Chapter 14: Growth Hormone — What It Does, How It Declines, and the Science of the "Youth Hormone"

"All scientific work is incomplete — whether it be observational or experimental. All scientific work is liable to be upset or modified by advancing knowledge. That does not confer upon us a freedom to ignore the knowledge we already have, or to postpone the action that it appears to demand at a given time." — Austin Bradford Hill, "The Environment and Disease: Association or Causation?" (1965)

Overview

Growth hormone is where this book gets difficult, and it is worth saying why up front.

Most of Part III is easy to be fair about because there is so little to weigh: a molecule with promising rodent data and no completed human trial gets a straightforward ❌ on the confident claim. Growth hormone is not that. It is an approved medicine that transforms the lives of children who cannot make it. Its absence in adults produces a real, characterized, treatable syndrome. It demonstrably changes body composition in older people who take it. And it is the foundation of an anti-aging industry that has been selling something the evidence does not support for three and a half decades.

All of that is true at once. If you finish this chapter thinking growth hormone is a fraud, I have failed. If you finish it thinking the anti-aging use is supported, I have failed worse.

Every reasoning tool this book has built — the difference between mechanism and evidence, between a surrogate and an outcome, between replacing something absent and overriding something intact, between "no evidence of effect" and "evidence of no effect" — is load-bearing here. You cannot get growth hormone right by being skeptical, and you cannot get it right by being generous. You can only get it right by being precise about which claim you are evaluating.

Some of you are reading this because a clinic emailed you. Some of you already have a vial in the refrigerator. This chapter is written for you too: the actual state of the evidence, an honest account of what is unknown, and a clear picture of what a clinician can do that a book cannot. What you will not get is a lecture, and you will not get a protocol either.

In this chapter, you will learn to:

• Describe growth hormone and draw its axis, including the brake most axes do not have
• Explain why a single GH blood level is close to meaningless, and what is measured instead
• Separate GH's direct effects from those it produces through IGF-1
• State the approved indications and explain, in Chapter 5's language, why they are replacement
• Describe somatopause without deciding a question the field has not decided
• Read the 1990 Rudman study as a reviewer would, and say what it did and did not show
• Name the reasoning failures that turned twelve men into a global market
• State the cancer question carefully enough that neither an alarmist nor a salesman could quote you
• Use acromegaly as evidence, and say where that analogy stops working
• Hold two different ratings for one molecule without flinching at either

Learning Paths

All five paths read §14.5, §14.6, and §14.9 in full. They are the clearest worked example in the book of a real finding becoming an unreal claim, and the skill transfers to every compound in Part III.

💊 GLP-1 — §14.2 and §14.7 matter most; GH is an insulin antagonist, and the interaction between a growth axis and a metabolic one is not a footnote. §14.4 previews an argument you will meet again about whether an age-related decline is a disease. 🏋️ Performance — this is your chapter; §14.5–§14.8 are the core. Read §14.5 twice. The Rudman result is the most-cited and least-read study in the performance space, and knowing its last sentence changes most conversations you will have about it. 🔬 Science — read straight through. §14.1 and §14.2 set up the secretagogue chapters, and §14.8 is the book's best example of reasoning from a natural experiment. 💄 Cosmetic — §14.2 and §14.7. "Growth hormone for skin" claims are common, and both the mechanism and its limits live there. 🏥 Clinical — §14.3 is largely review, but read §14.4 and §14.9 closely; they are where patients arrive from, and §14.9 ends with the specific things worth asking about.

14.1 The molecule and the axis it sits in

Human growth hormone is 191 amino acids, about 22 kilodaltons, in a single chain, secreted by specialized cells — somatotrophs — in the anterior pituitary gland. It is stabilized by two disulfide bonds, the cysteine staples from Chapter 1. Its generic pharmaceutical name is somatotropin, and per the naming conventions in §1.8, that -tropin stem is telling you exactly what it is: a pituitary hormone.

At 191 residues it is well past the conventional peptide/protein line, and Chapter 1's size table listed it under "protein." Yet every endocrinology text in the world calls it a peptide hormone, and so does this book. That is not sloppiness. It is a reminder that the convention is a convention. Nothing about how growth hormone is made, secreted, or degraded differs in kind from what a nine-residue peptide does. It is longer, and that is all.

The hormone does not act alone. It sits inside a control system, and almost every argument in this chapter is really an argument about that system rather than about the molecule.

THE GROWTH HORMONE AXIS — an accelerator AND a brake

                        HYPOTHALAMUS
                   ┌──────────┴──────────┐
                   │                     │
                 GHRH               SOMATOSTATIN
            (growth hormone           (growth hormone
             releasing hormone)        INHIBITING hormone)
                   │                     │
                 (+) accelerator       (−) brake
                   │                     │
                   └──────────┬──────────┘
                              ▼
                    ANTERIOR PITUITARY
                       (somatotrophs)
                              │
                              ▼
                    GROWTH HORMONE  ──── direct actions ────►  fat, muscle,
                     (191 aa, 22 kDa)                            liver, bone
                              │
                              ▼
                          LIVER  (and local tissues)
                              │
                              ▼
                           IGF-1  ──── indirect actions ────►  growth plate,
                    (insulin-like growth                        muscle, most
                       factor 1, ~70 aa)                        tissues
                              │
                              │  negative feedback
                              └──►  inhibits pituitary GH release
                                    AND increases somatostatin

  SECRETION IS PULSATILE: bursts, mostly at night, near-undetectable between them.

Walk the diagram from the top. The hypothalamus sends two opposing signals to the pituitary: GHRH — growth hormone releasing hormone — tells the somatotrophs to release, and somatostatin, as its name announces, tells them to stop. The pituitary integrates the two, and what comes out is not a steady stream but a series of bursts.

Growth hormone then acts along two routes. Some effects are direct: it binds growth hormone receptors on fat cells, liver cells, and others. Others are indirect, mediated by IGF-1 — insulin-like growth factor 1 — which the liver produces in response and releases into the circulation, and which many tissues also make locally. IGF-1 then closes the loop, suppressing pituitary release and increasing somatostatin.

Two features deserve pulling out, because everything downstream depends on them.

First: the axis has an explicit brake. Most endocrine axes work with an accelerator and negative feedback from the end product. This one has that and a dedicated inhibitory hormone on its own schedule. That architecture has a practical consequence Chapter 15 is built on: you can raise growth hormone either by pressing the accelerator or by lifting the brake, and those are different interventions with different consequences. Several compounds sold as "growth hormone alternatives" work on one or the other, and knowing which is the difference between understanding them and repeating their marketing.

Second: secretion is strongly pulsatile. Bursts, predominantly at night with deep sleep, and between them the circulating level falls to near-zero. A healthy young adult and a healthy older adult, sampled at the wrong moment, can both read as "undetectable." A single random growth hormone measurement is therefore nearly uninterpretable — not imprecise, not noisy, but close to meaningless on its own. This is why the field measures IGF-1, stable through the day because it is largely protein-bound, or performs dynamic stimulation testing, which provokes a release and measures the response.

Hold on to that. A practice that diagnoses "low growth hormone" from one morning blood draw has told you something important about itself before it has told you anything about you.

🧬 The Molecule — why growth hormone had to come from human beings

Insulin has an unusual property among hormones: the pig version works in people. For sixty years, diabetes was treated with insulin extracted from cattle and pig pancreases — not identical to ours, but close enough to bind the human receptor and do the job.

Growth hormone is not like that. Its receptor is species-specific to a degree insulin's is not, and bovine or porcine growth hormone does essentially nothing in a human being. Before the mid-1980s, if you wanted to treat a child who could not make growth hormone, there was exactly one source of the active molecule on Earth: human pituitary glands, collected at autopsy.

National programs existed to collect cadaver pituitaries, pool them, and extract enough hormone to treat a small number of children. Supply was the binding constraint on the entire field; doses and eligibility were both rationed.

Some preparations were contaminated with prions and transmitted Creutzfeldt-Jakob disease to recipients, with incubation periods measured in decades. Distribution of cadaver-derived growth hormone was halted in 1985. Chapter 3's case study covers the episode in full — but two lessons run straight through this chapter. Natural does not mean safe; the contaminated material was as natural as a substance can be. And long incubation periods mean short safety records prove little; every recipient looked fine for years, and the years were not the relevant timescale.

Then came recombinant production. Growth hormone became something a bioreactor could make in unlimited quantity, at a price that fell over time. The clinical consequence was that children stopped being rationed. The other consequence — the one this chapter is about — is that for the first time in history there was more growth hormone in the world than there were patients with growth hormone deficiency. The anti-aging industry did not become plausible in 1990 because of a study. It became possible in the mid-1980s because of a supply. The study came along at exactly the right moment to point the supply somewhere.

🔍 Check Your Understanding

1. A clinic offers a "growth hormone panel" consisting of one blood draw at 9 a.m. Using only §14.1, explain why that number cannot establish a deficiency.
2. The growth hormone axis has both an accelerator and a brake. Name each, and say what it means that a drug could target either one.
3. Why did growth hormone have to come from human cadavers when insulin could come from pigs?

14.2 What growth hormone actually does

The name is a problem. "Growth hormone" describes what the molecule does in children, and it has led generations of adults to assume that in an adult body it must do nothing much, or that anything it does must be a kind of growth. Neither is right.

Growth hormone has two categories of action, and keeping them separate resolves most of the confusion in this field.

The direct actions

Growth hormone binds its own receptor on target cells and produces effects that do not require IGF-1 as an intermediary.

Lipolysis. Growth hormone promotes breakdown of stored triglyceride and release of free fatty acids. This is the main reason fat mass falls when growth hormone is administered. It is not subtle and it is not disputed.

Insulin antagonism. Growth hormone opposes insulin's action in peripheral tissues and promotes hepatic glucose output. In plain terms, growth hormone raises blood glucose. Endocrinologists call it a counter-regulatory or diabetogenic hormone and mean it literally. This single fact explains a large fraction of §14.7's risk profile and essentially all of the metabolic disease in §14.8. If you remember one non-obvious thing about growth hormone, make it this one.

Fluid and sodium handling. Sodium and water retention — which is why edema and the pressure-related syndromes, puffiness and carpal tunnel, appear so reliably when exposure rises.

Protein sparing. Nitrogen retention, and a shift toward using fat rather than protein as fuel. Combined with lipolysis, that is the body-composition effect in a sentence: less fat, more retained lean tissue.

The IGF-1-mediated actions

Growth hormone stimulates the liver — and, importantly, many tissues locally — to produce IGF-1. Most of the classical growth effects belong to IGF-1 rather than to growth hormone itself.

Longitudinal bone growth. At the growth plate in a child's long bones, IGF-1 drives the proliferation of cartilage cells that lengthen the bone. This is why growth hormone deficiency in childhood presents as short stature, and why treating it works.

Cell proliferation and survival. IGF-1 is a mitogen: it tells cells to divide, and it inhibits apoptosis, the orderly programmed cell death that removes cells the body no longer wants. Note that sentence and keep it. It is the entire mechanistic basis of the cancer concern in §14.7, and it is also — read from the other direction — the entire mechanistic basis of the repair and recovery claims that fill Part III.

Tissue maintenance and anabolism in adults. IGF-1 signaling supports muscle protein synthesis, bone turnover, and the maintenance of a range of tissues throughout adult life.

Notice something awkward about that list: growth hormone and IGF-1 do not always point the same direction. Growth hormone raises glucose; IGF-1, being structurally and functionally insulin-like, tends to lower it. The system's net metabolic effect depends on the balance between the two, which depends on exposure, timing, and the state of the liver. "More growth hormone" is not a single direction of travel, and anyone who talks about the axis as though one dial controls one outcome has oversimplified it into uselessness.

🧬 The Molecule — IGF-1, and why it is the number everyone measures

IGF-1 is a single-chain peptide of about 70 amino acids, structurally similar to proinsulin — which is where the "insulin-like" comes from, and why it has measurable insulin-like effects on glucose.

The property that makes it clinically indispensable is boring and enormously useful: almost all circulating IGF-1 travels bound to carrier proteins, principally IGFBP-3, in a complex that greatly extends its lifetime. Where growth hormone spikes and vanishes over minutes, IGF-1 sits at a level that changes over days. The axis contains its own integrator: IGF-1 is, roughly, a running average of recent growth hormone exposure. That is why:

• IGF-1 is the standard screening test, not growth hormone itself.
• IGF-1 is how treatment is monitored, in both deficiency and acromegaly.
• IGF-1 is the number in the epidemiology. When you read that "higher IGF-1 is associated with cancer risk," the study measured IGF-1 because measuring growth hormone in a large cohort would have been close to meaningless (§14.1).

One caution for §14.3: IGF-1 is influenced by more than growth hormone. Nutrition, liver function, thyroid status, systemic illness, and age all move it. A low IGF-1 is a reason to investigate; it is not by itself a diagnosis, in either direction.

14.3 Deficiency: the approved indications

Growth hormone deficiency is a real disease, it is diagnosable, and treating it works. That is not a concession made to be fair before the debunking starts. It is the strongest evidence in this chapter, and it is where the ✅ lives.

In children

A child who cannot produce adequate growth hormone does not grow at the expected rate. The presentation is growth failure — falling across height percentiles over time, which is far more informative than any single height measurement — often with increased body fat, sometimes with delayed bone age and, in severe congenital cases, hypoglycemia in infancy. The cause may be congenital or acquired: a pituitary or hypothalamic tumor, cranial irradiation, surgery, trauma. Sometimes no cause is found.

Treated, these children grow. That result is unambiguous, has been replicated for decades across many countries, and is the reason the drug exists.

Regulators have also approved growth hormone for several conditions that are not growth hormone deficiency, where it improves growth outcomes anyway: Turner syndrome, Prader-Willi syndrome, children born small for gestational age who fail to catch up, chronic kidney disease before transplantation, SHOX gene deficiency, and — more contentiously — idiopathic short stature, where the child is simply short with no identified pathology. That last one is worth noticing precisely because it is contested: it is where treating a measurement rather than a disease has an approved foothold, and the debate about it rehearses in miniature every argument in §14.4.

In adults

Adult growth hormone deficiency is a distinct clinical entity, usually arising from pituitary disease — an adenoma, or the surgery or radiation used to treat one — from traumatic brain injury, or as continuation of childhood-onset deficiency.

The syndrome is characterized, not vague: increased fat mass with a central distribution, reduced lean mass, reduced bone mineral density with increased fracture risk, an unfavorable lipid profile, reduced exercise capacity, and impaired quality of life on validated instruments. Replacement improves body composition, bone density, lipids, and quality-of-life scores.

You should already be uncomfortable, and you should be, for exactly the right reason. That list of symptoms sounds a great deal like aging. Hold that thought for one more section; §14.4 is entirely about it, and the resemblance is not an accident but is also not what it appears to be.

💊 In the Clinic — how deficiency is actually diagnosed

Because a random growth hormone level is uninterpretable (§14.1), diagnosis proceeds differently from most hormone testing, and the sequence is worth recognizing.

First, a reason to suspect it. Adult growth hormone deficiency is not diagnosed out of the blue in a healthy person with a symptom list. It is diagnosed in people with a structural or historical reason to have pituitary failure — a known lesion, pituitary surgery, cranial irradiation, significant head injury, or other pituitary hormone deficiencies already documented. Pre-test probability does the heavy lifting.

Second, IGF-1. Stable, interpretable, age-referenced. A clearly low IGF-1 in someone with pituitary disease is strong evidence. A normal IGF-1, however, does not exclude adult growth hormone deficiency — genuinely counterintuitive, and the reason testing rarely stops here.

Third, dynamic stimulation testing. Provoke the pituitary and see what it can do. The insulin tolerance test has historically been the reference standard, though it is demanding and cannot be used in everyone; glucagon stimulation is a common alternative; and as of this writing an orally administered growth-hormone-secretagogue test is available in some jurisdictions. The point of all of them is identical: measure a response, not a snapshot.

Fourth, the rest of the pituitary. Deficiency rarely arrives alone; thyroid, adrenal, and gonadal axes are assessed, and imaging usually follows.

None of that is a protocol you could self-administer, and it is not offered as one. It is offered so you can tell a diagnostic workup from a sales funnel. A workup starts from a reason, uses a provocation, and checks the neighboring axes. A funnel starts from a symptom questionnaire and one blood draw.

📊 Evidence Rating

The claim: Growth hormone replacement improves growth outcomes in children with documented growth hormone deficiency, and improves body composition, bone mineral density, lipid profile, and quality of life in adults with documented growth hormone deficiency.

The rating: ✅ Strong clinical evidence (as of this writing, 2026)

Why: Decades of randomized and controlled human trials across multiple countries, consistent in direction, supporting regulatory approval in every major jurisdiction, with a characterized safety profile and long-term post-marketing surveillance registries — and, in children, an endpoint that is not a surrogate at all but the thing itself.

What would change it: Long-term follow-up demonstrating that treated patients suffer net harm outweighing these benefits — for example, a consistent and replicated excess of malignancy or cardiovascular mortality attributable to treatment rather than to the underlying pituitary disease. Surveillance studies of this kind exist and are ongoing; as of this writing they have not overturned the indication, though some analyses have raised questions that remain debated (§14.7).

Now apply Chapter 5's vocabulary deliberately, because the next five sections turn on it.

Growth hormone in documented deficiency is a replacement. Something the body should be producing is absent, and it is being supplied from outside. Chapter 5's rule predicts what replacement looks like: durable benefit, because you are restoring a signal the system is built to receive and has not habituated to. That is what the deficiency literature shows.

Growth hormone in a healthy adult is an override. The system is intact, functioning, and regulating itself. Adding hormone from outside does not restore anything; it pushes an operating control loop away from where it is currently sitting, and Chapter 3 told you what a control loop does when you do that. Exogenous growth hormone raises IGF-1; IGF-1 suppresses pituitary release and increases somatostatin; endogenous production falls. That is expected physiology, not a malfunction, and not a sign that anything has gone wrong. It is the loop doing precisely what a loop does.

Same molecule. Two entirely different pharmacological situations. If you take nothing else from this chapter, take the habit of asking which one you are looking at.

14.4 Somatopause: what declines, and what that means

Growth hormone secretion falls with age. This is not controversial and it is not marketing; it is one of the more reliable endocrine observations in the literature.

The commonly cited figure is a decline of roughly 14% per decade after about age 30, and that number should be treated as approximate — it summarizes studies using different populations and measurement approaches, and what is declining is not a single quantity. More precisely: the amplitude of secretory pulses falls substantially with age, more than the frequency does; total daily secretion falls; circulating IGF-1 falls in parallel; and the nocturnal burst associated with deep sleep shrinks, alongside the age-related decline in slow-wave sleep itself.

The name for this pattern is somatopause, coined by analogy with menopause — and the analogy is doing argumentative work before any evidence has been presented. Menopause is a discrete, near-total, relatively abrupt cessation of ovarian hormone production with a clear physiological threshold. Somatopause is a gradual, partial, decades-long decline that never reaches zero. Parallel names invite parallel reasoning, and that invitation should be declined until someone earns it.

The resemblance that drives everything

Here is the observation that makes the anti-aging argument feel compelling, stated at its strongest. Compare the adult growth hormone deficiency syndrome from §14.3 with ordinary aging:

That is a striking table, and a person who assembled it in 1988 and concluded that aging is partly a growth hormone deficiency state was not being stupid. They were doing something recognizable as science: noticing a pattern and forming a hypothesis. The hypothesis is reasonable. It is also, after thirty-five years, unproven.

Why the resemblance does not settle it

Four problems stand between the table and the conclusion, and each one is a general lesson.

Correlation, in a system where everything changes at once. Aging is not a single process with a single dial. Sex hormone production falls. DHEA falls. Sleep architecture degrades. Physical activity declines, usually a great deal. Muscle mass falls, reducing the metabolic demand that helps drive the axis. Chronic inflammation rises. Every one of those is correlated with every other one, and picking growth hormone out of that tangle and assigning it causal responsibility requires more than noticing that it fell too.

The arrow may point the other way. Body fat suppresses growth hormone secretion — a well-established physiological relationship, not a speculation. So in an older adult with more adipose tissue and lower growth hormone, the body composition may be causing the low growth hormone rather than the reverse. The observation is symmetric; the causal story is not given by the data.

A decline is not an absence. The deficiency syndrome in §14.3 describes people whose production is severely impaired or gone. A healthy 70-year-old with an age-appropriate reduction is a different physiological state, not a milder version of the same one. Treating them as points on one continuum is the move that turns a replacement argument into an override, and it is usually made silently.

Lower may be adaptive. This is the part the field genuinely has not resolved, and I am not going to resolve it here. Reduced growth hormone and IGF-1 signaling is associated with extended lifespan in multiple model organisms — the long-lived dwarf mouse strains are the standard example, and reduced insulin/IGF-1 signaling is one of the most reproducible longevity manipulations in laboratory biology. In humans, people with Laron syndrome — inherited insensitivity to growth hormone, with correspondingly very low IGF-1 — are short-statured and have been reported in studied cohorts to have strikingly low rates of diabetes and cancer. Meanwhile, adult growth hormone deficiency is a real syndrome with real morbidity, and treating it helps.

Those two bodies of evidence point in opposite directions, and both are real. Anyone who tells you the question is settled — in either direction — is telling you about their confidence rather than about the literature.

⚠️ Hype Check — "your growth hormone has crashed since you were twenty"

The claim, in its usual form:

"By the time you're 50, your growth hormone levels have fallen by more than half from their youthful peak. That's why you've gained fat, lost muscle, and stopped recovering. Restoring youthful levels reverses it."

What's true in it. The decline is real, and the magnitude is in the right neighborhood. Growth hormone secretion does fall substantially across adult life, IGF-1 falls with it, and the body composition changes described do occur. Nothing in the first two sentences is fabricated. That is what makes this claim durable — it opens with facts.

Where it fails. Every failure is in the word that's, and then in the word restoring.

"That's why" asserts causation from a correlation, in a system where dozens of things changed simultaneously and where the causal arrow is known to run in both directions. It is the tangle problem, stated as a conclusion.

"Restoring youthful levels" smuggles in the replacement frame. You are not restoring anything; the pituitary is working. Chapter 5's rule applies by name: this is an override of an intact system, not replacement of an absent one, and those have different expected trajectories.

There is also a quieter move in the phrase "youthful levels." IGF-1 reference ranges are age-adjusted, because IGF-1 legitimately differs by age. A clinic that compares a 60-year-old's IGF-1 to a 25-year-old's range and pronounces it "low" has not made a clinical finding. It has chosen a comparison group, and the choice is the conclusion.

Verdict: the decline is real, the mechanism is plausible, and the causal claim is unproven. "Your growth hormone has declined" is a fact. "Therefore raising it will make you younger" is a hypothesis that has now been tested, and §14.6 covers how it did.

14.5 The Rudman study, read as a paper

In July 1990, the New England Journal of Medicine published a paper by Daniel Rudman and colleagues that is, by any reasonable measure, one of the most consequential twelve-person studies ever conducted.

Here is the finding, in the words this book will use every time it appears:

Twelve men over 60, six months of growth hormone versus untreated controls: increased lean mass and decreased fat mass. It did not measure strength, function, or long-term safety.

That last sentence is not a criticism appended by a skeptic. It is a description of the study's design, and it must accompany every mention of the result, because leaving it off is precisely how a finding became an industry.

What the design was

The men were over 60 and were selected in part for having low IGF-1 levels, which matters for how far the result generalizes. Twelve received growth hormone for six months. A larger group of similar men served as controls and received no treatment — not placebo. Body composition was measured before and after.

Take those design features one at a time, the way a reviewer would.

Twelve treated subjects. Small studies are not automatically bad — for a large, consistent biological effect on an objectively measured variable, twelve can be plenty. But they give imprecise estimates, are vulnerable to chance imbalance between groups, and cannot detect uncommon harms at all. A twelve-person study is structurally incapable of finding a problem that occurs in one person in fifty.

Untreated controls, not placebo controls. Everyone knew who was getting the drug — not blinded in the modern sense, and less unusual in 1990 than it would be now. For body fat measured by an instrument, unblinding matters less than it would for a symptom questionnaire, but it is not nothing: expectation changes behavior, and behavior changes body composition. Men who know they are receiving an anabolic hormone may train and eat differently.

Six months. Long enough to detect a body composition change. Not remotely long enough to say anything about the outcomes people actually care about: fractures, disability, cardiovascular events, cancer, independence, death.

Body composition as the endpoint. This is the decisive one. Lean mass and fat mass are surrogate endpoints — measurements standing in for outcomes we care about, on the assumption that moving them moves the outcomes. Chapter 5 is entirely about why that assumption fails often enough to be dangerous. Nobody wants lean mass for its own sake. People want to be strong, to climb stairs, to not fall, to not break a hip, to stay in their own home. The study measured the proxy and not one of the goals.

What it established, and what it could not

It established, credibly, that administering growth hormone to older men for six months changes body composition in the expected direction. That is a real finding. It has been replicated many times since. Growth hormone does what §14.2 says it does, and anyone arguing that the Rudman result was a fluke or an artifact is wrong.

It could not establish that the men were stronger, because strength was not measured. It could not establish that they functioned better, because function was not measured. It could not establish that they were healthier, or that they would remain so, because six months is not that timescale and twelve people is not that sample. And it could not establish safety, because a study of that size and duration cannot.

Rudman and colleagues did not claim otherwise. The paper reported a change in body composition in a small sample over six months, which is what it found. The sentence the study did not contain is the sentence the industry was built on, and the journal itself would later note publicly that the article was being misrepresented in advertising for anti-aging products.

🔬 Read the Study

```text FIGURE 14.1 — "Twelve men, six months, and thirty-five years of consequences" [real published trial]

THE STUDY Rudman D, et al. Effects of human growth hormone in men over 60 years old. New England Journal of Medicine, 1990. Twelve men over 60, selected in part for low IGF-1, received growth hormone for six months; a comparison group of similar men received no treatment. Body composition was the measured outcome.

THE QUESTION Does administering growth hormone to older men change body composition?

WHAT IT SHOWS It does. Lean body mass increased and fat mass decreased in the treated men relative to the untreated controls over six months. The direction is exactly what the mechanism in §14.2 predicts — lipolysis plus nitrogen retention — and the result has been reproduced repeatedly in the decades since. This is a real biological effect, not a statistical accident.

WHAT IT Strength. Function. Endurance. Bone fracture. Falls. Independence. DOESN'T Cardiovascular events. Cancer. Mortality. Quality of life on a validated instrument. Anything at all beyond six months. Anything about people who are not older men with low IGF-1. Anything about harms occurring in fewer than roughly one in twelve people, which is nearly all of them. And, because the controls were untreated rather than given placebo, it cannot fully separate the drug's effect from the effect of knowing you are receiving it.

THE VERDICT A valid small study that answered the question it asked. The question it asked was about a surrogate endpoint. Its conclusions do not extend past six months, past twelve men, or past body composition — and every single popular use of this paper extends them past all three.

THE LESSON A study is not "good" or "bad." It has a scope, and the scope is set by its design: who was studied, for how long, against what comparator, measuring what. Rudman 1990 is a competent piece of work that has been asked, for thirty-five years, to answer questions it was never built to address. When you see a paper cited for a claim, do not ask whether the paper is trustworthy. Ask whether the claim is inside the paper's scope. ```

14.6 The industry that twelve men built

How does a six-month study of twelve men become a global market? Not through fraud, mostly. Through a sequence of individually small reasoning steps, each of which is recognizable and each of which this book has already named.

Step one: a real finding. The body composition change is genuine, and that is what makes the whole edifice stable. A claim built on nothing tends to collapse; a claim built on a real result that has been stretched can survive indefinitely, because every challenge can be met by pointing at the real result.

Step two: drop the final sentence. Cite "increased lean mass and decreased fat mass" and omit "it did not measure strength, function, or long-term safety." The citation remains technically accurate. The impression it creates does not.

Step three: upgrade with mechanism. Growth hormone builds tissue. It drives IGF-1. IGF-1 promotes proliferation and inhibits apoptosis. Therefore more growth hormone means more repair, more muscle, better recovery, younger tissue. Every link in that chain is real biology, and the chain as a whole is exactly what Chapter 5's third frozen rule forbids: never upgrade a rating with mechanism. Roughly nine in ten compounds entering human trials with a coherent mechanism never reach approval. Mechanism is necessary and never sufficient.

Step four: substitute the surrogate for the outcome. "Increases lean mass" becomes "builds muscle" becomes "makes you stronger" becomes "reverses aging." Each step feels like a paraphrase. Together they are a claim about function derived from a study that did not measure function.

Step five: reframe override as replacement. "Restoring youthful levels," "optimizing," "correcting an age-related deficiency." Every one of those phrases imports the deficiency frame from §14.3 into a population that does not have a deficiency — and imports the ✅ along with it.

Step six: sell to a population that wants it. Recombinant supply made the drug available in quantity; an aging, affluent generation supplied demand. And because the body composition change is visible in a mirror within months, the market got something better than evidence: personal confirmation. A user who loses fat and gains lean mass has experienced exactly what the study found, which feels like proof of the claims the study did not test.

That last point is worth dwelling on, because it is why this particular hype is so hard to dislodge. With most unsupported compounds, the user's experience is ambiguous and easily attributed to placebo. With growth hormone, the visible effect is real and the claimed benefit is unproven, at the same time. The user is not imagining anything. They are simply drawing a conclusion about function, health, and aging from a body-composition observation — which is exactly the inference the field has been unable to validate.

What the later evidence actually showed

The Rudman study was not the last word. Randomized trials of growth hormone in healthy older adults accumulated through the 1990s and 2000s and were eventually pooled. The picture from systematic review — most prominently a synthesis published in the mid-2000s in a major internal medicine journal — was consistent and unflattering:

• Body composition changed in the expected direction. Lean mass up, fat mass down. Rudman replicated.
• Measured strength and functional capacity did not convincingly improve. The surrogate moved and the outcome did not follow.
• Adverse effects were substantially more common in treated participants — soft tissue edema, arthralgia, carpal tunnel syndrome, gynecomastia in men, and impaired glucose metabolism.

That is not a null result. It says: the biological effect is real, the clinical benefit did not appear where it was looked for, and the costs did. Notice what it does not say. It does not say growth hormone does nothing, and it does not prove that no benefit exists at any exposure, over any duration, in any subgroup — those trials were still mostly short and modest in size, and not designed to detect fracture or mortality differences. No evidence of effect on function is not evidence of no effect on function. But after three decades of looking, the burden has shifted decisively, and it now sits with whoever wants to make the claim.

⚠️ Hype Check — the clinic email

The claim, in a form thousands of people receive every week:

"Peptide therapy for optimization and healthy aging. Our physician-supervised protocols use bioidentical growth hormone to restore your levels to a youthful range — clinically shown to increase lean muscle, reduce body fat, improve sleep and recovery, and reverse the effects of aging at the cellular level."

Work through it phrase by phrase; nearly every one is doing something.

"Peptide therapy" — a category term borrowing credibility from insulin and semaglutide, exactly as Chapter 1 warned. It carries no evidentiary weight whatsoever.

"Physician-supervised" — true, and irrelevant to whether the claim is supported. Supervision addresses safety monitoring, not efficacy. It is a reassurance placed where a citation should be.

"Bioidentical" — recombinant growth hormone is the human sequence, and it is the same molecule in the approved product, so the word distinguishes nothing. It is there to evoke naturalness, which Chapter 1 established predicts nothing about safety and nothing about benefit.

"Restore your levels to a youthful range" — the replacement frame applied to an override, plus the age-adjusted-versus-youthful reference range choice from §14.4.

"Clinically shown to increase lean muscle, reduce body fat" — accurate, and the only supported clause in the paragraph. It is also a surrogate.

"Improve sleep and recovery" — not established in this population. Note how it sits immediately after the supported clause, inside the same "clinically shown to" umbrella.

"Reverse the effects of aging at the cellular level" — unfalsifiable as written. No measurement would settle it, which means it cannot be wrong, which means it is not a claim about the world.

One factual note, offered as information rather than legal advice: United States federal law treats growth hormone unusually. Distribution for uses other than those approved is addressed by criminal statute, which is not how most prescription drugs are handled. A clinic here is not in the same regulatory position as one prescribing an approved drug off-label, and it is reasonable to ask a provider how they understand their own legal footing.

Verdict: one supported surrogate claim, wrapped in category-borrowing, a reframed comparison group, and an unfalsifiable finish.

📊 Evidence Rating

The claim: Growth hormone administered to healthy adults without documented growth hormone deficiency slows or reverses aging, or produces meaningful improvements in strength, function, healthspan, or well-being.

The rating: ❌ Hype outpaces evidence (as of this writing, 2026)

Why: The demonstrated effects in this population are on body composition, a surrogate endpoint; randomized trials that measured strength and function did not convincingly show improvement in them; adverse effects were consistently more frequent in treated participants; long-term safety in healthy adults has not been established at all; and the acromegaly literature (§14.8) gives a substantive reason for concern about chronically elevating this particular axis.

What would change it: An adequately powered, randomized, blinded, placebo-controlled trial in healthy older adults, running for years rather than months, with prespecified functional and clinical endpoints — measured strength, gait speed, disability, falls, fractures, hospitalization, independent living, cardiovascular events, malignancy, mortality — demonstrating benefit that outweighs the observed harms. That trial is entirely possible to run. It has not been run, and the reasons it has not are as much economic and regulatory as scientific.

Read this rating correctly. ❌ describes the evidence for this claim, not the molecule. Growth hormone plainly does things: Rudman measured real changes, and they replicate. The rating says that the confident anti-aging claim is not supported by human outcome data — not that nothing happens when you take it. Those are different statements, and the second one would be false.

14.7 Risks, including the one everyone asks about

Every drug has risks, and the honest way to present them is by exposure and by population rather than as a list that applies equally to everyone.

The common and expected effects

These follow directly from §14.2's physiology and are seen across contexts, more frequently at higher exposure:

Fluid retention and edema. Sodium and water retention; swelling of hands, ankles, and face. Generally reversible when exposure is reduced.

Arthralgia and myalgia. Joint and muscle pain, often described as stiffness. Common enough in trials of healthy older adults to be one of the main reasons participants withdrew.

Carpal tunnel syndrome and paresthesia. A direct consequence of soft tissue swelling in a confined space — the median nerve compressed at the wrist. Numbness and tingling in the hand, also generally reversible.

Insulin resistance and raised glucose. Because growth hormone is counter-regulatory, it opposes insulin. In someone with normal glucose handling this may show as a modest rise; in someone with prediabetes or diabetes it can be clinically significant. This is the risk most consistently underweighted in consumer discussion, and it is the one with the clearest mechanism.

Gynecomastia in men, and hypertension in some, have both appeared in trials of healthy older adults.

The risks of the approved use

This matters for the dossier below: the ✅ indication carries risks too. A rating of "strong clinical evidence" means benefits have been demonstrated to outweigh harms in that population, not that harms are absent.

In children on approved therapy, monitored risks include slipped capital femoral epiphysis, progression of pre-existing scoliosis, benign intracranial hypertension (usually early and usually reversible), and effects on glucose metabolism. In adults, the edema, arthralgia, and carpal tunnel effects above are the common ones, and glucose is watched. All of these are managed by monitoring — clinical follow-up, IGF-1 measurement, glucose assessment — which is one of the concrete things a prescribing clinician provides that a vial does not.

Separately, and covered in full in Chapters 19 and 34: growth hormone obtained outside a regulated supply chain carries an entirely different category of risk — wrong identity, wrong concentration, contamination, non-sterility. A counterfeit is not a lesser version of the drug. It is an unknown substance.

🩺 Safety and Risk — the cancer question, stated as carefully as I know how

This is the question everyone asks, and it deserves an answer that neither reassures falsely nor alarms falsely. It requires four sentences that must all be held at once.

One: the mechanism is real and it points the wrong way. Growth hormone and IGF-1 promote cell proliferation and inhibit apoptosis — programmed cell death, which is among the body's main defenses against a cell that has begun behaving badly. A signal that says "divide, and do not die" is, at the level of mechanism, the kind of signal you would not choose to amplify without a reason.

Two: the observational association exists. Higher circulating IGF-1 has been associated in epidemiological studies with increased incidence of several cancers, with prostate, breast, and colorectal cancer among the most frequently reported. These are real published findings from real cohorts.

Three: association is not causation, and here the confounding is substantial. IGF-1 tracks nutritional status, body size, height, insulin levels, physical activity, socioeconomic position, and general health. Every one of those is independently related to cancer risk. Reverse causation is also live — early disease can alter metabolic markers before diagnosis. No causal claim is established, and the honest state of the field is that we have a consistent association whose interpretation is genuinely contested.

Four: not knowing is not the same as knowing it is fine. There is no adequately powered long-term randomized trial of growth hormone administration in healthy adults with cancer incidence as an endpoint, and there may never be one. The absence of such a trial means the question is open, not closed in the reassuring direction. Long-term surveillance of adults treated as children for deficiency exists — large European follow-up efforts among them — and the results have been mixed and debated, with the difficulty that people treated for pituitary disease differ from the general population in ways that are hard to adjust away.

Where that leaves you. For an approved indication, the risk-benefit judgment has been made repeatedly by regulators with access to the full dataset, and monitoring exists for a reason. For a healthy adult contemplating chronically elevating a growth axis for a benefit that has not been demonstrated on any outcome anyone cares about, the calculus is different: an unproven benefit against an unquantified but mechanistically plausible risk, in a system where the closest thing we have to a long-term experiment is acromegaly. That is the next section, and it is the strongest evidence in the chapter.

Do not let anyone quote the first half of this callout as proof of danger, and do not let anyone quote the second half as an all-clear. The four sentences go together.

14.8 Acromegaly: the natural experiment

We cannot run a forty-year randomized trial of elevated growth hormone in humans. Nobody will fund it, no ethics committee would approve it, and the participants would be dead of other causes before it read out.

But nature ran it anyway, without consent and without a control group, and it has been running for as long as there have been people. Acromegaly is chronic growth hormone excess, almost always caused by a benign hormone-secreting adenoma of the pituitary. When it begins before the growth plates close, the result is gigantism and extreme height. When it begins in adulthood, the bones cannot lengthen, and the excess expresses itself as thickening and overgrowth instead.

This is the most informative evidence available about what chronic elevation of this axis does to a human body over decades, and it is worth knowing in some detail.

What decades of excess growth hormone produce:

Soft tissue and skeletal overgrowth. Enlargement of the hands and feet — ring size and shoe size changing in adulthood is a classic presenting history. Coarsening of facial features: protrusion of the jaw, prominence of the brow, increased spacing of the teeth, enlargement of the tongue and nose. Thickened skin. The changes are gradual, which is why diagnosis is often delayed by years — people around the patient adapt to a face that changes slowly.

Arthropathy. Joint disease is among the most disabling features. Cartilage overgrows abnormally and then degenerates, and the damage is often permanent even after the hormone excess is corrected. Carpal tunnel syndrome is common — the same soft-tissue mechanism as §14.7, sustained for years.

Cardiomyopathy. The heart is a muscle and responds to a growth signal like one: ventricular hypertrophy, diastolic dysfunction, eventually systolic failure and arrhythmia. Hypertension is common. Cardiovascular disease has historically been the leading cause of death in acromegaly.

Metabolic disease. Insulin resistance and diabetes are frequent, following directly from the counter-regulatory physiology in §14.2 — the same effect that produces a modest glucose rise in a trial participant, running continuously for twenty years.

Sleep apnea, from upper airway soft tissue enlargement, is common and adds to the cardiovascular burden. Colonic polyps are increased, with surveillance colonoscopy a standard part of management — a real signal on the proliferation question from §14.7, in the population with the most extreme exposure.

Reduced life expectancy. Untreated or poorly controlled acromegaly is associated with mortality above that of the general population. Critically, that excess mortality is substantially reduced when the hormone excess is brought under biochemical control — the closest thing this literature has to a dose-response demonstration, and what makes the causal story credible rather than merely correlational. Lower the exposure, lower the harm.

Reading the analogy correctly

Now the part that separates careful reasoning from motivated reasoning, in both directions.

What acromegaly does establish. Chronic elevation of growth hormone and IGF-1 in humans, sustained over years, causes serious harm across multiple organ systems. This is not merely an association; the dose-response relationship with biochemical control makes it about as causal as observational human data gets. The direction of risk from chronically elevating this axis is not speculative.

What acromegaly does not establish. It involves exposure far above anything produced in an anti-aging context, sustained continuously for decades, usually with a pituitary tumor present that has its own local effects, and with a secretion pattern abnormal in shape as well as amount — the tumor does not respect the pulsatile rhythm from §14.1. It is not a prediction of what happens to a healthy 55-year-old with a modestly raised IGF-1.

If someone tells you that using growth hormone will give you acromegaly, they are wrong, and you should say so. The correct statement is more careful and more useful: acromegaly establishes the direction of concern and the organ systems at stake; it does not establish the magnitude of risk at lower exposures, and nothing else does either. That is precisely why §14.7's question is open.

Then there is the mirror image, which keeps the section honest. Laron syndrome is the opposite natural experiment: inherited insensitivity to growth hormone, producing very low IGF-1 signaling for a lifetime. Affected individuals are short-statured, and studied cohorts have been reported to show remarkably low rates of diabetes and cancer alongside other health issues. Two natural experiments, one at each extreme of the same axis, pointing opposite directions — and neither tells you what the optimum is for a normal adult, because the optimum is not visible from the endpoints of a range.

🔬 Read the Study

```text FIGURE 14.2 — "How to read a natural experiment" [constructed teaching example]

THE STUDY A composite of the acromegaly outcome literature: cohort studies and national registries following patients with growth hormone excess over years to decades, comparing outcomes in those whose hormone levels were brought under biochemical control against those whose were not, and against general population rates. Constructed here as a teaching composite; no single trial is being cited, and no specific numbers are given because none should be invented.

THE QUESTION What does a lifetime of elevated growth hormone and IGF-1 do to a human being?

WHAT IT SHOWS A consistent multi-system burden — joint disease, cardiomyopathy, hypertension, insulin resistance and diabetes, sleep apnea, colonic polyps — and excess mortality relative to the general population. Crucially, it shows that outcomes IMPROVE when the excess is controlled. That gradient is what elevates this from "sick people have high hormone levels" to "the hormone level is doing the damage."

WHAT IT Anything quantitative about lower exposures. These patients had DOESN'T far higher and continuously abnormal secretion, usually with a pituitary tumor, for decades. The study population cannot tell you the risk at a modest, intermittent, externally administered exposure. It also cannot separate hormone effects from tumor effects and treatment effects in every case, and it is observational throughout — nobody was randomized to acromegaly.

THE VERDICT The strongest available human evidence about chronic elevation of this axis, and it is evidence of harm. It bounds the DIRECTION of risk convincingly and the MAGNITUDE at lower exposures not at all.

THE LESSON Natural experiments are how we learn about exposures we cannot ethically create. They are enormously valuable and they always come with a scope restriction, usually about dose. The disciplined move is to extract the qualitative conclusion — this axis, chronically elevated, damages these organs — and refuse the quantitative extrapolation. Both halves are required. Taking only the first half produces alarmism; taking only the second produces dismissal. ```

14.9 Two ratings for one molecule

Everything in this chapter converges here. Growth hormone is the book's cleanest demonstration of the sixth frozen rule — one molecule, many ratings — and it is worth seeing both side by side.

Same molecule. Same receptor. Same physiology. Two ratings, and neither is a compromise. Now walk each frozen rule through it, because this is the chapter where they all fire at once.

Rule 1 — a rating attaches to a claim, with a population and an endpoint. Neither row says "growth hormone." Each names who is being treated and what is being measured. Strip out the population and the rows contradict each other; leave it in and they are two different facts.

Rule 2 — ❌ describes the evidence, not the molecule. Growth hormone unquestionably does things in healthy adults; Rudman measured them, and the measurements replicate. The ❌ says the confident anti-aging and functional-improvement claim is not supported by outcome data in this population. Anyone who reads it as "growth hormone doesn't work" has misread it — and if you use it that way in an argument, you have handed the other side a valid objection for free.

Rule 3 — never upgrade with mechanism. Growth hormone has the most compelling mechanistic story in Part III. It builds tissue. It drives IGF-1. IGF-1 is anabolic and anti-apoptotic. The decline with age is real and measured. Every link holds. And the anti-aging rating is still ❌, because roughly nine in ten compounds entering human trials with mechanisms this good never reach approval, and because thirty-five years of trials looked for the functional benefit and did not convincingly find it. If mechanism could upgrade a rating, this would be the compound it upgraded. It doesn't, and that is the rule earning its keep.

Rule 4 — never downgrade with distaste. The ❌ has nothing to do with growth hormone's associations — bodybuilding, celebrity, expensive longevity clinics, the general aesthetic of the space. Strip all of that away and the rating is unchanged, because it rests on endpoint selection and trial results. If your reasons for a rating would evaporate when the social context changed, they were never reasons.

Rule 5 — date-stamped and falsifiable. Both ratings carry "as of this writing, 2026," and both name what would change them. The ❌ has a specific escape route: a long, adequately powered, blinded trial with functional and clinical endpoints. That trial is technically feasible. Its absence is a fact about incentives — the drug is off-patent, the endpoints take years, no commercial sponsor has an obvious reason to fund it — not a fact about biology.

Rule 6 — one molecule, many ratings. Demonstrated, and the two rows are not even the complete set. Growth hormone in HIV-associated lipodystrophy, in short bowel syndrome, in critical illness — each is a separate claim with a separate evidence base, and one of them notably produced a trial in critically ill patients where treatment was associated with worse outcomes than placebo. A molecule-level rating would have to average all of that into one symbol, and the average would be worthless.

If you are already using it

Some of you are. Here is what is actually useful, without moralizing and without a protocol.

Know which claim you are relying on. If you have documented deficiency, you are in the ✅ row and the evidence behind your treatment is strong. If you do not, you are in the ❌ row, and the honest description is that you are taking a real drug with real effects for a benefit that has not been demonstrated on any outcome that has been measured. That is a decision a person is entitled to make. It is not the same decision the ✅ row describes, and it is worth being clear with yourself about which one you have made.

The monitorable risks are monitorable. Glucose and HbA1c, because of the insulin antagonism. IGF-1, because it is the integrator and knowing where you sit against an age-referenced range beats any symptom. Blood pressure. Any new numbness or tingling in the hands, new joint pain, swelling. Age-appropriate cancer screening, kept current — not because a causal link is established, but because that is the reasonable response to an open question.

A clinician adds four things a book cannot. They can order and interpret the testing that distinguishes the two rows above — the single most consequential question in your situation. They know your history, including what makes a growth-axis intervention a worse idea: active malignancy, diabetes, untreated sleep apnea, certain cardiac conditions. They can watch a trend rather than a snapshot. And they can discuss what happens on stopping, including the suppression of endogenous production in §14.3 — expected physiology rather than damage, but worth understanding beforehand.

An endocrinologist is the right specialist, and it is reasonable to seek one whose practice is not built around selling the intervention. That is not a comment on anyone's integrity. It is a comment on how hard it is for anyone to evaluate a question they are compensated for answering one way. Chapter 39 is about making that conversation productive.

🔍 Check Your Understanding

1. Write out the two ratings from this chapter in full, each with its population and endpoint. Then explain to yourself why they do not contradict each other.
2. Growth hormone has an excellent mechanistic story. Which frozen rule prevents that from raising the anti-aging rating, and what would raise it instead?
3. A friend says the ❌ rating proves growth hormone is useless. Correct them in two sentences, without overcorrecting into endorsement.

📋 Your Evidence Dossier

This chapter fills Field 9 — Risks.

Field 9 is where most dossier entries go soft, in two predictable ways: listing risks only for compounds you are skeptical of, and listing mechanisms of possible harm rather than documented harms — which produces an entry that sounds rigorous and contains no information.

Field 9 has four parts, and all four are required for every entry, including the ones you rate ✅.

FIELD 9 — RISKS
  9a  Common / expected      effects seen frequently, mechanistically predictable,
                             usually manageable or reversible
  9b  Serious / monitored    less common, clinically significant, the reason
                             follow-up and testing exist
  9c  Unknown / open         what has NOT been established, and why — the honest
                             gaps, stated as gaps rather than as reassurance
  9d  Supply-chain           risks that belong to the product rather than the
                             molecule: identity, purity, sterility, concentration

Worked demonstration — growth hormone

Note that the demonstration is split by claim, because §14.9 was not a rhetorical flourish. Risks attach to exposures, and the exposures differ.

FIELD 9 — GROWTH HORMONE (approved deficiency indications)     [worked demonstration]

  9a COMMON      Fluid retention and edema. Arthralgia and myalgia. Carpal tunnel
                 syndrome and paresthesia (soft-tissue compression). Injection-site
                 reactions. Largely exposure-related and generally reversible.

  9b MONITORED   Impaired glucose tolerance and frank hyperglycemia (GH is a
                 counter-regulatory hormone). In children: slipped capital femoral
                 epiphysis, progression of pre-existing scoliosis, benign
                 intracranial hypertension. In all: periodic IGF-1 and glucose
                 assessment, and clinical follow-up, are part of the therapy —
                 not optional extras.

  9c OPEN        Very long-term malignancy risk. Large follow-up studies of adults
                 treated as children exist; results have been mixed and debated,
                 and separating treatment effects from the underlying pituitary
                 disease is genuinely hard. Not settled in either direction.

  9d SUPPLY      Low, within a regulated supply chain. Recombinant product,
                 pharmacy-dispensed, cold-chain handled.

  ⇒ NOTE: this entry belongs to a ✅ claim. A ✅ does NOT mean "no risks." It means
    benefits have been demonstrated to outweigh these risks IN THIS POPULATION.
    An entry with an empty Field 9 is an incomplete entry, not a safe compound.


FIELD 9 — GROWTH HORMONE (healthy adults, anti-aging use)      [worked demonstration]

  9a COMMON      The same list as above, and in randomized trials of healthy older
                 adults these occurred MORE often than in controls — edema,
                 arthralgia, carpal tunnel, gynecomastia in men. Some trials saw
                 withdrawals for these reasons.

  9b MONITORED   Insulin resistance and impaired glucose metabolism, consistently.
                 Hypertension in some. In a population that is not being monitored
                 by anyone, "monitored risk" becomes "unnoticed risk."

  9c OPEN        Long-term safety in healthy adults is NOT established — there is
                 no adequately powered multi-year randomized trial with clinical
                 endpoints. Cancer risk is unresolved: mechanism is adverse
                 (proliferation, apoptosis inhibition), observational IGF-1
                 associations exist, confounding is substantial, causation is not
                 established, and the closest long-term human evidence —
                 acromegaly — shows real multi-system harm at much higher exposure.
                 Write it as an open question, because it is one.

  9d SUPPLY      HIGH, and frequently the dominant risk in practice. Product
                 obtained outside a regulated channel may be counterfeit, wrongly
                 concentrated, non-sterile, or not GH at all. GH is a large
                 recombinant protein that is difficult to make correctly; it is a
                 poor candidate for informal manufacture.

Notice what the split reveals. The molecule and the pharmacology are identical in both entries. What differs is the population, the monitoring, the supply chain, and — decisively — what sits on the benefit side of the ledger to weigh the risks against. In the first entry, demonstrated benefit. In the second, a surrogate.

Your turn

For each peptide in your dossier, fill all four parts of Field 9. Three rules:

1. Do it for your ✅ compounds too. If your semaglutide entry has an empty risk field, you have built an advertisement, not a dossier.
2. Distinguish 9c from 9a. "Unknown" and "no reported problems" are different findings, and conflating them is the most common error in consumer risk writing. Write "not studied" when it is not studied.
3. Cite where each item came from — a label, a trial, a review, a mechanism. Items whose only source is mechanism belong in 9c as concerns, not in 9a or 9b as effects. Chapter 5's rule applies to harms exactly as to benefits: mechanism is necessary and never sufficient.

Conclusion

Growth hormone is 191 amino acids from the anterior pituitary, released in nocturnal pulses under opposing hypothalamic control — GHRH pressing the accelerator, somatostatin holding the brake. It acts directly on fat, liver, and muscle, and indirectly through IGF-1, which serves as the stable, measurable record of what the axis has been doing. Because the hormone itself spikes and vanishes, a single blood level tells you almost nothing, and the field measures IGF-1 or provokes a response instead.

In children who cannot make it, replacing it lets them grow. In adults whose pituitaries have been damaged, replacing it improves body composition, bone density, lipids, and how they feel. Those are strong, replicated, approved findings, and they carry their own monitored risks, as every effective drug does.

Growth hormone also declines with age — roughly 14% per decade after 30, approximately — and the resulting picture resembles the deficiency syndrome closely enough to have launched an entire industry when twelve men in 1990 gained lean mass and lost fat over six months. That study did not measure strength, function, or long-term safety, and the thirty-five years since have demonstrated what happens when a surrogate endpoint is allowed to stand in for the outcomes people actually want. When later trials went looking for strength and function, they mostly did not find them — and they did find more edema, more joint pain, more carpal tunnel, and worse glucose.

Whether the age-related decline is pathology or normal physiology is not settled, and this chapter did not settle it. Model organisms with reduced growth signaling live longer; people with a lifetime of excess growth signaling get joint disease, cardiomyopathy, diabetes, and a shortened life. Both are real, both point away from "more is better," and neither tells you what a healthy adult's optimum is.

The chapter's real lesson is the pair of ratings. ✅ for replacement in documented deficiency. ❌ for anti-aging in healthy adults. One molecule, one receptor, two claims, two states of evidence. Anyone who insists on a single verdict about growth hormone — enthusiast or debunker — has thrown away the only information that would let you reason about your own situation.

Chapter 15 takes the step the market took too. If injecting growth hormone is expensive, restricted, and suppresses your own production, why not stimulate the pituitary to make more of its own? That is the logic behind the growth hormone secretagogues — sermorelin, tesamorelin, the GHRP family, ipamorelin, MK-677 — and it is a genuinely better idea in several specific respects. Whether "more physiological" translates into "more effective" or "safer" is a different question, and the axis diagram in §14.1 already contains most of what you need to anticipate the answer.

Key Terms

Growth hormone (GH) — a 191-amino-acid, ~22 kDa single-chain peptide hormone from the anterior pituitary. Generic pharmaceutical name: somatotropin.

Somatotropin — the generic drug name for growth hormone; the -tropin stem denotes a pituitary-hormone-like molecule.

Somatotroph — the anterior pituitary cell type that produces and secretes growth hormone.

GHRH (growth hormone releasing hormone) — a hypothalamic peptide that stimulates pituitary GH release. The accelerator.

Somatostatin — a hypothalamic peptide that inhibits pituitary GH release. The brake; its presence makes this axis architecturally unusual.

IGF-1 (insulin-like growth factor 1) — a ~70-amino-acid peptide produced largely in the liver in response to GH, mediating most of GH's growth effects and providing negative feedback to the pituitary.

IGFBP (IGF binding protein) — a carrier protein that binds circulating IGF-1 and greatly extends its lifetime, which is why IGF-1 is stable enough to be measured usefully.

Pulsatile secretion — release in discrete bursts rather than continuously. GH secretion is strongly pulsatile and mostly nocturnal, which is why a single random measurement is nearly uninterpretable.

Dynamic stimulation testing — provoking hormone release and measuring the response; used to diagnose GH deficiency because a static level cannot.

Growth hormone deficiency (GHD) — inadequate GH production. In children, growth failure; in adults, a characterized syndrome of altered body composition, reduced bone density, adverse lipids, and impaired quality of life.

Adult growth hormone deficiency — GHD arising in or persisting into adult life, usually from pituitary disease, surgery, radiation, or head injury.

Somatopause — the age-related decline in GH secretion and IGF-1, commonly cited as roughly 14% per decade after age 30. The analogy to menopause is rhetorical, not physiological.

Replacement versus override — supplying a hormone that is absent versus adding to one being produced normally. Replacement tends to produce durable benefit; overriding an intact system tends to produce adaptation.

Surrogate endpoint — a measurement standing in for an outcome people care about. Lean and fat mass are surrogates; strength, function, fractures, and independence are outcomes.

Lipolysis — breakdown of stored triglyceride and release of fatty acids. A direct GH effect and the basis of its fat-mass changes.

Insulin antagonism (counter-regulatory action) — GH opposes insulin peripherally and promotes hepatic glucose output, raising blood glucose. The mechanism behind its metabolic risks.

Apoptosis — programmed cell death. IGF-1 inhibits it, which is central both to tissue maintenance and to the cancer question.

Acromegaly — chronic GH excess in adulthood, usually from a pituitary adenoma: soft-tissue and skeletal overgrowth, arthropathy, cardiomyopathy, insulin resistance, sleep apnea, reduced life expectancy. The natural experiment for chronic GH elevation.

Gigantism — GH excess beginning before growth plate closure, producing extreme stature.

Laron syndrome — inherited GH receptor insensitivity: very low IGF-1 signaling, short stature, and — in studied cohorts — reported low rates of diabetes and cancer. The mirror-image natural experiment.

Recombinant — produced by genetically engineered cells rather than extracted from tissue. Recombinant GH replaced cadaver-derived material in the mid-1980s and removed the field's supply constraint.

Spaced Review

1. (Ch. 3 + Ch. 14) Exogenous growth hormone raises IGF-1, which suppresses pituitary GH release and increases somatostatin, so endogenous production falls. A user reports this as "the drug broke my pituitary." Using Chapter 3's account of feedback, explain what is actually happening, and say whether the same reasoning would apply to a person receiving replacement for documented deficiency.

2. (Ch. 4 + Ch. 14) Growth hormone is 191 amino acids and is given by injection. A product is advertised as an oral growth hormone "activator" that delivers the same benefits. Using Chapter 4's material on delivery and degradation, name two distinct questions you would ask before evaluating any clinical claim about it — and say which one is answerable from chemistry alone.

3. (Ch. 5 + Ch. 14) State the Rudman finding in full, including its final sentence. Then identify which of Chapter 5's frozen rules is violated by each of these three statements: (a) "Rudman proved GH reverses aging"; (b) "GH builds tissue and drives IGF-1, so it must improve recovery"; (c) "GH is a bodybuilding drug, so the anti-aging evidence is obviously junk."

4. (Ch. 14) A 58-year-old with no pituitary history is told by a clinic that her IGF-1 is "low for optimal function" and is offered growth hormone. Name three specific things wrong with the diagnostic reasoning, drawing on §14.1, §14.3, and §14.4 — one from each.

5. (Ch. 5 + Ch. 14) Acromegaly demonstrates serious harm from chronic GH elevation. Explain why that evidence supports the ❌ rating on the anti-aging claim without supporting the statement "taking growth hormone will give you acromegaly." What is the difference between establishing a direction of risk and establishing a magnitude?