You can make someone worse far more easily than you can make them better
Durante décadas, la aspirina diaria fue presentada como un escudo silencioso contra las enfermedades del corazón, adoptada por millones como si la prevención fuera siempre sinónimo de protección. Grandes ensayos clínicos con más de 32.000 participantes han revelado ahora lo contrario: en personas sanas, la aspirina no reduce infartos ni accidentes cerebrovasculares, pero sí provoca hemorragias peligrosas y, en algunos casos, mayor mortalidad. Este giro no condena a la medicina, sino que ilumina una verdad incómoda sobre cómo las prácticas médicas se consolidan no por evidencia, sino por tradición, intuición e incentivos económicos. La pregunta que queda suspendida es cuántas otras recomendaciones rutinarias comparten el mismo origen frágil.
- Millones de personas tomaron aspirina diaria durante años confiando en una promesa de protección que los datos nunca respaldaron con solidez.
- Los ensayos ARRIVE y ASPREE demostraron que el riesgo real —hemorragias cerebrales, sangrados gastrointestinales, mayor mortalidad— superaba cualquier beneficio cardiovascular en adultos sanos.
- El problema no es solo la aspirina: mamografías anuales, suplementos vitamínicos y otras intervenciones preventivas persisten en la práctica estándar pese a evidencia acumulada de daño.
- Los incentivos financieros y la inercia institucional empujan a los médicos hacia la intervención, no hacia la prudencia, distorsionando el juicio clínico.
- La exigencia de evidencia rigurosa —ensayos clínicos masivos con datos reales que demuestren que los beneficios superan los riesgos— sigue siendo la única brújula confiable para la medicina preventiva.
Durante décadas, tomar una aspirina al día pareció una certeza médica inamovible. Millones de personas la incorporaron a su rutina con la misma naturalidad que una vitamina, convencidas de que prevenir era siempre mejor que curar. Luego llegaron los datos.
Dos grandes ensayos clínicos —ARRIVE, con más de 12.000 participantes en siete países durante seis años, y ASPREE, con casi 20.000 personas en Estados Unidos y Australia— llegaron a la misma conclusión perturbadora: la aspirina diaria no protegió a personas sanas de infartos ni derrames cerebrales. Sí causó daño medible. Hemorragias cerebrales, sangrados gastrointestinales, y en el caso de ASPREE, una mayor tasa de mortalidad en el grupo que tomaba aspirina frente al grupo placebo.
Este giro no es una condena a la medicina, sino una radiografía de cómo funciona en realidad. La medicina no es ciencia pura: es una empresa híbrida donde la evidencia comparte espacio con la tradición, la intuición y los incentivos económicos. La lógica de intervenir en personas sanas para prevenir enfermedades suena razonable, pero esconde una trampa: introducir riesgo en un sistema que funcionaba bien puede hacer más daño que bien con una facilidad que suele subestimarse.
El caso de la aspirina no es una anomalía. Mamografías anuales, suplementos vitamínicos y múltiples pruebas de detección forman parte de la práctica médica estándar pese a evidencia creciente de que perjudican a más personas de las que ayudan. Estos hábitos persisten en parte por tradición y en parte porque son rentables: los centros médicos se benefician económicamente de los procedimientos de detección, y los médicos son evaluados según cuántos pacientes convencen de someterse a ellos.
El estándar debería ser el mismo que se aplicó a las vacunas contra el COVID-19: ensayos masivos seguidos de datos reales que demuestren que los beneficios superan ampliamente los riesgos. La aspirina diaria para personas sanas no supera esa prueba. Tampoco muchas otras recomendaciones rutinarias. Cuando un médico sugiere un medicamento o procedimiento a alguien que está bien, la pregunta correcta es simple: ¿dónde están los datos del ensayo clínico? La vieja máxima médica —si no está roto, no lo arregles— puede ser, paradójicamente, el consejo más basado en evidencia que la medicina tiene para ofrecer.
For decades, the advice seemed settled: take an aspirin a day, and your heart will thank you. Millions of people followed this guidance, swallowing the small white tablet as casually as a vitamin, trusting that prevention was always better than cure. Then, in recent years, large clinical trials began to tell a different story—one that exposes a deeper problem in how medicine actually works.
Two major studies, one called ARRIVE and another called ASPREE, followed tens of thousands of healthy people across multiple countries over years. ARRIVE tracked more than 12,000 participants from seven nations for over six years. ASPREE enrolled nearly 20,000 people in the United States and Australia. Both arrived at the same unsettling conclusion: daily aspirin provided no measurable protection against heart attacks or strokes in people who had never experienced either. But it did something else. It caused real, measurable harm. Dangerous bleeding emerged as a consistent risk—cerebral hemorrhage, gastrointestinal bleeding, bleeding that could kill. In the ASPREE trial, more people in the aspirin group died during the study period than those taking a placebo.
This reversal is not, as some might frame it, an indictment of medicine itself. Rather, it is a window into how medicine actually functions—and how it often fails. Medicine is not purely a science. It is a hybrid enterprise, blending scientific evidence with tradition, intuition, financial incentive, and the deeply human desire to help. Physics has no need for a "science-based physics movement" because physics is already science. Medicine requires one because medicine is something else entirely.
The pattern is recognizable once you see it. Doctors recommend interventions to healthy people based on a logic that sounds reasonable: prevent disease before it starts. But this logic contains a hidden trap. It assumes that doing something is always safer than doing nothing. In reality, when you intervene in a healthy person's body, you are introducing risk into a system that was functioning. You can make someone worse far more easily than you can make them better if they were never sick to begin with. For years, physicians believed they should wait for proof that dietary fat was harmful before removing it from dietary guidelines. The safer course would have been to leave people eating as they had for millennia, and change nothing without solid evidence.
The aspirin case is not isolated. Screening tests for cancer, annual mammograms, vitamin supplements—these interventions are woven into standard medical practice despite accumulating evidence that they cause more damage than benefit for many people. Mammograms continue to be recommended annually even as evidence mounts that they may harm more women than they help. Vitamin studies have found that people taking supplements actually developed cancer more frequently than those taking placebos. These practices persist partly because they are tradition, partly because they are profitable. Medical centers benefit financially from screening procedures. Individual physicians often receive ratings and reviews from their health systems based on how many patients they convince to undergo these tests. The incentive structure points toward intervention, not restraint.
The standard for preventive treatment should be the same as it is for COVID-19 vaccines: enormous clinical trials, followed by large-scale real-world data showing that while harms may occur, they are rare enough that benefits overwhelmingly outweigh risks. Daily aspirin for healthy people fails this test. So do many other routine recommendations. Yet the system continues to push them, driven by habit and financial logic rather than evidence.
The lesson is not that medicine should never treat healthy people. It is that when it does, the evidence must be rigorous and the bar must be high. If you are well and a doctor recommends any medication or procedure, ask for the clinical trial data. Ask whether the benefits have been proven to outweigh the risks. Ask whether the recommendation is based on science or on tradition, intuition, or the financial interests of the institution recommending it. The old medical wisdom—if it is not broken, do not fix it—may be the most evidence-based guidance medicine has to offer.
Notable Quotes
It is difficult to make a healthy person better, but it is not so difficult to make them worse.— Dr. Gilbert Welch, author of 'Less Medicine, More Health'
Medicine is not exactly a science—it is a combination of things, with some science in the mix.— Faye Flam, the article's author
The Hearth Conversation Another angle on the story
Why did doctors recommend daily aspirin for so long if the evidence was already building against it?
Because medicine does not move like science does. A recommendation becomes standard practice, gets embedded in guidelines, and then inertia takes over. Doctors believed they were being cautious and preventive. No one was lying. They just were not waiting for proof before acting.
But these were large trials—32,000 people. That is substantial evidence. Why did it take so long to reverse course?
The trials are recent. ARRIVE and ASPREE were conducted over years, and the results took time to accumulate and be understood. But also, reversals are uncomfortable. They suggest past practice was wrong. There is resistance to that.
The article mentions that doctors get rated based on how many patients they screen. That seems like a perverse incentive.
It is. The system rewards volume and intervention, not restraint. A doctor who convinces ten patients to get mammograms looks more productive than one who tells ten patients they do not need one. The financial incentive and the medical incentive are misaligned.
So what should a healthy person actually do when a doctor recommends something preventive?
Demand the evidence. Ask for the clinical trials. Ask whether the benefits have been proven to outweigh the harms. If the answer is "it is standard practice" or "it is better to be safe," that is not enough. COVID vaccines passed that test. Daily aspirin did not.
Is there ever a time when doctors should act without solid evidence?
Yes. When someone is dying and there is no other option. Early AIDS treatment was like that. You might try something unproven because the alternative was certain death. But for prevention in healthy people, that logic does not apply. There is always the option of doing nothing.
What worries you most about how this plays out?
That people will lose trust in medicine entirely, and that would be wrong. Medicine does tremendous good. The problem is not medicine. It is medicine without evidence, applied to people who do not need it, driven by incentives that have nothing to do with health.