through last summer, more by now. That’s simply horrendous.

• The diseases we’d beaten. Measles came roaring back with 2,144 cases in 2025, the most since 1992, three dead, and elimination status de facto lost. The counts just keep climbing in 2026. Whooping cough hit about 35,000 cases in 2024 and 28,000 in 2025, with at least 13 deaths, most of them infants.

• The world. The new massive Ebola outbreak, from a strain that has no approved vaccine or treatment, is burning through several African nations, with hundreds already dead. It has been declared a global emergency much later than it could have been nipped in the bud. It arrives precisely as the administration fed USAID, in Elon Musk’s phrase, “through the woodchipper.” A Lancet model ties the global aid cuts to more than 14 million avoidable deaths by 2030, a third of them children.

• There’s much, much more, but I am really trying to keep this in my lane. Of course there are also for-profit, taxpayer-funded, literal concentration camps in America in which we have hunger strikes and Unites States Senators being pepper-sprayed. Didn’t think you’d read sentences like that for our 250th did you? We could all write books of them.

Is the present and looming health catastrophe the worst the country has ever had? In some ways it is. Tuskegee and eugenics were more grotesque in kind; nothing here matches that cruelty of intent. But in the sheer scale of preventable death and suffering this seems unmatched. Every prior catastrophe on this list was eventually corrected by institutions like the CDC, the FDA, the NIH, the ethics boards, the disease detectives. But this is the first time they have set about thoroughly dismantling the corrective machinery itself. That is what makes it qualitatively different, and quantitatively devastating. You can survive a disease. It is much harder to survive the destruction of your own immune system. That is what the surveillance networks, the advisory committees, the research pipelines, the institutional knowledge, and the aid programs were: the body’s defenses, built over a century, dismantled in eighteen months of tantrums and trauma. A year ago I wrote about why these project 2025 disciples set about dismantling everything… and over 36K people have read that now.

Five things we got gloriously right

We need hope. We need moral elevation.

We need to recall progress and historic wins even as we are surrounded by present losses.

1. The Framingham Heart Study (1948). Researchers in a Massachusetts mill town followed about 5K people who weren’t sick yet and invented the most useful idea in primary care: the risk factor. You know, like blood pressure, cholesterol, smoking, diabetes, the risk-assessing vocabulary I use 100 times a day? The entire paradigm by which primary care succeeds, the ounce of prevention truly weighing in at a pound of cure, the fact that we can modulate our health in evidence based ways… it all starts with discerning true risk factors. Before this, doctors pretty much just treated diseases. After this lightbulb, we gave much more credence to prevention.

2. The 1964 Surgeon General’s report on smoking. Adult smoking fell from 42% to under 12%. A 2014 analysis in JAMA credits the tobacco-control era it launched with saving roughly eight million American lives — more than any drug, surgery, or device in my lifetime.

8 million lives.

Enough said.

3. The community health center (1965). In the heart of the civil rights movement, Drs. Jack Geiger and Count Gibson opened the first neighborhood clinics on a radical premise: a poor person deserves the right to health care like anyone else. Sixty years later, some 1,400 health centers care for roughly 30 million Americans. Where they open, birth weights rise, hospitalizations decrease, and mortality drops. Communities with them did better during the pandemic as they expanded hours of care, educated their communities, created outreach programs that facilitated testing, and distributed vaccines.

4. The polio vaccine (1955), the pediatric vaccine schedule in general, and mRNA Covid shots in particular. First, the polio vaccine. It emptied the summer wards of paralyzed children, down from the previous 13,000-20,000 a year before the shot arrived. Asked who owned the patent, Dr. Jonas Salk answered that the people did: "Could you patent the sun?"

Second, the pediatric vaccine schedule. “Among (US) children born during 1994–2023, routine childhood vaccinations will have prevented approximately 508 million cases of illness, 32 million hospitalizations, and 1.13 million deaths, resulting in direct savings of $540 billion and societal savings of $2.7 trillion.” It’s excruciating to picture one child dying, but then multiply that times 1.13 million and you can see why we work so passionately to oppose anti-vax ideologies.

Third, mRNA Covid shots. “More than 2.5 million (Covid) deaths were averted… Eighty-two percent were among people vaccinated before any infection, 57% were during the Omicron period, and 90% pertained to people 60 years or older. Sensitivity analyses suggested 1.4 to 4.0 million lives were saved.” The mRNA shots were built on Nobel-winning work done largely here in Philadelphia at UPENN by Drs. Karikó and Weissman.

5. PEPFAR (2003). A bipartisan commitment to treat HIV across Africa that went on to save more than 26 million lives— the largest stand any nation ever took against a single disease. George W. Bush committed the United States to treating HIV across Africa, and it passed with rare bipartisan force. “CDC’s role in PEPFAR had benefits beyond HIV/AIDS. Its expertise and the systems it helped build in each country aided the response to other disease threats such as Marburg or mpox (and Ebola), ultimately providing health security to the United States.” Well done… but now executed in the literal sense by the current administration. Sorry, I couldn’t help myself on that one. To make it up, I’ll list simply list some more honorable mentions, and let the triumphs stand even as we know they are presently being undermined:

• Water fluoridation, first piloted in Grand Rapids in 1945 and expanded to protect teeth across every income bracket

• Medicare and Medicaid, which in 1965 opened the door to care for tens of millions of older, disabled, and low-income Americans overnight

• The American-led eradication of smallpox, declared complete in 1980

• Newborn screening, the heel-prick test pioneered by Robert Guthrie that now catches dozens of treatable conditions at birth

• The Pap smear and the dramatic decline in cervical-cancer deaths that followed, not to mention the HPV vaccine’s incredible success at preventing cancer in the first place going on right now

• Federal seatbelt and auto-safety standards that turned the car from a death trap into something survivable

• The March of Dimes model of mass public fundraising that bankrolled the polio fight

• The Clean Air and Clean Water Acts, which quietly saved more lungs and lives than most heroics that ever happened in a hospital.

The assessment and plan

What can we do now? What kind of birthday present might our country want for her 250th?

Let’s not treat that as a rhetorical question. Hopelessness and despair are crappy gifts and exhausting feelings. Exhausted people obey. So let’s acknowledge the exhaustion, but then keep getting up. We’re not the first to watch a free society get dismantled, and the scholars who studied the last century left us a playbook. Erica Chenoweth found that no movement mobilizing just 3.5% of a population (about eleven million Americans) has ever failed. And Dr. Timothy Snyder’s On Tyranny reads at times as if it were written for those of us in medicine.

Snyder’s first lesson: do not obey in advance. Most authoritarian power is freely given by people guessing what the regime wants before it asks — the family doc who quietly stops educating and recommending routine vaccines because a committee of cranks downgraded it. Don’t. The recommendation changed; the immunology did not. When a frightened parent asks about a shot, we can’t just send them a study. We have to ask what scares them, and listen, because people change their minds for people they trust, not for arguments they lose.

His fifth applies here, too: remember professional ethics. The professions are a firewall, and it’s mostly holding. When RFK Jr’s committee began rewriting the childhood schedule, the vast majority of doctors didn’t fall in line. Instead the AAP, AAFP, ACOG, and AMA, over a million clinicians, kept the real recommendations. Even health insurers we love to hate, covering 200 million Americans, pledged to keep paying for vaccines at no cost through 2026. And the states stopped waiting: the West Coast Health Alliance, a Northeast collaborative, and a bipartisan bloc of fifteen governors now issue their own science-based guidance.

A few more. Believe in truth, and investigate. Tuskegee ended not because the system corrected itself but because one person took the story to a reporter. Keep standing up and out, because the moment one person breaks ranks the rest discover they can too. Practice corporeal politics, because 3.5% isn’t a tweet, it’s people in the street, again and again.

What turns this into a chain reaction is the oldest medicine we have, the one thing this regime cannot defund: moral elevation — the lift in the chest we feel watching someone do the right thing at cost to themselves. The suite of feelings people may experience when witnessing an instance of moral beauty. Jefferson named it 250 years ago, Jonathan Haidt revived it recently, and it’s as real as a reflex. Decency, witnessed, is contagious. One act becomes two, then four. A nation, like my 103-year-old patient, is the grand sum of such acts.

So let our 250th birthday be less a pugilistic, profiteering, grotesque carnival on the White House lawn than a renewal of vows to the science that lets us see risk before it harms us, and to the idea that every life is worth defending. Improving. Healing. From the newborn’s first heel-prick to the centenarian who has lived 41% of the whole American experiment.

Snyder’s hardest lesson is his last: be as courageous as you can. This won’t be undone by one hero. It will be undone the way every American correction has, by ordinary people of common decency refusing, together, to obey in advance. Refusing to normalize the historic plundering of our nation and our world’s health.

Be one of them.

Then be the reason your lovely neighbor becomes one, too.

Subscribe now

Ensitrelvir/Xocova is now approved in the US for post-exposure prophylaxis in anyone 12 and older who’s been exposed to someone with Covid. That’s a first. We’ve had pills to treat the infection for years (Paxlovid and molnupiravir). We’ve had IV remdesivir. Our Doctor/President/Jesus-figure recommended a disinfectant like bleach, and hydroxychloroquine, and many others have touted horse de-wormer. But ensitrelvir is the first oral one cleared in the U.S. that can actually keep us from getting sick after the coronavirus has already made it into our houses.

Ensitrelvir is a SARS-CoV-2 main protease inhibitor. It gums up an enzyme the virus needs to copy itself, the same target Paxlovid hits. The difference is a longer half-life, so it’s a lot less pills taken over 5 days, and it skips needing to add ritonavir, the booster behind Paxlovid’s metallic-mouth side effect and much of its forbidding interaction list. But more on that shortly.

The prophylaxis data, which genuinely surprised me

The approval rests on SCORPIO-PEP, a global trial that enrolled household contacts of people with symptomatic Covid and started them on the drug within 72 hours of the sick person’s first symptoms. That’s key. Among contacts who took at least one dose, 2.9% came down with Covid by day 10, versus 9.0% on placebo. That’s a 67% drop in the odds of catching it from the person coughing across the kitchen table.

I will add that the 9% household spread (secondary attack rate) is much lower than what we see in real life. Previous studies estimate that about 30-40% of household contacts usually pick up Covid. Why the difference? This trial required both PCR confirmation and at least one symptom lasting 48 hours or more, filtering out the subclinical and short-duration infections that Omicron routinely produces in an already-immune population. When the trial measured any PCR-confirmed infection regardless of symptoms, the placebo rate rose to 21.5%, much closer to what we'd expect, suggesting transmission was happening at a familiar pace. Also, households enrolled in a clinical trial are counseled on exposure precautions like masking, separate rooms, better ventilation, etc. Therefore the 9% control arm number undersells how useful this drug could be in a higher-risk setting, like a nursing home outbreak or in a household with higher risk family members for whom a Covid infection presents a significantly bad risk.

More than 98% already had antibodies, from vaccination, prior infection, or both. This was not a naïve population. It was us, the immunologically experienced majority, and the pill still worked, on top of the immunity we already carry. For people with risk factors for severe disease, the gap was wider still with about a 70% reduction in spread.

Another small catch: protection tracked with the five-day course. Once the drug cleared, the treated group lost its edge against any new exposure, with fresh infections accruing at roughly the placebo rate. So it’s important to use other measures while people are sick in the house, like ventilation, beyond just the 5 days of Xocova during the most infectious period.

In the exam room, the limiting reagent won’t be the science. It will be the clock. Getting a real diagnosis in the household, a prescription written, and the first dose swallowed inside 72 hours is a tall order for a working family with one car and three jobs. Some people might prep in advance, the way I talked about with influenza and baloxavir (Xofluza, another weird Xo-named drug).

It’s not an on-label treatment in the US

Here’s where I’d gently correct the natural assumption that a new antiviral means a new way to treat Covid. In the U.S., ensitrelvir is approved only for prevention. The treatment trials in vaccinated, Omicron-era patients (SCORPIO-HR in outpatients, STRIVE in the hospitalized) reliably knocked the virus down on lab measures but failed to move the symptoms that actually matter to a sick person. In hospitalized patients there were even some worrying safety signals and no reduction in mortality. So it’s a preventive, not something I’d reach for once symptoms start. Japan approves it for both. We don’t.

Long Covid? A hopeful maybe

It’s already being floated as a long Covid preventive, and there is a signal, but I want to stay realistic about it. The randomized data showed reductions in post-Covid symptoms that never reached significance, with confidence intervals that all crossed zero. A large Japanese observational study was more encouraging, roughly a 14% lower risk, with a number-needed-to-treat around 26. Promising, and plausible if less virus early means less long-term damage, but not proven. Here are some stats on viral load reduction, if you too like that sort of thing:

The PLATCOV trial (Lancet Infectious Diseases, 2026) provides the first direct randomized comparison of ensitrelvir vs. nirmatrelvir-ritonavir (Paxlovid) in 604 low-risk outpatients (93% vaccinated) with early Covid-19 disease. This trial used ensitrelvir for treatment of Covid.

• Viral clearance half-life: ensitrelvir 5.9 hours vs. nirmatrelvir 5.2 hours vs. no drug 11.6 hours. This is the key metric for those concerned about viral loads and kicking this virus out the body faster.

• Ensitrelvir accelerated clearance 82% faster than no drug; nirmatrelvir was 116% faster

• Ensitrelvir was 16% slower than nirmatrelvir, failing to meet the non-inferiority margin of 10%.

• Viral rebound: 5% (ensitrelvir) vs. 7% (nirmatrelvir) vs. 7% (no drug) — no significant difference

• Bad taste: 1% with ensitrelvir vs. 26% with nirmatrelvir-ritonavir. This is a big deal actually, as Paxlovid can really taste terrible and people often stop taking it mid-course!

• Symptom resolution: Restricted mean symptom duration was identical at 5.4 days for both antivirals vs. 5.9 days for no drug

Do lower viral loads matter for each infection, for each year, for the cumulative exposures we will be getting for the rest of our lives in terms of Covid infections and trillions of coronaviruses running through our bodies? I think they will. I like lower viral loads. In my younger days I always hated parties if my apartment got trashed by the guests.

And finally, in a meta-analysis of all drugs tested in the PLATCOV platform (1,157 patients), nirmatrelvir and ensitrelvir had the largest antiviral effects among all small molecules, exceeding remdesivir and molnupiravir. All of this matters for future variants that might still come bulldozing through. We’ve been lucky so far that SARS CoV-2 has been stuck on the Omicron branch of the family tree for a few years.

The “no ritonavir ingredient like Paxlovid, so no interactions” story is way oversold. Ensitrelvir is itself a strong CYP3A inhibitor, so we still need to check for drug-drug interactions. Darn it. And it can cause fetal harm, so it’s a no-go in pregnancy, with breastfeeding paused through treatment and for two weeks after.

Bottom line

For those of us still quietly trying not to catch a gratuitous case of Covid (yes, I’m the guy still masking on the cough-filled plane and trying to enjoy my hard-earned vacations without whatever viral illness), a well-tolerated pill that cuts the odds by two-thirds after a known exposure is a solid option to add to the kit.

Would I stock up and take it if a family member were sick? Maybe. For higher risk family members? Seems prudent.

For interested patients I’ll be watching the 72-hour clock and their rapid antigen tests, checking the medication interaction list, prescribing when people want it, and reminding Americans that it prevents rather than treats Covid well.

This is the world we practice in now. A recent industry survey found that doctor recommendations and social media are the top sources patients use for supplement information. Another survey from Pew Research Center found that 75% of Americans most highly value medical training, transparency about conflicts of interest, and ease of understanding… with health care providers rated highest (even above AI for now). So here is the best I can do for magnesium. The version I would give in the office if only I had more time. I make no money from commissions or selling stuff.

The strongest case for supplementing is with genuine magnesium deficiency. Certain people are far more prone to it than the general population, where low magnesium levels still run about 3–10%. Antacids like omeprazole cause deficiency in roughly 20% of long-term users, and diuretics used for treating conditions like high blood pressure waste magnesium through the kidney. Poorly controlled type 2 diabetes (10–30% prevalence), alcohol use disorder, and chronic diarrhea or malabsorption are classic culprits, as are certain chemotherapies and older folks on five or more medications. If you fit one of these, the conversation shifts from wellness trend to genuine medicine. Nonetheless, we will briefly review the evidence for magnesium and sleep, migraines, blood pressure, insulin sensitivity, cognitive function, anxiety, and muscle cramps to name a few.

Read more

A patient asked me today whether his atorvastatin was making him feel old before his time — achy muscles, a little foggy, tired by afternoon. He had gone down a few fear-inducing, online, anti-statin rabbit holes. Nonetheless, it’s a fair question, and a version I hear often enough that it deserves a thoughtful, referenced answer.

The short version: some of what is attributed to statins probably isn’t from the statin. And here’s the genuinely surprising part — at the cellular level, statins may actually be doing the opposite of aging people.

On the muscle symptoms

Yes, atorvastatin can cause muscle aches. The FDA says so, and the data backs it up… but with important fine print. A large meta-analysis of 19 placebo-controlled statin trials found about a 7% relative increase in muscle pain or weakness in the first year. That sounds meaningful until you do the absolute math: roughly 11 extra events per 1,000 person-years, meaning only about 1 in 15 muscle complaints in statin users was actually caused by the drug. The rest? Aging, arthritis, life. Another reference about this at the end of the post.

There’s also a well-documented nocebo effect (knowing a medication can cause muscle pain makes you more likely to notice and report it).

On the brain fog

The FDA label includes rare post-marketing reports of memory issues, and many patients worry about their cognitive abilities getting any worse. But when researchers actually studied cognition in randomized controlled trials, they found no difference between statin and placebo groups. The American Heart Association has been direct about this: cognitive complaints are common in middle-aged and older people whether they’re on a statin or not.

On diabetes

Statins do modestly increase diabetes risk — about 0.2% of patients — and it’s almost entirely in people who already have metabolic syndrome features. Worth knowing, not worth panicking about. Higher dosages and certain statins have higher risks. I’ll point you to this deeper dive I wrote previously, including an analysis of concerns that statins might deplete natural GLP-1 levels.

Now for the part that surprises people

Here’s what the basic science shows: statins appear to slow biological aging at the cellular level. Fancy jargon and terms follow. They inhibit oxidative stress-induced cellular senescence through pathways involving SIRT1 and nitric oxide. Statin users show longer telomere length and lower telomere erosion rates with aging, particularly after 65. A recent transcriptomic analysis found statin use strongly associated with suppression of senescence-related pathways in the liver. Statins also accelerate DNA repair in vascular smooth muscle cells and reduce cellular senescence in atherosclerotic plaques.

This is not a claim that statins are longevity drugs. But it is a useful corrective to the idea that they’re quietly making people older, right?

The bottom line

If muscle symptoms are severe, or CK blood test levels are elevated, we investigate. If they’re mild and vague, we talk honestly about the math and probabilities, and about what the drug is actually doing for cardiovascular risk. Most of the time, the calculus favors staying on it.

The aches probably have another explanation, but a good conversation with the doctor is worthwhile. Muscle symptoms are the most common reason people stop taking statins, but true drug-caused muscle side effects are actually rare (estimated at only 1-2%) since most aches turn out to have other explanations like older joints, longer recovery from physical activity as we age, or the nocebo effect. The good news is that with some dose adjustment, timing changes, or switching statin formulations, most people can find a regimen they tolerate, and staying on some form of a statin matters — because stopping altogether is associated with real increases in heart attacks and mortality.

Have a great Friday night if you can, and let me know if you like this format, and want me to share more “quick” answers to real questions I get in the office :)

One of the most impactful things a clinician can do during a heat wave is simply communicate with people before and during the onslaught. A randomized trial of a preventive-message tool for older adults in Australia demonstrated significant uptake of protective behavioral strategies and a 63% lower risk of self-reported heat stress among those who received the intervention compared with controls. So here we go! I’m banging this out before my first patient comes in at 9 AM looking like “🥵” Much of this post has been informed by this paywalled resource about heat illnesses found in NEJM Evidence.

Here is the forecast for today in the US:

Last week we were were still wearing sweaters, with delightful sleeping weather dipping into the 40’s F. Not now. This is the kind of jump that should worry us more than even the long heat waves of August — not because the air itself is more dangerous, but because none of us have had time to get ready for it.

Heat acclimatization is real, and protective. Given a week or two of graduated exposure, the body adapts. Sweat glands open up the micro-hydrants earlier and produce more sweat at lower core temperatures, plasma blood volume expands, resting heart rate drops, cardiac strain falls. It takes at least one to two weeks of progressive exposure to build… and we are nowhere near that. We are walking into an oven our bodies haven’t been trained for yet this year, and the science is clear that this is when more bad things happen — even to the young and the fit.

A few practical things

Cooling, ranked.

• Air conditioning is the single most effective protective factor we have. If you don’t have AC at home, identify a place you can get to like a library, a community center, a mall, a friend’s apartment, even a public cooling center like in the city. Even if you are well off, have a plan in advance. You never know when the AC might fritz out.

• Fans help up to about 99°F; above that they can actually accelerate dehydration without cooling us meaningfully. And on humid days, fans work even less.

• Cool showers, self-dousing with a wet towel, and misting combined with a fan are all effective and can be repeated through the day.

• Pre-cooling before going outside if we have to go there can reduce risk: cold drinks, a cold shower beforehand, a damp shirt, a cooling vest for outdoor work.

• Pay attention to humidity, not just temperature. Sweat only cools us if it can evaporate, which is why 90°F at 70% humidity is more dangerous than 95°F in dry air. The number on the thermometer isn’t the whole story.

• The WetBulb Globe Temperature is a measure of the heat stress in direct sunlight, which takes into account: temperature, humidity, wind speed, sun angle and cloud cover (solar radiation). This map is wonky but a decent resource.

Hydration. For most of the day, plain water is fine. For sustained physical activity longer than an hour, we want a little sodium with our fluid. Sports drinks meet the threshold, and so does a little salty snack plus water. Plain water in very high volumes during long exertion can cause exercise-associated low sodium levels (hyponatremia), which is its own emergency. Don’t wait for thirst as it lags behind dehydration, particularly in older adults, where the sensation is blunted.

Move activity to the edges of the day. Plan outdoor exertion (if needed) before 10 AM and after 4 PM. Think loose, light-colored clothing. Sunscreen (sunburned skin can’t dissipate heat as well). Build in more rest breaks in shade than we think we need, especially if for outdoor workers and athletes practicing after school. The internal pressure to push through is one of the most consistent risk factors for catastrophic heat illness, especially in young athletes; be wary of any environment that frames perseverance as a virtue when bodies are telling people to stop.

If you’ve been recently sick, sit this one out. This is one of the most frequently missed risk factors. Recent febrile illnesses, particularly upper and lower respiratory infections, markedly raise the odds of serious heat illness. If you’ve had a cold, flu, Covid, stomach bug, or any fever in the past week, your thermoregulation is not where it normally is. Skip the workout. Stay in the air conditioning. This applies to children and athletes, too.

Check medications. A number of common medications impair sweating, blunt cardiovascular compensation, or push the body toward dehydration. Many commonly prescribed medications impair thermoregulation and increase heat risk. Patients might discuss their medication list with their clinician before the warm season. With help from Open Evidence I prepared the following quick review of drug classes and their mechanisms:

• Dehydration risk: diuretics, laxatives, alcohol

• Increased heat production: amphetamines, thyroid hormone replacement, stimulants/supplements (caffeine, creatine, ephedra)

• Impaired sweating: anticholinergics (oxybutynin, dicyclomine), first-generation antihistamines (diphenhydramine), topiramate

• Impaired cardiovascular compensation: beta-blockers, calcium channel blockers, antiplatelet agents

• Disrupted central thermoregulation: antipsychotics, lithium, SSRIs, tricyclic antidepressants, phenothiazines

• Reduced alertness to heat: benzodiazepines, alcohol

A study of older Medicare beneficiaries found that use of heat-sensitizing medications was associated with a 16–37% increased risk of heat-related hospitalization even in the absence of heat waves, with heat waves further increasing risk by another 21–33%. This is not a reason to stop any meds — it’s a reason to be more careful this week if you’re in the red zone, and worth a quick word with your clinician if you’re unsure.

Know the warning signs. Heavy sweating, cramps, headache, nausea, dizziness, weakness, a racing heart that doesn’t settle with rest.

Stop, get to a cool place, hydrate. Heat also dulls cognition and reaction time well before more obvious symptoms appear — if you’re driving, operating equipment, or supervising children, the loss of judgment can arrive before you notice it.

The red-flag finding is altered mental status — confusion, slurred speech, irritability, stumbling, agitation. That is heat stroke until proven otherwise, and it is a 911-level emergency.

Cool first, transport second. If you are with someone who has collapsed in the heat or has become confused, call 911, then begin cooling them immediately, before the ambulance arrives. The damage from heat illness is dose-dependent on time spent at high core temperature, and every minute matters. The gold standard is cold-water immersion — a bathtub, a kiddie pool, even a large cooler with ice water — anything that gets the body wet and cold all over. Failing that, cool wet towels rotated every minute or two, ice packs at the neck, armpits, and groin, and aggressive fanning to keep evaporation moving. Do not wait for the hospital to start cooling. This single principle (cool first, transport second) has a larger effect on survival than almost any other intervention in heat illness. More on heat-related illnesses and proper triage.

Check on someone. Call your parents. Knock on your elderly neighbor’s door. The people most likely to die in a heat wave are older adults living alone, and the most consistent predictor of who gets hurt is whether anyone is checking in. For example, a program of phone calls and home visits for adults over age 80 during Rome’s summer heat waves was associated with a proven drop in excess mortality. The intervention is genuinely that simple, and it works.

The forecast will turn cooler in a few days across much of the US.

Until then, treat the week with more caution than the calendar suggests.

Have a cooler heat wave day if you can!

In today’s post I’m going to make a confession that may cost me some professional credibility: I sometimes take those naps. Not every day, not for long, and not after 3 PM. But on my long day in the office, seeing patients and doing the rest of the job for up to 13 hours, I try to close my eyes and drift off for 20 minutes around 2:45 PM. I have always felt vaguely conflicted about this, in the way that physicians sometimes feel guilty about things that appear self-indulgent and delay the completion of never ending tasks. So I went looking for the science, and as usual, the science is more interesting and more complicated than the headlines about naps suggest.

Let’s start with what the evidence actually supports.

Short naps work. This part is pretty solid. For example, a meta-analysis of 60 studies, published in Sleep Medicine Reviews, found that afternoon naps produce meaningful improvements in:

• cognitive performance

• vigilance

• procedural memory

• declarative memory

• processing speed

The overall effect size held up across different age groups, nap durations, and whether or not subjects were sleep-deprived. The strongest signal was for vigilance, which matters if you’re a surgeon, a family doc tracking 5-25 chronic problems a visit, a nurse, a truck driver, or a parent. The optimal duration for maximizing alertness with minimal grogginess appears to be somewhere between 10 and 20 minutes. Long enough to restore, short enough to avoid the deep-sleep hangover (what researchers call “sleep inertia”) that can leave us feeling worse than before we lay down.

The sweet spot for timing is the early-to-mid afternoon, somewhere between 1 and 3 PM, which aligns with the circadian dip that most of us experience after lunch. This is not caused purely by lunch, by the way, though I swear carb-heavy lunches zonk me out. Rather, it’s a biological rhythm that happens whether we eat or not. Napping after 4 or 5 PM, on the other hand, encroaches on nighttime sleep quality, especially if we are prone to insomnia.

There are a few other findings worth noting. 30-minute naps specifically seem to improve memory encoding, not just alertness. Naps of 60 to 90 minutes, if we include slow-wave and REM sleep, can produce more durable cognitive gains, but come with a meaningful sleep inertia cost. Athletes show real physical recovery benefits from naps in the 40-to-90-minute range, including improved strength, reduced muscle soreness, and better attention. And in a controlled study where healthy young men were sleep-restricted to two hours and then given two 30-minute naps, the napping normalized elevated inflammatory markers (specifically interleukin-6 and urinary norepinephrine) that spiked in the non-napping group. However, I don’t recommend such a bizarre routine! But the researchers conclude that napping has stress-releasing and immune effects.

Now for the part that complicates the story.

The research on long or habitual napping paints a darker picture, and a new study just published in JAMA Network Open adds important detail. Researchers from Mass General Brigham tracked 1.3K older community-dwelling adults using wrist actigraphy (actual objective sleep measurements, not self-report) for up to 19 years. That’s a rare, long game study! They found that longer nap duration and more frequent naps were independently associated with greater all-cause mortality, even after adjusting for nighttime sleep quality and chronic health conditions. More strikingly, morning nappers had a 30% higher mortality risk than early afternoon nappers. But the authors suggest that morning napping (when healthy adults are typically alert) may signal underlying health problems, circadian disruption, or subclinical disease.

Here’s a table from the study, and I’ll break it down for you below:

These four survival curves track mortality over those 19 years in 1.3K senior adults, grouped by napping behavior as measured by wrist actigraphy. The curves show the extremes (10th percentile versus 90th percentile, short and sweet versus long and excessive).

Panels A and B show that people who napped longer and more frequently had higher mortality rates, both reaching statistical significance. Panel C showed that day-to-day variability in how long naps lasted had no relationship with mortality (p=.93). Panel D showed how morning nappers fared meaningfully worse than afternoon nappers, once again more likely because habitual morning napping signals an underlying health problem rather than a benign rest preference.

This is a pretty well-designed study, and the use of actigraphy rather than self-report is a genuine methodological improvement over most prior work. That said, it’s still observational, which means we can’t prove causation from it. The fundamental interpretive challenge with napping research (what the authors themselves call the “nap paradox”) is that sick people nap more. This echoes sleep. Fatigue, depression, cardiovascular disease, sleep apnea, early neurodegeneration, practicing family medicine on long Wednesdays: all of these increase the drive to nap. So when a study finds that people who nap more have higher mortality rates, it’s genuinely hard to disentangle whether napping is contributing to poor health outcomes or simply reflecting them. The JAMA authors adjusted carefully for known comorbidities, but residual confounding from unmeasured or subclinical disease almost certainly remains. We should take the findings seriously, but stop short of proclaiming them as proof that napping kills. My brain needs that nap sometimes, and I have to go with that sleep drive.

Beyond this timely JAMA piece, the broader meta-analytic literature needs to be considered as this was not the first study of napping. The broader literature seems to show a consistent dose-response relationship between nap duration and both cardiovascular disease and mortality: the longer the nap, the higher the risk.

For cardiovascular disease, the relationship follows a J-curve, with risk actually declining slightly from zero to about 30 minutes, then rising gradually, then climbing sharply beyond 45 minutes. Don’t worry about the technicalities if this is confusing, just follow the solid black “J” curve plotted versus “nap time” on the X-axis. The “1.00” on the Y-axis represents the average rate of cardiovascular disease. Below 1 is potentially good/less risk, above 1 is potentially bad/more risk. The dotted lines are the range of possible truth:

Looking at that best case scenario, lower dotted line, it is statistically possible that people taking a 30 minute nap have a lower risk of cardiovascular disease… but the much more likely solid black line shows us it’s not really significant.

For all-cause mortality, the relationship appears to be a statistically significant positive linear trend across the full range of nap durations, with roughly 4% higher mortality risk for every additional 10 minutes of daily napping. When researchers grouped all naps under 60 minutes together and compared them as a category to non-nappers, the result didn't reach statistical significance. But that categorical analysis likely obscured a real underlying trend rather than suggesting short naps are truly risk-neutral. Naps of 60 minutes or more crossed clearly into significance, with a rate ratio of 1.82 for cardiovascular disease and 1.27 for all-cause mortality.

Here’s that linear relationship between nap duration and all-cause mortality:

Need a break from study analysis, too?

So what do we do with all this?

For most healthy adults, the evidence supports short, occasional naps (10 to 30 minutes, in the early afternoon) as genuinely beneficial for the brain’s performance and not posing a meaningful risk for overall health. If you’re an athlete partying at the Olympics with all-nighters, or an average person who had a bad night of sleep, occasional longer naps may be appropriate, with the understanding that we’ll need some time to shake off the grogginess before operating heavy machinery, seeing patients, or smiling for the public.

The more important clinical message may be this: if we find ourselves needing to nap every day, for long stretches, especially if we’re napping in the morning, that’s worth paying attention to. It may not be laziness. It may not even primarily be poor sleep hygiene. It may be our bodies signaling something worth investigating, whether sleep apnea, depression, cardiovascular strain, or early cognitive changes. A nap is not just a nap. The literature suggests that sometimes it’s a symptom.

But other times a well-located, well-timed, who-cares nap can be an exquisite pleasure. The scientific literature does not acknowledge this enough, so I will provide this illustration for proof:

And finally, since we live in the age of the wellness influencer: “Sleepmaxxing” has apparently become a massive TikTok trend (I don’t watch this), amassing over 100 million posts on the platform, aimed primarily at Gen Z and focused on maximizing both the quantity and quality of sleep through various hacks and interventions. Common tools include blackout curtains, magnesium supplements, mouth taping (don’t!), sleep trackers, weighted blankets, white noise machines, and cooling mattress toppers. Napping occupies an ambivalent place in this ecosystem. On one hand, expert-endorsed sleepmaxxing strategies that are actually evidence-based include things like decreasing caffeine, increasing physical activity, reducing evening light exposure, and notably, limiting daytime naps. On the other hand, a parallel social media trend frames “sleep-positive” households as ones that encourage sleeping in and taking naps, explicitly pushing back against the idea that napping equals laziness.

Follow your instincts in addition to the decent science and questionable influence, as one size does not fit all. It should be noted that napping is not a human quirk — it is ancient and conserved across the animal kingdom. Virtually every species that sleeps also naps, and most mammals are naturally polyphasic sleepers, taking multiple shorter rest periods across the day rather than one consolidated nocturnal block. The seven to nine-hour monolithic sleep we treat as normal and indeed virtuous may itself be a recent artifact of industrialization and artificial light. But from an evolutionary standpoint, the afternoon urge to nap isn't a failure of discipline. It's an inheritance, and it must have some survival advantages:

I’m going to keep my 20-minute afternoon power naps once a week, and feel only mildly conflicted about them.

They help my patients if nothing else, bolstering my vigilance for another evening session as I see people in the office.

What does your intuition tell you about napping, and what are your good, bad, and somewhere in between habits?

Answering this question involves resisting my reflexive defensiveness regarding a medicine that has helped/saved many lives. But we also need to acknowledge a reflexive dismissiveness towards any Covid treatment in these “all-clear” post-pandemic days.

The NEJM study pooled results from two open-label platform trials (PANORAMIC in the UK and CanTreatCOVID in Canada) involving vaccinated outpatients “at elevated risk for severe disease.” But that characterization is not really fair. It's worth noting who wasn't in these trials. Patients were excluded if they had significant drug-drug interactions with Paxlovid, required dose adjustments for renal impairment, or were pregnant. In practice, these are often the oldest and most medically complex patients — the ones with the most medications, the most comorbidities, the most to lose from a prolonged Covid course. Changing up meds for a week can be done in many patients with the help of medication interaction checkers (instead of just telling them “no can do” like in the trials). The study population skewed female (65-69%), was younger than typical high-risk primary care patients (median age ~55), and was heavily vaccinated and relatively healthy by the standards of who actually ends up hospitalized. I wouldn’t say this was cherry-picking in any nefarious sense. Rather it's the pragmatic reality of trial design. But it does mean the results may underestimate benefit in the patients for whom we most often consider Paxlovid. A number needed to treat that looks unimpressive in a 55-year-old woman with one comorbid condition and no kidney disease looks quite different in a 72-year-old man on a statin, a blood thinner, and an immunosuppressant — who, notably, also couldn't have enrolled. The baseline risk of hospitalization or death in the studied population already was under 1%.

The headline finding was this: Paxlovid did not significantly reduce that primary endpoint of hospitalization or death compared to usual care.

And here is where I want to slow down, because this finding is being interpreted in ways that I think are incomplete.

Drs. Fauci and Lane, writing the accompanying editorial in the same issue of NEJM, make an important point in both directions. On one hand, the 89% relative risk reduction (hospitalization/death) gained by using Paxlovid seen in the original EPIC-HR trial applied to an unvaccinated population facing the delta variant. That’s not today’s world, and we shouldn’t pretend otherwise. On the other hand, a failed primary endpoint in an underpowered study is not the same thing as proof of no effect. When your baseline hospitalization rate is under 1%, you need tens of thousands of patients to reliably detect a difference in that outcome. You need to include more comorbidities and an older median age. These trials enrolled only a few hundred to few thousands. The absence of a statistically significant signal tells us the study couldn’t see one, not necessarily that one doesn’t exist.

Additionally, the secondary endpoints tell a different story. In PANORAMIC, patients on Paxlovid were 74% more likely to achieve early sustained recovery. Median time to complete recovery was 14 days with treatment versus 21 days. That’s a week of people’s lives. In CanTreatCOVID, patients were roughly twice as likely to recover faster. Viral load dropped more rapidly in the Paxlovid group. These are not trivial findings. They’re just not the ones that make the contrarian headlines.

Dr. Vincent Racaniello, virologist and TWiV legend, age 73, put it plainly: he would take Paxlovid “no question” to reduce symptoms if nothing else. Dr. Daniel Griffin (virologist and TWiV co-legend) gave a summary that was even simpler: patients were “about twice as likely to feel better sooner.” These are experts who have spent considerable time steeped in the data every week of the pandemic. Their read matters.

Furthermore, and I want to be honest about what we know and what we’re still piecing together, there is the question of what Paxlovid does downstream. This is where the viral load story becomes a bit speculative and a bit evidence based, and certainly outside the box.

The evidence across multiple viral illnesses is consistent: higher viral loads predict worse outcomes, and faster clearance matters. But it’s worth being precise about what that means, because the relationship differs meaningfully depending on the infection.

In chronic viral infections, the principle is clearest. In HIV, viral load is the central metric guiding treatment — earlier antiretroviral therapy produces better survival, less disease progression, and greater immune recovery. In chronic hepatitis B, HBV-DNA levels correlate directly with liver damage severity, cirrhosis risk, and hepatocellular carcinoma. Just ask the leaders who just trashed the birth dose recommendation for Hepatitis B. Actually, don’t do that. But in these chronic diseases, sustained viral suppression directly and durably changes trajectories.

Acute self-limiting viral infections are more nuanced. Oseltamivir (tamiflu) reduces influenza viral load dramatically, and cuts overall viral shedding more than tenfold. Yet in otherwise healthy outpatients, symptom duration shortens only modestly. The viral load reduction and the symptom benefit appear to travel somewhat independently. Where oseltamivir’s impact becomes unambiguous is in high-risk patients: in hospitalized elderly patients, it reduces mortality regardless of when it’s administered. It also helps reduce household spread. I wrote about this and baloxavir before this recent flu season.

Covid sits in an interesting, still-evolving, sometimes middle position. For most patients, live virus clears within two to three weeks. But a meaningful subset harbors persistent viral remnants like nucleic acids and antigens detectable months to years after infection. Some develop true viral reservoirs where subgenomic RNA, a marker of active replication, has been detected in tissue biopsies well beyond the acute phase. Patients with long Covid show virus-specific T cell levels dramatically elevated compared to those who fully recovered, suggesting ongoing antigenic stimulation rather than simple post-viral fatigue. A prospective study found that slower viral clearance after peak infection was associated with nearly a fivefold increased risk of long Covid in patients with five or more acute symptoms.

This points toward a plausible mechanism for some people: high early viral load → suboptimal clearance → reservoir establishment → chronic immune activation → long Covid. If that chain is real, intervening early before the reservoir forms matters in a way that’s categorically different from simply shortening a bout of flu. With Covid the therapeutic window appears narrow and front-loaded, which is precisely why the five-day prescribing clock matters. And in the larger of the two studies synthesized in NEJM last week, the median start time for Paxlovid was already on day 4, kind of a buzzer beater after peak viral load had already taken place.

What about preventing Long Covid? I am hopeful, but I think the honest answer is that we still don't yet completely know, and the data are genuinely mixed. Observational studies are suggestive but imperfect. The largest target trial emulation (drawing on nearly 450,000 patients in the N3C/RECOVER database) found no significant overall reduction in long Covid, but did find modest benefits in terms of cognitive symptoms and fatigue. A VA cohort study in an older, sicker population reported a more meaningful effect — a 26% relative risk reduction, with a number of patients needed to treat of around 22. You can dive into a whole bunch of previous studies here in a post I did about Paxlovid a while back.

Critically, the two published randomized trials that tested Paxlovid for treating established long Covid — rather than preventing it — were both negative. That distinction matters: the therapeutic window for interrupting viral reservoir formation appears to be early and narrow.

The 2026 European Respiratory Society guidelines still recommend antivirals during acute Covid for long Covid prevention, but with low-to-moderate certainty. Two ongoing prevention-focused randomized trials are still pending, and their results will be essential. We are, in other words, still mid-experiment.

So where does this leave us in primary care?

I’m not prescribing Paxlovid for every sniffly thirty-year-old who tests positive. They generally don’t even know they have Covid because most people don’t test or stay home. But for my middle age to older patients, for anyone with immunocompromise, diabetes, obesity, or significant cardiovascular disease, for anyone for whom a week less of illness actually matters, or for whom the downstream possible consequences of a prolonged viral course are genuinely concerning — I am still having this conversation. Still offering treatment if they want it, and even bringing up metformin if they are concerned about preventing Long Covid. Covid is still best considered on a case-by-case basis, which is, as Fauci and Lane conclude, exactly what the evidence now supports.

The high bar of only caring whether an antiviral reduces hospitalization and death has always been perplexing to me regarding Covid. Do we do the same with other treatments? The hospitalization rate for urinary tract infections is less than one percent, yet we gladly treat women with antibiotics to cut the duration of their symptoms and risk of associated complications. Men with erectile dysfunction have been prescribed meds like Viagra billions of times, despite that condition resulting in not even one hospitalization that I’m aware of.

The cost of Paxlovid can still be formidable. Paxcess can help some. There are drug interactions to figure out. The taste can be gross. I admit that I’m coming from a primary care angle… but we are the ones who see our patients long term, who have the most vested interest in their health over decades, and who pick up the sometimes subtle, sometimes obvious problems after Covid infections. So if we have a drug that reduces symptoms (and in my anecdotal experience it really does make a difference), reduces severe disease especially in higher risk populations, works best when given earlier than day 5, reduces viral load, speeds viral clearance, and is still being studied in randomized controlled trials to definitively answer the question of long Covid reduction…

See what I mean?

Keep the conversation going?

~

Beware the carb processing, and consequent glycemic index

This first article comes from STAT News. It was written by Dr. David Ludwig, a Harvard pediatrician and author of two recent books on food. He argues that the obesity epidemic is not caused by food being too tasty or “hyperpalatable.” He points out there’s little evidence that deliciousness drives overeating, noting that many binge foods (chips, pretzels, cereal) are actually fairly bland. Instead, he identifies fast-digesting carbohydrates as the real culprit — they spike blood sugar, then crash it, triggering cravings and a repeat cycle. The article mentions popcorn, potato chips, pretzels, fries, breakfast cereals, candy, sugary beverages, and corn syrup as examples, and other common fast-digesting carbs include white bread, potatoes, white rice, soda, and pastries.

He criticizes a broad “ultra-processed food avoidance” framework as imprecise, arguing it would make food less appealing without necessarily making it less fattening. He advocates for specifically targeting processed carbohydrates rather than tastiness or ultra-processed foods generally.​​​​​​​​​​​​​​​​

Putting it into practice

I want to expand on this article by pulling in the glycemic index framework as a way of understanding what Dr. Ludwig is talking about. Glycemic index (GI) ranks carbohydrates on a scale of 0–100 based on how quickly they raise blood sugar, with pure glucose scoring 100. Foods above 70 are considered high-GI and are the ones most likely to trigger the surge-and-crash cycle that Ludwig describes. This drives fat deposition and insulin resistance.

The simplest strategy might be pairing, not eliminating. Adding protein, healthy fat, or vegetables to carb-heavy meals slows digestion and blunts the blood sugar spike. For example, topping rice with chicken and snow peas , spreading almond butter on toast, or eating manchego cheese with crackers makes a big difference in how quickly our bodies can process the carbs and limit blood sugar spikes. Chicken, vegetables, almond butter, and cheese are the speed bumps.

Swapping refined grains for whole-grain versions (steel-cut oats instead of instant, whole wheat pasta instead of white) also lowers the glycemic impact. More on this in the second article below BTW! Eating whole fruit rather than drinking juice keeps the fiber intact, which slows absorption. Even meal order matters — eating vegetables and protein before carbohydrates can reduce post-meal glucose spikes significantly. I recorded a mini-podcast on meal sequencing if interested.

A side note on potatoes from a defensive 50% Irish man: despite their high GI, a major 2025 BMJ study found that baked, boiled, or mashed potatoes were not significantly linked to increased diabetes risk. French fries were the real problem, associated with a 20% higher risk at just three servings per week. So preparation method matters enormously. Keep the scrubbed skin on for extra fiber, and avoid deep frying.

There are many GI charts out there to memorize and consult, but here is a good one I found (from Arkansas Heart Hospital, far from Philly!). There are some really baffling yet confirmed numbers here, like Gatorade being worse than Coca Cola, ice cream lower than watermelon, and Russet baked potatoes being much worse in terms of GI than those French fries. Once again, beware of reductionist plots. GI is just one part of the overall healthy food consideration:

~

The second article comes from The Atlantic, and is written by Dr. Trisha Pasricha, a Harvard gastroenterologist who is making the rounds in support of her new book You’ve Been Pooping All Wrong. Her article in The Atlantic is not much about poop, however, but instead centers on the problems with whole grain labeling and reality.

The whole-grain myth

Dr. Pasricha begins with the simple premise of the whole-grain revolution: swap refined grains for whole ones, and sidestep the metabolic damage caused by blood sugar spikes.

For decades, that bargain seemed solid. Studies linked whole-grain consumption to lower rates of cancer, heart disease, and diabetes. But a closer look reveals the science and food engineering are considerably messier than grocery store labels suggest.

The first problem is definitional. There’s no agreed-upon standard for what actually makes a food “whole grain.” The FDA, the Dietary Guidelines, and the industry-sponsored Whole Grain Council all use different criteria, meaning the same product could qualify under one definition and fail another. Worse, ready-to-eat breakfast cereals (among the most heavily processed foods in the supermarket) are the category most likely to earn a whole-grain label, often while being high in sugar and low in fiber.

The deeper truth lies in the structure of the foods. Many commercial whole-grain products are, as one researcher puts it, “molecularly disassembled” — their bran, germ, and endosperm are sourced separately and then recombined. This processing destroys the physical architecture of the grain, causing it to be digested just as quickly as refined flour, spiking blood sugar at nearly the same rate. In one striking study, whole-wheat bread produced a higher blood sugar response than white pasta, because pasta’s dense, extruded structure resists digestion in a way that bread simply doesn’t. I think cooking to just al dente makes a difference, too. It makes sense that more dense foods simply take longer to disassemble, thereby slowing down their digestion and resulting glucose spikes.

I’ll expand her article, too, and add that the best practical examples of genuinely beneficial whole grains are those eaten in their intact kernel form — farro, barley, wheat berries, oat groats, and spelt among them. These are boiled and eaten as whole grains, not ground into flour and reconstituted, so their dense starch matrix remains intact and digests slowly in the way the research encourages. Even within a single grain, form is everything: oat groats and instant oats are the same species, but their metabolic effects diverge sharply based on processing. The rule holds consistently → it's not as much which grain we choose, but how much of its original structure survives by the time it reaches our plates and stomachs.

Processing also affects the gut microbiome. Intact grains deliver starch all the way to the colon, where bacteria can convert it into beneficial compounds. Finely milled grains are absorbed much earlier in the digestive tract upstream, robbing downstream gut bacteria of that fuel entirely.

The practical upshot: intact, minimally processed whole grains like oatmeal, brown rice, corn on the cob, etc. still appear genuinely beneficial. But most of what lines supermarket shelves may offer little advantage over the refined products they replaced.

Two very useful rules of thumb emerge from the article:

• Look for foods in which carbohydrate grams are no more than 10x the fiber grams.

• Run a simple mental experiment — would this food rapidly dissolve into mush if placed in water? If mush likely, our GI systems are going to make much the same, and absorb that carbohydrate→glucose quickly.

Here are four quick examples from my pantry for practice.

I recently discovered this pasta at the Whole Foods down the block. It meets the total carb to fiber ratio of less than 10X (36g/5g). Soaking in water would not result in mush, unless that water was boiling, but I’ll keep it al dente. The structure of the pasta feels pretty dense when chewing it, which is mechanically and intuitively better.

This is our go-to cereal, though we know that breakfast cereal in general is not all that great. But the carb/fiber ratio comes in at 47:7, which is <10. Unlike most products, Kashi actually tells a story behind the “whole grain wheat” in their Autumn Wheat cereal. According to the product description, Kashi starts with whole organic wheat berries, then steams, shreds, and weaves them before baking — which sounds closer to a structurally intact product, similar in concept to shredded wheat. That process is meaningfully different from grinding to flour and reconstituting. The mush test? Probably not great, as I’ve experienced this with milk in a bowl.

Sometimes I’ll go easy on the gluten. Jovial's brown rice pasta holds up pretty well structurally when cooking/eating it… but fails the carb to fiber ratio. 44 divided by 2 = 22 → much higher than 10. Doing a little research on the preparation method, it looks like the brown rice is milled into flour first, so the original kernel structure is lost. It's not an intact whole grain in the way that cooked brown rice or farro is. But the extrusion process partially compensates: pressing the dough through dies under pressure creates a compact, dense starch matrix that resists digestion and blunts blood sugar response significantly compared to the same flour baked into bread or crackers. Nota bene - the “bronze” dies, despite the artisanal marketing, are primarily a texture feature rather than a metabolic one. So Jovial seems like a genuinely better choice than most gluten-free alternatives, and better than rice-flour bread or crackers — but it shouldn't be conflated with eating intact whole grains.

We have a few great recipes using farro in our repertoire. But usually they include a good amount of preparation time. Interestingly, the ratio on this package exceeds the proposed 10 (35g divided by 3g). But farro is a good illustration of how nuance matters, because it is minimally processed and remains much more structurally intact than say a whole grain slice of bread with a spongy matrix that might technically have a higher fiber content. Soaking this cooked farro in water (broth) takes a decent time to turn into complete mush.

Take away

Presenting, synthesizing, and riffing on these two articles does not a dietary manifesto make… but I hope you come away from this post with some whole grain food for thought. I did. The takeaway from both articles is basically the same: the quality of our carbohydrates matters more than the quantity, and structure is the hidden variable that labels rarely capture.

A grain that arrives at our plates largely intact — chewy, dense, resistant to quick digestion — is doing something fundamentally different from one that has been milled, reconstituted, and pressed into a convenient shape, regardless of what the package claims.

The 10x carb-to-fiber rule and the mush test are imperfect but useful shortcuts in a food environment full of misleading signals and limited time for planning, researching, and executing. Breaking complex foodstuffs down into more simple components, then reassembling them back into supposedly whole grains, is kind of humptydumpty-ish, right?

But modern life demands we take carb shortcuts… so consider pairing carbs (including those made with “whole grains”) with vegetables, healthy fats, proteins, and meal sequencing when possible.

From the earliest age of childhood, who among us has not looked upon a crystal with a sense of wonder, scientific curiosity, and even magic. Crystals have long been suspected of having mysterious powers… but in terms of those formed by cholesterol, evidence increasingly points towards a dark power. They can trigger our human immune systems. One only need ask a sufferer of gout (a different kind of crystal disease involving urate) how the reaction to crystals can bring us to our knees. I’ll leave gout for another day, but in this post I want to synthesize and present what cholesterol actually does inside our arteries, and more importantly, what we might do about it.

We’ve been telling an incomplete story about cholesterol for decades.

The usual story is one about plumbing. Cholesterol builds up, arteries clog, flow slows, heart attack can happen. This makes intuitive sense to both doctor and patient and plumber. Fix the sludge in the pipes, lower the disease. The culprit was simple, the solution was simple, and it fit neatly on a prescription pad and lifestyle improvement educational sheet. But only considering a slow clogging hypothesis failed to explain why most sudden heart attacks occur in people without critically blocked arteries detected in primary care or cardiology check ups. In fact, the majority (up to 80%) of all heart attacks arise from plaques that were found to be non-obstructive on prior testing, highlighting that plaque biology often matters more than blockage severity.

It’s why many patients are left confused after feeling great one day and then having a heart attack the next.

What’s happening inside our coronary arteries isn’t just a passive clogging. It’s a slow burning battle of attrition. And the damage isn’t coming from cholesterol alone. Much of it is coming from our own immune systems, as we try to fight a great war on crystals.

Read more

Nonetheless, in Covidlandia we believe:

• Covid infections can still bang us up sometimes — on the exterior and under the hood. It’s wise to game out the before, during, and after.

• Prevention lies along a continuum. It can include varying risk tolerance with Covid rates, fair to quality masking, situational nuance, staying up to date with yearly or q6 month boosts, maybe some nasal sprays, and protecting others when we are sick.

• Treatment can/should include antivirals like Paxlovid, especially for people who are at higher risk… which in America is the majority of us. I personally like metformin, too.

• Recovery from Covid infections needs to be mindful of long Covid potential, symptoms, and caution with return to vigorous physical activity.

• We don’t shame anyone who wants to mask, to get a shot or not at this point, who wants to keep talking about Covid or take a break. But we have no tolerance for Covid trivialization or pandemic revisionism, nor authoritarian agendas in the house of science and medicine. Too many have died alone, suffered mightily, fought through personal ruin and professional burnout to play that. Dr. Robby knows what we are talking about.

  

And so without further delay, here is our Covidlandia update for April, 2026!

What’s the scenario in the US right now?

Not too shabby. Covid is still out there for the catching, but the winter wave has crested. Based on sewage analysis, might not be a good time to take a vacation in Tennessee or Mississippi, where it is estimated that 1 out of every 25 people is infectious with Covid illness right now. But where I practice in Jersey and Philly, we are sitting at about 1 in every 100 people. Might want to check out Idaho, as they are the best in the country at 1 in every 269. Looks beautiful:

I wore an N95 on the planes to and from a great international vacation last week (despite ICE disgraceful hyperfunding and TSA disgraceful underfunding). On the flight home we were surrounded by a very ill family — literally constant productive coughing, with the little girl even vomiting twice. Poor little one.

But no one in my family got sick this week. Thanks 3M Aura. You are now like 16 for 16 on flights, batting a thousand. Wish I had known about you my whole life, as half of my vacations involving flights were previously marred by illness!

The BA.3.2 variant seems to be no cause for alarm

The Global Virus Network is among others monitoring BA.3.2, an Omicron subvariant informally dubbed “cicada” for its re-emergence after a quiet period for this branch of the coronavirus evolutionary tree. BA.3.2 might have been percolating and accumulating mutations in an individual unable to fully clear their infection. Early data show immune escape consistent with SARS-CoV-2 evolution, but no increased severity or sustained transmission growth. So it might lead to easier infections, but hopefully not worse outcomes than typical. Staying current on vaccinations and routine precautions remain the appropriate responses… and considering this “cicada” nickname odd and insulting to real cicadas.

The BA.3.2 variant might become the target of updated Covid shots next season, but the WHO has not made a decision yet. Kids might be 5X more likely to be infected by this variant than others, with some interesting speculation as to why.

Spring boosters

If you got an updated vaccine in the fall/winter like me and you are under age 65, then you are probably all set. But if you are over age 65 or have immunosuppression, it is actually recommended you get a second shot/boost.

As of this linking to the CDC Covid shot page on 4/12/2026, their guidance seems pretty legit and in basic agreement with more stable organizations like AAFP, AAP, ACOG, IDSA, etc.

If you want that recommended second shot:

• For the 2025–2026 dose #2:

  • Moderna [Spikevax], Novavax, or Pfizer-BioNTech: wait 6 months (minimum interval 2 months) after your first 2025–2026 dose

  • Moderna mNexspike: wait 6 months (minimum interval 3 months) after your first 2025–2026 dose

I know that some people have reactions to these shots, and rarely some people have had serious adverse events. It’s good to make your own decisions in consultation with your doc. I personally got Novavax in the late fall, and consistent with clinical trials and my patients’ anecdotal experiences, this brand is usually much better tolerated for people who react. I’ve decided that I like the idea of hedging my bets with the IgG4 antibody class switching that seems to be occurring with repeated mRNA shots. This IgG4 deep dive was a bear to write, but many told me it was helpful and far exceeded RFK Jr’s grasp on vaccinology.

FYI - no Covid illness this fall, winter, or early spring since my shot.

How well did this season’s Covid shots work?

Now, if you’re looking to the CDC to help answer the question of how well this season’s Covid shots performed — bad news. It seems like our ideology-driven federal leadership wants to bend yet another answer to a fools’ gold standard. I don’t feel like writing about this, so I’m just going to quote a source that had the stomach for it (CIDRAP):

Today, the Washington Post reported that Jay Bhattacharya, MD, PhD, director of the NIH and former acting director of the CDC, delayed the publication of a study in Morbidity and Mortality Weekly Report showing the COVID-19 vaccine significantly reduced the chances of hospitalization and emergency department (ED) visits last season. The report was to be published on March 19. According to anonymous sources, Bhattacharya had issues with the methodology, but sources told the newspaper the methodology is the same one used to assess the efficacy of flu vaccines. According to the unpublished report, which was shared with the Post, from September to December last year, people who had received an updated seasonal COVID vaccine reduced their likelihood of ED and urgent care visits by 50% and the likelihood of COVID-associated hospitalizations by 55%.

That’s not how you run the show. That’s how you get vulnerable people to leave during intermission, or unnecessarily die. And it’s not just this guy. We can assume this administration will keep hacking away at science and medicine whenever they can. From STAT News:

After a courtroom defeat, Trump administration health officials have revised the governing documents for a key federal vaccine panel to broaden its membership, increase its focus on potential harms of vaccines, and empower allies of health secretary Robert F. Kennedy Jr.

The new charter for the Advisory Committee on Immunization Practices echoes language frequently used by anti-vaccine advocates, suggesting for example that ACIP should study “the cumulative effects of vaccines and their constituent components.”

Stop. Just stop.

From the Rosalyn Carter-founded vaccinateyourfamily.org:

These ACIP charter changes are yet another example of efforts aimed at creating confusion around vaccines, which ultimately will reduce our community protection against dangerous vaccine-preventable diseases.

Nonetheless, we do have some estimates of current vaccine effectiveness from other sources:

• A VA system preprint study (not yet peer-reviewed) found the updated Pfizer LP.8.1 vaccine was roughly 57% effective against emergency room/urgent care visits and 54% against outpatient visits about four weeks post-vaccination. This performance mirrors last season's KP.2 vaccine. Further duration of protection was not assessed.

• Looking back to last year, a CDC study recently published in JAMA Network Open found the 2024-25 COVID vaccine was 40% effective against hospitalization and 79% effective against ending up on a ventilator or dying among 8,493 hospitalized adults. Notably, protection held steady through six months — and actually trended higher over time. Effectiveness scaled with severity: 46% against supplemental oxygen need, 60% against ICU admission, and 79% against death. Results were comparable across variants. Bottom line was this: while the vaccine offers modest protection against hospitalization, it provides robust defense against the worst outcomes.

• During the same time, Covid was associated with an estimated 290,000 to 450,000 American hospitalizations (could have cut this by up to 40%) and 34,000 to 53,000 American deaths (could have cut this by up to 80%). How is that acceptable?

New review circles back on metformin to prevent long Covid

I’ve written about this several times, and though metformin treatment during Covid infections is quite rarely done, it remains a fairly evidence-based option. In the esteemed journal Clinical Infectious Diseases we find a recent review of multiple randomized trials and real-world data finding that metformin — a cheap, widely available “diabetes drug” — reduces the risk of developing long Covid by roughly 40–60% when started during acute infection. The effect has been replicated across several studies. Metformin may also accelerate viral clearance.

Importantly, this is for prevention of long Covid, not treatment of established long Covid. Treatment doesn’t seem to work. But the medication fluvoxamine was recently shown to help with Long Covid fatigue in the same study showing no benefit for metformin as treatment.

Metformin’s low cost, strong safety profile, and compatibility with Paxlovid make it a serious candidate for outpatient acute Covid management, especially in 2026 as we see long Covid much more commonly than death/severe outcomes in family practice. And yet consensus treatment guidelines want even more evidence before making metformin mainstream. GI side effects can be intense for some. Not me x 1.

Previous deeper dive on cool, overlooked, broad metformin possibilities here.

Another antiviral, finally?

In mid-June the FDA is supposed to get around to maybe approving ensitrelvir, an antiviral with proven benefit for treatment of Covid, and also for post-exposure prophylaxis to prevent transmission in households. I’ll do a special post about this when/if approved here in the US (like it already is in other countries such as Japan). The delay here has been ridiculous. This medication has some nice advantages when compared with Paxlovid.

Taking antibiotics for Covid

A lot of people don’t test when sick, and some ask quickly for antibiotics like Z-paks. A UCSF-led Nature Microbiology study of 1,164 hospitalized Covid patients found that azithromycin rapidly disrupted the upper airway microbiome — depleting commensal bacteria while enriching potentially dangerous pathogens like Staphylococcus and Klebsiella. Antibiotic resistance genes activated within one day of treatment and remained elevated 7–10 days after stopping the drug. The antibiotic showed no anti-inflammatory effect and no impact on viral load. Bottom line: azithromycin caused measurable, persistent microbiome harm with zero clinical benefit for COVID-19. Another study recently showed that microbiome disruptions from antibiotics can last for years. Take ‘em only when you need ‘em, in consultation with your doc.

Stimulant prescribing doubled during Covid Pandemic

A Canadian study in CMAJ found new stimulant prescriptions, mostly for ADHD/ADD, doubled among Ontario adults during the pandemic, with young adults and women driving the surge. Prescribing shifted toward nurse practitioners and shorter diagnostic timelines. Researchers flag concerns about potential misdiagnosis fueled by social media and telehealth. But…

One driver the authors didn’t consider that I intuitively do: long Covid brain fog can be nearly indistinguishable from ADD symptomatically, and the demographics of the prescribing surge — younger adults, women, compressed diagnostic timelines — overlap substantially with long Covid’s profile. Some portion of these “new ADD” cases may actually represent unrecognized post-infectious cognitive impairment. I see brain fog and memory problems more frequently in my patients since the pandemic… although anecdotally/hopefully I think this is decreasing in frequency with time. I’ll have to email the study authors sometime. Will let you know what they think!

Influenza seems to have been worse than Covid this year

For the second straight winter, flu has outpaced Covid as the dominant respiratory threat. Part of this stems from the fact that we’ve had two unusually nasty flu strains in a row, including H3N2 this season. Don’t forget however, that Covid circulates year-round, including summer spikes. But even in Covidlandia we can admit that widespread immunity from infection or vaccination has blunted Covid’s severity. However, experts caution that these unusual flu strains also reduced flu vaccine effectiveness, and it's too soon to declare a lasting new normal.

Maybe we will still have Covid shots next year?

With the staggering incompetence and chaos in whatever remains of the once world class ACIP, it is difficult to report timely and consistently on what’s coming down the pike. But it was recently reported that the federal vaccine advisory panel (ACIP) has shelved efforts to end recommendations for mRNA Covid shots, reportedly amid concerns about political fallout ahead of midterm elections.

That’s good, and shows there is still power in public opinion, criticism from medical professionals, and tanking approval ratings. You may recall another win: Court Order Presses Pause on New ACIP Committee and Changes to Childhood Vaccination Schedule. Also, Dr. Robert Malone was coming hard for Covid vaccines with RFK Jr, so it’s good news for vaccine choice and availability that he recently quit.

Why might we still have a choice/opportunity to get Covid vaccines this fall? Have the new advisors caught up with the previously fired experts after some diligent studying?

The move away from direct criticism of the vaccines comes as Republicans face plummeting polling numbers ahead of the November midterm elections, along with continuous criticism from medical professionals.

Dr. Jake Scott wrote a very convincing piece about just how compromised the ACIP really is, and how it serves a political project more than following any notion of evidence. Quote:

…the institutions responsible for vaccine policy have been reorganized around conclusions rather than evidence, and the consequences are already visible in case counts, in professional society withdrawals, and now in a 45-page federal court opinion finding that the process was likely illegal.

It’s all quite unacceptable, just like annihilating entire civilizations, and embarrassing the Easter Bunny.

Covid photo gallery

An interesting collection of Covid related radiology images can be found here. Just two to break up the reading:

Is there an evidence based treatment for loss of smell?

One of the great promises of AI is answering patient questions at the point of care while leveraging professional training, clinical experience, and millions of clinical trials that a single human mind cannot command. One patient asked me about ideas to help with his persistent loss of smell. The clinically validated AI tool from OpenEvidence (I think it’s just available to clinicians?) helped me cook up an outside-the-box regimen for him to try. It is reportedly helpful up to 90% of the time, if not totally curative. We’ll see.

Sperm and Covid

One way to get men of a certain age to care about a topic can be its effect on sperm and fertility. A Vaccine umbrella review of 14 studies found that Covid infection is associated with reduced sperm count, motility, and semen quality in men. These effects persist 90+ days after recovery. Female fertility and assisted reproduction outcomes appear largely unaffected. Critically, Covid vaccination showed no impact on fertility in either sex. Evidence quality is rated as low, and more longitudinal research is needed... but this builds on previous similar evidence.

Crummy Paxlovid/antiviral prescribing (and taking) rates

Despite being at highest risk for severe Covid, only 16–38% of Americans aged 65+ received antiviral prescriptions — with rates dropping significantly during low-incidence periods like we are in now. A weird psychology. Wider provider and patient education about antivirals and vaccination remains urgently needed in this population. Here we are. Here you are, still hanging in there :) One more to go…

I like to wear a good mask when in the ER, pharmacy, or hospital - especially during waves

Turns out that’s not irrational. A Kirby Institute study sampling air in a Sydney hospital's ER and ICU during two Covid waves (November 2023–July 2024) found SARS-CoV-2 RNA in 40% of aerosol samples — with 80% positive samples in the ER. Crucially, positive samples appeared up to a week before a formal outbreak declaration, suggesting aerosol monitoring could enable earlier detection like we do with sewage analysis. Authors conclude that ventilation alone is insufficient; HEPA filtration and respiratory protection (i.e. masking) are essential during outbreaks. Or whenever you feel like it in Covidlandia.

Take home

I hope you found value in this edition! It does take time and effort collecting, curating, and presenting this stuff so I hope it helps. This spring edition was made easier by listening to Jonny Meister’s curation of the blues on WXPN here in Philly on Saturday night as I typed away. The blues are worth a therapeutic listen in these troubled times. Meister’s program has been going strong for more than 40 years. My Dad used to fill our house with a mess of blues on many a Saturday night. But I digress:

Take good care out there :)

Speak truth to power and patients.

O Covidlandia, we stand on guard for thee.

I returned from vacation to my work as a doctor on Monday. That cognitive gear shift was hard enough, not to mention the staggering 300+ test results, patient messages, calls, emails, and forms to sign against the dark backdrop of a United States president talking shit about committing unthinkable atrocities and violence against civilians in our name when the whole world including us should be looking up at the moon in solidarity and wonder.

Each of the 300+ items I had to catch up on while seeing patients in the office were not really items at all. Each was connected to a human life, a person requesting help and attention, someone feeling vibrant or lonely or scared or sick or frustrated or all of the above. I stayed at work late last night for one such old soul, and old friend really. While I was away her health took a sudden turn for the worse, and she went from home to the hospital and then home again, but now on hospice — all within three breathtaking days. Incredulously, I read through her hospital records, the expert consultations, the dismal final consensus. I called her and we discussed her situation. I thought of a few long shot ideas that might right the ship, that might save her life. We got some emergency consultations, tried one last strategy still consistent with her wishes, but tonight as I called her before leaving work she said a stoic, courageous, incredible goodbye. Family is hurriedly traveling to her bedside.

I’ve known this patient for two decades I think. In saying my goodbye I used a word I almost never do in a professional capacity. Love. She responded that when the urgently consulted specialist asked her why her primary had requested one more try, she replied because he loves me.

I don’t think I’ll ever forget how that feels. The raw emotion. The burst of sadness. The indomitable light pouring from her very being as she spoke with unfiltered gratitude about the radiant life she has lived.

When I think about the hundreds of children already killed, the thousands of people, the millions more being threatened with deadlines and timelines that strike an unholy parallel with this single beautiful human life on the brink, it is almost too much to bear. Numbers are numbing, threats are theater, bombs and blasts drown out the wailing misery of human life destroyed.

I want competent peacemakers intent upon making peace on this fragile planet.

I want righteous leadership intent upon making civilization right.

I want every life departing this world to have been respected, loved, cared for, appreciated.

Have you ever wished you could save the world?

If not, what about saving a single beloved person?

Often we cannot.

And yet we can always choose to dignify human life.

To be reverent in the reckoning with it.

Those traveling moments, it turns out, might be among the better things we can embrace and encourage for our brain health and longevity. And there is supporting data. I am often asked about this or that supplement, but few people ask me if they should travel more. And yet I think we know intuitively that there is something special about scrambling our routines for the joys and Griswold pains of the holiday roads.

What the research actually says

A prospective cohort study following over 6,700 older adults found that those with at least one tourism experience in the prior two years had a 31% lower risk of cognitive impairment and a 59% lower risk of dementia compared to those who hadn’t traveled even after adjusting for age, income, education, and baseline health. There was a dose-response relationship, meaning more travel, more benefit.

A separate cohort study found that older travelers had a 27% lower all-cause mortality risk than non-travelers.

And the Multiple Risk Factor Intervention Trial — a rigorous nine-year follow-up of over 12,000 middle-aged men — found that more frequent vacationers had a 17% lower all-cause mortality risk and a striking 32% lower cardiovascular mortality risk.

As always, we need to consider the quality of the evidence. These are observational studies. It would be farcical to design a “gold standard” randomized placebo controlled trial of people traveling versus pseudotraveling! Also, healthier, wealthier people travel more, and residual confounding is real. But the associations are consistent, biologically plausible, and the effect sizes are comparable to other things we routinely encourage like exercise, social connection, and cognitively stimulating activity.

Travel is a full-spectrum cognitive workout

Reading about the Resplendent Quetzal bird versus tiptoeing around a cloud forest trying to spot one are categorically different experiences for the brain. Maybe we find an armadillo trundling around instead. Novel environments force active engagement. We are navigating, problem-solving, updating our mental maps, and doing it without the comfortable autopilot of routine. This is exactly the kind of stimulation associated with building cognitive reserve: the brain’s accumulated resilience against the inevitable wear of aging.

We can read a new menu in an unfamiliar language and then explode into a multidimensional experience of mouth, nose, and brain while eating a new food, a new cuisine. We can struggle to figure out a complex transit system. We might navigate a night market by smells and shadows. These are not trivial mental exercises. They are neurologically compelling ones.

The body comes along for the ride

We don’t emphasize this enough: travel makes us walk. Museums extract a surprising mileage from us. Even a modest hike to a viewpoint, or a morning spent wandering a neighborhood instead of a cubicle-lined hellscape, puts us on our feet in ways that ordinary days at home/work often don’t. The physical activity guidelines have moderate-to-strong evidence for physical activity improving global cognition and executive function in adults over 50. Travel doesn’t have to mean trekking Patagonia to deliver that benefit. It means getting out of the car though.

Small talk is underrated neuroscience

Striking up a conversation with the woman running the trattoria about where the bread comes from. Asking a fellow traveler on a train platform how their trip has gone. Brief, unexpected social exchanges may feel trivial but they are not — they activate the same social processing networks that sustained loneliness quietly degrades. Another study found that leisure travel was specifically associated with reduced loneliness and fewer depressive symptoms, both of which are established accelerants of cognitive decline. Meeting people unlike ourselves, even briefly, turns out to be protective. Uplifting.

Adversity as cognitive inoculation

The missed train. The weather that reversed our plans. The AirBnB that wasn’t what the photos suggested. We tend to experience these as pure inconvenience, but there’s something worth noting. Navigating real-world disruption with limited resources is a form of stress inoculation. Leisure activities reduce “allostatic load” (the cumulative physiological toll of chronic stress) through active coping, positive affect, and recovery. The problem we solved in a foreign city is not just a story for later. It’s a deposit into a resilience account. Augmented confidence in dealing with adversity.

On awe

There is something that happens when we stand at the edge of a canyon, or watch a storm come across a lake, or find ourselves suddenly alone on a hiking trail surrounded by nothing but trees, bird song, and nature. Researchers are increasingly taking awe seriously as a distinct positive emotion with measurable effects on inflammation markers, prosocial behavior, and subjective wellbeing. Travel, particularly in natural settings, is one of the most reliable awe-delivery systems we have. It is not frivolous, and it can be quite economical to deliver.

A note on access and ability

Travel does not require a passport, a long-haul flight, or a vigorous constitution. A national park a couple hours from home. A weekend in a town or city we’ve never explored. A river trail we’ve driven past for years. The cognitive benefits of novel environments, physical engagement, and social exposure do not scale with airline miles. There is emerging evidence that lower-socioeconomic-status individuals may derive relatively greater cognitive benefit from leisure engagement than their wealthier counterparts, likely because baseline stimulation is lower and the contrast effect is larger. You don’t need to go far. You need to go somewhere genuinely new to you.

A prescription?

No clinical guidelines yet specifically recommend travel for longevity or dementia prevention that I know of. But travel satisfies nearly every criterion that guidelines do endorse: physical activity, cognitive stimulation, social engagement, stress recovery, exposure to novelty. It is, in that sense, a multidomain lifestyle intervention disguised as something people actually want to do instead of kale, cold plunges, and hours of yoga.

(I actually like kale, avoid cold plunges, and sometimes scream in pain while doing yoga.)

The risks are real and worth knowing

I’ve made a case for travel as medicine. But medicine has side effects and risks, too. As always, pre-travel medical consultation matters, especially for older adults or those with chronic illness. Cardiovascular events account for nearly half of travel-related deaths among Americans abroad. A conversation with your doctor before a big trip is not excessive caution, it’s prudence.

There is genuine evidence that vacation travel and holiday periods carry elevated cardiovascular risk, particularly for vulnerable patients. A study of 92 patients who suffered a heart attack while traveling abroad found that 21% of those events occurred in the first two days of the trip (the period of maximal disruption, adjustment, and exertion).

Compared to hospitalized controls, vacation heart attack was significantly more common among those traveling by car and those staying in less comfortable accommodations. Maybe it’s worth a splurge to stay a notch above the youth hostel. The physical and emotional demands of early travel like long drives, unfamiliar beds, and logistical stress appear to be the culprit, not the vacation itself. It makes me want to have a chill first day after researching this, instead of only slowing down the pace towards the end like I usually do.

Holiday periods more broadly carry elevated risk. A Swedish nationwide study of over 283,000 myocardial infarctions found that Christmas Eve carried the highest heart risk of any day in the calendar year, and both Christmas and midsummer holidays showed significantly elevated rates, with the greatest excess risk among patients over 75, those with diabetes, and those with prior coronary artery disease. And a recent South Korean study found that out-of-hospital cardiac arrest incidence was 9% higher on the first weekday after a holiday period compared to ordinary weekdays, driven by sympathetic nervous system reactivation, catecholamine release, and the abrupt return to structured demand after rest. Coming home from a good vacation/holiday can be rough in many ways.

Air travel adds its own specific risk profile. The absolute incidence of symptomatic venous thromboembolism in the month following a flight over four hours is roughly 1 in 4,600 flights, rising 18% for each additional two hours aloft. And high risk passengers can face up to a 5% risk of lower extremity blood clots per long-haul flight.

And then there are the slippery rocks that can injure toes. It’s good to have a buddy after these injurious situations, and perhaps keep on hiking if it doesn’t hurt!

None of this argues against travel. It argues for preparation and being a little cautious. The contributing factors like overindulgence in food and alcohol, sleep disruption, dehydration, excessive exertion beyond usual levels, and delayed care-seeking are largely modifiable.

For patients with significant cardiovascular risk factors, known coronary artery disease, diabetes, prior blood clots, or other specific concerns, a pre-travel conversation with your physician is not a boring formality. I often pull up the CDC Travelers’ Health website and go country by country with people regarding specific regional risks, vaccines, precautions, and then personal considerations.

The end

Well, if nothing else, I hope this post gives you one more reason to consider a little trip or a grand adventure. We should expect adversity, disappointment, elation, and transcendence. Less people discuss the experience in terms of longevity medicine or cognitive exercise, but travel obviously touches both realms in a multidimensional way.

What was one of the best trips of your life?

Any hopes or dreams for future travel?

Are you a meticulous planner, or a spontaneous seeker?

Did you know that armadillos are this cute?! That they have little cone-shaped ears?

In last week’s post I presented the newest way primary care docs might have evidence-based, nuanced conversations about heart risk, statin medications, and implicitly the amount of Kerrygold Irish butter to be slathered on warm Irish soda bread with currants. The former can be informed by the PREVENT calculators, while the latter should be informed by the quality of the soda bread and whether ‘tis St. Patty’s Day.

So in these two adjacent posts I’m going to explain BAC testing that can be done with mammograms and a little AI help, while tying this in with the PREVENT calculator. Should women be doing both?

But first, a very important aside. This photo was taken while this year’s soda bread was cooling. It is not a great photo. But I think we can all agree she still looks grand, worthy of eating with tea and abandon and butter:

And I certainly did that.

BAC

Here is a striking fact to open with: in a 2019 American Heart Association survey, women were more than twice as likely to name breast cancer as the leading cause of death in women, even though heart disease claims nearly seven times more women than breast cancer does. Compared to a decade earlier, women were 64% less likely to correctly identify cardiovascular disease as their greatest common threat. Education campaigns, it seems, have been losing ground.

This perception gap matters, because it explains something else: nearly 70% of American women 45 and older are up to date with their mammograms, while fewer than 40% know their own cholesterol level.

So what if there were a way to look under the hood at both issues at the same time, towards bridging the gap between breast cancer screening and cardiovascular risk assessment?

The railroad tracks

When a radiologist reads a mammogram, they look for possible signs of cancer. But breast tissue is threaded with small muscular arteries, and those arteries can calcify. The pattern, when present, is distinctive —> thin parallel lines tracing the vessel wall, like a miniature railroad track, or eventually white conduits shaped like spaghetti noodles. That calcified information has been there for years, but rarely was it commented upon or standardized for interpretation.

Stay on the left side of this graphic borrowed from the study. You can trace the tracks in black and white images. On the right is the “artificial intelligence model heat map” calculating the score:

These findings are called breast arterial calcification, or BAC, and are quantifiable.

The case for paying attention was substantially strengthened just last week by a large study published in the European Heart Journal, led by researchers at Emory University and validated at the Mayo Clinic. Across 124K women undergoing routine screening mammography (a racially diverse cohort with a median seven years of follow-up) an AI model was used to precisely measure BAC burden. The findings were hard to dismiss.

BAC was detected in 16 to 21% of women. Those with the highest BAC burden had a major adverse cardiovascular event risk incidence more than eight times higher than those with no BAC at all. Namely, this risk rose from about 6 events per 1,000 person-years in the zero-BAC group to nearly 49 per 1,000 person-years in the severe-BAC group.

Major adverse cardiovascular events included a composite of heart attack, stroke, heart failure, and all-cause death. This is abbreviated as MACE by the way. That increased risk held across every individual component of the composite.

And crucially, after the researchers adjusted for PREVENT scores (see last week’s post), BAC remained an independent predictor of outcomes. Women with severe BAC had about 3X the risk of any major cardiovascular event compared to those with zero BAC, even with PREVENT already in the model. 8X versus 3X isn't a contradiction: the hazard ratios reflect what BAC adds after age, blood pressure, cholesterol, and kidney function are already accounted for — which is precisely the point of how this BAC testing might add value to your heart conversation with your primary care doc.

And finally, an aside for the doctors and highly motivated readers in the house… ok to just get the gist of this:

• Mild BAC (>0-10 mm²): adjusted HR 1.28 —> 28% higher risk

• Moderate BAC (>10-25 mm²): adjusted HR 1.79 —> 79% higher risk

• Severe BAC (>25 mm²): adjusted HR 2.80 —> almost 3X risk

A quick detour into pathology

To understand what BAC is telling us, we need to understand what it isn’t telling us. A breast artery scan is not the same as a coronary artery scan.

The calcification that builds up in coronary arteries (the kind a coronary artery calcium (CAC) scan detects) forms inside atherosclerotic plaque in the inner lining of the vessel wall. It is the end product of cholesterol-driven, inflammatory, plaque-building disease. It narrows the channel blood flows through. This is called intimal atherosclerosis.

Mammography, by contrast, images the medium-sized muscular arteries threading through breast tissue. These vessels are far less susceptible to classic plaque formation. Instead, they are prone to calcification of the middle muscular layer. This is called medial atherosclerosis, and is a fundamentally different process driven less by cholesterol deposits and more by metabolic and blood pressure stress accumulating in the arterial wall over years. Classically, this does not lead to “blockages.” In that technical sense it has historically been called “benign.”

But benign is doing a lot of work in that sentence. Medial calcification is strongly correlated with vascular aging, and is accelerated by diabetes, hypertension, chronic kidney disease, and obesity. These are also the conditions most powerfully linked to heart failure.

This is why BAC is not the same as a CAC score, and why it shouldn’t be treated as one. The Emory study found, for instance, that smoking (a powerful driver of atherosclerosis) was not associated with BAC. That finding would make no sense if BAC were simply a proxy for coronary plaque. But it makes good sense if BAC reflects a distinct but related pathway of vascular injury driven by metabolic and hemodynamic stress rather than by tobacco-induced endothelial damage we see in the coronary arteries.

The two lanes BAC opens

This intimal versus medial pathological distinction has a practical implication that Dr. Lori Daniels, cardiologist at UC San Diego, laid out in a helpful accompanying editorial last week in the same journal. She proposes that BAC detection should trigger not one clinical response but two distinct, parallel pathways.

• The first is the coronary artery disease “ASCVD prevention lane.” BAC should prompt formal PREVENT score calculation with a primary care doc if one hasn’t been done, a fasting lipid panel, and maybe LDL-targeted therapy like statins. If the PREVENT score is in the borderline range (roughly 4.5–7.5%), a high BAC finding is sufficient grounds to steer the conversation toward statin initiation, using the same shared decision-making framework that would apply to an elevated coronary artery calcium score. Blood pressure should be optimized to the new 2025 ACC/AHA target of below 130/80 mmHg. Smoking cessation if applicable is always advisable. If the cholesterol picture is borderline and people still want more variables, more data like blood tests for high-sensitivity CRP and Lp(a), or a formal CAC scan can help further define risk. The practical goal of this lane is to get the patient onto the guideline-directed lipid and blood pressure therapies that reduce atherosclerotic events.

• The second is the heart failure prevention lane. Because BAC reflects arterial stiffening driven by hypertension, diabetes, chronic kidney disease, and obesity, the women most likely to have significant BAC are also the women most likely to be on a riskier trajectory toward heart failure. Technical terms: heart failure with preserved ejection fraction (HFpEF) is a form of heart failure where the ventricular muscle becomes stiff rather than weak. It disproportionately affects women. So this lane calls for aggressive blood pressure control (the single most modifiable driver of HFpEF), weight management maybe including GLP-1 agonist therapy when indicated, optimization of blood sugar with particular attention to meds like Jardiance and other SGLT2 inhibitors (which carry a specific HF-reduction benefit independent of blood sugar control), and kidney function monitoring. Probably should check an echocardiogram looking for early diastolic dysfunction in which a stiff heart has trouble relaxing and filling with blood in between beats.

In the discussion it is suggested that BAC scores should not necessarily or reflexively trigger coronary imaging, panic, and lots more cardiac testing in the absence of symptoms. It should trigger a thoughtful, structured primary care visit that works both lanes in parallel. This is going to take a lot of time and talking! I’m going to have to schedule some follow up telemedicine sessions for this sort of thing.

Brief intermission

I’m going to drop a few easier posts soon, promise! But this stuff is timely and I wanted to grind it out for you.

Can you find a little paper critter? The one my then 8 year-old daughter made for me to take into my office years ago?

Back to the show…. almost there!

Where BAC fits in the PREVENT framework

If you read my earlier post on PREVENT, you’ll remember that the calculator produces a number — a 10-year risk percentage — but guidelines are explicit that the number is a starting point, not a verdict. For patients in the borderline zone, clinicians are supposed to layer on “risk-enhancing factors” like family history of premature heart disease, elevated Lp(a), high-sensitivity CRP, coronary artery calcium score.

Family history, notably, does not go into the PREVENT calculator at all. This blind spot in all of these calculators has always troubled me. Family history was not something reliably recorded and therefore easily mined from the datasets informing these calculators. Therefore family history has been banished to this second tier, applied with clinical judgment rather than mathematics. A woman with a PREVENT score of 5% but a father who had his first heart attack at 52 is not the same as a woman with a PREVENT score of 5% and no family history — but the number alone cannot tell us that.

BAC adds to this same second tier. It is objective anatomical evidence, not a calculated risk estimate. Like a CAC score, it tells us what has already been happening inside a patient’s vessels. And so in some ways it is both future risk and present evidence.

Even among women PREVENT had already classified as low risk, those with moderate or severe BAC had significantly worse outcomes. The calculator said they were safe. The arteries disagreed.

40 million mammograms a year

One of the most underappreciated arguments for BAC reporting is logistical rather than clinical.

This Emory study noted that complete PREVENT data was available for only 6-28% of study participants. That gap is a reflection of reality. Running the PREVENT calculator requires having a recent cholesterol panel, blood pressure, BMI, kidney function, HbA1c, and medication history all simultaneously available and documented. For many women, especially those not yet engaged in cardiovascular care, that data doesn’t exist in a usable form. Evidence of this showed up even in the comments after my last post, showing real world difficulties plugging and playing.

Screening mammography, by contrast, is performed on approximately 40 million American women annually. No additional appointment, no blood draw, no extra radiation. Tempting.

Should we pay for the add-on?

Some radiology practices are now charging patients out-of-pocket, I think it’s around $100–$150 to use AI software for formal BAC scoring of their existing mammogram.

The logical answer, for now, is not recommending this as a universal standard. The evidence base, while compelling and rapidly growing, is still largely retrospective in nature. Professional society guidelines have not yet established standardized thresholds for BAC reporting or clinical management protocols. Critically, a negative BAC finding does not necessarily provide meaningful reassurance. Even in the Emory dataset, most cardiovascular events occurred in women without detectable BAC! A normal result does not override all the other risk factors.

What is reasonable, today? Perhaps finding out how much BAC on the mammogram is visible, bringing that information to a primary care visit framed by one’s PREVENT score, family history, and full cardiometabolic risk profile. In women with moderate or severe BAC, especially those in PREVENT’s borderline zone or with family history, that finding really deserves a clinical response and an elevation of a patient’s game.

The bottom line

PREVENT was a major advance in cardiovascular risk assessment. It removed race coefficients. It added heart failure. It extended the age range down to 30, and it could look out 30 years ahead. But it still runs on population equations that cannot see inside an individual’s vessels. It cannot capture the arterial stiffness accumulating in a 47-year-old woman with well-controlled diabetes and borderline blood pressure. It cannot capture the vascular aging visible on her routine mammogram. BAC can.

It would seem there are two types of women who might pay for this sort of additional information on mammograms (if we assume people can afford this, which these days is a dicey assumption):

• Women who want a more complete, non-invasive risk assessment in addition to everything else we can already measure, discuss, and game out.

• Women who are not convinced they need aggressive clinical management by their doctors or aggressive personal lifestyle changes to address modifiable risk factors — but are willing to be convinced otherwise.

I’ll give the last word to that editorialist Dr. Lori Daniels:

Both lanes begin with primary care-driven identification and optimization of conventional risk factors, with selective cardiology referral for higher-risk individuals. This approach avoids reflex coronary imaging while still enabling targeted testing when clinically indicated. It also acknowledges a practical reality highlighted by the authors: complete data to calculate PREVENT were available in only a minority of patients. Screening mammography, in contrast, is already ubiquitous, making BAC a scalable, opportunistic prompt to initiate formal CVD risk assessment rather than a replacement for it.

Best of luck with your decisions and discussions :)

I do know this: when in doubt, if presented with an irresistible Irish soda bread, especially on St. Patrick’s Day, I personally will be slathering on the butter.

Ah, the good old days… but the good old days weren't always good.

The formula worked reasonably well, but it had blind spots. It was built from data collected primarily on White and Black Americans. It didn’t account for kidney disease. It didn’t factor in diabetes control. It couldn’t predict heart failure. And it was known to overestimate risk in some groups and underestimate it in others — by as much as 60 to 90% in some subsequent studies! Millions of people were either overtreated or undertreated as a result.

That formula — the Pooled Cohort Equations — has now been mostly replaced.

Meet PREVENT

In 2023, the American Heart Association introduced a new risk calculator called PREVENT, which stands for Predicting Risk of Cardiovascular Disease EVENTs. Kind of cheating to get that acronym, right? But it is admittedly more memorable than “PROCDE,” which might be confused with proctology guidelines.

I previously dove into PREVENT specifically as it pertained to blood pressure when the new guidelines were released last year (and it was one of my most read posts!), but I want to revisit it now with even more clarity. I also want to explore what the calculator means for optimizing cholesterol, and who should consider statins.

{Addendum April 2026 - if we have had a Coronary Artery Calcium scan (CAC), we can get even more valuable predictive information by plugging that score + more into the MESA Calculator. }

PREVENT was built from health data on more than 6.5 million Americans (without known heart disease) from diverse racial, ethnic, socioeconomic, and geographic backgrounds. It removes race as a biological input — because race is more of a social construct, not a consistent biological variable — and replaces it with optional data about your zip code and community environment which track more directly with risk. It works for adults as young as 30. And critically, it looks not just ten years into the future, but thirty.

When I plug my data into the calculator, statistically speaking I have a 1.1% risk of serious cardiovascular disease in the next 10 years, and 9.5% risk over 30 years. Given stress levels and other intangibles including family history, my guess is that my true risk is a bit north of that.

How useful are these numbers?

In 2025, like I wrote about, the American College of Cardiology and American Heart Association updated the national guidelines for high blood pressure management and formally embedded PREVENT as the recommended risk tool. This wasn’t a minor tweak. It changed who gets treated, when they get treated, and how doctors think about the relationship between blood pressure, cholesterol, kidney function, and heart disease as a unified system.

This tool is designed for primary prevention discussions, i.e. conversations with your doctor before events like heart attacks, heart disease, and heart failure have already developed.

It cannot be validly applied to patients who have already experienced a cardiac event. Those patients — in what clinicians call secondary prevention — are already considered high-risk by definition, and their treatment decisions are governed by a different set of guidelines and tools. In secondary prevention, the focus is on preventing a recurrent event rather than a first one. And yet the broader philosophy of managing blood pressure, cholesterol, blood sugar, and kidney health together remains just as important. The overlaps are many, but the treatment decisions become much easier in secondary prevention — essentially, just do it.

Ok, let’s run through primary prevention first.

Read more

It happened again. I had every intention of getting to bed by 11 PM. The plan was to sleep deeply, awash in bodily repair mechanisms, my brain’s glymphatic system running like the dishwasher downstairs. Upon awakening I would feel great, rested, like a fully charged lithium ion battery, with reduced risks of cardiovascular disease, cancer, and dementia. I would even feel virtuous.

But instead I saw patients until 5:30 PM. I went shopping for groceries on the way home. I picked up my daughter from the train station in Philly. I cooked a healthy (late) dinner and ate it with my family sitting around the sacred kitchen table. I cleaned up the kitchen, tidied up the house, dropped an 80 point Scrabble/Crossplay word on some unfortunate online opponent name Ravi, stubbed my formerly improving ingrowing big toenail, ate the last brownie in the house (yes, I did it!), “worked out” for 20 minutes, took a shower, brushed and flossed, and plopped into bed around half past midnight. I did fall asleep quickly, as usual.

Is it worth squeezing more out of my days in this relatively healthy middle age, risking some future calamity and enduring my present sleep guilt? Or should I restructure everything around executing that 11 PM bedtime commandment that would deliver me unto ~8 hours of precious sleep?

What do you do? What did you do? What are your priorities? Want to examine the whole waking-hours-of-life-versus-investment-in-unconscious-sleep thing a bit more critically with me? With no guilt? More data?

We can tell a lot about a culture by what it makes us feel guilty about. Right now, ours has decided that sleeping 6.5 hours instead of 8 is practically self-harm. A kind of reverse Puritan work ethic. In the Puritan worldview, time was a precious resource to be used for labor and the glory of God; therefore, sleep was restricted to only what was strictly necessary for health. That meant more like six hours, and at most seven to eight on the weekends.

Open any wellness/health education feed and the message is relentless: seven to nine hours are the sacred numbers, non-negotiable, the biological floor beneath which our immune systems crumble, our hearts strain, and our brains quietly entangle. Matthew Walker’s Why We Sleep (despite being extensively critiqued for scientific overreach) helped lodge this terror firmly in the cultural imagination. Walker was featured on Rogan, Attia, and became a Google sleep scientist. But that critique of his work (and this whole subject) dropped some pretty smart bombs:

Now, looking at Figure 1, we can see that people who reported sleeping just below 7 hours a night had the lowest mortality. If we interpret this in light of the last quote (people over-reporting the amount of sleep they get by about 48 minutes), people who have the lowest mortality (might) actually sleep 6 hours a night.

If you feel crummy about your poor sleep, and feel bad about your mission impossible to improve it — dive into that critique, if only for the psychological balm.

I’m not advocating 6 hours of sleep for all, but here’s the contrarian case: the actual data are far more forgiving than the rhetoric, the difference between 6.5 and 7.5 hours is clinically mild for most healthy adults, and when we do the arithmetic on waking hours, the wellness gospel may be asking us to trade nearly two years of our most capable, conscious life for a mortality benefit that is, at best, modest and, at worst, possibly statistical noise.

Let’s look at the numbers. But first, please know that the point of this post is mostly to justify my own failure to consistently hit the golden seven hours I really want. Also, people are individuals, and their sleep needs vary from person to person. Also, some people really struggle with insomnia, and I hear that in the office literally every day over the past 22 years of seeing patients. So please do not read this as a parallel gospel of sleep according to me!

What the major studies actually show

Most of the evidence I could find out there showed “U” and “J” shaped curves when plotting the relationship between mortality and sleep duration; i.e. there were problems at both extremes of sleep.

J-shaped curves. I’m going to borrow this excellent table from a meta-analysis of prospective cohort studies published in the journal Scientific Reports, presented as a screenshot with some important interpretations to follow:

That hockey stick is pretty reassuring to me. Here is how to read this. The graph is saying that, in these pooled studies, the lowest risk of dying (from any cause) is around ~7 hours of sleep per night, and risk goes up if you sleep much less or much more.​

• The X axis (left to right) is hours of sleep at night.​

• The Y axis (up and down) is relative risk of death.​

• “Relative risk = 1.0” is the baseline (the reference point), and numbers above 1.0 mean higher risk than that baseline.​

What the J-curve/hockey stick shape means:

• The solid black curve makes a “U or J shape” depending on your observational preference: risk is lowest near ~7 hours, a bit higher at shorter sleep, and then rises sharply once you get past roughly 8 hours.​

• The two dashed black lines around the solid line are the study’s “margin of error” (called the 95% confidence interval): the true average could plausibly be anywhere between those dashed lines.​

The practical takeaway:

• Around 7 hours/night is the “sweet spot” in this dataset.​

• But sleeping 5 hours looks roughly similar to sleeping 8 hours, and perhaps even better in this pooled dataset​! Check out that red box if you don’t believe me! It helps reduce any optical illusion.

• But the big message is: the farther you get from ~7 hours—especially toward 9–11 hours—the higher the observed risk.

More U-shaped curves. A different study by Yin et al. (another meta-analysis of prospective cohort studies) quantified the risk per hour of deviation from 7 hours:

• For all-cause mortality, the pooled relative risk (RR) was 1.06 per hour below 7 hours a night (6% higher risk)

• 1.13 RR per hour above 7 hours (13% higher risk). Again, too much sleep often worse than not enough. Causal, or just association? Read on.

More big cohort data. Yang et al. analyzed over 284,000 US adults linked to the National Death Index. Hazard ratios above 1.00 indicate increased risk of the outcome, in this case that outcome is death. So a 1.10 hazard ratio means about 10% higher risk.

Compared to 7-hour sleepers, the hazard ratios for all-cause mortality were:

Here goes that trend again. Eight-hour sleepers had higher all-cause mortality risk than six-hour sleepers. That finding is consistent across other datasets I checked, including the UK Biobank’s 407,500-participant cohort and the 116,632-person PURE study spanning 21 countries.

Sleep duration and cancer, cardiovascular disease, and dementia

The associations reported between not enough sleep and these fearful conditions create a lot of havoc, anxiety, and indeed suffering for people. Here is a summary of the actual evidence though, and I personally found it more reassuring than alarm-clocking .

For cancer, the overall trend is not that strong but certainly confusing. The relationship between sleep duration and cancer risk shows that U-shaped or J-shaped pattern, with both very short (<6 hours) and very long (≥9-10 hours) sleep durations associated with modestly increased cancer risk compared to 7-8 hours, though overall pooled analyses failed to confirm any significant association. Some evidence links long sleep duration to increased colorectal cancer risk and cancer mortality, while short sleep shows variable associations that differ by population, sex, and cancer type. It’s kind of all over the place from my reading. For example, in one study women showed particularly high risk with persistently short sleep patterns, but another showed potentially decreased breast cancer risk. Others showed sleep-cancer mortality relationships were stronger in men than women. Most importantly, proving that interventions to normalize sleep duration actually reduce cancer risk remains to be established through randomized trials.

For cardiovascular disease, the data are also nuanced, but also show that hockey/U/J-shape. Check out this table composed using OpenEvidence and a primary source from Journal of the American College of Cardiology… and then the explanation below:

The table shows that cardiovascular risk increases progressively as sleep duration deviates from the optimal 7 hours per night. The relationship is asymmetric—each hour of sleep above 7 hours increases risk more steeply than each hour below. For example, sleeping 10 hours is associated with approximately 43% higher all-cause mortality and 64% higher stroke risk compared to 7 hours, while sleeping 4 hours is associated with “just” about 19% higher all-cause mortality and 16% higher stroke risk. Thinking about my 6.5 hours lately, for example, I only see about a 5% discount on my luck. Now I will foreshadow the idea of confounding variables with all these studies, so once again I am not trying to be a less-sleep cheerleader. But I am still trying to put my sense of 7-9 hours/night common failure in perspective!

For dementia, the picture is again U‑shaped but with a notable emphasis on midlife short sleep. Large cohort studies show that sleeping ≤6 hours in midlife is associated with roughly a 30% higher hazard of dementia compared with 7 hours, an effect that persists even after adjusting for mental health and lifestyle. Conversely, long sleep (≥9 hours) is also associated with even higher dementia risk, and some evidence suggests that prolonged sleep may be an early marker of neurodegeneration rather than a benign choice. Thus, the lowest risk appears to cluster around 7 hours (again!), with both ends of the distribution carrying a moderate increase in hazard.

The arithmetic of waking hours

Ok, so I’ve convinced myself that failing to consistently hit 7 hours is not the end of the world. I should still shoot for that, but I’m not convinced that anything above 7 (or my close to 6.5) is really worth the sacrifice of consciousness all the time.

Sleeping 6.5 hours instead of 8 hours yields 1.5 extra waking hours per day. Over a year, that’s 547 extra hours — roughly 23 full days.

Over the 30 years between ages 30 and 60, that amounts to nearly 690 additional days of conscious life — close to two full years.

What’s the downside risk?

The epidemiologic evidence suggests that, averaged across entire populations (including the chronically ill, the sleep-apneic, the shift workers, and the economically stressed), I might be gambling that 2 years worth of consciousness (stolen from sleeping 6.5 versus 8 hours) against 0.5 to 1.2 years of statistically added life expectancy if I got ideal sleep. That time is appended to the end of life. You know, those years more likely to involve physical and medical challenges. Granted, the whole game falls apart for the person who actually suffers an early, devastating medical problem statistically attributed to less sleep and loses decades. But that one person happens between one in every 150-250 (NNH) based on these very imperfect studies.

That is not a nihilistic statement about old age. Many people cherish those years profoundly, and through their disabilities and degenerations still consider them among the most fulfilling and happy years. But it is a very different proposition than the usual terror-framing of “sleep less now, automatically die younger in your prime.” The data simply do not support large, consistent, early penalties for healthy adults sleeping 6.5 hours.

I think!

The confounders the headlines ignore

The published hazard ratios are not the clean, causal numbers they appear to be.

Self-report bias. Almost all of this research relies on self-reported sleep duration, which correlates poorly with objective measurement. People tend to overestimate their time asleep.

Reverse causation and confounding. Short sleep clusters tightly with poverty, rotating shift work, psychiatric illness, chronic pain, and caregiving burdens. When researchers adjust aggressively for these variables, associations weaken. The Whitehall II cohort found that after stress adjustment, short sleep’s mortality association significantly attenuated. Long sleep’s excess risk largely disappears when you remove the sick-and-sleeping-more effect.

Quality matters more than duration. The combination of poor quality and short duration is pretty conclusively tied to higher cardiovascular risks. But a consistent, good-quality 6.5 hours almost certainly carries far less risk than a fragmented, anxious 8 hours as we are worrying about those same 8 hours. It should be noted that sleep regularity (going to bed and waking at consistent times) may be a stronger mortality predictor than duration alone.

Who this logic applies to

None of this is an argument for five-hour nights, stimulant dependency, and fighting through daytime exhaustion wearing it as a badge of Puritan productivity.

Below 6 hours, especially in older adults and those with cardiometabolic disease, the causal evidence does find consistent meaningful risk. Severe sleep restriction is genuinely harmful — and if you need an alarm to drag yourself out of bed every morning, feel foggy by early afternoon, or rely on caffeine to remain functional, your 6.5 hours may not be a restorative 6.5 hours. I’m pretty sure I need more like 7, but I’m not hating on myself as much having gone down this rabbit hole of relative risks and scientific uncertainty. The evidence that 6.5 versus 7-8 hours makes a meaningful difference in individual health trajectory is genuinely thin.

A more honest conversation about sleep

The wellness industry has done something subtly cruel with sleep science: it has taken real but modest population-level associations, stripped out the confidence intervals and the confounders, and repackaged them as individual-level moral imperatives. Sleep eight hours or you are damaging yourself! Your mitochondria! Every night under the threshold is a withdrawal from our biological savings account.

This generates anxiety, guilt, and hypervigilance around sleep, all of which, ironically, make sleep worse. Here are some ways I collected that can make it better.

A more compassionate and evidence-aligned message might be this: aim for those classic 7 hours if we can do so without sacrificing exercise, relationships, meaningful work, or the activities that make a life feel well-lived. If we naturally settle at less hours (hopefully above 6), feel genuinely well, and live healthily, the data offer decent reassurance about tradeoffs. If our sleep is fractured, non-restorative, or leaving us functionally impaired regardless of its duration, that warrants attention as a physiological signal.

But the goal was never to maximize hours spent unconscious. Look around. The world is not doing so well, is it? This chaos not only interferes with our peaceful sleep, but it also reminds us that none of us know how the future here on Earth might turn out. This might be as good as it gets. I want to be awake to help make home cooked dinners and eat them with my family around a sacred kitchen table, then drop another 80 point Scrabble/Crossplay word on Ravi, stub my formerly improving ingrowing big toenail again, eat the last brownie in the house (yes, I’ll do it again!), and now it’s time to go to bed.

Good night!

I’m afraid tonight’s SOTU Address will not set the health and science record straight. I read this excellent post entitled What Has Been Lost: One Year Into the Trump 2.0 Administration, by Dr. . I felt her compilation deserved a complementary speech, a rough draft grounded in the facts and references she and others have documented.

So here we go.

~

Members of Congress, I have the high privilege and distinct honor of presenting to you the President of the United States:

My fellow Americans — the right kind of Americans, anyway — the State of the Union, in terms of health and science, is strong. Possibly the strongest, greatest, most potent state any union has ever been in. America is hot, they say. Doctors and scientists are telling me this. Very smart people. Incredible people. Like ICU nurses. Some of them still have jobs. Lives. Many of them can still feed their families.

So let us begin with food, because I want to talk about hunger.

In just one year, we heroically cut $186 billion from food stamps over ten years for Americans who are poor and trying to hang in there. The program known as SNAP will now be called SNAPPED. We have defunded nutrition education for poor families, because frankly, if you can’t afford food, you mercifully shouldn’t be learning about it. We eliminated the Farm-to-School program, because children eating local vegetables is clearly a woke/artichoke conspiracy. We also eliminated the Farm-to-Food-Bank program connecting food banks to fresh, local produce from farmers because it turns out there’s no actual money for us in a food “bank.” We also halted USDA food deliveries to food banks during a period of peak demand. We terminated the Food Security Survey Report, meaning we will no longer be able to measure food insecurity — which, if you think about it, is the most efficient solution to food insecurity we have ever devised. Apply that to job reports, climate science, infectious disease tracking, and more.

We proposed slashing WIC benefits for pregnant women, infants, and children. Congress, being weak like women, infants, and children, fully funded it anyway. Oh well.

And we are thrilled to announce that eighteen states can now restrict what poor people put in their grocery carts. Nothing says freedom like the government controlling your shopping list, provided you are sufficiently poor. We also threatened to withhold remaining SNAPPED funds from Democratic-led states entirely, because basic nutrition for hungry Americans should be a political privilege.

On food safety, we have achieved historic breakthroughs, specifically, historic lows in foreign food inspections. We reduced surveillance of foodborne pathogens from eight to two bugs, because frankly, diarrhea and sepsis are hard to pin on specific sources without surveillance. Did you catch that video I posted of me dumping liquid feces on Americans from a fighter jet? We cut one-third of FDA food safety staff, some of whom, we admit, we had to quietly rehire to keep the whole safety system from udder collapse. We paused milk quality testing during this ongoing bird flu outbreak when it spread to cows, because what’s wrong with a little flirtation with an even worse pandemic than Covid? Want unpasteurized milk?

We withdrew a proposal to reduce salmonella in poultry after years of careful development, because chickens have suffered enough bureaucratic interference, and that Irish immigrant Typhoid Mary has been gone for a long time. We eliminated food safety advisory committees, silenced food toxicologists, suspended the Food Emergency Response Network’s early detection program, and proposed permanent increases in slaughterhouse line speeds over the objections of every expert who knew what a slaughterhouse was.

We also revoked 52 food “standards of identity”— pesky legal definitions that determine what a product must actually contain to carry a familiar name like 'mayonnaise' or 'cheddar cheese.’ This allows manufacturers to quietly swap in cheaper, lower-quality ingredients while keeping the same trusted label, stripping away decades of consumer protection in the name of reducing 'regulatory burden,' with the costs falling heaviest on those least equipped to scrutinize an ingredient list. Two senior FDA officials resigned in protest, stating they could no longer protect the food supply under current conditions. Chickens! We consider this draining the swamp for Big Food and new additives (just don’t tell MAHA that). How do you think Happy Meals still cost $4.50?

Now, let us talk about science — a field we have approached with the boldness of someone who has never trusted a thermometer, nor taken a science class beyond obligatory ones in high school.

We terminated over 7,800 research grants. 25,000 research personnel were dismissed. Those are people. We froze $1.81 billion in NIH awards for cancer, HIV, and ALS research. We cut $258 million in HIV vaccine funding, setting back global efforts by an estimated decade. Unbelievable. We cut $500 million in mRNA vaccine development contracts. We canceled a $600 million contract for avian flu vaccine development. Crossing our fingers has worked so far. Pandemic preparedness is for those who believe in pandemics, who still think Covid was actually a thing. Did you know after my first term The Lancet estimated that 40% of U.S. Covid deaths in 2020 could have been averted with a more science‑driven response? We cut $500 million in gun violence prevention research that showed evidence of reducing gun deaths. We eliminated the CDC’s Division of Oral Health. We eliminated the Newborn Screening Advisory Committee, because babies and parents should be surprised. We eliminated the National Institute for Occupational Safety and Health, gutting decades of workplace safety research — though after nine months of chaos, we quietly reinstated the staff in January, which we mention here only because omitting it would be considered a preventable accident. Get this: we are now aiming for research cuts of 40% for NIH, nearly 57% for NSF, and about 24% for NASA in a single fiscal year.

We withdrew from the World Health Organization because, frankly, the world part felt insulting. America first, am I right?

We abruptly cut, then restored, nearly $2 billion in mental health and addiction funding through SAMHSA, creating what experts described as “chaos and confusion,” and what we describe as “agile governance.” We barred NIH subawards to foreign institutions, cutting off international research partnerships, because science should not cross borders any more than our preferred migrants should. We pushed back on community water fluoridation and began steps to end federal support for it, because fluoride chemophobia trumps healthy teeth in low-income children.

8,000 public-facing federal health webpages were removed including datasets on youth risk behavior, reproductive health, environmental hazards, and more. If you cannot find the data, the problem does not exist. Remember that!? The EPA refused to release a completed report on the toxicity of PFAS chemicals, suppressing the work of its own scientists after cutting $15 million in PFAS research. Climate reports were scrubbed from federal websites. References to human-caused climate change were removed from the EPA’s own homepage. We set plans to dismantle the National Center for Atmospheric Research, because weather, like so many things, has become too inconvenient. We won’t let the EPA interfere anymore with unlimited CO2 and methane. These heat-trapping gases will hasten the demise of your world, and let us billionaires move on to our more comfortable fortress retirements that we’ve been constructing all along! Jeffrey… Oops, did I say that out loud? Look over there!

We have also made extraordinary progress in the field of health communication. All federal health guidance is now reviewed by political appointees before publication, because career scientists with thirty years of experience often say things we find inconvenient. We sent threatening letters to our top American medical journals. We made up our own medical journal. Also, terms like “equity,” “diversity,” and “systemic racism” have been purged from federal research. Federal scientists who published inconvenient findings resigned. We wish them well.

Together with the Secretary of Health and Human Services we publicly claimed that Tylenol causes autism, despite there being no evidence. We are going on a hunch, a potential windfall for lawsuits, and guessing that correlation here proves causation. That’s what they call gold standard science. We later walked back the claim. A little. But the progress, of course, had already been done. We prefer to think of the whole debacle as a conversation starter. Acet.. aceta… aceto… minnow… minophen. More like a hard word contest you might use at a beauty pageant. The Secretary also made false claims about Covid deaths and vaccine safety during a Senate Finance Committee hearing, along with similar statements and teasers by our minions. RFK Jr. remains the Secretary, and I made good on my promise to let him go wild on health. That means your health.

On vaccines, we are proud to announce that we have changed the pediatric schedule amidst an ongoing, historic measles outbreak, which is a remarkable coincidence. Did you know that a measles infection wipes out 30-70% of our previous immunity to all infectious diseases, not just measles? We fired all seventeen career members of the CDC’s immunization advisory committee and replaced them with hand-picked individuals, many of whom have major conflicts of interest, which at least makes them honest. We ended the longstanding hepatitis B birth vaccine, a decision backed by decades of evidence that we found burdensome… and then awarded a $1.6 million grant to Danish researchers to study it further, a grant that Africa’s CDC subsequently canceled as unethical. We removed Covid vaccines from the routine schedule for healthy children and adults under 65, and pregnant women. “Our doctors get more money if somebody dies from Covid. You know that, right? … So what they do is they say, ‘I’m sorry, but everybody dies of Covid.’” Next we’re trying to cancel RSV vaccines that crushed infant RSV hospitalizations last year. We canceled key FDA and CDC vaccine advisory meetings including flu strain selection sessions, and updated the CDC’s autism-and-vaccines webpage to imply that the link had not been disproven. This contradicts decades of scientific consensus, but consensus, like fluoride, rhymes with dense us.

Environmentally, we launched what the EPA called “the biggest deregulation day in U.S. history.” Deregulation means more money at all environmental cost, until that short term math catches up to us after people like me are gone. We stopped counting lives saved when setting air pollution rules, because lives are difficult to monetize and complicate policy considerably. We approved PFAS pesticides that will contaminate soil for generations. We established an actual email address allowing industrial polluters to personally request presidential exemptions from Clean Air Act regulations on mercury, arsenic, and other toxic emissions. Plus another email address just for oil company executives, too. We eliminated the EPA’s Office of Research and Development, slashed the EPA budget by $300 million, and gutted the environmental justice teams protecting communities already bearing disproportionate pollution burdens. You know what those communities look like *wink*. We vetoed a clean water project in Colorado that would have provided clean tap water to tens of thousands of people. We are killing wind and solar and battery projects everywhere we can. China is thrilled. “If you have a windmill anywhere near your house, congratulations, your house just went down 75 percent in value. And they say the noise causes cancer. You tell me that one, OK?” We proposed narrowing the Clean Water Act’s definition of protected streams and wetlands, and removing the regulatory definition of “harm” under the Endangered Species Act — because if harm cannot be legally defined, harm cannot be legally proven. We consider this elegant like an elephant.

But I digressed from health. The environment and our health are two separate, unrelated issues, right? Lead might be a good segue out of here. When a lead contamination crisis emerged in Wisconsin, the CDC could not adequately respond. We had cut the relevant staff. We regret the timing. Not really. Did you even hear about that one?!

On healthcare broadly: we passed the One Big Beautiful Bill, which includes the largest cuts to Medicaid in American history — over $900 billion — set to trigger after the beautiful midterms, with ICE thugs strategically deployed, and the SAVE Act hopefully disenfranchising millions of women. There is nothing more beautiful than a Bill, Clinton aside, and nothing more big than eliminating the healthcare of the poor and middle class. ACA subsidies were not extended, meaning millions face sharply higher premiums heading into 2026. Actually, you’re all paying higher premiums now, as lower risk people stop pooling risk with their own insurance payment contributions. “Experts” told me this would happen. We reversed Medicare drug price negotiation, allowing pharmaceutical companies to raise prices without limits, which they have found very liberating.

Remember I said “We will have Healthcare which is FAR BETTER than ObamaCare, at a FAR LOWER COST — BIG PREMIUM REDUCTION. PEOPLE WITH PRE EXISTING CONDITIONS WILL BE PROTECTED AT AN EVEN HIGHER LEVEL THAN NOW.” Millions of my fellow Americans are today facing steep health insurance premium hikes in 2026, with employer costs rising 10%, ACA benchmark plan premiums up 26%, and out-of-pocket payments for Obamacare enrollees expected to spike 114% due to the expiration of enhanced federal subsidies.​​​​​​​​​​​​​​​​

I launched TrumpRx, which functions like a coupon aggregator so people can pay brand name prices for generics they could find much cheaper elsewhere. Here’s a shiny golden eagle carrying a gilded logo. Feel better?

Did I already mention that we cut suicide prevention grants and community mental health programs, despite rising mental health needs? We shortened ACA enrollment periods, making it harder to sign up for insurance, and we froze $10 billion in childcare funding in Democratic states. A judge has temporarily blocked this, but we remain optimistic. That same $10 billion is coincidentally what I’m suing our own country for, so that your tax dollars can pay me for the prior release of my tax records.

You, the American people, are going to pay for a hospital ship to be sent to Greenland, where the population is suffering from universal access to free health care. Think about that.

We terminated the CDC’s Division of Reproductive Health, cut Planned Parenthood from Medicaid, withheld Title X family planning funding, reinstated the Global Gag Rule, terminated all USAID global family planning grants, and rescinded emergency room requirements to provide abortion care when a pregnant patient’s life is at risk. We proposed eliminating state maternal mortality review committees… ultimately we did not, but we feel the proposal sent a message to women and doctors.

Globally, our health aid cuts are estimated to have contributed to approximately 300,000 deaths, including over 200,000 children. Imagine that. We froze $43 billion in USAID funding without notice, then eliminated USAID entirely, with projections suggesting fourteen million additional deaths by 2030. Does that seem Christian? We cut $1 billion from GAVI, the global vaccine alliance. We withdrew from sixty-six United Nations bodies covering health, climate, and science. We paused PEPFAR funding for global HIV/AIDS relief, creating service disruptions in the world’s most vulnerable communities. We declined to commemorate World AIDS Day. Guess why. We’ve been busy!

I could keep going, but honestly I’m getting tired. So in conclusion, the State of this Union’s health and science has never been more authoritatively focused. We have streamlined public health by removing much of it. We have improved science communication by reducing the science and the communication. We have protected freedom, particularly the freedom to get measles, Covid, cancer, hepatitis, HIV, and to lose your insurance. There is freedom to go hungry, breathe polluted air, and drink PFAS-contaminated unflouridated water without federal interference.

Experienced researchers are leaving federal agencies at tremendous rates. Many of the world’s brightest students are now avoiding America, as if we had concentration camps full of immigrant-looking people and separated children. Let’s not have a moment of silence for an ICU nurse, one of the VA system and healthcare’s own. Many of our top brains are being “poached” by other countries, too. Say that funny word with me: poached. Sounds ridiculous, doesn’t it? Scientific integrity has been replaced with Bobby’s ideas. Programs designed to keep Americans fed, healthy, and safe are being dismantled at a speed that continues to exceed even the best expectations. DOGE that.

This will take many years, perhaps decades, perhaps never, to rebuild.

God bless you, and may God bless these United States, the ones where I knew I could shoot somebody in the middle of the street and not even lose voters, OK?

Are you still with me America?

Thank you for your inattention to this matter.

I get it.

Now, imagine a single blood test that could look for dozens of cancers at once, long before symptoms appear, and tell where in your body to look. That’s the promise of multi‑cancer early detection (MCED) tests—a new generation of screening tools that have drawn huge excitement, intense marketing (see Super Bowl ads for example), and now, some sobering results.

Two MCED tests are currently marketed in the U.S.:

• Galleri, made by GRAIL

• CancerGuard, made by Exact Sciences

Both are sold as “laboratory‑developed tests”, meaning they are available by prescription even though they are not yet FDA‑approved. And as of early 2026, neither has been proven to reduce cancer deaths. That is the key context for everything that follows.

This post walks through what these tests do, what the data actually show, what the latest big trial in England found, and how to think about risks and benefits if you’re considering one. I’ve been working on this post since before the Super Bowl, but decided to drop it now, one day after the bombshell announcement by GRAIL yesterday about their clinical trial results.

Would I get one of these tests? Yearly? How will I reply to patients who ask for my opinion about these blood tests? Will I order one for them as requested?

Here are the evidence based details, you know, examined.

Why MCED tests exist at all

We all know by now that traditional cancer screening is disease‑by‑disease: mammograms for breast cancer, colonoscopy or stool tests for colorectal cancer, Pap tests for cervical cancer, PSA for prostate cancer, low‑dose CT scans for lung cancer in high‑risk smokers. These tests save lives, but they cover only a handful of cancers.

Meanwhile:

• Most cancer deaths come from cancers that have no routine screening test—like pancreatic, ovarian, liver, esophageal, and some head and neck cancers.

• These cancers are often diagnosed at Stage IV, when cure is rare and treatment is mainly about control, not eradication.

MCED tests try to fill that gap. They aim to:

• Detect many cancers at once from a blood sample.

• Catch at least some of the currently “unscreened” cancers at earlier, more treatable stages.

• Potentially complement, not replace, existing single‑cancer screening.

In theory, that could be transformative. In practice, as we just got a glimpse from the GRAIL trial results yesterday, the real picture is more complicated.

Galleri: the flagship MCED test

Galleri looks for fragments of tumor DNA that might be floating in the blood—“cell‑free DNA”—and focuses on methylation patterns, a type of chemical tag on DNA that differs between healthy and cancer cells. Using machine learning, the test tries to answer two questions:

1. Is there a cancer signal in the blood?

2. If yes, what is the likely Cancer Signal Origin. For example, pancreas vs ovary vs lung?

If Galleri says “Cancer Signal Detected,” it also reports a predicted tissue or organ of origin, which then guides imaging and other tests. This is a key feature, as just getting a maybe somewhere result is a mess. Read on.

How accurate is it?

There are three key numbers for any screening test:

• Sensitivity – how often it finds cancer when cancer is truly present.

• Specificity – how often it correctly says “no cancer” when there really is no cancer.

• Positive Predictive Value (PPV) – if the test is positive, how likely is it to really be cancer?

From large Galleri studies:

• PATHFINDER and PATHFINDER 2 (prospective real‑world screening trials) show:

  • Sensitivity for cancers caught within 12 months of testing: about 40% for all cancers, ~74% for the 12 deadliest cancers.

  • Specificity ~99.6%, meaning only ~4 false positives per 1,000 people tested.

  • PPV ~62%: when Galleri was positive, about 6 in 10 people truly had cancer.

  • Cancer Signal Origin accuracy around 90%+.

In plain English:

• Galleri is very good at avoiding false alarms.

• When it says “cancer signal,” it’s more often right than wrong.

• But it misses most cancers (100% - 40% for all cancers = 60% missed), especially the earliest ones.

Catching a chance early is obviously the goal of any cancer screening, as earlier detection leads to earlier treatment and better results. How early are cancers found with Galleri? In earlier case‑control data, Galleri’s sensitivity by stage was roughly:

• Stage I: ~17%

• Stage II: ~40%

• Stage III: ~77%

• Stage IV: ~90%

So the test is best at finding more advanced cancers and least sensitive at Stage I—the exact opposite of what many people assume. Here is what that looks like, in a graph I made with help from Perplexity:

What that big NHS‑Galleri trial just showed

The NHS‑Galleri trial in England was the high‑stakes test for this technology. It enrolled ~142,000 adults ages 50–77, randomized them to:

• Usual National Health System screening
  vs.

• Usual screening plus annual Galleri testing for three years.

The main goal/primary endpoint was to show fewer late‑stage (Stage III–IV) cancers in the Galleri arm on the logic that if Galleri works, some cancers that would have presented as Stage IV could be “shifted” to Stage II or III instead. That would obviously be good!

Topline results:

• The trial did NOT achieve its primary endpoint: there was no statistically significant reduction in total Stage III–IV cancers in the Galleri arm vs control.

• GRAIL emphasizes several secondary signals:

  • A “favorable trend” toward fewer Stage III–IV cancers within a pre‑specified group of 12 deadly cancers (but not statistically significant for the primary endpoint).

  • A substantial reduction in Stage IV diagnoses for those 12 deadly cancers—>20% fewer Stage IV cases in years two and three of screening—and similar trends across all cancers.

  • An increased detection of Stage I–II cancers in those deadly cancers in the Galleri arm.

  • Fewer cancers diagnosed through emergency presentations, which are linked to worse outcomes.

  • Test performance (PPV, specificity, CSO accuracy) was similar to previous North American trials; no major safety issues.

Independent experts that I read about quickly last night and this morning are more blunt:

• STAT News and the New York Times both describe this as a failed trial in terms of its primary goal—Galleri did not significantly reduce late‑stage cancer diagnoses overall.

• My Substack friend Dr. : “There’s no such thing in clinical trials as a favorable trend, especially in a trial that has got immense statistical power.”

• Cancer researcher Dr. Adewole Adamson: “The study failed. End of story”.

• Cancer researcher Dr. Richard Houlston calls Galleri “not a very good test” and says the results “don’t support rollout within the American health care system.”

So… we’re left with a nuanced picture, like always! Why is nothing simple? Nuance is ripe for exploitation by other people offering cheap and simple answers, so it’s important to chew on this for a while:

• There may be meaningful stage shifting for some of the worst cancers (fewer Stage IV, more Stage I–II), but…

• The trial was designed to show a clear reduction in overall Stage III–IV cancers and did not.

• And we still have no mortality data—no proof yet that lives are actually being saved. That’s what we care the most about. That’s what earns a screening test a place in the primary care repertoire, and the United States Preventative Services Task Force (USPSTF) guidance for doctors and patients. Recall that RFK Jr. is quietly yet actively undermining the USPSTF as we speak. Recall what I said about opportunists swooping in to do the easy work of approving all sorts of vibe-driven tests and supplements and questionable treatments. Follow the money, and demand real expertise.

CancerGuard: the new competitor

A patient asked me just last week about a new test that was marketed to him after he received his routine colon cancer results from a Cologuard test (approved, and a good screening test for colon cancer that uses a stool sample). Should he pay the extra money to do this new blood test for multiple other types of cancer? Should I order it for him?

CancerGuard is another MCED test offered by the company who brings us Cologuard (Exact Sciences). It uses what called a multi‑omic approach:

• DNA methylation (like Galleri, using technology that preserves DNA sequence)

• Blood proteins associated with cancer.

The test is aimed at adults aged 50–84, and it screens for 50+ cancer types and subtypes—but excludes breast and prostate cancers (which are covered by other Exact Sciences products).

Key differences vs Galleri:

• CancerGuard does not provide a specific cancer signal origin. A positive result triggers a broad imaging workup (which probably triggers a contrast CT of chest/abdomen/pelvis ± PET‑CT) rather than a more targeted search. That’s a lot of radiation, and potential to find other random stuff. See below.

• It is cheaper—about $689 vs Galleri’s ~$949 list price.

• It is considerably less validated in real‑world screening populations.

CancerGuard’s evidence

The headline numbers come from case‑control studies:

• Development study: sensitivity 64.1%, specificity 97.4%.

• Independent validation: sensitivity 55.6%, specificity 97.4%.

• For six of the most lethal cancers (e.g., pancreatic, lung, ovarian), sensitivity 68%, and over one‑third of Stage I–II cancers detected.

However, all of this is from people already known to have cancer vs matched controls, not from large numbers of healthy people being screened. Exact Sciences explicitly warns that real‑world sensitivity may be lower than these case‑control estimates. The company is running additional prospective studies, but nothing at the scale of PATHFINDER 2 or NHS‑Galleri yet.

How do Galleri and CancerGuard compare?

Here’s a simplified comparison based on current data and manufacturer disclosures. Once again, I deployed Perplexity for help visualizing this data:

A few themes emerge to help us tease out the difference here:

• Galleri is far better studied in real‑world screening; its strengths and weaknesses are clearer.

• CancerGuard is cheaper and may ultimately benefit from multi‑omic design but currently lacks the same level of prospective validation.

• Neither has shown mortality benefit, and Galleri has just failed that key trial meant to show a robust reduction in advanced cancers in the NHS.

Benefits: what these tests can actually offer right now

Despite the new negative trial, and my own reluctance to dive into all of this, I do want to acknowledge that there are real potential benefits to weigh.

• Detecting cancers with no other screening tests

  Both tests can pick up cancers like pancreatic, ovarian, liver, and some head and neck cancers, for which there are no routine screening tests for average‑risk people. Some of these are exactly the cancers we most fear, because they usually present late and are often lethal. Even catching a modest fraction of them earlier could matter for some individuals.

• Very low false positive rates (especially Galleri)

  Galleri’s specificity around 99.5–99.6% means very few people get a false alarm. That’s a major advantage over some traditional tests (like mammography, where 10% of screens can be called abnormal).

  CancerGuard’s specificity (~97.4%) is somewhat lower but still high compared to many traditional screenings.

• Helpful guidance when positive

  Galleri’s Cancer Signal Origin prediction (correct >90% of the time) can help focus the diagnostic work‑up—targeted imaging rather than a whole‑body search. CancerGuard’s strategy is more broad‑brush (CT/PET‑CT), but its makers offer navigators and some financial support for follow‑up imaging. However, the financials of follow up testing are opaque to me, and are likely variable between insurance plans covering or not. It might all be out of pocket after the random screening test which was not indicated for chasing down a symptom or problem.

• Signals of earlier detection in deadly cancers

  Even in the NHS‑Galleri trial, which failed its primary endpoint, there appears to be:

  *A reduction in Stage IV cancers, and

  *A rise in Stage I–II cancers within the 12 deadliest cancers.

For most cancers, moving from Stage IV to Stage III can make a difference; for others, the gain is modest. Without full, disease‑by‑disease data and, crucially, survival data, we can’t yet quantify what this means in terms of lives saved.

Risks and limitations

Galleri’s sensitivity is lowest for Stage I cancers, catching maybe 1 in 6 at that earliest stage. That means a lot of true cancers are missed, especially early ones. CancerGuard’s case‑control data suggest better early‑stage performance, but we don’t yet know how that holds up in screening populations.

A negative MCED test therefore does not mean you’re cancer‑free. It simply means the test didn’t see a signal that day.

Real‑world reports show that some patients (and some telehealth companies) are treating Galleri as a kind of “all‑clear” when it’s negative, even delaying mammograms or colonoscopies. That’s dangerous. That was the outrage among people in the know who watched the Super Bowl ad. Given the modest sensitivity and the new NHS‑Galleri result, a negative MCED test should never be a reason to skip proven screening.

Are you an anxious person? That needs to be considered. A positive MCED test can launch weeks or months of:

• Waiting for scans

• Further testing and biopsies

• Living with the worry: Is there cancer somewhere in me that we just haven’t found yet?

Some people wind up with a positive blood test but no detectable tumor, a situation one breast cancer researcher called the “Damocles syndrome”—a sword hanging over your head with no clear next step. Even if false positives are rare, each one is psychologically and financially costly.

So far, no MCED test has shown that its use reduces cancer deaths in a randomized trial. NHS‑Galleri was an important step, designed to show stage shift… but even that primary goal was not met. Demonstrating mortality benefit will require longer follow‑up and possibly different trial designs.

Without mortality data, earlier detection could still reflect lead‑time bias. This happens when we simply know about cancer being present longer without actually living longer.

Costs

For most of us non-billionaires, costs remain a consideration. Here is the best summary I could come up with, though prices often change and different sources quote different exact numbers:

• Galleri costs roughly $949 (sometimes discounted to ~$700–$749 through certain providers or programs).

• CancerGuard is priced around $689.

• Most insurers and Medicare do not currently cover these tests.

In February 2026, the President signed the Nancy Gardner Sewell Medicare MCED Screening Coverage Act, which will eventually allow Medicare to cover FDA‑approved MCED tests that show clinical benefit—but not before around 2028, and likely with careful conditions. After the NHS‑Galleri result, experts are urging that coverage, if granted, be tied to clinical trials or registries so we keep learning rather than just paying.

So… should we get one?

At this point, no major medical society recommends routine MCED testing for everyone. The American Cancer Society’s guidance emphasizes shared decision‑making: a careful doctor‑patient conversation about what’s known, unknown, and what matters most to you personally.

Reasonable situations to consider an MCED test might include:

• Being 50 or older

• Having strong risk factors for cancers without standard screening (like significant family history, or a known hereditary cancer syndrome)

• Understanding that the test is experimental, with no proven survival benefit

• Not using a negative result as an excuse to skip mammograms, colonoscopies, or other recommended tests

• Being able to afford the cost (and possible follow‑up costs) without giving up other important care

For many others, especially average‑risk adults, the most cautious and evidence‑driven stance is:

• Focus on doing the proven screenings well and on time. Your good old-fashioned primary care doc is all about this.

• Pay attention to symptoms and seek prompt evaluation

• If you opt into an MCED test, treat it as participation in an evolving area of medicine—not as a guaranteed, life‑saving intervention.

A balanced way to explain this to patients

SO here is what I’m settling in with in terms of how I should answer my patients’ questions about those new blood tests for cancers. I’m thinking I might send them a link to this post, but some will prefer less of a deep dive. For them, perhaps I’ll say:

• “These blood tests are a promising research frontier. They can sometimes catch cancers we’d otherwise miss, and they are quite good at avoiding false alarms.”

• “However, they miss a lot of cancers, especially at the earliest stage, and the best large trial we have so far did not show a clear reduction in advanced cancers overall.”

• “We also don’t yet know whether they help people live longer, and they can start a chain of follow‑up tests, cost, and anxiety.”

• “If you’d like to consider one, we can talk through your personal risk, your values, and what you’d do with a positive or negative result. But whatever you decide, we should still keep up with regular mammograms, colon screening, Pap tests, and other proven tools.”

How does that sound?

For now, MCED tests sit in a gray zone: more than hype, but less than the revolution their marketing sometimes implies. They’re probably a glimpse of the future, but not yet the future fully realized.

As patients and clinicians, the most important thing is to keep two ideas in our heads at once:

1. The need for better early detection, especially for deadly cancers we currently find too late, is absolutely real.

2. The evidence we have on today’s MCED tests, especially after the NHS‑Galleri trial dropped yesterday, is not yet strong enough to treat them as a standard, life‑saving screening for everyone.

Ok, I’m done! I hope this deep dive was helpful. I relied heavily on sources like GRAIL, Exact Sciences, NY Times, STAT News, Perplexity, and other random sites found through an exhausting crunch of sources. Best of luck with your decisions, and please just consider this post one more educational input that informs an ultimate discussion with your doctor… who will not have time to go into the weeds like we just did together. That’s why I first started writing Examined in a quest to expand upon vital and overlooked ideas your family doctor might share — if only we had more time.

Enjoy your weekend if you can :) Significant snow predicted here in Philly tomorrow!

In that moment he was justifying having written a book about lofty philosophical and neuroscientific ideas, even as we are just scraping by with horrific politics, governance, and societal decline. Do we really have the luxury of reading a book about consciousness in a time of titanic clashes— democracy versus fascism, centibillionaires versus the working class, climate collapse versus unthinkable climate change denial?

Absolutely.

I’m going to riff on the idea of consciousness as it pertains to our physical, mental, and holistic notions of health, in addition to gaming out a better way to be a patient and a doctor in a room together. This is part one.

And if you like where this essay goes, I will also follow Michael Pollan’s answer about how our consciousness is being polluted by our politics in 2025-26, and how that absolutely impacts our collective health and wellbeing. This would be part two.

As a family doctor I have the privilege and peril of being a jack of all trades, a synthesis seeker, a crackpot philosopher when I feel up to it.

Let’s go.

The gap

Have you ever walked into a doctor’s appointment as a conscious being, maybe suffering, or at least aware that something is wrong in ways you may not have words for or that cannot be measured/scanned/poked and prodded… yet the visit proceeds as if only the measurable parts of you matter?

I want to ask a narrower version of Pollan’s question: What would medicine look like if we actually took consciousness seriously? What follows is my attempt at an answer. Not like PhD-level philosophy, but like something we could use in an exam room, as part of a therapeutic relationship, as a re-humanization paradigm.

A vignette:

A woman walks into her doctor’s office with fatigue, diffuse pain, and a sadness she can’t name. Her labs are normal. Her scans look good. She is told she is fine. Drink plenty of water.

But she does not feel fine.

She is conscious after all, aware of her pain, her dread, her felt sense that something is wrong. And that intuitive awareness is the one thing modern medicine is least equipped to address. We have extraordinary tools for measuring what happens in the body, a catalogue of pills for every problem, but a limited clinical framework for engaging with what it feels like to be in that body. We often outsource that, or try our best within the 20 minutes we have for an impossibly long meeting agenda.

And yet every symptom a patient reports, every fear they carry into the exam room, all of it is mediated by consciousness. It is the medium through which all suffering is experienced and all healing is potentially felt, and we are fools not to discuss this explicitly more often.

Please allow me to try.

What consciousness actually is (the short version)

Consciousness is what it’s like to be you, right now. The philosopher David Chalmers called explaining this the “hard problem.” We are not talking about how the brain processes information, but why any of it feels like anything at all. You know, like why we are not automatons and zombies.

Three dimensions matter clinically:

• Phenomenal consciousness is the raw feel of experience: the sharp burn of a paper cut, the heaviness in our chest at bad news.

• Access consciousness is what we can put into words for our doctor: the pain is a six out of ten, it started three days ago, it’s worse after eating.

• Interoceptive consciousness is perhaps the most important for medicine. It is our awareness of our body’s internal states, our heartbeat, our breath, our gut, the felt sense of our physiological being. Interoceptive consciouness carries information that lab tests can approach but never fully capture. And many people themselves have lost an intuitive, embodied understanding of it entirely.

What generates these dimensions of consciousness?

The most clinically useful theory of consciousness generation right now is something called predictive processing. The idea is that the brain doesn’t passively receive reality. It actively constructs a best guess. This in part explains why two patients with identical pathology can have radically different experiences, why the placebo effect genuinely works, and why some chronic pain persists long after tissue has healed. Sometimes the brain’s prediction simply hasn’t been updated.

Biochemistry supports what the theory suggests. There is no symptom that is purely physical and no symptom that is purely mental. Over 90% of the neurotransmitter serotonin is produced not in our brains but in our guts. Chronic stress cortisol structurally remodels the brain. Inflammatory cytokines cross the blood-brain barrier and alter consciousness itself. The patient is not a mind riding in a body. The patient is a bodymind, one conscious biochemical being.

Now let me disclaim something important. None of this replaces good reductionist medicine. When someone has bacterial pneumonia, the actionable answer is antibiotics, not embodied awareness. The Western biomedical model’s power lies in its ability to isolate treatable mechanisms: the fractured bone, the occluded artery. But that lens is a method, not the whole truth. Every intervention we make ultimately serves a conscious being whose experience of illness and recovery is the true measure of whether we’ve helped.

The honest problem

If you’re a fellow physician reading this, you may be thinking: This is lovely, but I have twenty freaking minutes with each person.

I know. I freaking know.

The compressed visits, the documentation burden, the panel sizes, the clinical decision-making volume make consciousness-informed medicine nearly impossible to consistently practice. We are lucky to make it to the bathroom to urinate without feeling several kinds of pressure at once.

And if you’re a patient, you will never have a physician who speaks to you about consciousness per se, though hopefully you’ve experienced the related corollaries of empathy and compassion. You will certainly have twenty minutes with someone who is already thirty minutes behind schedule. And unless you are feeling vulnerable, or have a good doctor-patient relationship with a trusted doctor, you might not even want someone attending to your intimate, conscious experience.

And yet most of us do want that human connection, that humanization, that holistic analysis and resonance with a helpful soul. I do as both a patient and a physician.

So here’s what we might do on our own first.

When we are patients

There are some concrete, easy-ish things we can start doing today that tune us into our consciousness.

Rebuild our ability to listen. Try a five-minute body scan: eyes closed, attention moving from head to feet, not diagnosing, just noticing what’s tight, heavy, or numb. Most of us have learned to override the body’s signals, treating them as interruptions rather than information. This is a practice of paying attention again. Meditation emphasizes this sort of awareness, acknowledging but not following unpleasant sensations.

Keep a real journal when struggling. Not just symptoms, but context: mood, sleep, stress, any grief or conflict, what we ate, whether we moved. No single office visit can reveal the patterns that emerge over weeks.

Trust the signal when labs are normal. If our tests look fine but we feel terrible, we shouldn’t accept that as proof nothing is wrong. It may mean our conscious experience is detecting something our current tools haven’t captured yet. Advocate calmly, persistently, and without apology. And yet the paradox here is that sometimes a normal work up, an MRI that looks good, can be therapeutic in and of itself. More on that later.

Name the emotional dimension. Even silently: I’m afraid this won’t get better. I’m angry no one believes me. I’m grieving the life I had before this started. Naming isn’t wallowing. It’s the moment unconscious suffering becomes conscious and therefore workable.

The deeper tools

Now, are you ready for the rabbit hole?

I don’t think the crux of this whole framework is just paying attention to our bodies. It’s developing the capacity to watch ourselves paying attention. Researchers call this metacognition. Contemplative traditions have been teaching it for centuries.

Most of us experience this fleetingly: the moment we notice we’re anxious rather than just being anxious. The moment we catch ourselves catastrophizing and think, there I go again. That small gap between experience and the awareness of experience is, neurobiologically, one of the most powerful therapeutic instruments a human being possesses. The square of the square in mathematics.

Here’s how we might deploy it deliberately.

With physical symptoms: Most people encounter a sensation and immediately leap to a story about it. This headache means something terrible. This chronic fatigue means I’ll never feel normal again. Meta-awareness lets us insert a step between sensation and narrative. The sensation is data. But the story is prediction. And as predictive processing tells us, the story our brains construct about a symptom can amplify or even generate suffering far beyond what the original signal warrants. Have you ever noticed that just by turning the conscious spotlight on a symptom, a twitch, an ache, a worry, we magnify the experience as if through a magnifying glass? There is brief utility in that, but there is also risk.

A moderating practice is both simple and difficult: What am I actually feeling right now, and what am I adding to it with my mind?

With mental health: A person who is depressed and identified with their depressive thinking might become trapped inside the content of consciousness. Thoughts like I’m worthless, nothing will ever change, it’s hopeless feel like hard facts. A person who can observe that same thinking with a step back, who can notice that their mind is generating hopeless thoughts and that it tends to do this when they’re exhausted and isolated, has not eliminated the depression. But they have fundamentally changed their relationship to it. The thoughts are still present, but they have been reclassified from truth to weather. This is a predictor of recovery across anxiety, depression, PTSD, and addiction actually, and a good counselor can evoke and harness this sort of reckoning.

With chronic illness: Patients with fibromyalgia, long Covid, autoimmune conditions, menopausal syndromes, and chronic fatigue often endure the additional injury of not being believed. Meta-awareness can give them a kind of internal authority that doesn’t depend on external validation. When we have carefully and consistently observed our own body, when we know that our fatigue deepens predictably after certain exposures and that our pain has a rhythm our labs can’t see, we carry an evidentiary record that no dismissive encounter can erase. We are not guessing or making stuff up. We are reporting from direct, disciplined observation of our own consciousness.

*The common thread: meta-awareness does not change what is happening. It changes our relationship to what is happening. And that shift, from being captured by experience to being present with it, turns out to be one of the most consequential chess moves a human being can make in life. It is the difference between being lived by your illness and living with it, even above it. Mindfulness meditation is one way to learn some of these techniques.

For the doctors in the house

No new philosophy degree required. No billing codes.

Swap one closed question (Are you depressed?) for one open one (What’s been weighing on you?) and watch how often the answer reframes the entire clinical picture. Changes the therapeutic nature of the encounter itself.

When labs are normal but the patient isn’t feeling it, say so honestly: Your tests look reassuring, and I also hear that you don’t feel right. Both can be true. Let’s keep looking, and keep the door open in this conversation.

We all know there is an art and a science to doing this job well. Just like schools with underfunded budgets, the first thing cut is usually art.

And protect your own consciousness. One breath before you open the door. A self scan of body signals and cold hands. One moment of genuine eye contact before the keyboard. A precious shred of small talk before the complex problems. These aren’t luxuries. A burned-out, dissociated physician is operating with the same diminished integrative capacity we’d recognize as pathology in our patients.

Where this may all be going

Artificial intelligence is being rolled out like a steamroller. It is more aptly described as alien intelligence, yet it fundamentally seems to work on predictive cogwheels like human consciousness may be deploying.

As artificial intelligence increasingly handles the pattern recognition, differential diagnosis, and algorithmic decision-making that have traditionally defined clinical expertise, scrutinizing this parallel intelligence becomes urgent. Existential.

AI can read an imaging study faster than any radiologist. It can predict hospital readmissions more accurately than any risk calculator. What it cannot do, and what it may never be able to do, is authentically meet another consciousness with presence. It cannot recognize the irreducible particularity of a patient’s suffering, or hold the space in which a person becomes more than their diagnoses. It has not suffered, triumphed, had a first kiss or a broken heart.

As medicine’s technical functions become automated, the physician’s core competency will shift toward exactly what machines cannot replicate. We need to use AI not to become even more dehumanized technicians, but to free up attention for what has always mattered most: the human capacity to bear witness to another’s consciousness with compassion and humility and professional expertise.

In this future, understanding consciousness won’t be a philosophical luxury.

It just might be the defining clinical skill.

The End

Michael Pollan wrote his new book because he believes consciousness is worth defending. I plan to read it (which translated means that I plan to listen to the audiobook on my pithy-thought-generating, expletive-laden-battle of a morning commute).

In the exam room, I argue it's worth practicing with consciousness. Two people sit across from each other, each carrying an inner life and light that no chart can hold, and for a few minutes they have the chance to actually meet. Perchance to dream. That meeting, small and unglamorous and perpetually rushed as it is, remains the most powerful clinical tool either of them will ever have access to. More powerful than sitting at home, DIY, alone in front of a screen, typing fretful thoughts into a search field or chatbot prompt.

We must do that searching… but we must not forget our luminous and human consciousness, nor the vital connections to other people, ourselves, and the deep well of collective possibility and understanding contained within.

Read more