What Adam Is Reading
Nine Reversals, One Post
The X post in full, what each episode actually was, and what fixed it. A companion to this week's Anti-Anti-Science section.
Source check · 10 episodes · October 2026

The post below went around X in late October 2026. It is reproduced in full so you can read it before you read anything I say about it. Below the post, each episode gets a card with three parts. How the practice arose and who pushed it. What the later data showed. And what changed as a result, because that last part is the half of the story the post leaves out.

Each card carries a verdict on the post's version of events, using the usual WAiR scale. Solid means the post has it right. Mostly Solid means the facts are right and the framing is doing extra work. Embellished means the post's version does not survive the sources.


The post
Sama Hoole (@SamaHoole) on X, October 29, 2026. Reproduced for commentary.

The year is 1873. Your wife tires easily and reads too much. The doctor prescribes the rest cure: six weeks flat in bed, no books, no visitors, fed on milk, forbidden to sit up without his permission. The man who devised it is the most famous nerve specialist in America, and the women come out of it quieter, which is taken as the cure.

The year is 1915. The hospital has a new way of delivering your child. Morphine, then scopolamine, and the mother wakes two days later with a baby she cannot remember having. They call it twilight sleep. The women are strapped into padded cots so they do not hurt themselves while they scream, and the straps are the modern part. The babies arrive blue and slow to breathe.

The year is 1945. The truck comes down your street fogging DDT, and the children run behind it in the white cloud because the man on the back says it is harmless. At the beach it is sprayed over the bathers on their towels. It goes on the cattle, over the barn, and across the kitchen table.

The year is 1948. Your son has trouble with his ears. The doctor slides a rod tipped with radium up each nostril and holds it there for the best part of ten minutes, and you bring him back three more times. Hundreds of thousands of American children get the same treatment. The cancers arrive in their fifties.

The year is 1962. The health visitor tells you breastfeeding is old-fashioned and the formula is scientific. By 1972 fewer than one American baby in four is breastfed at all. The formula is skimmed milk and sugar, with the milk fat taken out and vegetable oil put in.

The year is 1971. The family planning clinic fits a small plastic crab called the Dalkon Shield. Its tail is a bundle of fibres that wicks bacteria up into the womb. 2.8 million are fitted in America before it is withdrawn, and the company goes bankrupt under the claims. It had been sold as safe and superior.

The year is 1983. The doctor says your arteries are the butter's fault. The dripping tin goes in the bin, the yolks go to the dog, the margarine goes in the fridge, and you have the heart attack anyway, on skimmed milk, having done everything right.

The year is 2001. Your 15-year-old is low, and the doctor prescribes Paxil, because the trial in the journal says it is effective and well tolerated in adolescents. The raw data, reanalysed 14 years later, shows more suicidal thinking on the drug than on the sugar pill. The company pays $3 billion to settle. Nobody rings the house.

The year is 2009. The government has spent around £500 million stockpiling Tamiflu for the pandemic, on the strength of the manufacturer's trials. When the trials are finally prised loose five years later, the drug shortens flu by about half a day and nothing shows it keeps anyone out of hospital. The stockpile goes out of date in a warehouse.

The year is 2026. The doctor has ten minutes and a screen. The number is high, so the tablet is for life. The weight is high, so the injection is for life. The meat is bad for the planet and the oil is good for the heart, and the person in the chair is as certain as everyone who ever sat in it.

Every one of those was settled, by a committee, in a journal, with a doubter somewhere being laughed at. The rest cure, the straps, the fog, the rod, the formula, the crab, the margarine, the trial, the stockpile. All of it certain, all of it printed, all of it done to somebody who had asked for help.

You were never the one who was wrong. You were the one it was done to, and the only person in the story who did as they were told every single time.

The apology comes a generation late, in a footnote, addressed to nobody.

The next one is being drafted now.

One note on the author. Sama Hoole is a UK fitness blogger who advocates animal based diets and writes against veganism and seed oils. This is one of a series of posts in the same format (an earlier version opened with Bayer selling heroin as a children's cough syrup in the 1890s). The medical history is the runway. The 1983 and 2026 paragraphs are the destination.


The episodes, one at a time
1
The rest cure (1870s to 1920s)
How it arose

Silas Weir Mitchell, a Philadelphia neurologist and the leading American nerve specialist of his day, devised the treatment for "neurasthenia" and hysteria. The theory held that overstimulation depleted nerve force, so the prescription was weeks of enforced bed rest, isolation, a milk diet, and no reading or writing. It was adopted widely in elite practice on Mitchell's reputation and his case reports. There was no controlled evidence.

What later data showed

Patients often deteriorated or became dependent, and the most famous of them, Charlotte Perkins Gilman, wrote The Yellow Wallpaper (1892) about the experience. The treatment faded as psychiatry moved toward psychological and social models in the early twentieth century.

What changed

Nothing regulatory. This predates any framework for evaluating a treatment. The rest cure is the clearest case on the list of authority standing in for evidence.

Solid
The post has the facts right. The dates, the milk, the ban on books, the quieter women.
References

APA Monitor, "Go rest, young man" (2012) · Anne Stiles, "The Rest Cure, 1873 to 1925," BRANCH · "The Rest Cure," PMC

2
Twilight sleep (1910s to 1920s)
How it arose

Developed at the Freiburg women's clinic in Germany (morphine plus scopolamine, producing amnesia rather than true analgesia). American women's magazines and a feminist campaign brought it to the United States in 1914 as the modern, painless birth. Hospitals adopted it quickly. The evidence was case series from Freiburg and testimonials.

What later data showed

Mothers were disoriented and restrained, labor was prolonged, and newborns had respiratory depression from the opioid. A prominent advocate, Frances Carmody, died in childbirth in 1915 and the movement collapsed in the press, though scopolamine persisted in obstetrics into the 1960s. It was displaced by regional anesthesia and better neonatal care.

What changed

Obstetric anesthesia became its own discipline with its own outcome measures, including neonatal scoring (the Apgar score arrived in 1952).

Solid
Accurate. "The straps are the modern part" is editorializing, and fair.
References

Embryo Project Encyclopedia, "Twilight Sleep" · Leavitt, "Birthing and Anesthesia: The Debate over Twilight Sleep," Signs (1980)

3
DDT fogging (1940s to 1972)
How it arose

DDT stopped typhus and malaria epidemics among troops and civilians in the Second World War, and Paul Müller won the 1948 Nobel Prize for discovering its insecticidal properties. Early toxicology showed low acute toxicity in humans. Civilian use followed immediately, including fogging trucks, aerial spraying, and the famous 1945 photos of bathers at Jones Beach. Persistence and bioaccumulation were not yet understood.

What later data showed

DDT and its metabolite DDE accumulate in fat and up the food chain, thinned raptor eggshells, and persist for decades. Occupational and cohort studies found associations with some cancers, and IARC classified DDT as probably carcinogenic to humans in 2015. Acute human harm from the fogging itself was rare, which is why the man on the truck could say it was harmless and be, in his narrow sense, right.

What changed

Rachel Carson's Silent Spring (1962), then the EPA cancellation of most DDT uses in 1972, two years after the agency was created. Environmental persistence became a standard part of pesticide review.

Solid
The imagery is accurate. The harm was mostly ecological and delayed, which the post leaves vague.
References

EPA, "DDT: A Brief History and Status" · IARC Monographs, DDT and lindane, Lancet Oncology (2015)

4
Nasal radium (1940s to 1960s)
How it arose

Nasopharyngeal radium irradiation was developed at Johns Hopkins in the 1920s to shrink lymphoid tissue around the eustachian tube, treating recurrent ear infections and hearing loss. The military used it for aviators and submariners with barotrauma. Civilian use spread in the 1940s and 1950s on small case series showing symptom improvement. Late radiation risk at these doses was not characterized.

What later data showed

Cohort studies in Maryland and the Netherlands found elevated risk of head and neck cancers and some other late effects in treated patients, though the absolute numbers were small and the studies were limited. No registry exists. The CDC and NCI estimate that hundreds of thousands to over a million Americans were treated. NCI concluded there was no basis for routine screening of exposed people.

What changed

The practice stopped as antibiotics and tympanostomy tubes replaced it and as radiation dose standards tightened. The CDC published patient and physician fact sheets in the 1990s after a Senate hearing.

Mostly Solid
The procedure is described accurately. "The cancers arrive in their fifties" states a modest, uncertain excess risk as a certainty.
References

CDC, "Facts About Nasopharyngeal Radium Irradiation" · MMWR, Nasopharyngeal radium irradiation (1996)

5
Formula over breastfeeding (1950s to 1970s)
How it arose

Commercial formula was marketed as modern and scientific, pediatric authorities often endorsed it, and hospital routines (separating mother and baby, scheduled feeds) made breastfeeding harder. Breastfeeding initiation in the United States fell to about 22 percent in 1972, the low point. Formulas of the era were built on skim milk with added vegetable oils and carbohydrate, which is roughly what the post describes.

What later data showed

Breastfeeding is associated with fewer infections, less sudden infant death, and some longer term benefits. The strongest evidence (the PROBIT cluster randomized trial and the 2016 Lancet series) shows large benefits in low income settings and more modest ones in high income settings. Formula itself is not a toxin, and its composition has improved repeatedly.

What changed

The WHO International Code of Marketing of Breast milk Substitutes (1981), the Baby Friendly Hospital Initiative (1991), and FDA formula standards under the Infant Formula Act (1980). US initiation rates are now above 80 percent.

Mostly Solid
The numbers hold. The harm is implied rather than stated, and for most babies in rich countries it was modest.
References

Institute of Medicine, Nutrition During Lactation, historical trends · Victora et al., Lancet breastfeeding series (2016)

6
The Dalkon Shield (1971 to 1974)
How it arose

A.H. Robins bought the device from its inventors and marketed it as safer and more effective than the pill. The multifilament tail string, which wicked bacteria into the uterus, was never tested for that. Devices did not require premarket approval at the time, so the only check was the company's own.

What later data showed

High rates of pelvic inflammatory disease, septic abortion, infertility, and some deaths. Roughly 2.8 million were placed in the United States before Robins suspended sales in 1974. Litigation drove the company into bankruptcy in 1985 and a trust paid claims for decades.

What changed

The Medical Device Amendments of 1976, which created the device classification system and premarket review that still governs implants. Every IUD since has had to prove itself before sale.

Solid
Accurate throughout. The post omits that this is the episode that created device regulation.
References

AMA Journal of Ethics, "FDA Device Oversight from 1906 to the Present" · Dalkon Shield, overview and litigation history

7
Butter, margarine, and the low fat era (1970s to 1990s)
How it arose

Ancel Keys's ecological data, early cohort studies, and American Heart Association guidance tied saturated fat to coronary disease. Guidelines told people to cut butter and eggs. Industry responded with partially hydrogenated vegetable oils, which were cheap, shelf stable, and loaded with trans fat.

What later data showed

Trans fats raised cardiovascular risk more than the saturated fat they replaced. Replacing saturated fat with refined carbohydrate did nothing. But the core claim survived. The 2020 Cochrane review of 15 randomized trials (about 59,000 participants) found that cutting saturated fat reduced cardiovascular events by about 17 percent, with no effect on total mortality. The error was the substitute, not the advice to cut butter.

What changed

Mandatory trans fat labeling (2006) and the FDA determination removing partially hydrogenated oils from the food supply (2015, effective 2018). Dietary guidelines now say what to replace saturated fat with.

Embellished
"You have the heart attack anyway, having done everything right" implies butter was vindicated. The randomized trials say otherwise. This is the paragraph the author cares about most and the one with the least behind it.
References

Hooper et al., Cochrane review, saturated fat and cardiovascular disease (2020) · Mozaffarian et al., "Trans Fatty Acids and Cardiovascular Disease," NEJM (2006)

8
Paxil in adolescents, Study 329 (2001)
How it arose

GlaxoSmithKline's Study 329, published in 2001 in the Journal of the American Academy of Child and Adolescent Psychiatry, reported paroxetine as effective and well tolerated in adolescent depression. The paper was ghostwritten, the primary endpoints had failed, and the positive findings came from outcomes added after the fact. It was cited widely and the drug was promoted off label for teenagers.

What later data showed

The 2015 reanalysis of the raw data under the RIAT initiative (Restoring Invisible and Abandoned Trials), published in BMJ, found no efficacy advantage over placebo and more suicidal ideation and behavior on paroxetine. GSK's 2012 settlement of $3 billion was the largest health care fraud settlement in US history. It covered off label promotion of Paxil and Wellbutrin and the withholding of safety data on Avandia, so the figure is not Paxil's alone.

What changed

FDA black box warnings on antidepressants in young people (2004), the ClinicalTrials.gov results reporting requirement (2007), and the data sharing norms that made RIAT possible. Study 329 is the standard example in every course on trial transparency.

Mostly Solid
The core is right. "$3 billion" is a composite across three drugs, and "nobody rings the house" skips the 2004 black box, which was the system ringing, badly.
References

Le Noury et al., "Restoring Study 329," BMJ (2015) · US Department of Justice, GSK settlement press release (2012)

9
Tamiflu stockpiles (2009 to 2014)
How it arose

Governments stockpiled oseltamivir for pandemic preparedness on the strength of a 2003 industry funded meta-analysis claiming reduced complications. Eight of its ten trials were unpublished. The UK spent about £424 million, the US about $1.5 billion. During the 2009 H1N1 pandemic a Japanese pediatrician pointed this out in a comment on the Cochrane review, and the reviewers asked Roche for the data.

What later data showed

After a four year campaign with BMJ, Roche released the full clinical study reports in 2013. The 2014 Cochrane review, built on about 160,000 pages, found Tamiflu shortened symptoms by about 17 hours in adults, with no good evidence of reduced hospitalization or complications and with more nausea and vomiting. The trials were never designed to answer the hospitalization question, which is why "nothing shows" is the accurate phrasing and "shows nothing" would not be.

What changed

The European Medicines Agency's clinical data publication policy (2015), the AllTrials campaign, and a UK Public Accounts Committee inquiry into stockpiling decisions. Regulators now expect full study reports, not summaries.

Mostly Solid
The numbers are close (£424 million, not £500 million; 17 hours, not quite half a day). The post leaves out that reviewers inside the system are who exposed it.
References

Jefferson et al., Cochrane review of neuraminidase inhibitors (2014) · BMJ, the Tamiflu campaign · The Conversation, "Tamiflu cost us £424m" (2013)

10
The tablet, the injection, and the oil (2026)
What stands behind it

Statins have been tested in dozens of randomized trials with heart attacks and death as the endpoints. The Cholesterol Treatment Trialists' meta-analysis of 26 trials and 170,000 participants found a 22 percent reduction in major vascular events per 1 mmol/L of LDL lowering. Intensive blood pressure control reduced mortality in SPRINT (9,361 participants). Semaglutide reduced major cardiovascular events by 20 percent in SELECT (17,604 participants with established cardiovascular disease and obesity, median follow up over three years). None of the nine earlier episodes had evidence of this kind before it reached patients.

What is still open

The long term effects of GLP-1 drugs in people who start them young and stay on them for decades are not yet known, because no one has been on them for decades. That is a real known unknown and the honest place to put skepticism. The seed oil claim has no randomized evidence behind it at all. The post treats these as one category.

What changed

Every mechanism the earlier episodes produced (preregistration, data access, device review, postmarket surveillance, results reporting) is in force for these therapies. That does not make them immune to reversal. It makes a reversal findable.

Embellished
The entry has no reversal of its own, so it is placed at the end of the list to borrow one from the other nine.
References

Cholesterol Treatment Trialists' Collaboration, Lancet (2010) · SPRINT Research Group, NEJM (2015) · Lincoff et al., SELECT trial, NEJM (2023)


What the list has in common

The episodes are independent. Different decades, different drivers (reputation, marketing, wartime urgency, incomplete toxicology, ghostwriting, unpublished trials). They do not form a chain and they do not show a standing conspiracy.

What they share is a process feature. Early or limited data (case series, small trials, industry controlled analyses, theory) got scaled into practice or policy before larger, longer, independent evidence existed. The problems appeared with enough patients and enough time.

Selection. Nine failures across 150 years, from a period that also produced insulin, penicillin, the polio vaccine, dialysis, and a steep drop in cardiovascular mortality. A list of crashes without a count of flights.

Hindsight. Several of these looked reasonable with the tools available at the time. Knowing the outcome does not make the original decision reckless by the standards of its day. It does make the standards of its day the thing to fix.

Settled is not the same as unrevisable. Clinical claims are provisional. The response to past reversals has been better methods, and each card above ends with one.

So What

Seven of the ten entries ended with a regulation, a data policy, or a review standard that did not exist before the failure. The post reads that record and concludes you should trust the institutions less. The record says the institutions improved at exactly the rate they were forced to measure.

Verdicts are on the post's version of events, not on whether the harm occurred. All ten harms, or claimed harms, were checked against the sources linked in each card. Where the post's numbers differ from the sources, the card says so.