A silent cost that was never on the prescription label

"Doctors are men who prescribe medicines of which they know little, to cure diseases of which they know less, in human beings of whom they know nothing." — attributed to Voltaire

Voltaire wrote that in the 18th century. The medicine has changed. The last part has not.

Every evening, the same pill. Doing what it was prescribed to do. You trusted that was the whole story. But the symptoms did not care about the whole story. They were there in the morning when you reached for something and your hand felt wrong. They were there at noon when the tingling moved through your feet without asking permission. They were there at night when your face felt like it belonged to someone else. So you looked. Between appointments, in the hours the system was not available, you read and searched and found people describing the same things in different words, all of them carrying the same unanswered question. You were not being reckless. You were doing what any person does when their body is sending a signal the appointments keep returning unopened.

It had been documented. Published. Known. And in ten years of prescriptions, not once mentioned.

I. There is a name for what happened to you

There is a word for illness caused by the treatment itself.

Iatrogenesis. From the Greek: iatros, physician. Genesis, origin. Physician-originated. Medicine coined this term. Not its critics. Not alternative practitioners looking for ammunition. The medical establishment named this phenomenon itself, built it into clinical taxonomy, documented it in journals, and continued largely as before.

That last part is the piece worth sitting with.

Iatrogenesis is not a fringe accusation. It is a formally recognized category of clinical risk. It covers surgical complications, medication errors, hospital-acquired infections. It covers the prescription that treats one thing while quietly degrading another. It covers ten years of a documented interaction between a drug and a nutrient that never made it from the research into the room where the prescription was written.

This is not about doctors who do not care. The system that produced this gap was not built by negligent people. It was built by people optimizing for what the frame they were given allowed them to see. That frame, as the previous piece argued, was designed to track targets. Blood glucose. Cholesterol. Stomach acid. What moved the target in the right direction was success. What the target's management cost the biology surrounding it was a different question.

The gap between what was documented and what was mentioned is not a mystery. The research existed. The gap is about design. The system was built to manage the condition. It was not built to monitor what managing the condition costs. Those are two different jobs. A system built to move a target number has no built-in mechanism for asking what moving that number takes from everything else. Only one of those jobs has an infrastructure.

II. Ten years on the same prescription, and half the story was never told

Metformin is one of the most prescribed drugs in the world. More than 100 million people take it daily for type 2 diabetes. It works. Blood glucose comes down. HbA1c improves. The target moves in the right direction. By every metric the system was built to measure, it succeeds.

The infrastructure for measuring that success is built, staffed, and running in every clinic in the world.

The infrastructure for measuring what that success costs was never built.

What it costs, in a significant number of people who take metformin long-term, is vitamin B12. Not dramatically. Not immediately. Quietly, across months and years, through two mechanisms running simultaneously in the same body.

The first mechanism runs through what you eat. Every time you consume meat, dairy, or eggs, your body begins a precise delivery process. Think of B12 as a package that requires a specific carrier to travel through the digestive system. That carrier is a protein called intrinsic factor, produced in the stomach. The package and the carrier bind together and travel to a specific receiving port deep in the lower small intestine. That port only accepts authenticated deliveries: packages that arrive bound to their carrier, confirmed by a calcium-dependent signal. When the signal fires correctly, the port opens, the package is accepted, and B12 enters the body.

Metformin interferes with the authentication signal. The packages arrive, bound to their carriers, but the port cannot confirm them. They are turned away. Every meal. Every day. For as long as the drug is taken.

The second mechanism runs through the gut itself. The intestine contains trillions of bacteria, and before metformin those bacteria showed little particular interest in B12. Metformin changes their behavior at the genetic level, switching on collection machinery they were not previously running. The bacteria move to the front of the queue. What they absorb, the body cannot. Every meal that should have contributed to your B12 reserve contributes to theirs instead. The drug prescribed to manage your blood glucose is simultaneously redirecting a nutrient your nervous system depends on away from the body that needs it.

Two systems failing the same supply simultaneously. Neither producing any signal in a standard diabetes panel. The blood glucose numbers improve. The drug is working. The B12 reserve is running down.

The body holds reserves. The liver stores enough B12 to sustain the nervous system for several years before the deficit becomes clinically visible. A large multi-center randomized controlled trial tracking patients across 27 clinical sites found that roughly one in five long-term metformin users showed combined low or borderline-low B12 levels. Each additional year of use increased the odds of deficiency by approximately 13 percent.

When the reserves run low enough, the body starts sending signals. Numbness. Tingling. Fatigue that sleep does not fix. Weakness in the hands. In the feet. In the face. In the tongue.

These are also the symptoms of diabetic neuropathy.

This is the trap. B12 deficiency damages nerves through demyelination, the gradual breakdown of the protective sheath surrounding nerve fibers. Picture electrical wire losing its insulation: the signal does not disappear immediately, it becomes unreliable, misfiring, fading. Diabetic neuropathy damages nerves through a different process: the chronic effect of elevated blood glucose on the tiny blood vessels that supply those same fibers. Cut off the blood supply long enough and the fiber dies from the outside. Demyelination works from the inside. Two different causes. Two different processes. Nearly identical experience.

But not completely identical. And the differences are worth knowing.

B12 deficiency neuropathy tends to appear in the upper limbs earlier than diabetic neuropathy does. It frequently affects the tongue and the mouth, which diabetic neuropathy rarely does. It often comes accompanied by cognitive symptoms: brain fog, difficulty concentrating, memory gaps, that are not characteristic of diabetic nerve damage alone. Diabetic neuropathy typically follows what clinicians call a stocking-and-glove pattern, beginning symmetrically in the feet and moving upward, tracking closely with the history of glucose control over time.

These are not diagnostic certainties. A B12 test is the only way to know. But they are the questions worth bringing into the room. They are the differences the system has no protocol for raising unless someone asks.

And without the test, without the question, the damage accumulates inside the same frame being used to interpret it. The neuropathy is noted. The progression is documented. The drug managing the condition that is being blamed for the progression continues unchanged. The infrastructure for asking what that management costs does not exist. So no one asks.

Here is what the research says plainly, and it belongs here because precision about what is known and what is not is part of what earns trust: replacing B12 after the damage has become structural stabilizes the nerve function. It does not reverse it. The window between when the depletion begins and when the damage is permanent is the only window that matters. Once it closes, you are managing what remains.

No definitive clinical guideline currently requires routine B12 monitoring for patients on metformin. The research recommending it has existed for years. The recommendation lives in the journals. It has not traveled into the appointment.

That distance is the same distance Section I described. Here it has a face. Here it has a timeline. Here it has a cost that can be measured in nerve fibers.

III. It was not just your drug, and it was not just your nutrient

Metformin is not the only drug with a second ledger.

Statins lower cholesterol by blocking an enzyme called HMG-CoA reductase. Think of this enzyme as a factory that produces several things from the same raw material. The statin shuts the factory down to reduce one product: cholesterol. But the same factory also produces coenzyme Q10, a molecule that sits at the center of how every cell in the body generates energy. The mitochondria, the structures inside each cell responsible for converting fuel into usable power, depend on CoQ10 the way an engine depends on oil. Not to run. To run without destroying itself. Between 10 and 15 percent of statin users develop muscle symptoms ranging from persistent aching to, in severe cases, breakdown of muscle tissue itself. The honest note: clinical trials testing CoQ10 supplementation have produced mixed results. The depletion is mechanistically real and well documented. Whether supplementation fully compensates remains an open question. That uncertainty belongs in the conversation your doctor is not having with you.

Proton pump inhibitors suppress acid production in the stomach. They are prescribed for reflux, ulcers, gastric protection. What they also do, through a mechanism research has not yet fully characterized, is impair the body's ability to absorb magnesium from food. Magnesium participates in over 600 enzymatic reactions in the human body. It governs muscle contraction, nerve signal transmission, blood sugar regulation, and the electrical system that keeps the heart beating in rhythm. Think of magnesium as the body's quiet infrastructure: invisible when it is working, catastrophic when it is not. Documented cases in the clinical literature include cardiac arrhythmias severe enough to require hospitalization, generalized seizures, and tetany. The depletion is not dose-related, meaning there is no safe threshold below which the effect disappears. And it does not resolve with magnesium supplements while the drug continues. The deficiency is refractory to supplementation until the PPI is stopped.

Beta blockers regulate heart rate, blood pressure, arrhythmia. They work by blocking beta-adrenergic receptors, the same receptors the pineal gland uses at night to receive the signal that triggers melatonin production. Block the receptor and the signal does not arrive. The pineal gland produces less melatonin. Melatonin is not simply a sleep hormone. It is the conductor of the circadian system, governing the timing of immune function, cellular repair, hormonal release, and inflammatory regulation across a 24-hour cycle. Stephen Porges spent decades mapping the autonomic nervous system as the master coordinator of the body's recovery states. The polyvagal framework he developed shows that rest and repair are not passive. They are actively orchestrated. A drug that suppresses the molecule conducting that orchestra is not simply affecting sleep quality. It is entering the regulatory architecture that governs when and how the body rebuilds itself.

Oral contraceptives have been associated with depletion of B6, folate, riboflavin, and zinc across decades of research, in a drug taken continuously by hundreds of millions of women, with routine nutritional monitoring never becoming standard care.

Four drug classes. Four documented patterns of depletion. Each one running quietly beneath the metric the system was built to track. Each one producing consequences that will be attributed to something else if no one is looking at the right number.

The prescription treats the condition. The second ledger treats itself.

IV. While the numbers looked fine, the body was keeping score

The system tracks the target. Blood glucose. Cholesterol. Stomach acid. Heart rate. Each one measured, managed, adjusted at every appointment. Each drug assessed by one question: did the number move in the right direction?

It did. That is not the argument.

The argument is about what is not being measured while the number moves.

In 1993, the neuroendocrinologist Bruce McEwen named something that had been accumulating in clinical data for decades without a framework to hold it. He called it allostatic load. The concept is precise and worth understanding properly because it reframes everything the previous sections described.

The body does not simply respond to demands and return to baseline. It maintains stability through continuous adjustment. When a demand arrives, whether physical, chemical, or psychological, the body mobilizes: cortisol shifts, inflammatory markers change, cardiovascular parameters adjust, immune priorities reallocate. Think of it like a household managing an unexpected expense. The budget flexes. Other things get quietly defunded to cover the cost. The lights stay on. The adjustment is invisible from the outside.

The problem arrives when the adjustment never stops. When the demand is not a single unexpected expense but a permanent change to the budget. When the body is always mid-adjustment, drawing continuously from the same reserve to cover costs that keep arriving.

Allostatic load is the measure of that cumulative cost. Not a single marker. The pattern across multiple systems simultaneously. McEwen's research showed that higher allostatic load predicted, with statistical significance across years of follow-up, mortality, cardiovascular disease, cognitive decline, and functional deterioration, independent of what any single marker was showing at the time of measurement.

The body was keeping a running total. The standard appointment was reading individual line items.

Here is the extension that medicine has not yet built into clinical practice.

A drug taken every day is not a neutral event sitting apart from that account. It is an entry. It makes demands on specific biological systems. It requires metabolism, processing, adaptation. When that drug also depletes a nutrient those systems depend on, the demand compounds. The mitochondria running on less CoQ10. The nervous system maintaining itself on a shrinking B12 reserve. The heart managing its rhythm with less magnesium. The circadian system rebuilding itself with less melatonin. Each one a withdrawal. Each one entered into the same account. None of them visible in the metric being tracked.

Gabor Maté spent a career watching what happens when that account finally becomes visible as a diagnosis. His patients arrived with names for their conditions. The system treated what could be measured. It did not ask what the account looked like in the years before the diagnosis arrived. It did not have a protocol for asking.

What Maté found, documented across hundreds of cases in When the Body Says No, is that the body cannot separate its costs into neat columns. Chronic stress and chronic pharmacological burden do not draw from different reserves. They draw from the same one. The neuroendocrine systems. The inflammatory regulation. The capacity for adaptation that gets spent on one demand is not available for the next. The person managing a difficult decade on multiple medications is not managing separate situations in separate biological compartments. They are managing one reserve with multiple simultaneous claims on it. And the prescription has no view of the balance.

Traditional Chinese medicine has a concept for what McEwen measured and what Maté observed. Jing is the body's constitutional reserve, the finite biological essence that governs the capacity for sustained function and regeneration. It is not identical to allostatic load, but it illuminates the same reality from a direction Western medicine abandoned when it separated the body into manageable targets. Jing can be depleted. Chronic depletion without restoration produces what we would now recognize as accelerated deterioration, the kind that arrives earlier than it should in people who have been managing something difficult for a long time.

The question TCM asks about every intervention, including herbal ones, is a question the standard prescription does not ask about pharmaceutical ones: does this add to the reserve, or draw from it?

Not as mysticism. As a framework for a question modern medicine knows it needs to answer but has not yet built the infrastructure to answer at scale.

The system tracks whether the target moved. The account keeps running regardless.

V. The research always had the answers, just never a way into the room

This is not an argument for stopping your medications.

That needs to be said plainly, because the territory this node has covered can produce a kind of alarm that leads somewhere counterproductive. The drugs described in the previous sections are managing real conditions with real consequences. Uncontrolled blood glucose causes damage. Unmanaged cholesterol carries risk. The prescription is not the enemy.

The argument is narrower and more specific than that.

The system was built to ask one set of questions. Whether the target moved. Whether the drug is working. Whether the side effects are tolerable. These are the right questions for the frame the appointment was built inside. They are not the only questions a person managing a chronic condition on long-term prescriptions should be asking.

The additional questions are not exotic. They are not from outside medicine. They are sitting in the same journals the prescriptions came from, written by the same researchers whose work informed the clinical guidelines. The distance is not between conventional and alternative. It is between what the system was built to transmit and what it was not.

If you have been on metformin for more than two years, a B12 level is a standard blood test. If it has never been ordered, that absence is information. If your levels are borderline rather than clearly deficient, methylmalonic acid and homocysteine levels can detect depletion at an earlier stage than B12 alone. These are not specialist investigations. They are available in any standard laboratory. The research recommending them has existed for years. The clinical guideline requiring them does not.

If you are on a statin and you have unexplained muscle aching, persistent fatigue, or weakness that does not track with your activity level, CoQ10 depletion is a documented mechanism worth raising. The evidence for supplementation is mixed enough that certainty is not available. The conversation still belongs in the room.

If you are on a proton pump inhibitor long-term, magnesium is worth checking. The symptoms of deficiency overlap with enough other conditions that they will be attributed to something else if no one is looking at the specific number. Muscle cramps, irregular heartbeat, fatigue, anxiety: each one has a dozen other explanations. The system will find one. Unless someone asks for the test.

If you are on a beta blocker and your sleep has changed, your recovery from exertion feels different, your capacity to bounce back from a difficult week has shifted, the melatonin connection is worth understanding. Not as a reason to stop the drug. As a reason to ask what else might support the system the drug is affecting.

None of this requires a different doctor. It requires different questions walking into the same room.

The person who searched between appointments, who read and looked and found people describing the same things in different words, was not being reckless. They were doing what becomes necessary when the body is sending a signal the system was not built to receive. That searching was not a failure of trust in medicine. It was the only rational response to a gap the system has no infrastructure to close.

The question the appointment asks is whether the drug is working. That is the right question for the system that built it.

It is not the only question your body is waiting for someone to ask.