The companion catalogue. Every variable worth tracking, sorted into the three categories that take three different instructions — and for every switch, both costs named explicitly. Because the point of a map is to show you which gaps are worth filling, and which “gaps” are actually decisions in disguise.
Chains and switches was the distinction. This is the map. And building it out surfaced two refinements worth having. First: there is a third category — variables that are neither weak links to strengthen nor positions to choose, but tanks to fill. Call them gauges. Second, and more useful: several things almost everyone files as chains are switches, and mis-filing them has a measurable cost. Oxidative stress is the most instructive case in the entire field, and we’ll get to why.
The categories are distinguished by curve shape — and by a question that turns out to be a remarkably clean discriminator: is there a real debate about the target?
A gauge isn’t a chain, because more is better fails past repletion — extra vitamin D above sufficiency does nothing, and eventually harms. But it isn’t a switch either, because nothing opposing is being suppressed; there’s no arm you’re trading away, and no decision to make. The instruction is entirely different from both: test, fill if low, stop. Most supplement waste lives here — people applying chain logic (“more is better”) to a tank that’s already full.
No trade-offs anywhere in this list. This is the gap-filling column: find the one furthest behind, improve it, and the ceiling rises for everything else. They also interlock, which is what makes this a network rather than a checklist.
Does: How readily cells take up glucose. Sits upstream of inflammation, fat storage, and mitochondrial function — and is bidirectionally linked to both.
Lever: Visceral fat, activity, sleep, fibre, muscle as a glucose sink.
Does: Background immune activation. Drives CD38 — which leaks NAD⁺ — plus atherosclerosis and insulin resistance. The hub node of the whole network.
Lever: Visceral fat, sleep, gut integrity, omega-3s, exercise.
Does: ATP throughput plus quality control (mitophagy). Every tissue depends on it; the brain most of all at ~20% of energy on 2% of mass.
Lever: Exercise (biogenesis), NAD⁺ pool, mitophagy windows, membrane lipids.
Does: An endocrine organ, not storage. Secretes inflammatory signals and drives insulin resistance — the first link in the chain above.
Lever: Energy balance, activity, sleep; pharmacological options exist.
Does: VO₂max is among the strongest mortality predictors ever measured — rivalling or exceeding smoking, diabetes, and hypertension as a risk marker.
Lever: Zone 2 base plus intervals. Nothing else substitutes.
Does: The particle count that actually deposits into artery walls. Causal in atherosclerosis — not merely associated.
Lever: Diet, fibre, body composition; medication where indicated.
Does: Mechanical load on arteries, kidneys, and brain. A quiet driver of vascular dementia and renal decline.
Lever: Sodium/potassium, body composition, fitness, sleep, alcohol. *Has a distant floor — but behaves monotonically across the range anyone is actually in.
Does: A readout of methylation health and a vascular risk marker — the branch point where the NAD⁺ methylation cost surfaces.
Lever: B12, folate, B6, TMG/betaine; treat the cause, not the number.
Does: A glucose sink, a metabolic reserve, and fall protection. Strength predicts mortality independently of mass.
Lever: Resistance training. Note: the outcome is a chain — but some levers on it (systemic GH) are switches.
Does: In older age a hip fracture is a mortality event, not an inconvenience. Silent until it isn’t.
Lever: Loading, protein, vitamin D status, oestrogen/testosterone status.
Does: Barrier integrity governs how much inflammatory signal leaks systemically; the microbiome modulates immunity and metabolism.
Lever: Fibre diversity, fermented foods, avoiding needless antibiotics.
Does: Clearance capacity for everything else you do — including every compound you take. Degrades silently.
Lever: Blood pressure, glucose, alcohol, drug load, hydration.
Does: Glymphatic clearance, GH pulses, testosterone production, glucose handling, appetite regulation. The cheapest lever with the widest reach.
Lever: Consistency, light timing, temperature, alcohol, screens.
Does: Periodontal disease is a genuine, ongoing systemic inflammatory source — with associations to cardiovascular and cognitive outcomes.
Lever: Actual dental care. Unglamorous and underrated.
Does: Sensory loss is an independent, well-evidenced input to cognitive decline — partly via reduced engagement and cognitive load.
Lever: Protection, testing, correction. Hearing aids are a cognition intervention.
Does: Among the most robust mortality predictors in the epidemiology, and almost never on a supplement list. Builds cognitive reserve.
Lever: Relationships, novelty, purpose, learning.
Does: Tobacco, excess alcohol, and air pollution are the three exposures with the clearest dose-dependent damage across nearly every system.
Lever: Removal. There is no optimum above zero for tobacco.
Run this column and ask, honestly, which one you’re worst at. For most people it isn’t exotic — it’s sleep, cardiorespiratory fitness, visceral fat, or a dentist appointment. Every one of these is free, well-evidenced, and interlocked with the others. This column is not the warm-up act for the pharmacology. It is the larger share of the entire win.
These have a wide flat optimum. Correcting a genuine deficit produces real benefit; adding more once you’re replete produces nothing, then eventually harm. The instruction includes the word stop, which is what makes them a separate category.
Does: A hormone with receptors throughout the body — immune modulation, neurotrophic factors, calcium handling. Deficiency tracks with poor outcomes.
Watch: Correcting a deficit matters; topping up when replete doesn’t. Real toxicity at the high end.
Does: Methylation substrate. Deficiency causes irreversible neurological damage — which is why this one is worth actually testing.
Watch: High folate can mask B12 deficiency while nerve damage progresses. Test both.
Does: Cofactor in hundreds of reactions; blocks the NMDA calcium channel, damping excess glutamate signalling.
Watch: Deficiency is common and under-tested. Excess is mostly a GI problem.
Does: Immune function, testosterone synthesis, wound healing. Deficiency measurably impairs all three.
Watch: Chronic excess blocks copper absorption — a real deficiency you can create by over-supplementing.
Does: Oxygen transport and mitochondrial function. Deficiency is fatigue and anaemia.
Watch: The narrowest window here. Excess drives oxidative damage; haemochromatosis is not rare. Never supplement blind.
Does: Thyroid hormone synthesis and conversion. Both deficiency and excess cause thyroid dysfunction — in both directions.
Watch: Narrow windows. This is where “more must be better” does real damage.
Does: DHA is structural brain material; EPA is anti-inflammatory. There’s a target range, not a maximise.
Watch: Plateau above target; bleeding risk at very high intakes.
Does: It’s the ratio, not sodium alone — and most people are low on potassium rather than merely high on sodium.
Watch: Both extremes of sodium associate with harm. Kidney disease changes the rules entirely.
Don’t supplement a gauge you haven’t measured. Iron is the clearest case — blind supplementation can drive oxidative damage in someone who was never deficient. Zinc taken indefinitely creates a copper deficiency. Selenium and iodine have windows narrow enough that enthusiasm becomes the problem. A gauge is the one place where “it probably can’t hurt” is reliably false.
Here is what a switch entry has to include that a chain entry doesn’t: what you pay at each end. Not “pros and cons” — the two ends of a dial, each with its own gain and its own bill. If you can’t name the cost at both ends, you haven’t understood the variable.
The master toggle — every other switch here ultimately reports to it. Reciprocally wired: each arm phosphorylates and disables the other.
You gain: Autophagy, mitophagy, mitochondrial biogenesis, DNA repair, metabolic flexibility. The machinery behind every lifespan-extending intervention in animals.
You pay: Nothing gets built. Sustained hard, you lose tissue — and muscle mass is itself a mortality predictor.
You gain: Protein synthesis, muscle, tissue repair, wound healing, anabolic tone.
You pay: Autophagy shut off. Damaged proteins and mitochondria accumulate — which is a large part of what ageing is at the cellular level.
Rides the mTOR arm via a kinase cascade. The liver integrates GH pulses into tonic IGF-1 — so pulsatility protects the GH arm, not this one.
You gain: Permissive for autophagy. Laron syndrome (IGF-1 deficiency) shows near-total protection from cancer and diabetes.
You pay: Sarcopenia, frailty, poor repair; low IGF-1 in older adults tracks with higher mortality. Laron individuals don’t obviously live longer — they die of other things.
You gain: Visceral fat reduction (the one demonstrated win), repair, anabolic tone, subjective vitality.
You pay: Autophagy suppressed body-wide; tonic IGF-1 in every tissue; cancer-associated signalling; worsened insulin sensitivity. Acromegaly shortens life.
Not on the mTOR switch — a steroid acting via the androgen receptor as a transcription factor. It sets the gain rather than flipping the toggle. U-shaped in humans.
You gain: Nothing worth having. This end is a deficit, not a longevity strategy.
You pay: Sarcopenia, metabolic syndrome, insulin resistance, frailty, bone loss, low mood — and higher all-cause mortality in men.
You gain: Muscle, strength, bone, libido, drive, anti-catabolic protection.
You pay: Polycythemia, left ventricular hypertrophy, cardiovascular strain, axis suppression, fertility loss. Castrated males live longer across species — the androgen cost is real.
The most direct handle on the master toggle, and the only one that is free. Asymmetric clocks: mTOR flips on in 1–3 hours; autophagy needs many hours of scarcity.
You gain: Autophagy, insulin sensitivity, metabolic flexibility, AMPK signalling, ketone production.
You pay: No building. Prolonged, it costs muscle; and it is genuinely harmful in the context of some medical conditions, medications, or a disordered-eating history.
You gain: Nutrient repletion, training fuel, muscle protein synthesis, hormonal support, recovery.
You pay: mTOR on, autophagy off — and every flip resets the cleanup clock to zero. This is why grazing all day never runs the maintain program.
People file this as a chain. It isn’t — leucine is one of the most direct mTORC1 activators you can swallow.
You gain: Permissive for autophagy; the fasting/cleanup arm can actually run.
You pay: Sarcopenia, poor recovery, worse bone, worse satiety, muscle loss with age.
You gain: Muscle preservation, satiety, bone, recovery, better body composition.
You pay: Chronic mTOR activation — the build arm never releases. “High protein all day, every day” is a decision about this switch that most people never notice making.
The best counterexample in the field — almost universally mis-filed as a chain. It’s hormetic: the signal is the adaptation trigger.
You gain: Less oxidative damage to DNA, proteins, and lipids.
You pay: Blunted training adaptation — demonstrated, not theoretical: high-dose vitamin C and E supplementation has been shown to reduce the benefits you get from exercise. Also impairs immune killing and mitochondrial biogenesis signalling.
You gain: A genuine adaptive stimulus — mitochondrial biogenesis, antioxidant enzyme upregulation, resilience.
You pay: Oxidative damage to mtDNA, membranes, and proteins; accelerated cellular ageing when chronic.
Hormetic, like ROS — and for the same reason. The stimulus is a stress; the adaptation is the response to it.
You gain: Full recovery, no injury risk, no cortisol load.
You pay: No adaptation. Deconditioning, sarcopenia, and losing the single best chain-lever you have.
You gain: More stimulus — up to the point where recovery can still service it.
You pay: Overreaching, chronically raised cortisol, HPG axis suppression (lowered testosterone), impaired immunity, injury. “Train more for T” inverts here.
The stress hormone people try to minimise — but a flat cortisol curve is a clinical problem, not an achievement.
You gain: Less catabolism, less anxiety.
You pay: Adrenal insufficiency; no cortisol awakening response; poor stress tolerance; inability to mount an acute response.
You gain: Acute mobilisation — alertness, glucose availability, the response you need in a real emergency.
You pay: Muscle catabolism, visceral fat deposition, insulin resistance, hippocampal damage, immune suppression, disrupted sleep.
A textbook U — both ends have names, and both are diseases.
You gain: Nothing beneficial. Lower metabolic rate is not a longevity strategy in humans.
You pay: Fatigue, weight gain, cognitive slowing, depression, dyslipidaemia, cold intolerance.
You gain: Nothing beneficial — the “energy” is a symptom.
You pay: Atrial fibrillation, bone loss, muscle wasting, anxiety, heat intolerance, cardiac strain.
Routinely treated as a women’s variable and as men’s enemy. Wrong on both counts — men need it too, and it’s a switch in everyone.
You gain: In men, less aromatisation-related gynecomastia.
You pay: Bone loss, joint pain, poor libido, adverse lipids, vascular and cognitive costs — in men as well as women. Crushing estradiol on TRT is a classic own goal.
You gain: Bone protection, vascular and cognitive support, libido.
You pay: In men: gynecomastia, water retention, mood. In women: proliferative and thrombotic risks depending on context.
Not a chain in either direction — which is why “boost your immune system” is an incoherent goal.
You gain: Less autoimmune and inflammatory burden.
You pay: Infection susceptibility, and degraded cancer surveillance — your immune system is a tumour-clearance mechanism.
You gain: Pathogen clearance, cancer surveillance, tissue repair signalling.
You pay: Autoimmunity, chronic inflammation, and the CD38-driven NAD⁺ leak that runs straight through your metabolic loop.
The purest illustration that “more” isn’t a plan — and a caution against any intervention promising to lengthen them.
You gain: Replicative senescence — which is, among other things, a tumour-suppressing mechanism.
You pay: Cellular senescence, tissue exhaustion, ageing phenotypes.
You gain: Greater replicative capacity, delayed senescence.
You pay: Associated with increased cancer risk in several analyses — more divisions means more chances for a mutation to matter.
The most reliable way to get this wrong isn’t bad reasoning — it’s vocabulary. Several everyday words cover a chain and a switch at the same time, so a sentence that’s true of one half gets applied to the other.
Oxidative stress. Everyone files it as a chain: ROS bad, antioxidants good, less is better. It’s a switch — and mis-filing it has a demonstrated cost, not a theoretical one. High-dose vitamin C and E supplementation has been shown to blunt the training adaptations from exercise: people took antioxidants to reduce oxidative stress and got measurably less benefit from their workouts. They applied chain logic to a switch and paid for it in the currency they were trying to earn. If you want one example of why this taxonomy isn’t philosophy, that’s it.
1. Audit the chains first. All eighteen. Find your worst. It’s free, it’s well-evidenced, and it’s probably not the thing you were reading about.
2. Measure the gauges, then correct only what’s low. Then stop. Re-test rather than escalate.
3. Only then approach the switches — and approach each one as a decision: name what you’re buying, name what you’re paying, and name for how long. If you can’t state the cost, you haven’t made a decision; you’ve made a purchase.
And the reason for the order isn’t moral, it’s arithmetic: the chain column is the longest, the best supported, and the only one that’s free. The switch column is the shortest, the least settled, and the most expensive. Working them in the reverse order — which is what the entire content ecosystem encourages — optimizes the smallest, weakest, costliest fraction of the map.
Your variables come in three kinds, and each takes a different instruction. Chains (18 of them — insulin sensitivity, inflammation, mitochondria, visceral fat, VO₂max, muscle, ApoB, blood pressure, sleep, gut, homocysteine, kidney/liver, bone, oral health, hearing/vision, social connection, cognitive engagement, toxin exposure): monotonic, no trade-off — improve, always. Gauges (8 — vitamin D, B12/folate, magnesium, zinc, iron, iodine/selenium, omega-3, sodium/potassium): a cliff at the low end, a wide flat optimum, a distant ceiling — test, fill if low, stop, and never supplement one you haven’t measured. Switches (12 — mTOR↔AMPK, GH/IGF-1, testosterone, estrogen, thyroid, cortisol, feed↔fast, protein, ROS, training stress, immune activation, telomeres): reciprocal or U-shaped, both ends carry a bill — decide a position. The clean discriminator: is there a real debate about the target? No for chains and gauges; yes for switches — and the debate is real, not ignorance. Beware the words that straddle two categories (“inflammation,” “body fat,” “oxidative stress,” “stress”), and note the case that proves it: antioxidants blunt training adaptation — chain logic applied to a switch, paid for in the exact currency you were chasing.
This article is for educational purposes only and is not medical advice, diagnosis, or treatment, and does not recommend any compound, dose, supplement, protocol, diet, or fasting regimen for any individual. It is a conceptual framework, not a testing or treatment plan. Hormonal agents including GH-axis drugs, testosterone, thyroid hormone, and corticosteroids are prescription compounds with significant effects, contraindications, and monitoring requirements, and are not appropriate for self-directed use. Supplements are not inherently safe: several named here (notably iron, zinc, selenium, iodine, and vitamin D) cause real harm in excess, and none should be taken on the basis of assumed deficiency. Fasting and caloric restriction are not appropriate for everyone and can be harmful in the context of certain medical conditions, medications, eating-disorder history, pregnancy, or age. The claim that optimizing health markers extends human lifespan is a hypothesis, not an established finding. Testing, interpretation, and any intervention belong with a qualified physician who can assess your individual situation.
This lesson relates to these health systems — health works as a connected system, not isolated topics.
Prerequisite: Chains and Switches: Why “Optimize Everything” Isn’t a Plan