The five sisters reminisce about everyday life on the farm — filling tick mattresses, butchering hogs — the rhythms of a world that no longer exists. By today’s standards, theirs were difficult lives, but they seem to have been happy, until they lost both parents within a year of each other, leaving the girls orphaned at ages ranging from seven to twenty. Their subsequent lives were turbulent, but as one of the sisters puts it, “My childhood according to kids today was a mess but I lived thru it…”

From a medical perspective, their history is striking. In addition to the five sisters who survived, three other siblings died in infancy, and both parents died when quite young — a mortality rate of 50 percent in a single family.

One sibling is described as succumbing to “summer complaint” — a quaint but deceptive term for what we would now call acute gastroenteritis, typically caused by contaminated milk and poor sanitation, and peaking in summer months due to the absence of refrigeration. The cause of death of the other two infants is unspecified, but it is nearly certain they also fell prey to infectious disease, either gastrointestinal or respiratory.

Their mother, Sarah, died at 43 of a chronic illness of unspecified duration. She is described as having “spells,” typically in the evenings, that left her in a “stupor” the following day. She died either during one of those spells or shortly after, and her death seems to have been unexpected. Based on these limited descriptions, epilepsy seems the most likely diagnosis.

Their father, Henry, died a year later at 53. My grandmother, who later became a nurse, noted that he had “Bright’s Disease,” an older term for renal failure, and mentioned that she was surprised he didn’t “go into a coma,” by which she presumably meant a uremic coma. The likely culprits in that era were untreated chronic hypertension and post-streptococcal glomerulonephritis.

Three deaths from acute illnesses which were almost certainly preventable, and two from conditions that would be manageable, had this family lived a century later.

Thomas Hobbes, writing in Leviathan in 1651, described the life of man as “solitary, poore, nasty, brutish, and short.” Looking at this family’s history, it’s hard to argue with him — at least for their era.

Life expectancy data from the United Kingdom, which has the oldest such records and tracks closely with U.S. data from the mid-19th century onward, shows centuries of average lifespans hovering between 35 and 40 years. Then, beginning around 1800, a sustained and dramatic rise.

What drove that rise is worth examining carefully, because the answer is not what most people assume.

It wasn’t, in the main, what we would call “modern medicine”, at least not at first. Much of the early improvement came from sanitation: cleaner water, better waste removal, and a growing public health sensibility that preceded formal acceptance of germ theory by decades. Pasteurization of milk became common in the United States in the early twentieth century, followed by widespread refrigeration. And while general anesthesia arrived in the 1840s, it would take another generation before surgeons reliably washed their hands before operating.

Edward Jenner established that inoculation with cowpox provided protection against smallpox in 1796. Other vaccines followed in the 20th century: diphtheria (1923), tetanus (1924), pertussis (1926), polio (1955), measles (1963). These were not small things.

Against that backdrop, “modern medicine” arrived quite late. Insulin, the first transformative non-vaccine medication, was used clinically for the first time in 1922. Penicillin became widely available only during World War II. Modern ventilators and cardiopulmonary bypass came in the 1950s. Dialysis for chronic renal disease, which might well have bought Henry many more years, didn’t become commonplace until the 1970s. CT scanning was routine by the 1980s; MRI followed a decade later.

What I inherited as a surgeon is, in its most meaningful sense, barely a century old. And much of what came before it — the sanitation revolution, the vaccines, pasteurized milk — is now so thoroughly taken for granted that we’ve largely forgotten it was ever an achievement.

Which brings me to where we are now.

If you look at that life expectancy curve, we appear to be at something of a plateau. I think that’s an accurate read. Through prevention and successful intervention, we have largely conquered the acute illnesses and injuries that claimed lives like those in this family. The low-hanging fruit has been harvested.

What remains is harder. Chronic diseases, such as cancer, heart disease, and dementia, may be manageable, and incremental advances likely. But dramatic breakthroughs are much more difficult, because the nature of progress has changed. Eradicating smallpox was huge, immediate, and visible. Its effects were obvious within a generation. The early gains of modern medicine came with large effect sizes, obvious benefits, and a simplicity of cause and effect that is genuinely rare now.

The trials we run today are different by necessity. We are working at the margins of diseases that are complex, chronic, and multifactorial. Effect sizes are smaller. Follow-up periods are longer. The signal is harder to separate from the noise — not because the science has gotten worse, but because the questions have gotten harder.

This is precisely why evidence-based medicine matters more today, not less. When the benefit is self-evident — when the child with diabetes gets insulin and lives — you don’t need a controlled trial. When a therapy offers a modest relative risk reduction in a surrogate endpoint over a five-year follow-up, rigorous methodology is the only reliable guide we have.

I’m reminded of what one of the sisters wrote, explaining why she wanted to set these memories down in the first place: “just so they are not laid aside and forgotten.” She was talking about farm life, about family, about a world that had passed. But I find myself thinking the same thing applies here. It is easy to take for granted what clean water, safe food, and effective vaccines have done for human health — precisely because those gains are now invisible, absorbed into the baseline of ordinary life.

The acute has been largely conquered. What lies ahead is the chronic, the incremental, the difficult to demonstrate. We should be clear-eyed about that — and hold the evidence to exactly the standard such difficulty demands.

In case anyone would like to peruse the letters from the five sisters:

Five Sisters Memories

What I kept encountering was the same pattern: confident advice resting on thin evidence, and very little help for people trying to understand what actually matters. Mainstream media tends to reward what is novel and provocative. Healthcare influencers often make claims that are, at best, weakly supported and, at worst, self-serving. And now we have a newer wrinkle: AI-generated deepfakes of clinicians promoting products and dispensing bad advice.

There are many smart people doing excellent work evaluating the medical literature for clinicians. There are also public-facing commentators offering generally sound advice, but their focus is often necessarily broad. I’ve only found a handful of people who take a deeper dive into medical research and present it in a way that genuinely helps non-experts think more clearly, but they take some effort to find.

That’s the space I want to occupy going forward.

This isn’t about chasing every new study, diet, screening test, or supplement making headlines. It’s about helping people recognize when evidence is strong, when it’s weak, and when uncertainty is being papered over with confidence. We’ll touch on medical history, research design, and medical literacy. We may debunk a few persistent myths. But the goal is practical: to help you navigate the noise so you can make better healthcare decisions.

I’ll start by writing here on Substack. Some of this may eventually move into audio or video, because voice can sometimes convey nuance that text cannot. I may even have to reengage with social media, which gives me some pause. The focus, however, will remain the same.

Clarity over certainty.

And if you ever see my likeness enthusiastically hawking moisturizers or supplements, rest assured: that’s a deepfake, not me.

Over the past few decades, a series of whistleblowers and trailblazers have exposed the shortcomings of modern research — and offered concrete ways to fix them. Here’s a brief history of metascience, and why it matters to all of us: scientists, laypeople, and journalists alike.

In my last post, Consensus-Followers, Reformers, and Contrarians, I noted that much of our published scientific research is unreliable. While this view hasn’t yet become mainstream, it is gaining traction within the scientific community. That recognition has given rise to the field of metascience, or meta-research — essentially, the scientific study of science itself.

I admitted I’m a late-comer to this field of skepticism, but I’ve enjoyed diving into its history. I believe it’s vital that we all become better purveyors of scientific research. So at the end of this piece, I’ll share a brief history of this school of skepticism.

That history — and the subject of metascience more broadly — is aimed mainly at clinicians and academics. If the recommendations of the thinkers I’ll cite were followed, it would help us better parse studies we see in journals or at professional meetings, and design and conduct stronger research in the first place.

But what about outside the clinic or the lab? The rest of us are constantly bombarded by “breaking medical news” from mainstream outlets, algorithm-driven social media posts, and “healthcare influencers” whose motives often extend beyond your well-being. How can you know which research deserves attention and which to ignore?

And what about journalists, tasked with writing quick blurbs on new studies? Asking the right questions is essential for deciding whether a study is newsworthy — and how to present it responsibly if it is. The way these stories are covered doesn’t just satisfy curiosity; it shapes patient decisions, physician practices, and even public policy.

I’ll circle back soon to how to assess science reporting in everyday life — and how journalists can do a better job of it. To whet your appetite: just yesterday my wife flagged an article on hormone replacement therapy and Alzheimer’s risk. The coverage is better than most, yet it still highlights how tricky it is to properly evaluate research and report it responsibly.

Hormone Replacement Therapy Timing Linked to Alzheimer’s Disease Risks

A Brief History of Metascience

• 1972 – Archie Cochrane publishes Effectiveness and Efficiency, a short book arguing that medicine should be guided by randomized trials and judged by both effectiveness and efficiency. It became a manifesto for evidence-based medicine.

• 1980s – Iain Chalmers, Murray Enkin, and Marc Keirse lead the charge for systematic reviews. Their Effective Care in Pregnancy and Childbirth (1986) pulled together every trial they could find in obstetrics. This approach led directly to the founding of the Cochrane Collaboration in 1993, named in honor of Archie Cochrane, and dedicated to synthesizing medical evidence across fields.

• 1992 – The Evidence-Based Working Group coins the term “EBM” in JAMA (Evidence-Based Medicine), led by David Sackett and Gordon Guyatt. Their aim: to make clinical decisions explicitly grounded in critically appraised evidence.

• 1994 – Douglas Altman publishes The Scandal of Poor Medical Research in BMJ. His blunt message: “We need less research, better research, and research done for the right reasons.”

• 2005 – John Ioannidis publishes Why Most Published Research Findings Are False in PLoS Medicine. His probability model showed how low study power, weak prior odds, and bias make many “statistically significant” findings unreliable. The paper became one of the most cited in modern science.

• 2006 – Richard Smith, longtime editor of BMJ, published The Trouble with Medical Journals in the Journal of the Royal Society of Medicine, noting publication bias, conflicts of interest, poor peer review, and a tendency to favor flashy results over solid science.

• 2009 – Iain Chalmers and Paul Glasziou published Avoidable Waste in the Production and Reporting of Research Evidence in The Lancet. They estimate that as much as 85% of medical research is wasted — through asking the wrong questions, using poor methods, failing to publish results, or reporting them inadequately.

• 2010s – The Replication Crisis hits home.

  • 2012 – Amgen Cancer Reproducibility Study
    C. Glenn Begley and Lee Ellis reported in Nature (Raise Standards for Preclinical Cancer Research) that Amgen scientists tried to replicate 53 “landmark” preclinical cancer studies — and succeeded in only 6. The failures highlighted how weak methods, bias, and poor reporting undermine even high-profile biomedical research.

  • 2015 – Open Science Collaboration
    Brian Nosek and colleagues reported in Science (Estimating the Reproducibility of Psychological Science) that of 100 psychology studies replicated, only 36% produced significant results again, and effect sizes were often smaller.

  • 2015 – Vinay Prasad and Adam Cifu publish Ending Medical Reversal, showing how many accepted treatments collapse under stronger evidence.

• Today – Metascience (or meta-research) has matured into a formal field, with centers like the Cochrane Collaboration, the Center for Open Science, and the Meta-Research Innovation Center at Stanford (METRICS). Special mention also to the founders and collaborators at the Substack, Sensible Medicine, which I have found to be an invaluable resource. Their combined mission is to study how science is done, expose systemic flaws, and promote reforms for more reliable research.

Medicine has always contended with more unknowns than knowns. Scientific research is our best tool for finding answers—but its usefulness depends entirely on the quality of the research performed.

When my patients make make some spurious connection between some past event and a current problem, I bring out one of my favorite stupid jokes: “Humans are natural scientists—we’re just not naturally good ones”. We chuckle, but we physicians and researchers aren’t immune from faulty reasoning either.

In the 1970s, Archie Cochrane argued that medicine was too often guided by tradition rather than evidence. Two decades later, Douglas Altman warned about the misuse of statistics in research, saying: “We need less research, better research, and research done for the right reasons.”

In 2005, John Ioannidis’s essay Why Most Published Research Findings Are False crystallized those warnings. It showed how small studies, poor methods, selective reporting, and funding bias made much of medical knowledge provisional at best, and sometimes simply wrong.

This self-skepticism of science has become a field unto itself, variously termed metascience or meta-research. It hasn’t yet reached the mainstream, but its influence is growing. I plan to return to this in future posts to make the concepts that separate good research from bad accessible to a lay audience—and to help readers guard against uncritical reporting in the media.

My own awakening came later, with Vinay Prasad and Adam Cifu’s 2015 book Ending Medical Reversal. Their premise was simple: interventions often enter practice on weak evidence, only to be abandoned when stronger studies overturn them. Reading it, I kept thinking, “I used to do that,” or “I remember when that used to be done.” It was humbling.

For years I might have described myself as a consensus-follower: capable of critiquing research rigorously when I had to, but usually willing to accept the prevailing consensus. Politicized issues—climate, gender, COVID—shook me from that “dogmatic slumber,” to borrow Kant’s phrase. I’m now firmly in the reformer camp. I believe science is essential, that we’re not doing it well enough, and that we can do it better.

In addition to consensus-followers, who tend to trust the published literature, and the small but growing ranks of reformers, we have the contrarians. Contrarians are distrustful of the published literature, which is understandable, but that skepticism tends to lead them to reject science altogether, often replacing it with ideology dressed up as evidence.

Those who departed the CDC this week may have been consensus-followers, preferring the caution of following established guidelines. That’s not a failing; it’s the natural posture of most professionals when evidence is uncertain. They may have not been the reformers we really need to move us forward, but their status-quo bias still would have been useful as a bulwark against contrarian forces.

Robert F. Kennedy, Jr., is clearly in that contrarian camp. While I can agree with him on some broad goals—like making America healthier, or questioning unnecessary additives in our food— too often he leaps from reasonable doubt to wholesale rejection. Earlier this year, for example, he announced a halt to funding for mRNA research, then posted hundreds of pages of published “scientific studies”, most irrelevant or misleading. That isn’t reformist critique; it’s contrarian misuse of evidence, and it corrodes public trust.

And that brings us back to the real challenge. Agencies like the CDC don’t operate in a vacuum. They make decisions under pressure, with incomplete data, and under constant political scrutiny. Fixing science will require uncomfortable changes in the institutions that support it—from HHS to NIH to CDC.

Consensus-followers tend to preserve the status quo. Contrarians reject the system outright. What we need are reformers: skeptics whose goal is to make science better, not blindly defend it or blindly destroy it.

Science has never been about certainty. It is provisional truth, always open to revision. That fragility makes it vulnerable to politics—but it is also what makes it worth defending and improving.

His return is encouraging, especially given reports of support from Human Services Secretary Robert F. Kennedy Jr. and colleagues within the FDA. His reinstatement suggests reform is still possible. But the fact that he was pushed out at all is a stark reminder: science is fragile when it threatens entrenched beliefs and self-interest—whether political, ideological, or corporate.

Second Opinion: The MAHA Movement and Lessons of a Forced Resignation

AP News: Ousted FDA Vaccine Chief Vinay Prasad is Returning to the Agency

Science and ideology are not allies. They operate on opposing principles. Ideology thrives on certainty; science survives on doubt and curiosity. Despite popular phrases like “settled science,” science, by definition, is never settled. Its purpose isn’t to confirm dogma—it’s to challenge assumptions and test ideas until they fail. Any claim that cannot be falsified—like a belief in ghosts or reincarnation—may be interesting, but it isn’t science.

Science, when practiced honestly, follows the evidence. And good evidence requires rigorous study design, precise execution, and impartial analysis. Even then, the conclusions are always provisional, subject to revision when better evidence emerges.

Ideologies, by contrast, are systems of belief—sometimes religious, sometimes political, sometimes cultural. Their origins may be ancient or nebulous, but their influence is persistent. They shape our worldview, inform our values, and often resist contradiction. When challenged by evidence, we tend to dismiss the data or discredit the source rather than question our beliefs.

There is absolutely nothing wrong with a having a system of beliefs and opinions—in fact, they are part of what make us human. They are essential for making sense of our lives, helping us navigate meaning, morality, and identity. But ideological beliefs do not belong in the lab.

Science is a relatively new way of understanding the world. Ideology is ancient. Their tension is long-standing. In 1553, Nicolaus Copernicus published his mathematical and astronomical findings supporting a heliocentric view of the universe. Ninety years later, Galileo Galilei was tried for heresy by the Roman Catholic Church, forced to recant his belief that the Earth moves around the Sun, and spent the rest of his life under house arrest.

The details have changed, but the dynamic persists. In recent years, ideologies have once again hijacked scientific discourse. Topics like climate change, gender, and COVID have become ideological battlegrounds. To question prevailing narratives—or even suggest alternative interpretations—is to risk ridicule, censorship, and professional ruin.

Ideological interference in science is not confined to one side of the political spectrum. On issues like vaccines, for example, we’ve seen skepticism emerge from both the right and the left—sometimes for different reasons, but with similarly corrosive effects on scientific discourse.

Ironically, that convergence across traditional ideological boundaries could offer an opportunity. If both sides can acknowledge the damage ideology inflicts on science, there may be a path to common ground—one rooted not in shared beliefs, but in shared desire for good evidence and public benefit.

Science thrives when ideas are tested, not protected. It advances through doubt, not dogma. Ideology demands loyalty. Science demands humility.

When ideology seeps into science, discovery becomes propaganda. Only the “right” research gets funded. Inconvenient findings are buried. Conclusions are shaped before the data are collected. The result isn’t stronger policy—it’s weaker science, lost trust, and slower progress.

If we want science to serve the public good, it must follow evidence—not ideology. And if we want real reform, we must protect those like Dr. Prasad—not because they are always right, but because they are willing to ask the questions that ideology refuses to entertain.

Dr. Vinay Prasad is a well-respected hematologist-oncologist and tenured professor at the University of California, San Francisco who I’ve been following for a number of years. In May, he was appointed Director of the FDA’s Center for Biologics Evaluation and Research (CBER), and a few months later was elevated to a new agency-wide role as Chief Medical and Science Officer. He was an inspired choice for both positions—not only because of his clinical expertise, but also because of his principled critiques of low-quality evidence, performative science, and industry bias.

Despite continued support from Health and Human Services Secretary Robert F. Kennedy Jr., Dr. Prasad resigned from both roles at the FDA yesterday. According to reports, his resignation was not entirely voluntary, but driven by coordinated pressure from political and industry actors. Chief among them was conservative activist Laura Loomer, who absurdly labeled him a “progressive leftist saboteur” of the MAHA movement, and former Republican Senator Rick Santorum, who referred to him as “the man destroying @POTUS legacy for helping patients.”

The purported reasons for these attacks are baffling. Long before his FDA appointment, Dr. Prasad had been critical of COVID lockdowns, mask mandates, and vaccine mandates—positions I also held. Since his appointment, the FDA has modified its COVID vaccine guidance, limiting recommendations primarily to older adults and those with serious health conditions, which I believe is a wise and overdue correction. Perhaps more immediately relevant to his resignation, Dr. Prasad has consistently questioned drug approvals that rely on poor evidence—something I view not as obstruction, but as prudence.

Oddly, the decisions most proximate to Dr. Prasad’s departure both involved treatments for a rare neuromuscular disease, Duchenne muscular dystrophy (DMD). Earlier this month, the FDA rejected an application by Capricor Therapeutics for a biologic called Deramiocel, which consists of progenitor cells derived from donor hearts. That decision was sound: the initial study was poorly designed, underpowered, and failed to demonstrate either efficacy or safety. Importantly, the therapy was not shelved entirely—the FDA simply requested additional data.

Just a week later, the FDA formally asked Sarepta Therapeutics to pause all shipments of their gene therapy drug, Elevidys, amid safety concerns. In June 2023, the FDA had granted limited accelerated approval of Elevidys, despite a lack of evidence of clinical improvement (for those checking my work: the drug showed some benefit on secondary endpoints, but not on its primary one—but that’s a discussion for another time). Since then, despite ongoing study, clear clinical benefit remains elusive. Reports of adverse events—including several deaths—prompted the FDA to request the pause. Shipments were partially resumed on Monday, the day before Dr. Prasad’s resignation

.There are two ways one might view the FDA’s role in regulating pharmaceuticals, biologics, and medical devices. One view holds that the agency is unnecessary—that companies should be free to develop and market their products directly to patients, without regulatory interference. This is a libertarian view which I sympathize with, but think unwise. The other view, which I share, is that regulatory oversight is essential—but only if it is applied rigorously, transparently, and grounded in evidence. Today, we seem to have the worst of both worlds: an agency that postures as rigorous while quietly succumbing to political influence and industry capture.

The FDA, in this case, functioned less as an arbiter of evidence than as a pressure valve—venting public criticism, absorbing industry demands, and ultimately ejecting someone who insisted on applying the same standards across the board. That’s not reform; it’s regression. And it sends a chilling message to others who might try to bring integrity back to the regulatory space: you will not be protected, even if you’re right.

It’s also worth noting who replaced Dr. Prasad—at least in the interim. Dr. George Tidmarsh, now acting director of CBER, is a physician and researcher with an impressive résumé—but also a long career as a biotech executive. He helped bring multiple drugs to market as CEO or founder of companies like Threshold Pharmaceuticals and Horizon Pharma. His appointment, coming immediately after the forced resignation of a regulator known for challenging weak evidence and resisting pressure, suggests a sharp return to business as usual. That’s not a judgment on Tidmarsh personally, but it’s hard to ignore the optics—or the implications for the future of reform.

For the MAHA movement, this is a clarifying moment. It exposes the limits of trying to change institutions from within—at least without sustained public support. It also underscores the importance of building parallel sources of accountability: independent watchdogs, transparent evidence reviews, and public forums where dissent isn’t just tolerated but welcomed. MAHA was never going to be transformative overnight. But if it’s going to grow, it needs to show that what happened to Dr. Prasad wasn’t the end of something—it was the beginning of a much larger reckoning.

A useful overall metric is longevity. In my first 50 years, life expectancy rose from just under 70 years to nearly 79 years by 2010. This was driven largely by improvements in public health, advances in medical care, and declining smoking rates. However, beginning around 2014, life expectancy stagnated and then began to decline, due in large part to rising rates of chronic illness and so-called “deaths of despair,” including drug overdoses (especially opioids), suicides, and alcohol-related disease.

With the arrival of COVID-19 in 2020, American longevity declined even further and has yet to recover to pre-pandemic levels. Drivers of increased mortality include not only the novel coronavirus but also the aforementioned deaths of despair, alongside the growing burden of chronic conditions such as obesity, diabetes, hypertension, and cardiovascular disease.

The obesity rate when I was born in 1960 was about 13 percent. By 2020, 42 percent of Americans were obese. Childhood obesity has increased four- to five-fold since my youth. Driven largely by these rising obesity rates, we’ve seen commensurate increases in hypertension, diabetes, and cardiovascular disease.

Comparing rates of mental illness over time is more difficult due to the stigma that surrounded it in earlier eras and evolving diagnostic criteria. Still, it seems abundantly clear that we are now facing a mental health crisis on par with our physical health crisis. Americans today have a lifetime incidence of depression of 20–25 percent and anxiety of around 30 percent. Our youth have been particularly affected, with rates of anxiety and depression among teens and young adults having doubled over the past decade. Nearly 30 percent of teen girls have seriously considered suicide.

“Deaths of despair” was not a recognized concept in 1960, but deaths from drug overdoses, suicide, and alcohol-related disease were relatively uncommon. By the 2020s, overdose deaths had risen to over 100,000 per year. Suicide deaths have increased by about 35 percent in recent decades, with 49,500 recorded in 2022—the highest number on record. Alcohol-related deaths have also nearly doubled, now contributing to an estimated 140,000 deaths annually.

By any of the above metrics, we are not doing well. Turning the clock back to 1960—at least in terms of health outcomes—would be a welcome change. There are many factors that must be reviewed and addressed. Some are not directly medical, such as economic and social disparities, but others fall squarely into the health bucket, including the quality of our food supply, the adequacy of our public health system, and the effectiveness of our healthcare delivery.

So yes, I welcome a renewed focus on health through initiatives like MAHA. If the primary goal of our federal government is to protect Americans, then it is long past time to address these pressing problems. One significant hurdle this new effort will face is the lack of public trust in the scientific endeavors necessary for real progress. To a considerable degree, I sympathize with that skepticism—which I will explore further in future installments.