The Three-Legged Stool
A Framework for Evaluating Medical Interventions
The National Institute of Mental Health’s flagship study on ADHD medications, the largest clinical trial ever conducted on the disorder, followed 579 children for eight years. At fourteen months, the investigators declared victory for stimulants. At three years, the results reversed. Medication use had become, in the researchers’ own words, “a significant marker not of beneficial outcome, but of deterioration.” Children who remained on stimulants showed worse hyperactivity, higher delinquency scores, and were shorter and lighter than their unmedicated peers. By year eight, there were no differences between groups in grades, arrests, or psychiatric hospitalizations.
William Pelham, one of the principal investigators, eventually said what the data showed: “We had thought that children medicated longer would have better outcomes. That didn’t happen to be the case. There were no beneficial effects, none. In the short term, medication will help the child behave better, in the long run it won’t. And that information should be made very clear to parents.”
The NIMH did not make it clear. The drugs remain first-line treatment. Millions of children take them. This is not an isolated failure. It is a pattern.
Support Independent Research
This work remains free because paid subscribers make it possible. If you find value here, consider joining them.
What paid subscribers get: Access to the Deep Dive Audio Library — 170+ in-depth discussions (30-50 min each) exploring the books behind these essays. New discussions added weekly. That’s 100+ hours of content for less than the price of a single audiobook.
[Upgrade to Paid – $5/month or $50/year]
Get in touch Essay ideas, stories, or expertise to share: unbekoming@outlook.com
The Framework
Before any medical intervention—a drug, a vaccine, a procedure—three questions must be answered. First: Is this necessary? Is there a genuine problem requiring intervention, and does it require this specific intervention, given available alternatives including doing nothing? Second: Is it safe? What are the documented harms, both short-term and long-term, and have they been honestly assessed? Third: Does it work? Not in surrogate endpoints or statistical manipulations, but in outcomes that actually matter to the person receiving the intervention.
These three questions form a three-legged stool. If any leg fails, the stool cannot stand. The intervention should not be administered.
I developed this framework in 2022 while working through the evidence on childhood vaccination. The COVID response had shattered my trust in public health institutions, and I found myself asking questions I had never thought to ask. What I discovered was that not a single routine childhood vaccine had been licensed based on a trial using a true placebo control. The “safety” comparisons were between vaccines, or between vaccines and their aluminum adjuvants—never between a vaccine and an inert substance. The efficacy measurements relied on antibody levels, not disease prevention. The necessity claims rested on inflated disease risk and ignored natural immunity entirely.
Once I understood the pattern, I began seeing it everywhere. The three-legged stool that could not support childhood vaccines could not support antidepressants, or ADHD medications, or statins, or the birth control pill, or bisphosphonates, or the cascade of interventions imposed on laboring women. The same template—manufactured necessity, hidden safety signals, inflated efficacy—appeared in different costumes across the pharmaceutical landscape.
This essay applies the framework to seven medical interventions that affect millions of people. The purpose is not merely to catalog failures, though the failures are damning. The purpose is to demonstrate how necessity gets manufactured, how safety gets hidden, and how efficacy gets fabricated—so that readers can recognize the pattern when they encounter the next intervention they are told is necessary, safe, and effective.
A person in anaphylactic shock needs epinephrine. Their throat is closing. The alternative is suffocation. The necessity is immediate and undeniable, the efficacy measurable in minutes, and the safety risks accepted because the alternative is death. This is what genuine medical necessity looks like. The seven interventions examined here are something else entirely.
Explaining It to a Six-Year-Old
Before you take medicine, you need to know three things.
First: Do I really need this? Maybe your body can fix it by itself. Maybe rest or soup or time is all you need. Medicine is for when your body really can’t do it alone.
Second: Will it hurt me? Medicine that helps one thing can sometimes hurt another thing. You need to know what it might do—the good parts and the bad parts.
Third: Does it actually work? Not just because someone says so. Does it really make people better?
If you can’t answer yes to all three, you shouldn’t take it.
Think of it like a stool with three legs. If all three legs are strong, you can sit on it safely. But if even one leg is broken, the whole stool tips over and you fall.
Before you sit down, check the legs.
SSRIs: The Chemical Imbalance That Never Was
For thirty years, patients were told that depression results from a chemical imbalance—specifically, insufficient serotonin in the brain. Antidepressants correct this imbalance the way insulin corrects diabetes. The theory was elegant, intuitive, and repeated by every doctor who wrote a prescription, every advertisement that reached consumers, every health website that explained the condition.
Research never supported it. Joseph Schildkraut proposed the serotonin hypothesis in 1965, but by the mid-1970s, researchers Joseph Mendels and Alan Frazer had concluded the theory arose from inadequate evaluation of contradictory findings. Kenneth Kendler, coeditor of Psychological Medicine, summarized decades of research in 2005: “We have hunted for big simple neurochemical explanations for psychiatric disorders and have not found them.”
Several observations directly contradict the theory. Tianeptine, marketed as an antidepressant, actually lowers serotonin. Mirtazapine, another antidepressant, does not affect serotonin at all. When antidepressants do raise serotonin levels, this change occurs within one to two days—yet the drugs take weeks to show any effect on depression scores. Steven Hyman, former director of the National Institute of Mental Health, acknowledged in 1996 that patients do not start with chemical imbalances—the drugs create them.
Despite the absence of supporting evidence, surveys show that 80 percent of patients taking antidepressants believe their medication corrects changes that occurred in their brain. Ninety-two percent of American university students report having heard that depression is caused by a chemical imbalance. A 2019 review of 39 popular health websites across ten countries found that 74 percent attributed depression to chemical imbalance. The theory was abandoned by researchers while still being sold to patients.
The necessity leg fails because depression was reframed from an episodic condition that typically resolves on its own into a chronic disease requiring lifelong medication. When Brown University psychiatrist Michael Posternak tracked untreated depression in patients who relapsed but did not resume medication, 23 percent recovered within one month, 67 percent within six months, and 85 percent within a year. As Posternak noted, “if 85 percent of untreated patients recover within a year, it would be extremely difficult for any intervention to demonstrate a superior result.” The alternatives—psychotherapy, lifestyle changes, addressing underlying causes—were marginalized in favor of pharmaceutical solutions.
The safety leg fails spectacularly. In 2004, the FDA conducted a meta-analysis of 24 pediatric antidepressant trials. Children and adolescents taking antidepressants showed twice the rate of suicidal thinking and behavior compared to placebo. The probability of this result occurring by chance was 1 in 20,000. The finding led to black-box warnings—the FDA’s most serious safety alert. Depressed patients taking Paxil were 6.4 times more likely to develop suicidal thoughts than those taking placebo. If depression itself caused the suicidality, the placebo group should have fared worse. They fared better.
Additional harms include emotional blunting, sexual dysfunction that can persist permanently after discontinuation (Post-SSRI Sexual Dysfunction), severe withdrawal symptoms, and akathisia—a state of severe inner restlessness that has been linked to violence and suicide. Between 40 and 65 percent of people taking SSRIs experience some form of sexual dysfunction.
The efficacy leg fails on examination of the actual trial data. Irving Kirsch’s meta-analysis of FDA trial data found that the difference between antidepressants and placebo was clinically meaningless—about 1.8 points on a 52-point scale. The small statistical advantage appeared only in severely depressed patients and may have resulted from breaking of the blind (patients could tell they were on the active drug due to side effects). Only 286 patients completed the four-to-six week trials used for Prozac’s FDA approval. Publication bias ensured that positive studies were published while negative ones remained hidden.
The outcomes that matter tell the story. In 1955, approximately 355,000 Americans were hospitalized with psychiatric disorders—roughly 1 in 468 of the population. By 2007, nearly 4 million Americans were receiving federal disability payments due to mental illness—approximately 1 in 76. This represents a sixfold increase in disability rates during an era of massive expansion in psychiatric treatment. The STAR*D trial—the largest study of antidepressant effectiveness ever conducted—found that only 3 percent of patients who entered treatment remained well at the one-year follow-up.
All three legs fail. The stool collapses.
ADHD Medications: The Marker of Deterioration
In 1994, the American Psychiatric Association’s Textbook of Psychiatry made a quiet admission. After sixty years of prescribing stimulants to difficult children, the profession acknowledged that these drugs “do not produce lasting improvements in aggressivity, conduct disorder, criminality, educational achievement, job functioning, marital relationships, or long-term adjustment.” The statement appeared without fanfare, buried in a chapter that nonetheless recommended stimulant treatment.
The necessity leg fails because there is no biological test for ADHD. The diagnosis rests on behavioral observations—fidgeting, difficulty paying attention, acting without thinking—that describe a substantial portion of normal childhood, particularly among boys. The youngest children in any classroom are dramatically more likely to be diagnosed than the oldest, suggesting that immaturity is being pathologized as disorder. Boys are diagnosed at far higher rates than girls, raising the question of whether the diagnosis reflects genuine pathology or intolerance of typically male childhood behavior.
The 1998 NIH Consensus Conference on ADHD acknowledged: “We don’t have an independent, valid test for ADHD; there are no data to indicate that ADHD is due to a brain malfunction; existing studies come to conflicting conclusions as to whether use of psychostimulants increases or decreases the risk of abuse, and finally, after years of clinical research and experience with ADHD, our knowledge about the cause or causes of ADHD remains speculative.” The alternatives—smaller class sizes, more physical activity, later school start times, addressing family stress, teaching coping skills—require systemic changes. Pills require only a prescription.
The safety leg fails when you read the package inserts. Methylphenidate (Ritalin) is classified by the Drug Enforcement Administration as Schedule II—the most restrictive category for drugs with accepted medical use, reserved for substances with high potential for abuse that may lead to severe psychological or physical dependence. It shares this classification with morphine, oxycodone, and cocaine.
The physical effects documented in the literature include drowsiness, appetite loss, lethargy, insomnia, headaches, abdominal pain, motor abnormalities, facial and vocal tics, growth suppression, hypertension, cardiac arrhythmias, and sudden death. The emotional effects include depression, apathy, mood swings, crying jags, irritability, anxiety, and hostility. The psychiatric effects include obsessive-compulsive symptoms, mania, paranoia, psychotic episodes, and hallucinations. The FDA warned in 2023 that these drugs can lead to substance use disorder and addiction even when prescribed for an indicated disorder.
Nadine Lambert, a professor at UC Berkeley, conducted a 26-year follow-up of 492 children. Among those who had never smoked and never taken stimulants, 2 percent became cocaine-dependent as adults. Among those who both smoked and were treated with stimulants: 40 percent. Lambert presented these findings at the 1998 NIH Consensus Development Conference on ADHD—the data appear in the conference proceedings—but the full study was never published in peer-reviewed literature. NIDA discontinued funding for her research following the presentation. Lambert died in 2006. The findings remain the longest follow-up on stimulant treatment and substance abuse, and no subsequent research has contradicted or replicated them at comparable duration.
The efficacy leg fails despite the short-term reality that stimulants do increase focus. The MTA study—the NIMH’s flagship trial—showed this clearly at fourteen months. But by three years, medication use was “a significant marker not of beneficial outcome, but of deterioration.” By six and eight years, there were no differences between medicated and unmedicated children in school grades, arrests, or psychiatric hospitalizations. The drugs work for the immediate task the adults want accomplished. Whether they benefit the child’s life is another question, and the answer appears to be no.
The children themselves consistently report disliking the medications. Researchers found “a pervasive dislike among hyperactive children for taking stimulants.” Children on Ritalin rated themselves as less happy and more dysphoric. Observers noted they became “passive, submissive,” “socially withdrawn,” and sometimes “zombie-like.” One researcher noted the drugs reduced “curiosity about the environment.” Another observed that medicated children “lose their sparkle.”
All three legs fail. The stool collapses.
Statins: The Four-Day Benefit
The statin data, when converted from relative risk to life expectancy, reveal what the percentages obscure. Secondary prevention trials show an absolute mortality reduction of approximately 1.2 percent over five years for patients with established heart disease. Applied to life expectancy: treating 1,000 post-heart attack patients for five years prevents roughly 12 deaths, translating to an average gain of a few days per person treated. Cardiologist Aseem Malhotra calculated this at four to five days for five years of treatment—less than one day per year. This is the best-case scenario, for people with established heart disease taking the drugs for years. For primary prevention—people who have not had a cardiac event—the major trials show no statistically significant reduction in total mortality. The Cochrane Collaboration’s reviews have repeatedly found insufficient evidence of mortality benefit in primary prevention populations.
The necessity leg fails because the cholesterol hypothesis—the idea that dietary cholesterol raises blood cholesterol which causes heart disease—has never been established. Cholesterol is essential; every cell in the body requires it, the brain is particularly dependent on it, and the body manufactures it when dietary intake is low. The thresholds for “high cholesterol” have been repeatedly lowered, each revision creating millions of new patients. In the elderly, higher cholesterol levels correlate with longer life. The body produces more cholesterol when cells are damaged—it is part of the repair process, not the cause of the damage. Blaming cholesterol for heart disease is like blaming firefighters for fires because they are always present at the scene.
The safety leg fails before trials even begin. As Malcolm Kendrick documented in The Great Cholesterol Con, statin trials use a “run-in period”—everyone receives the drug for several weeks before randomization, and anyone who has an adverse reaction is excluded from the study. In plain English: they identify everyone who will have a problem and remove them before the trial starts, then declare the drug safe because it causes no problems in the carefully selected remainder. This is not safety testing. This is safety theater.
Documented harms include muscle pain and weakness affecting up to 20 percent of users, cognitive impairment (the brain requires cholesterol), increased diabetes risk, and CoQ10 depletion. Some researchers have noted that statins may cause the arterial calcification they are supposed to prevent. The alternatives—addressing diet, stress, exercise, inflammation, and metabolic health—require lifestyle changes rather than pills but address root causes rather than biomarkers.
The efficacy leg fails when you examine absolute versus relative risk. Patients are told statins reduce heart attack risk by 30-40 percent—the relative risk reduction. The absolute risk reduction is 1-2 percent. You need to treat 100 people for five years to prevent one cardiovascular event. The number needed to treat reveals what the percentage obscures: for 99 out of 100 people, the drug provides no benefit while exposing all 100 to its risks. For women in the major trials, reductions in cardiovascular deaths did not translate to overall mortality benefit—deaths from other causes offset the gains.
All three legs fail. The stool collapses.
Bisphosphonates: The Bones That Break Anyway
In 1994, the WHO redefined osteoporosis based on bone mineral density scans, using the bone density of a healthy 35-year-old woman as the reference standard. By this definition, any woman whose bones had decreased from their youthful peak—which describes virtually every woman over 50—could be diagnosed with osteopenia (pre-osteoporosis) or osteoporosis. The condition “osteopenia” did not exist before this redefinition. Internal Merck memos reveal company excitement about the new diagnostic category and the market it would create for their bone drug Fosamax.
The necessity leg fails because bone density is not bone strength. The T-score measures quantity of mineral, not quality of bone architecture. A woman with lower density but good bone quality may have stronger, more fracture-resistant bones than a woman with higher density but compromised structure. The natural decrease in bone density with age was reframed as pathology requiring pharmaceutical intervention. Bone is a living tissue; it remodels constantly, breaking down old bone and building new. This is not failure—it is maintenance. As one PhD commenter noted: “These drugs basically kill osteoclasts. But these drugs give you a disease called osteopetrosis—you’re not rebuilding bone because you’re no longer resorbing it to create new bone. You’re literally exchanging a natural phenomenon with a disease.”
The alternatives—weight-bearing exercise, adequate vitamin D, calcium from food rather than supplements, vitamin K2, and addressing the metabolic factors that actually cause bone weakness—work with the body’s natural processes rather than against them. Wolff’s Law states that bones adapt to the loads placed on them; the way to build stronger bones is to use them, not to chemically freeze them in place.
The safety leg fails catastrophically. Bisphosphonates can cause osteonecrosis of the jaw—the jawbone literally dying, leading to pain, loose teeth, and exposed bone. They can cause atypical femur fractures—the thigh bone snapping with minimal or no trauma, in an unusual location, in patients taking drugs supposedly to prevent fractures. The drugs make bones denser but more brittle, like a dried twig that snaps rather than bending. They accumulate in bone tissue for decades after discontinuation. The very fractures they claim to prevent, they may cause through a different mechanism.
The efficacy leg fails on the numbers. The absolute fracture reduction is 1-2 percent. You need to treat 50-100 women for years to prevent one hip fracture—while exposing all of them to risks including the fractures caused by the treatment itself. The endpoint that matters is fractures prevented. The endpoint measured is bone density increased. These are not the same thing. Studies show that bisphosphonate-treated bone is mechanically weaker, with increased microcrack accumulation and no improvement in bone microarchitecture.
All three legs fail. The stool collapses.
The Birth Control Pill: The Leg That Partially Stands
The birth control pill differs from the other interventions examined here in one important respect: it does prevent pregnancy. The efficacy leg, evaluated against its stated purpose, holds. This makes the pill a useful test of the framework—demonstrating that the three-legged stool can produce nuanced verdicts, not just blanket condemnations.
The necessity leg fails because fertility is not a disease. A healthy biological function has been reframed as a condition requiring pharmaceutical management. Non-hormonal alternatives exist—fertility awareness methods, barrier methods—that do not involve daily consumption of synthetic hormones for years or decades. The pill is often prescribed to teenage girls for acne or painful periods, conditions that might resolve naturally, respond to dietary changes, or indicate underlying hormonal imbalances that deserve investigation rather than suppression.
The safety leg fails on documented harms that appear on the FDA-required package insert itself. The boxed warning states that cigarette smoking increases the risk of serious cardiovascular events. Contraindications include high risk of arterial or venous thrombotic diseases, history of blood clots, stroke, coronary artery disease, and breast cancer. The warnings section addresses blood clots (3-4 times the risk of non-users), liver disease, hypertension, gallbladder disease, and depression. These are not rare complications or fringe concerns—they are printed on the label.
A massive Danish study tracking nearly half a million women found that those on hormonal contraceptives had a 70 percent higher risk of depression compared to non-users. For teenagers, the risk was even higher—adolescent girls on the pill were 80 percent more likely to be prescribed antidepressants. The same researchers found that pill users had double the risk of suicide attempts and triple the risk of completed suicide. The risk was highest in the first two months of use—exactly when young women are most vulnerable and least likely to connect their emotional changes to their new prescription.
Multiple studies have confirmed that women on the pill are attracted to different types of men than when cycling naturally—the pill reverses normal mate preferences, causing women to prefer men with similar immune system markers rather than different ones. Women who meet their partners while on the pill often experience a change in attraction when they stop taking it. We are conducting a mass experiment on human bonding without acknowledging it. Women on the pill report lower sexual satisfaction, decreased libido, and difficulty achieving orgasm—cruel ironies for a medication marketed as enabling sexual freedom.
One leg holds; two legs fail. The stool still collapses. One broken leg is sufficient.
Pitocin: The Cascade Begins
Pitocin is synthetic oxytocin, used to induce or augment labor. The cascade it initiates is predictable and has been documented repeatedly: Pitocin leads to epidural (because the artificial contractions are unbearable), epidural leads to more Pitocin (because anesthesia slows labor), which leads to fetal distress (because the baby cannot recover between relentless contractions), which leads to cesarean (to rescue the baby from the emergency the interventions created). One in three American women now delivers by cesarean section. Michel Odent, who pioneered natural childbirth approaches, achieved cesarean rates around 6 percent.
The necessity leg fails because labor is not a disease requiring management. Due dates are estimates based on population averages—most women will go into labor by 41 weeks without intervention. “Failure to progress” is often caused by the hospital environment itself: stress hormones (cortisol, adrenaline) work against oxytocin, the hormone that opens and softens. Animals seek dark, quiet corners to labor. We strap women to beds under fluorescent lights and wonder why labor stalls. Friedman’s curve—the standard for “normal” labor progress—was developed decades ago under conditions that no longer apply, yet hospitals start pushing Pitocin when labor does not match an arbitrary timeline. The alternatives—patience, movement, privacy, support, darkness, freedom to eat and drink—require time rather than drugs.
The safety leg fails because Pitocin contractions are not like natural ones. Natural contractions build gradually, peak, release—with breaks between for mother and baby to recover, to oxygenate. Pitocin contractions slam in relentlessly, without the natural feedback loop that prevents overstimulation. Your body does not recognize these contractions as its own, because they are not. Babies get squeezed harder and longer than nature intended. Their heart rates drop—of course they do, they can barely catch their breath. Fetal distress becomes the manufactured emergency that justifies the cesarean. The synthetic hormone does not cross the blood-brain barrier like natural oxytocin does, so the hormonal cascade that produces bonding, euphoria, endorphins, and natural pain relief does not occur properly. Women get all the pain with none of the purpose.
The efficacy leg depends on what you measure. Pitocin does cause contractions. But if the goal is “healthy mother, healthy baby” in the fullest sense—not just alive, but untraumatized, bonded, able to breastfeed—the cascade of interventions it initiates represents failure. Cesarean sections carry their own risks: hemorrhage, infection, injury to surrounding organs, complications in future pregnancies, and difficulty bonding. Women who birth at home with midwives, where patience replaces Pitocin, have better outcomes on metrics that matter.
Two legs fail; one is questionable. The stool collapses.
Childhood Vaccines: The Stool That Started It All
This is where the framework was born. Working through the evidence on childhood vaccination in 2022, I kept encountering claims that collapsed under examination. The COVID response had taught me to question official narratives; the childhood vaccine schedule taught me that official narratives had been failing for decades.
Prevnar 20, Pfizer’s pneumococcal vaccine, exemplifies the template used across the childhood vaccination program. The FDA’s package insert admits that “an opsonic antibody titer that is predictive of protection against invasive pneumococcal disease or pneumococcal pneumonia has not been established.” For seven of the twenty strains targeted, approval was based solely on antibody production—not disease prevention. Surrogate endpoints are legitimate when validated against clinical outcomes in the target population. The insert admits this validation has not been established. The vaccine was approved based on demonstrating that it produces antibodies, without establishing that those antibodies prevent disease.
The necessity leg fails because the diseases targeted have been declining since before vaccines were introduced, natural immunity provides robust protection, and the actual risk to healthy children in developed countries has been systematically exaggerated. The package insert lacks comprehensive pre-vaccine disease incidence data, provides no age-stratified disease burden, and completely ignores natural immunity considerations. Most people exposed to pneumococcus remain asymptomatic carriers—their immune systems handle it without intervention. The insert uses broad language about “pneumonia and invasive disease” without quantifying actual risk or distinguishing mild from severe presentations. Fear of rare worst-case scenarios drives acceptance of interventions whose necessity has never been demonstrated for the average healthy child.
The alternatives—a healthy immune system supported by nutrition and reduced toxic exposures, breastfeeding, natural immunity from exposure, improved sanitation—address underlying susceptibility rather than targeting individual pathogens with injections of aluminum adjuvants and other excipients. These alternatives are not considered because they cannot be patented and because the infrastructure of pediatric medicine is built around the vaccine schedule.
The safety leg fails because the pivotal trials for routine childhood vaccines do not use inert placebos. The pattern is consistent across the CDC schedule. The MMR-II package insert describes trials comparing the vaccine to earlier measles vaccines—not saline. The DTaP (Infanrix) insert describes trials using whole-cell DTP as the control. Prevnar 20’s trials used Prevnar 13 as the comparator. Gardasil’s trials used aluminum-containing placebo. The Hepatitis B vaccine given on day one of life was tested against other vaccines in most trials, with limited saline-controlled data of short duration.
This is not cherry-picking; it is the licensing standard. When attorney Aaron Siri pressed Paul Offit—considered the world’s leading vaccine authority—to name a single routine childhood vaccine licensed on the basis of trials with an inert placebo control, Offit could not produce one. Each vaccine’s safety rests on comparison to another vaccine or adjuvant whose safety was established the same way. The chain never reaches an inert baseline.
The Prevnar 20 trials illustrate what this means in practice. Serious adverse events occurred in 8.2 percent of recipients—a category that includes death, life-threatening conditions, hospitalization, and permanent disability. This rate was deemed acceptable because the Prevnar 13 control arm showed similar rates. The safety conclusion is not “this vaccine is safe” but “this vaccine is no more harmful than the previous vaccine”—whose safety was established by the same circular method. The package insert states that the vaccine was not tested for carcinogenic or mutagenic potential, or for impairment of male fertility. Female fertility was tested only in a limited rabbit study. These are not gaps—they are choices. The studies that would answer the most important safety questions were never conducted.
The 1986 National Childhood Vaccine Injury Act removed product liability from vaccine manufacturers. Unlike every other product on the market, vaccines cannot be sued for injury no matter how defective or harmful. The market incentive for safety was eliminated. Parents who believe their children were harmed must petition a special vaccine court, where the burden of proof lies on the injured party, the proceedings are sealed, and the manufacturer bears no cost regardless of outcome. This is not how safety works in any other industry.
The efficacy leg fails because antibody production is a surrogate endpoint, not a clinical outcome. Antibodies are not immunity. The studies showing vaccinated populations experiencing outbreaks of the diseases they were vaccinated against suggest the gap between surrogate endpoint and clinical outcome. The unvaccinated populations that have been studied—including practices like Dr. Paul Thomas’s, which tracked health outcomes in vaccinated versus unvaccinated children—show dramatically different health profiles: lower rates of allergies, asthma, ADHD, autism, and chronic conditions. But these comparisons are not conducted by the agencies responsible for vaccine safety, because those agencies have no interest in results that would undermine the program.
All three legs fail. The stool collapses.
The Pattern
Seven interventions. Seven collapsed stools. The same failure modes appear in different costumes. What becomes visible is not seven independent failures but a single system operating across domains.
How Necessity Gets Manufactured
Disease definitions expand. Depression becomes a chronic condition requiring lifelong treatment rather than an episodic experience that typically resolves. ADHD criteria broaden until normal childhood behavior qualifies. The DSM adds disorders with each revision, never removing them, expanding the territory of pathology into ordinary human experience. Grief lasting more than two weeks can now be diagnosed as depression.
New conditions are invented. Osteopenia did not exist until Merck needed a market for bone drugs. “Pre-diabetes” creates millions of patients from people whose blood sugar falls within normal variation. “Pre-hypertension” does the same for blood pressure. The prefix “pre-” is remarkably useful for transforming healthy people into patients.
Normal variation gets pathologized. Bone density naturally decreases with age, but the comparison to a 35-year-old standard makes this decrease look like disease. Cholesterol levels that were normal become dangerous as thresholds lower. The youngest children in a classroom are most likely to be diagnosed with ADHD—their “disorder” is being younger than their classmates.
Screening creates patients. The DEXA scan finds “pre-osteoporosis” in women who feel fine and would never have fractured. The PSA test finds prostate abnormalities that would never cause symptoms. Mammograms detect cancers that would never progress. The screening itself generates the necessity for intervention.
Fear marketing amplifies rare worst-case scenarios while ignoring baseline risk. Parents are told about children dying of diseases without being told how many healthy children in developed countries actually die of those diseases. The rare tragedy becomes the representative case, driving acceptance of interventions whose necessity rests on inflated risk.
How Safety Gets Hidden
Trials are designed to minimize detection of harms. Run-in periods exclude people who react badly before randomization begins. Short trial durations—often weeks or months—miss long-term effects that may not emerge for years. Vulnerable populations—children, the elderly, pregnant women, people with multiple conditions—are excluded from trials, then prescribed the drugs after approval based on data that does not include them.
Active comparators replace true placebos, so the new drug only needs to be no worse than an existing drug whose safety was never established. The vaccine pyramid scheme extends this logic to absurdity: each vaccine is compared to another vaccine, and the chain never reaches an inert baseline. The new drug causes the same harm as the old drug, so both are declared safe.
Adverse events get reclassified. If enough people have a symptom, it becomes a “known side effect”—something to manage rather than a reason to reconsider the drug. Side effects become conditions requiring their own medications. Post-market surveillance relies on passive reporting systems that capture perhaps 1 percent of actual adverse events, allowing manufacturers to claim low rates of harm while most harm goes uncounted.
Liability shields remove the market pressure for safety. Vaccines enjoy complete immunity from product liability lawsuits. For other drugs, settlements are calculated as a cost of doing business—if the profits exceed the settlements, the drug remains on the market regardless of the harm it causes.
How Efficacy Gets Manufactured
Surrogate endpoints substitute for clinical outcomes. Antibody levels stand in for disease prevention. Bone density stands in for fracture reduction. LDL cholesterol stands in for heart attacks. Depression scores stand in for recovery and functioning. The surrogate is easier to measure, shows effects faster, and makes the drug look effective even when it does not improve the outcomes patients actually care about. Whether the surrogate actually predicts the clinical outcome is assumed, not demonstrated.
Relative risk replaces absolute risk in communications. “30 percent reduction” sounds impressive; “1 percent absolute reduction, meaning 99 out of 100 people get no benefit” does not. The framing is not accidental. Patients are given the number that makes the drug look most effective; the number that would enable genuine informed consent is buried in technical literature.
Publication bias ensures positive studies get published while negative ones stay in file drawers. The FDA may see all the trials, but doctors and patients see only the selected subset that supports the drug. Industry-funded trials are systematically more likely to show favorable results than independent trials—not because the drugs work better when industry pays, but because industry controls the design, analysis, and publication decisions.
Statistical significance gets confused with clinical significance. A difference can be statistically real—unlikely to have occurred by chance—but too small to matter to any individual patient. Antidepressants beat placebo by 1.8 points on a 52-point scale. This is statistically significant and clinically meaningless.
The Iatrogenic Cascade
Perhaps most insidiously, each intervention creates conditions requiring further intervention. The SSRI causes emotional blunting, which leads to dose increases or additional medications. The weight gain from one psychiatric drug leads to diabetes medications. The stimulant causes mood instability, which generates a bipolar diagnosis, which leads to antipsychotics with their own devastating effects on metabolism, movement, and lifespan. The Pitocin causes fetal distress, which leads to emergency cesarean. A vaccine can trigger an autoimmune response, which requires immunosuppressants. The three-legged stool does not just fail—it generates customers for additional failed stools.
Joseph Biederman, the Harvard psychiatrist who established the diagnosis of childhood bipolar disorder, explained his approach in a 2009 deposition: “The conditions that we see in front of us are reconceptualized.” He had previously, he noted, “successfully led the medical profession to conceive of ADHD as a chronic illness,” and now he would do the same for bipolar disorder. He received $1.6 million from pharmaceutical companies between 2000 and 2007. The child diagnosed with ADHD receives stimulants. The stimulants induce mood instability. The mood instability generates a bipolar diagnosis. The bipolar diagnosis leads to antipsychotics. At each stage, the drugs create the symptoms that justify the next diagnosis and the next prescription. Every child who enters this sequence is, from a business perspective, a reliable long-term customer.
Using the Framework
Before accepting any medical intervention, ask three questions:
Do I need this? Is there a genuine problem requiring intervention? Does this problem require this specific intervention, given available alternatives including doing nothing, lifestyle changes, or addressing root causes?
What are the harms? What does the package insert actually say? What adverse events occurred in the trials? How long were the trials, and what might emerge later? Was a true placebo used, or was the comparison to another active drug?
Does it work? What is the absolute risk reduction, not the relative risk reduction? What is the number needed to treat—how many people must take this intervention for one person to benefit? Are the outcomes measured (surrogate endpoints) the outcomes I care about (clinical outcomes)?
Ask your doctor specific questions: What is the absolute risk reduction? What is the number needed to treat? What happens if I do nothing? What is the track record in people like me? Have you read the actual studies, not just the summaries? Would you give this to your own family member?
Learn to recognize non-answers: “This is the standard of care” tells you what others do, not whether it works. “The studies show...” without specifics is an appeal to authority. “We’ve prescribed millions of doses” demonstrates market success, not clinical benefit. “The benefits outweigh the risks” assumes both have been honestly assessed.
The burden of proof belongs to the person proposing the intervention, not to you for declining it. You are not required to prove an intervention is harmful before refusing it. The proposer is required to prove it is necessary, safe, and effective before you accept it.
The three-legged stool is not a tool for reflexive refusal. It is a tool for genuine evaluation. Some interventions pass. Emergency surgery for internal bleeding passes—the necessity is undeniable, the efficacy is measurable in survival, and the safety risks are accepted because the alternative is death. Epinephrine for anaphylaxis passes. Suturing a wound passes. The framework produces “yes” as well as “no.”
But the framework demands evidence, not assertions. It demands outcomes that matter, not surrogate endpoints. It demands honest assessment of harms, not trials designed to hide them. It demands comparison to alternatives, not false dichotomies between the intervention and doing nothing.
The seven interventions examined here affect hundreds of millions of people. Each was presented as necessary, safe, and effective. Each fails the basic test that any intervention should pass before being administered to a human being.
The pattern is visible now. The next time you are offered an intervention—any intervention—you have the questions to ask. The three-legged stool either stands or it doesn’t. Do not sit on one that cannot hold your weight.
References
The Framework
The three-legged stool framework was developed in 2022 and first published in “Necessity” (Lies are Unbekoming, April 2023), later expanded in “The Three-Legged Stool That Cannot Stand: How Prevnar 20 Exemplifies the Childhood Vaccine Fraud” (July 2025).
SSRIs
Moncrieff, J. et al. (2022). “The serotonin theory of depression: a systematic umbrella review of the evidence.” Molecular Psychiatry, 27, 3243-3256. [The definitive review establishing that the chemical imbalance theory lacks empirical support.]
Kirsch, I. et al. (2008). “Initial Severity and Antidepressant Benefits: A Meta-Analysis of Data Submitted to the Food and Drug Administration.” PLoS Medicine, 5(2), e45. [The meta-analysis showing 1.8-point difference on 52-point scale.]
Food and Drug Administration (2004). Meta-analysis of pediatric antidepressant trials. [Basis for black-box warning; showed doubled suicidality risk.]
Posternak, M.A. et al. (2006). “The naturalistic course of unipolar major depression in the absence of somatic therapy.” Journal of Nervous and Mental Disease, 194, 324-329. [85% recovery within one year without medication.]
GlaxoSmithKline (2006). “Dear Healthcare Provider” letter on Paxil and adult suicidality. [6.4 times increased suicidality risk.]
Hyman, S. (1996). “Initiation and adaptation: A paradigm for understanding psychotropic drug action.” American Journal of Psychiatry, 153, 151-162. [Former NIMH director acknowledging drugs create chemical imbalances.]
Teicher, M.H. et al. (1990). “Emergence of intense suicidal preoccupation during fluoxetine treatment.” American Journal of Psychiatry, 147, 207-210. [Early documentation of SSRI-induced suicidality.]
Breggin, P. & Breggin, G. (2014). Talking Back to Prozac. Open Road Media. [Detailed analysis of Prozac FDA approval trials; source for 286 completed patients figure.]
Whitaker, R. (2010). Anatomy of an Epidemic. Broadway Books. [Source for disability statistics and long-term outcome data.]
ADHD Medications
MTA Cooperative Group (1999). “A 14-month randomized clinical trial of treatment strategies for ADHD.” Archives of General Psychiatry, 56, 1073-1086. [Initial positive findings.]
Jensen, P. et al. (2007). “3-Year Follow-Up of the NIMH MTA Study.” Journal of the American Academy of Child & Adolescent Psychiatry, 46, 989-1002. [Medication as “marker of deterioration.”]
Molina, B. et al. (2009). “MTA at 8 Years: Prospective Follow-Up of Children Treated for Combined-Type ADHD in a Multisite Study.” Journal of the American Academy of Child & Adolescent Psychiatry, 48, 484-500. [No differences in grades, arrests, hospitalizations.]
American Psychiatric Association (1994). Textbook of Psychiatry. [Admission that stimulants produce no lasting improvements.]
National Institutes of Health (1998). “Consensus Development Conference Statement: Diagnosis and Treatment of Attention Deficit Hyperactivity Disorder.” November 16-18, 1998. [No valid test, no brain malfunction data, speculative causation.]
Richters, J. et al. (1995). “NIMH Collaborative Multisite Multimodal Treatment Study of Children with ADHD.” Journal of the American Academy of Child & Adolescent Psychiatry, 34, 987-1000. [Planning documents acknowledging no demonstrated long-term efficacy.]
Lambert, N. (1998). Presented at National Institutes of Health Consensus Development Conference on ADHD, November 16-18, 1998; findings included in conference proceedings. [26-year follow-up showing elevated cocaine dependence in stimulant-treated children who smoked; not published in peer-reviewed journal.]
Food and Drug Administration (2023). “FDA Updating Warnings to Improve Safe Use of Prescription Stimulants.” Drug Safety Communication, May 11, 2023. [Warning on addiction and overdose death.]
Sleator, E. (1982). “How Do Hyperactive Children Feel About Taking Stimulants and Will They Tell the Doctor?” Clinical Pediatrics, 21, 474-79. [Pervasive dislike among children.]
Sroufe, A. (1973). “Treating Problem Children with Stimulant Drugs.” New England Journal of Medicine, 289, 407-13. [Children believing in “magic pills”; reduced curiosity.]
Breggin, P. (2001). Talking Back to Ritalin. Perseus Publishing.
Statins
Cholesterol Treatment Trialists’ Collaboration (2010). “Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials.” Lancet, 376(9753), 1670-1681. [Source data for absolute risk reductions.]
Cholesterol Treatment Trialists’ Collaboration (2012). “The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease.” Lancet, 380(9841), 581-590. [Primary prevention meta-analysis.]
Taylor, F. et al. (Cochrane Collaboration, 2013). “Statins for the primary prevention of cardiovascular disease.” Cochrane Database of Systematic Reviews. [No significant mortality benefit in primary prevention populations.]
Malhotra, A. Various publications and presentations on statin absolute risk. [Life expectancy calculations from ARR data.]
Kendrick, M. (2008). The Great Cholesterol Con. John Blake Publishing. [Run-in period methodology; comprehensive analysis of statin evidence.]
Ravnskov, U. (2000). The Cholesterol Myths. New Trends Publishing. [Critique of cholesterol hypothesis.]
Bisphosphonates
World Health Organization (1994). “Assessment of fracture risk and its application to screening for postmenopausal osteoporosis.” WHO Technical Report Series 843. [Redefinition establishing T-score criteria.]
Odvina, C.V. et al. (2005). “Severely suppressed bone turnover: a potential complication of alendronate therapy.” Journal of Clinical Endocrinology & Metabolism, 90(3), 1294-1301. [Atypical fractures in bisphosphonate users.]
Shane, E. et al. (2010). “Atypical subtrochanteric and diaphyseal femoral fractures: report of a task force of the American Society for Bone and Mineral Research.” Journal of Bone and Mineral Research, 25(11), 2267-2294.
Mercola, J. (2022). “Why You Should Avoid Osteoporosis Medications.” [Summary of bisphosphonate risks and alternatives.]
Birth Control
Skovlund, C.W. et al. (2016). “Association of Hormonal Contraception With Depression.” JAMA Psychiatry, 73(11), 1154-1162. [Danish study: 70% increased depression risk, 80% for adolescents.]
Skovlund, C.W. et al. (2017). “Association of Hormonal Contraception With Suicide and Suicide Attempts.” American Journal of Psychiatry, 175(4), 336-342. [Double suicide attempts, triple completed suicides.]
Roberts, S.C. et al. (2008). “MHC-correlated odour preferences in humans and the use of oral contraceptives.” Proceedings of the Royal Society B, 275(1652), 2715-2722. [Mate preference reversal.]
FDA Prescribing Information for combined oral contraceptives. [Boxed warnings, contraindications, depression warnings.]
Pitocin
Lederman, R. et al. (1978). “The relationship of maternal anxiety, plasma catecholamines, and plasma cortisol to progress in labor.” American Journal of Obstetrics and Gynecology, 132(5), 495-500. [Stress hormones inhibiting labor.]
Odent, M. (2013). Childbirth and the Evolution of Homo Sapiens. Pinter & Martin. [6% cesarean rates with physiological approach.]
Buckley, S.J. (2015). “Hormonal Physiology of Childbearing.” National Partnership for Women & Families. [Synthetic vs. natural oxytocin differences.]
Declercq, E.R. et al. (2013). “Listening to Mothers III: Pregnancy and Birth.” Childbirth Connection. [Intervention cascade documentation.]
Childhood Vaccines
MMR-II (Measles, Mumps, and Rubella Virus Vaccine Live) Package Insert. Merck & Co. [Trial comparator: earlier measles vaccines, not inert placebo.]
Infanrix (DTaP) Package Insert. GlaxoSmithKline. [Trial comparator: whole-cell DTP vaccine.]
Gardasil (Human Papillomavirus Quadrivalent Vaccine) Package Insert. Merck & Co. [Trial comparator: aluminum-containing placebo.]
Engerix-B (Hepatitis B Vaccine) Package Insert. GlaxoSmithKline. [Limited saline-controlled data; most trials used other vaccines as comparators.]
Prevnar 20 (Pneumococcal 20-valent Conjugate Vaccine) Package Insert. Pfizer Inc. [Trial comparator: Prevnar 13; no established protective antibody threshold; 8.2% serious adverse events; no carcinogenicity/mutagenicity/fertility testing.]
Siri, A. & Offit, P. Exchange documented in “A Shot in the Dark” series (Candace Owens, 2025). [Offit unable to name routine childhood vaccine licensed with true placebo control.]
National Childhood Vaccine Injury Act of 1986, 42 U.S.C. §§ 300aa-1 to 300aa-34. [Manufacturer liability protection.]
Thomas, P. & Margulis, J. (2020). The Vaccine-Friendly Plan. Ballantine Books. [Vaccinated vs. unvaccinated health outcomes in pediatric practice.]
The Pattern
Biederman, J. (2009). Deposition testimony in Risperdal litigation. [”Reconceptualized” conditions; $1.6 million from pharmaceutical companies.]
Frances, A. (2013). Saving Normal: An Insider’s Revolt Against Out-of-Control Psychiatric Diagnosis. William Morrow. [DSM expansion critique from DSM-IV task force chair.]
Gøtzsche, P. (2013). Deadly Medicines and Organised Crime: How Big Pharma Has Corrupted Healthcare. Radcliffe Publishing.
Healy, D. (2012). Pharmageddon. University of California Press.
Additional Resources
Whitaker, R. & Cosgrove, L. (2015). Psychiatry Under the Influence. Palgrave Macmillan.
Jefferson, T. et al. (Cochrane Collaboration). Various systematic reviews on pharmaceutical interventions.
Abramson, J. (2004). Overdosed America. Harper Perennial.
Support Independent Research
This work remains free because paid subscribers make it possible. If you find value here, consider joining them.
What paid subscribers get: Access to the Deep Dive Audio Library — 170+ in-depth discussions (30-50 min each) exploring the books behind these essays. New discussions added weekly. That’s 100+ hours of content for less than the price of a single audiobook.
[Upgrade to Paid – $5/month or $50/year]
Get in touch Essay ideas, stories, or expertise to share: unbekoming@outlook.com
New Biology Clinic
For those of you looking for practitioners who actually understand terrain medicine and the principles we explore here, I want to share something valuable. Dr. Tom Cowan—whose books and podcasts have shaped much of my own thinking about health—has created the New Biology Clinic, a virtual practice staffed by wellness specialists who operate from the same foundational understanding. This isn’t about symptom suppression or the conventional model. It’s about personalized guidance rooted in how living systems actually work. The clinic offers individual and family memberships that include not just private consults, but group sessions covering movement, nutrition, breathwork, biofield tuning, and more. Everything is virtual, making it accessible wherever you are. If you’ve been searching for practitioners who won’t look at you blankly when you mention structured water or the importance of the extracellular matrix, this is worth exploring. Use discount code “Unbekoming” to get $100 off the member activation fee. You can learn more and sign up at newbiologyclinic.com
Unbekoming! My man! If there's a way to kick off a new year, this is it! Wow - an amazing piece! The intro on ADHD was phenomenal. When I practiced community pharmacy, giving three year olds METH but calling it medicine, was the straw that begin to break things for me. I think I discuss that here in the interview: https://unbekoming.substack.com/p/when-grief-became-a-prescription
This three legged stool is an AMAZING construct. It can pierce through so many lies and deception! I'm going to spread this work. Phenomenal work. That piece about the iatrogenic effect is tremendous!
Keep up the great work man. Truly sensational stuff with this one. Happy New Year and may this be the best one yet!
"We've prescribed millions of doses."
Oh, I've heard that one before.
Why would anyone accept that as evidence of safety?