Calibrated for Nothing

Two observations that appear to contradict, and don’t. Most people who have tried a nutritional supplement will tell you it did little or nothing. And the medical literature contains, across virtually every essential micronutrient, documented cases of dramatic response within hours or days. The two facts sit at opposite ends of a spectrum most readers do not know is a spectrum. This piece is the anatomy of the calibration that separates them.

Two shelves in the same library

Shelf one. The friend who took a magnesium tablet for a month and reports nothing changed. The colleague who started vitamin D last winter and cannot say whether it helped. The parent who was told to give the child a multivitamin gummy and observed no discernible effect. This is the majority experience of supplementation in the general population, and it is why the informal consensus in most social circles is that nutritional supplements are, at best, insurance, and at worst, marketing.

Shelf two. Wernicke’s encephalopathy: intravenous thiamine reverses ocular symptoms in hours, a phenomenon documented since the early twentieth century. Pellagra: niacin corrects dermatitis, diarrhoea, and dementia in days, established by Joseph Goldberger’s US Public Health Service work in the 1910s. Hypokalaemic paralysis: intravenous potassium restores muscle function within hours. Hypocalcaemic tetany: intravenous calcium in minutes. Xerophthalmia: vitamin A restores night vision in days. Beriberi: thiamine reverses cardiac and neurological signs in days, sometimes hours for the cardiac form. Pyridoxine-dependent neonatal seizures: intravenous B6 stops seizures in minutes. Scurvy: James Lind’s citrus trial aboard HMS Salisbury in 1747 produced improvement in six days.

These are not marginal cases. They are the phenomenology from which the entire concept of “vitamin” was named. Every essential micronutrient was identified because reversing its deficiency produced a clinical turning point on a timescale of days.

The two shelves describe the same molecules, in the same species. The distance between them is a distance of parameters.

The catalogue of rapid response, contemporary

Beyond classical vitamin science, the modern literature and clinical practice document rapid response across nearly every essential nutrient class.

Water-soluble vitamins. High-dose thiamine in dysautonomia, chronic fatigue, POTS, and Parkinson-associated fatigue, with response reported by Derrick Lonsdale (Cleveland Clinic) across five decades of clinical writing, and by Antonio Costantini and Roberto Fancellu in a series of Italian case reports in the 2010s, in days to weeks. Niacinamide in early schizophrenia (Abram Hoffer and Humphry Osmond, mid-twentieth-century Saskatchewan), with response reported within days in a documented subset. Pyridoxine in carpal tunnel syndrome (John Ellis, 1980s Texas). Methylfolate augmentation in depression not responsive to folic acid (Jonathan Alpert and Maurizio Fava at Massachusetts General). Intramuscular B12 in pernicious anaemia, with mood and cognitive improvement within days.

Fat-soluble vitamins. Vitamin A in measles reduces mortality within 48 hours, adopted as WHO protocol on the basis of Alfred Sommer’s Indonesian field trials of the 1980s. Vitamin K by injection reverses warfarin anticoagulation within hours, standard emergency-department practice.

Minerals. Magnesium by intravenous infusion abortive in acute severe migraine (documented in emergency-medicine literature), in acute severe asthma (British Thoracic Society guideline), in eclampsia (the Magpie trial and subsequent adoption), in torsades de pointes. Potassium replacement resolving hypokalaemic paralysis within hours. Zinc lozenges reducing rhinovirus symptom duration when taken within 24 hours of onset (Harri Hemilä’s meta-analyses at Helsinki). Sodium in postural orthostatic tachycardia syndrome producing same-day improvement in orthostatic tolerance. Copper in copper-responsive myeloneuropathy, described by Neeraj Kumar at the Mayo Clinic. Iodine in hypothyroidism where dietary deficiency is the cause.

Conditionally essential and adjacent molecules. Alpha-GPC on lower-body force output after a single dose (Daniel Bellar and colleagues at Louisiana-Lafayette, 2015). Citicoline on reaction time and frontal cognitive function (Marisa Silveri and Erin McGlade at McLean Hospital). Coenzyme Q10 resolving statin-induced myopathy within weeks. N-acetylcysteine as the acute antidote to paracetamol toxicity within hours. Glutathione by injection in Parkinson’s disease case reports (Gian Pietro Sechi at Sassari, David Perlmutter’s clinical writing). Glycine improving sleep depth in a single night. Creatine in cognitive performance under sleep deprivation, and in the emerging depression-augmentation literature.

Emergency medicine as the specialty that already knows this. Intravenous thiamine before glucose in suspected Wernicke. Intravenous magnesium in eclampsia and torsades. Intravenous calcium in tetany. Intravenous potassium in dangerous hypokalaemia. Intravenous vitamin K for anticoagulation reversal. N-acetylcysteine in paracetamol overdose. Hydroxocobalamin in cyanide poisoning. The specialty that treats nutrients as drugs when time is short is not asserting a different physiology from the outpatient clinic. It is using the same molecules on the same species. The tempo it observes is the tempo those molecules have. The outpatient clinic, faced with the same molecules used to correct chronic depletion, insists the tempo is months. The fractionation is administrative, not biological.

The case-report and self-experimentation literature. Beyond peer-reviewed trials, the community of self-experimentation (online forums, patient groups, orthomolecular clinical practice) documents rapid response daily. This literature is dismissed as anecdote by the evidence hierarchy that placed the randomised controlled trial at the top. Yet the RCT was never designed to detect rapid response. It was designed to compare group means at fixed endpoints in stratified populations. The mode of evidence that captures the rapid signal is precisely the mode the field’s synthesis discards.

The pivot

If you have taken a supplement and felt nothing, none of this is a description of a group you do not belong to.

The rapid-response literature describes what happens when dose, form, route, schedule, and observation window are set correctly. The null personal trial describes what happens when they are not. The two are not evidence about different kinds of people. They are evidence about different parameter choices.

The person whose thiamine deficiency resolved in three days on 300 mg of allithiamine and the person who took a B-complex gummy for two months and felt no different are not two different biologies. They are two different trials, one calibrated to detect the response and one calibrated not to.

The specifications on the bottle you bought were not written by the researcher who documented the rapid response. They were written by an industry whose margin does not depend on producing that response. The instructions your pharmacist repeated (one per day, with food) were not selected because they produce the therapeutic window. They were selected because they are the compliance profile a consumer will follow. Your doctor’s silence on the topic was not indifference to your suffering. It was the reflection of a curriculum that did not teach nutrient pharmacokinetics as a clinical subject.

You did not fail the trial. The parameter chain from the peer-reviewed literature to the bottle in your bathroom failed to preserve the information that would have made the trial meaningful.

The anatomy of the null trial

Ten places where the information is lost.

1. The dose

The Recommended Daily Allowance is calibrated to prevent overt deficiency in nearly all healthy people. It is a floor, not a therapeutic target. For most nutrients, the value at which functional improvement in the depleted begins is a multiple of the RDA, sometimes two-fold, sometimes ten-fold. Studies frequently use the RDA or a modest multiple because that is the framework the field has inherited. Bottles print the RDA because that is the dose the regulator recognises. The consumer takes the RDA because that is what is in the bottle. A trial conducted at the prevention threshold is a trial designed to sit below the therapeutic threshold.

2. The form

Magnesium oxide. Cyanocobalamin. Folic acid. Ferrous sulphate at doses that inflame the gut. Retinyl palmitate at doses that do not saturate the transporter. Cheap forms whose pharmacokinetic disadvantages, relative to the effective forms (glycinate, methylcobalamin, methylfolate, bisglycinate, retinol), are documented and understood in the specialty literature. The bottle uses the cheap form because it is cheap. The consumer does not know a form matters. A trial with the wrong form is a trial in which the molecule that reaches the tissue is not the molecule the response literature describes.

3. The schedule

The plasma half-life of most water-soluble nutrients is a matter of hours. Once-daily dosing cannot maintain a tissue concentration when elimination outruns supply. Trials give once-daily dosing because that is what compliance data support and what is easiest to monitor. Bottles say “one per day” because that is what consumers will do. The correctly-dosed schedule for a nutrient with a four-hour half-life is not the schedule the trial ran, nor the schedule the bottle recommends. It is the schedule the trial would have run if the researcher’s convenience had not been the design constraint.

4. The observation window

Trials measure outcomes at week four, week twelve, week twenty-four. Rarely at day three. Almost never at hour six. Effects that occur in the first days are not in the tables of the published literature, because they were not looked at. The aggregated impression of the field that nutrient effects are “slow and diffuse” is an artifact of when the measurements were taken. The consumer, having absorbed this expectation from science journalism, waits two months to notice a change and by then has stopped watching.

5. The endpoint

The trial measures the Hamilton depression rating scale. The nutrient acts on neuromuscular fatigue. The trial reports null. The nutrient did what it does; the trial did not measure it. A field that consistently mismatches endpoint to mechanism accumulates a literature of null results that describe nothing biological, only the choice of endpoint. Iron and mood, potassium and neuromuscular endurance, magnesium and sleep architecture are common examples of nutrient-endpoint pairs where the wrong outcome is asked about the right molecule.

6. Baseline heterogeneity

A trial of vitamin D on depression that enrols participants with baseline 25-OH-D between 15 and 80 ng/mL will report a dilute average effect. Within that trial there is likely a strong effect in the deeply deficient subgroup and no effect in the replete. The pooled result masks both. Pharmacological methodology does not stratify by baseline nutrient status because it inherited the drug-development model, in which the target population is defined by disease rather than by nutrient reserve. Nutrient trials that do stratify (as recent iron work has begun to do) recover the signal the pooled trials had lost.

7. Cofactor blindness

Vitamin D without magnesium and vitamin K2. Iron without vitamin C. B12 without folate. Methylation cofactors administered singly rather than as the interdependent system they are. The nutrient never acts alone in physiology; the trial acts on it alone because the pharmaceutical unit is one active molecule. The unit is not the biology.

8. Multi-system tempo

A single nutrient touches several tissues that respond at different speeds. Iron: haemoglobin over weeks, brain iron over months, cellular energy metabolism sometimes within days. Potassium: skeletal muscle within hours, cardiac rhythm within hours, cognitive fog on a slower curve. The consumer watches one system, usually the one the marketing highlighted, and misses the change in the others. When the change appears in the system watched, it is often at a tempo the consumer has already decided is not fast enough to be real.

9. Expectation shaped by the null literature

The consumer who has read that supplements “show small effects at twelve weeks” is prepared to conclude either that it is not worth trying, or that any change they notice is placebo. The literature’s own framing pre-installs the interpretive lens that dismisses the response the individual might otherwise perceive. Expectation, in this domain, is not decoration. It is the filter through which perception is validated or discarded.

10. Industrial inertia

The bottle at the RDA, in the cheap form, on a once-daily schedule, sold to the consumer whose null trial confirms the conclusion the literature had already suggested. Every actor in this chain is doing what its role incentivises. The regulator recognises the RDA. The manufacturer minimises cost. The pharmacist follows the label. The physician was not taught otherwise. The consumer trusts the chain. The chain is stable because the failure mode it produces (small perceived effect, low expectation, weak demand for change) does not threaten any party’s position.

What a correctly parameterised personal trial looks like

The therapeutic dose, not the RDA. The bioavailable form, not the cheapest form. The dosing schedule keyed to the half-life, not to the convenience of daily habit. The observation window that starts on day one and continues past week twelve, not one that begins at week twelve. Stratification by one’s own baseline status, obtained through testing at the thresholds the specialty literature supports, not the ones printed on the routine lab report. Attention to more than one system at more than one tempo. Awareness that cofactors matter, and that the single-nutrient bottle is a category error inherited from drug design.

The result is not always dramatic. Not every deficiency on the reader’s list is the one that will change their week. But when it is dramatic, the person who felt it is not exaggerating. They are reporting the response the rapid-response literature has always described, on the parameter chain that reproduces the conditions under which that literature was written.

What the null trial is evidence of

The null personal trial is evidence of one thing, and it is not the biological unresponsiveness of the person who ran it. It is evidence that the parameter chain running from peer-reviewed research to the consumer’s medicine cabinet is broken at multiple links, and that the broken chain is stable enough to be treated as the default. The reader who concluded from their own null trial that supplements are not for them was drawing the only conclusion the parameter chain permits. That conclusion is not the reader’s error. It is the parameter chain’s design.