First Opinion is STAT’s platform for interesting, illuminating, and provocative articles about the life sciences writ large, written by biotech insiders, health care workers, researchers, and others.
To encourage robust, good-faith discussion about issues raised in First Opinion essays, STAT publishes selected Letters to the Editor received in response to them. You can submit a Letter to the Editor here, or find the submission form at the end of any First Opinion essay.
The story
“Are blue zones real? Answering that question is harder than ever,” by Shelley Wood and Eric J. Topol
The response
I’ve spent the past 20 years writing for National Geographic, during which time I’ve written four major articles about blue zones, the world’s statistically longest-lived hot spots. I broke the story announcing their discovery, explained the methods used to validate them and suggested the lessons the rest of us might learn from them. At National Geographic the fact checkers occupy corner offices; editors force journalistic integrity.
The most surprising thing about STAT’s recent article questioning the validity of blue zones wasn’t that it featured critics. Scientific criticism is essential. Any serious idea that has attracted global attention should be scrutinized, challenged, and tested.
What surprised me was what the article left out. Readers were presented with a lengthy retelling of a sole critic who argues that blue zones are built on flawed demographic assumptions, questionable age validation, and statistical anomalies.
Yet nowhere in the article were readers informed that most of those same criticisms were directly addressed and disproven in a peer-reviewed paper published in The Gerontologist, one of the leading journals in aging research. Whether one agrees with that response is beside the point. The point is that it exists. When covering a supposed scientific controversy, it’s hard to justify giving so much space to one side while leaving out a major peer‑reviewed response that challenges whether there’s a real dispute at all.
This omission is not a minor detail. It fundamentally alters how readers understand the state of the evidence. The article leaves the impression that blue zones researchers have been confronted with devastating new criticisms to which there has been little meaningful response. That is simply not true. There are over a dozen peer reviewed articles by career demographers documenting the validity of blue zones; the STAT article only presents the arguments from a lone critic who continuously fails peer review.
More importantly, the article repeatedly conflates two very different subjects: blue zones research and Blue Zones Project. The distinction matters because one is a scientific inquiry into places where people have historically lived exceptionally long lives, while the other is a community well-being initiative designed to improve health outcomes in modern cities. The article spends considerable time questioning the first while implying doubts about the second. But those are entirely separate questions.
What has always struck me as odd about the recent attack on blue zones is 1) they compile misleading straw man arguments while conveniently omitting the fact career demographers spent years painstakingly verifying ages, and 2) the zeal with which the gullible popular media parrot the same misleading claims.
The authors of the STAT article seem to have fallen prey to the same clickbait.
Readers were not told that a peer-reviewed response to the criticism exists. They were not told that this isn’t an active scientific debate so much as one being presented that way. And they were given little information about the tangible, systems-level improvements in people’s lives that Blue Zones Project has produced in communities where it’s been put into practice.
The result is a story that asks whether blue zones are real while largely overlooking the evidence that matters most. For me, the most important question is how we can apply what we learn from the world’s healthiest populations to drive system‑level changes that help people live better, healthier lives.
The original blue zones work began with a simple observation: Certain populations appeared to produce unusually high concentrations of healthy older people. Working alongside respected demographers, epidemiologists, and researchers, we sought to understand what factors might contribute to that phenomenon. In all the blue zones, our central observations remain remarkably consistent: People tend to live longer when they move naturally throughout the day, maintain strong social connections, avoid smoking, eat mostly whole foods, and have a sense of purpose. Their secret? They live in environments that make these healthy choices easy or unavoidable. These principles are supported by decades of public health and epidemiological research that extends far beyond any single longevity hot spot.
Yet the larger omission in the STAT article concerns outcomes. If writers want to evaluate Blue Zones Project, then the relevant question is not whether academics continue to debate demographic records in Okinawa or Sardinia. The relevant question is whether communities that implement Blue Zones principles become healthier.
That question has actually been studied.
For years, Gallup measured well-being in communities where Blue Zones Project was implemented. In Fort Worth, Texas, Gallup documented substantial improvements in population well-being, healthy behaviors, and quality of life. During a period when many American communities struggled to move these indicators in a positive direction, Fort Worth became one of the strongest examples of large-scale improvement. Similar gains were recorded in California’s Beach Cities, where residents consistently reported levels of well-being that exceeded national averages and where long-term improvements were observed across multiple health and quality-of-life measures.
These findings are not promotional slogans. They are population-level measurements collected over years. They represent the type of evidence public health experts routinely seek when evaluating whether an intervention is working.
Yet readers of the STAT article learn almost nothing about those results.
Instead, they are encouraged to view blue zones primarily through the lens of a sloppy critique that has already been discredited.
It is also important to correct another common misconception. Cities do not pay Blue Zones Project for certification. Communities engage Blue Zones Project because they need support in improving the social and environmental conditions that shape health — the policies, schools, workplaces, food environments, social networks, and built environments that determine how easy or hard it is to live well in a given ZIP code. The objective is not branding. The objective is measurable improvement in population health and well-being by comprehensively making the healthy choice the easy choice. We are proud that the Blue Zones community model stands as one of the most meaningful innovations in population health.
That is the work Blue Zones has always been about. And that is the question readers deserve to see examined with the same rigor applied to the controversy.
— Dan Buettner
The response
My country of origin can be considered what is called a Blue Zone. There were and still are members of the population who lived to their 90s and beyond. I have had relatives in this category, but unfortunately it seems the next generation will not. Centenarians were visited by an official of government, and this event was published in the news, so records of these milestones exist and this is still ongoing.
The threat to this way of life is movement of people to and from these places. Some may call it progress. As a child, in my old country there were fewer than 20,000 cars, now it has increased to 180,000. Processed food has increased, but fresh food is still abundant. Martinique was mentioned in the article, but please note, the wider Caribbean including Barbados have lived like this for many years but may not in the future.
— Loretta Cuffley
The response
Leaving aside the controversy as to whether these particular locales qualify, why would one not expect population isolates exhibiting increased or decreased aging simply due to genetic variance alone? Such hot spots of related individuals with a specific haplotype may correlate with healthy lifestyles or may not, as shown by centenarians who used alcohol or were sedentary for much of their lives. The conclusion that lifestyle is responsible for an as yet unreplicated finding seems premature.
— Victor Reus, M.D., UCSF School of Medicine
The story
“Ending animal testing could set back xenotransplantation just as the field is poised for a breakthrough,” by Joshua Mezrich
The response
The road map for phasing out animal experimentation is grounded in two interrelated considerations: the moral recognition of animal suffering and the growing evidence that animal models are often unreliable predictors of human biology. Together, these ethical and epistemic critiques underpin current efforts to reduce and ultimately replace animal use in science. Xenotransplantation disrupts this trajectory.
As Mezrich notes, constraints on animal-based research may hinder xenotransplantation. However, the field is structurally dependent on extensive animal use across its entire translational pipeline. Rodents are used to study immune rejection, inflammatory pathways, coagulation incompatibilities, and graft survival. Gene-edited pigs are generated through genome engineering and somatic cell nuclear transfer, requiring oocyte donor animals, surrogate sows, and additional populations to establish and maintain breeding lines. Large pathogen-free herds are sustained for selective breeding and organ production. Before clinical application, non-human primates such as baboons and macaques serve as recipients in preclinical transplantation studies to assess graft function and long-term survival.
Across these stages, animals function simultaneously as experimental models, reproductive resources, breeding populations, and organ sources. If translated into clinical practice, this system extends beyond research and becomes embedded in health care delivery, where animal bodies are continuously produced and consumed as therapeutic infrastructure.
Beyond animal ethics, xenotransplantation raises questions of distributive and clinical justice. It is unclear which patients will receive animal-derived organs and which will receive human organs. If human organs remain clinically preferable, xenotransplantation may generate stratified access to transplantation. Concerns regarding zoonotic risk may also require long-term surveillance, monitoring, and behavioral constraints for recipients of animal organs.
The policy commitment to reduce and replace animal experimentation reflects an established ethical and scientific direction in which animal use should not constitute the default framework of biomedical research. Xenotransplantation departs from this trajectory by reestablishing animal use as a central and permanent component of medicine. It institutionalizes a system in which animals are deliberately engineered, bred, confined, and killed as renewable biological resources for human therapeutic ends.
This transformation represents a substantive reconfiguration of the human–animal relationship in medicine and sits in tension with contemporary science policy objectives. Although citizen interviews, focus groups, and deliberative studies on xenotransplantation are limited and not systematically updated, they consistently report moral resistance to animal instrumentalization, alongside concerns about genetic modification, intensive breeding, killing, and precautionary risks. Xenotransplantation therefore remains ethically contested and lacks stable social consensus, warranting explicit democratic deliberation rather than implicit normalization.
— Arianna Ferrari, Center for Innovation Systems and Policy, Austrian Institute of Technology
The story
“We published in Nature Medicine in 2025 for free. In 2026, it cost us $12,850,” by Elizabeth Selvin
The response
June 11’s First Opinion by Elizabeth Selvin represents common misunderstandings of the economics of scientific publishing today. We would like to clarify some points to further the conversation.
False: Not paying an article-processing charge (APC) to be published means being published without one was “free.”
Fact: There is a cost either way, and hybrid (subscription + free articles) journals like Nature Medicine need to split the difference.
False: The publisher is to blame for the “new” cost.
Fact: NIH announced in April 2025 that the policy would be effective July 31, 2025. The paper in question was submitted Aug. 7, 2025.
False: The cost was unavoidable.
Fact: If the cost of open-access fees is a financial hardship, there is a global shopping mall at your fingertips. Citations and press coverage for a paper in Nature Medicine pay dividends in academic credibility and further funding opportunities. Everyone knows that.
False: Paying $13,000 is too much.
Fact: The return on investment on a publication fee that is a percentage of an R01 grant is relative.
False: PubMed Central sans embargo is the way to go.
Fact: Congress allocates tax dollars for research grants. Reports of this research may turn into scientific papers. Publishers independently assess these reports and take financial risks for their authors. PubMed Central is a secondary U.S. taxpayer-funded platform created to support “public access” compliance. Like any tech platform, it is expensive to run and maintain. It hosts content from authors and publishers across the globe while U.S. taxpayers foot the bill. Some estimates suggest that the budget to run it has quadrupled in recent years. Like everything else, it is subject to political crosswinds.
We understand Selvin’s point of view. Blaming independent commercial publishers for trying to run their businesses while an untrustworthy U.S. government changes policies on a whim, however, is just the tip of a messy iceberg that op-eds like hers make messier. Comparing the “costs” of publishing in Nature Medicine to what she has gleaned from her time as a part-time editor for an American Diabetes Association journal is comparing apples and oranges. It is of no help to her fellow U.S. scientific colleagues who are getting squeezed.
— Kent Anderson and Joy Moore, co-authors, “How the Internet Disrupted Science” (Globe Pequot, August 2026)
The response
Selvin is right that publicly funded research should be widely accessible to support knowledge and discovery, a goal Springer Nature shares. However, the piece misses important distinctions, particularly around article processing charges (APCs). Publishing is both resource intensive and high value. Under a subscription model, institutions and libraries that subscribe to journals cover the cost. Publishing open access without embargo periods — which NIH policy requires — shifts those costs to APCs, which are typically covered by research funds. Our APCs are transparent and reflect the costs associated with publishing high-quality trusted research. APCs support editorial evaluation, independent review, formatting, and distribution, and ensure continuous access to publications. They also allow for investment in research integrity infrastructure, including advanced detection of fraudulent submissions — essential to ensuring published science can be trusted.
Springer Nature’s average APC is $2,500. For high-impact, selective journals like those in the Nature portfolio, costs are higher due to the time, expertise, and investment required. These journals receive a very high number of submissions, meaning specialist teams assess many more papers than are ultimately published, resulting in higher costs per article. Our authors retain copyright in their work. We are granted a license to publish by the author so we can ensure their work reaches the widest number of people and has the greatest impact. Licensing terms help protect authors’ work from reuse in ways that are not transparent or aligned with their intentions. Access to trusted research is central to scientific and societal progress. We are committed to partnering with researchers and the federal government to support this shared goal.
— Anna Troise, Springer Nature
The story
“Dr. Glaucomflecken wants to make a stink,” by Torie Bosch
The response
I just want to thank you for airing this story. Unchecked capitalism is making all of us poorer in actual wealth and also in services and protections that matter (unless we are in the 0.1%). Let’s link corruption, “lawful” or unlawful, with a poorer outcome for all of us because that is the truth. I am so sad that doctors are pressured or coerced by money to take jobs in unbridled capitalistic ventures that suck money from us all. I am even sadder that young docs assume this is how it will be. I am a psychiatrist in private practice for several decades save for a five-year stint at an aspiring startup. I told myself I could work there because it is a “good version” of what is coming anyway. Sadly, it probably is a better version. But any practice having non-medical stakeholders can’t also care about patients and docs more than giving that concern lip service. It’s like that old saying: You can have two out of three, but you can’t have all three. One example: A patient was charged a no-show fee even though her “excuse” was “I was sitting with my husband who died the day of my appointment.” If I didn’t see it with my own eyes I wouldn’t have believed it. Please, please, please let us as physicians push back on big business which is making us all poorer. Young people: Don’t assume it has to be this way.
— Jody Whitehouse
The story
“AI doctors should be licensed. Here’s a framework to do that,” by Alon Bergman
The response
Alon Bergman’s thoughtful essay highlights the urgency of building a coherent federal framework for autonomous clinical AI. Yet the Utah pilot and the broader national debate reveal a deeper governance gap that precedes questions of licensure or competency testing. Before we can responsibly license an “AI doctor,” we must first regulate what an AI system is allowed to be perceived as.
Recent events in Pennsylvania underscore this point. There, a consumer chatbot allegedly adopted the identity of a licensed psychiatrist and provided treatment‑style guidance under a fabricated license number. In Utah, the state’s regulatory sandbox authorized an autonomous system to renew prescriptions for nearly 200 medications, with plans to phase out physician review. These two cases appear different — one is misrepresentation, the other delegation — but they expose the same missing pillar: We lack a standard for how AI systems may present themselves, what authority patients are led to attribute to them, and who is accountable once a patient relies on that interaction.
This gap matters because AI‑mediated medical advice becomes unsafe long before we evaluate the accuracy of the advice itself. Without clear rules for identity, accountability, informed consent, and real‑world monitoring, patients may reasonably assume that an AI system is licensed, supervised, or federally validated when none of those conditions are true.
The result is a distortion of trust that neither state‑level experimentation nor federal device regulation is currently equipped to address. Bergman is right that the FDA’s static device framework cannot keep pace with adaptive models, and that a national competency standard is needed. But competency alone is insufficient. A licensing pathway must be paired with guardrails that define how AI systems communicate their role, how responsibility is shared between developers and deploying institutions, and how patients are informed in plain language, when an autonomous system is acting in place of a clinician.
Without these protections, even a highly capable model risks being deployed in ways that outstrip public understanding and erode trust. The promise of AI in addressing workforce shortages is real, particularly for routine, lower‑risk tasks. But safe integration requires more than demonstrating that AI can perform at or above the median USMLE score. It requires a governance structure that ensures patients understand who is caring for them, what these softwares can do, what the system is authorized to do, and who stands behind its decisions. Until that foundation is in place, the central question is not simply whether AI is competent, but what the AI leads the patient to believe and who is responsible for that belief.
— Ediriweera Desapriya, Department of Pediatrics, Faculty of Medicine, University of British Columbia
The story
“Tributes to Craig Venter and the genomics race are missing something important,” by Zachary Utz
The response
I loved the message and the balance of this article. It in no way diminishes Venter’s creativity and impact while acknowledging that hundreds of scientists, many NIH funded, but also from the U.K., Japan, France, etc., were critical. They also generated not just sequence data but shared technology, physical reagents, and genetic maps that built a scaffold onto which sequence data from many sources could be overlaid. Adding in the many non-U.S. contributors going back to at least the 1970s might even more fully complete the picture.
— Jeff Murray, University of Iowa
Source: www.statnews.com
