🌱 Welcome to Factnetize — honest, well-researched articles on tech, health and AI. Read more →
Health

The minimal health stack, ranked by evidence: an S-to-C tier list

Protein, creatine, D3, olive oil, omega-3, magnesium, collagen, K2, GlyNAC, zinc, boron — ranked S to C by the actual strength of the human evidence, with doses and every trial cited.

The minimal health stack, ranked by evidence: an S-to-C tier list

⚡ Key takeaways

  • This is a tier list, not a shopping list: supplements are ranked by how strong the human evidence actually is, S down to C.
  • Each entry gives you the short version first — dose, what it does, what to watch out for. The trial details sit in a separate grey box underneath.
  • S-tier (protein, creatine, vitamin D3 when deficient) has the largest, most repeated evidence. C-tier (boron, zinc, NAC solo) is plausible but thin.
  • This is general science, not personal medical advice — see the note at the end.

Ask ten biohackers for “the minimal stack” and you’ll get ten different lists, most sourced from a podcast rather than a paper. So this is the opposite approach: I pulled the actual randomized trials behind the commonly recommended supplements and ranked them S-tier down to C-tier, based on the strength of the evidence rather than the size of the marketing budget.

How to read this: every supplement below has two parts. First a short, plain-language summary with the dose and the one thing you need to know. Then a grey box — “The studies behind it” — with the trial names, numbers, and caveats, for anyone who wants to check the work. Skip the grey boxes entirely and you’ll still get the full picture.

The stack at a glance

Tier What For Dose Watch out for
S Protein (whey if needed) Keeping muscle after 50 1.0–1.2 g/kg/day Kidney disease → ask a doctor
S Creatine monohydrate Muscle & strength, 50+ 3–5 g/day Kidney disease → ask a doctor
S Vitamin D3 Correcting a deficiency 1,000–2,000 IU/day Test your level first
A High-polyphenol olive oil Cardiovascular events, inflammation 20–30 ml/day, ≥250 mg/kg Polyphenol level matters, not volume
A Omega-3 (EPA) High triglycerides, heart risk Trial dose: 4 g/day Blood thinners → ask a doctor
A Magnesium bisglycinate Sleep quality 250 mg elemental/day Works best if you’re low
B Creatine Brain/cognition 3–5 g/day Only holds up in older adults
B Creatine Artery stiffness 3–5 g/day Early evidence, older adults only
B Vitamin K2 (MK-7) + D3 Artery calcification 360–720 mcg/day MK-7 Warfarin → do not combine
B Collagen peptides Joints, bone, nails Target-specific: 40 mg UC-II to 15 g Not a whey substitute
B Vitamin C Required partner for collagen 50 mg is enough; more is harmless Deficiency rare in Europe
B Magnesium oxide Constipation Per label Different form than for sleep
B Magnesium (from food) Lower colorectal cancer risk Dietary intake Observational evidence only
B Glycine + NAC (GlyNAC) Cell energy (ATP) 2.4–7.2 g/day One research group so far
C Boron Free testosterone (men) 3–10 mg/day Very small studies
C Zinc Immune support 10–15 mg/day Stay under 40 mg total
C NAC (on its own) General antioxidant Evidence is medical, not general

What the tiers mean

  • S-tier — large trials or repeated meta-analyses. The kind of evidence guidelines get built on.
  • A-tier — solid trial evidence, but for a narrower group or a specific outcome.
  • B-tier — real trials pointing the same way, but small, short, or a modest effect.
  • C-tier — small pilot studies or mechanism plus animal data. Interesting, not settled.

The ranking is about strength of evidence, not how good the biochemical story sounds. A compound can have a beautiful mechanism and still land in C-tier if the human trials are small or contradicted.

S-Tier

Enough protein — the one that makes the rest work

1.0–1.2 g of protein per kg of bodyweight per day after ~50. Around 25 g per meal. Whey powder if you can’t hit it with food.

Strictly speaking protein is food, not a supplement — but it belongs at the top of this list because it’s the thing most people over 50 quietly fall short on, and because creatine barely works without it. Here’s the problem: as you age, your muscles become less responsive to the protein you eat. A 71-year-old needs roughly 140% more protein in a meal than a young adult to trigger the same muscle-building response. That’s called anabolic resistance, and it’s why the standard adult recommendation (0.8 g/kg) is now widely considered too low for older people.

Practical version: for an 80 kg person that’s 80–96 g of protein a day, spread across meals rather than dumped into dinner. If you eat plenty of meat, fish, eggs and dairy you may already be there. If you don’t — smaller appetite, skipped breakfast, mostly plant-based — a scoop of whey is a cheap, boring, effective way to close the gap. It isn’t magic; it’s just protein in convenient form. And it’s the foundation the creatine entry below actually depends on.

One important distinction: collagen powder does not do this job. It’s an incomplete protein — low in leucine, missing tryptophan — so it won’t drive muscle protein synthesis the way whey does. Collagen has its own separate uses (covered further down), but it is not a whey substitute.

📐 The studies behind it

The ESPEN Expert Group recommends at least 1.0–1.2 g protein/kg/day for healthy older people, rising to 1.2–1.5 g/kg/day for those who are malnourished or managing acute or chronic illness[47]. The PROT-AGE group adds a distribution rule: roughly 25 g protein per meal containing 2.5–2.8 g leucine, spread evenly through the day[48]. On the mechanism, older adults (~71 years) require about 140% more protein per meal than young adults to stimulate muscle protein synthesis at an equivalent rate, attributed to impaired mTOR signalling and reduced postprandial amino acid availability and muscle perfusion[49]. Observationally, elderly individuals consuming 1.2 g/kg/day lost 40% less lean body mass over three years than those at 0.8 g/kg/day[49], and intakes around 1.3 g/kg/day more consistently preserve appendicular lean mass under catabolic or training stress[50]. An indicator-amino-acid-oxidation study in older adults with diagnosed sarcopenia estimated requirements even higher, at 1.21 g/kg/day average and 1.54 g/kg/day recommended[51]. Caveat: people with kidney disease need individual medical guidance on protein intake.

Creatine monohydrate — muscle and strength

3–5 g per day. No loading phase needed.

This is the best-supported supplement on the list. From around age 50 you start losing muscle and strength every year; creatine combined with strength training measurably slows that down. It’s cheap, it’s been studied for decades, and it’s safe for healthy kidneys — if your blood test shows a bump in creatinine after starting, that’s a measurement quirk of how creatine breaks down, not damage. If you already have kidney disease, check with your doctor first.

📐 The studies behind it

The Devries & Phillips meta-analysis (2014, Med Sci Sports Exerc) found creatine plus resistance training beat resistance training alone for lean mass and strength in older adults[1], and it’s been reproduced repeatedly — including a 2021 meta-analysis in older females showing significant strength gains, especially with ≥24 weeks of training[2], and a 2025 review covering 357 older adults[3]. Muscle strength declines by about 1.5% a year between ages 50 and 60, and by roughly 3% a year after that[4]. On safety, a 2025 review in Frontiers graded kidney risk as “not supported” at Evidence Grade I for healthy people, based on trials up to 24 months at 5–20 g/day[5]; the creatinine rise is creatine spontaneously converting to creatinine, not renal dysfunction[6].

Vitamin D3 — but only if you’re actually low

1,000–2,000 IU per day. Test your blood level first.

Here’s the honest version: D3 reliably fixes the problems that a deficiency causes — bone, muscle, immune function. What it doesn’t appear to do is deliver extra benefits once your levels are already fine. The huge VITAL trial gave 2,000 IU daily to nearly 26,000 people and found no reduction in cancer or heart disease in a mostly non-deficient group. So the value is in correcting a real shortfall, not in taking it as a general upgrade. In Northern Europe, winter deficiency is common enough that testing is worth it. I’ve written separately about correcting a severe deficiency quickly, including how I overshot my own target while following careful advice.

📐 The studies behind it

VITAL (25,871 participants, 5.3-year follow-up, NEJM 2019) found daily 2,000 IU D3 did not significantly reduce invasive cancer or major cardiovascular events in a general, largely replete population[7], and a VITAL substudy found no effect on cognitive decline[8]. This is why the S-tier claim here is narrow: it’s about correcting deficiency, not about supplementation in already-sufficient people.

🧪 My own take: I aim for a blood level of 40–60 ng/mL rather than the ~20 ng/mL that most guidelines call “sufficient.” That cutoff was set for bone health specifically, and I’d rather sit in the range that researchers describe as optimal for broader health[45][46]. Practically that’s often 2,000–4,000 IU/day once you know your baseline. One thing worth separating clearly: that target range comes up in papers on the Brazilian “Coimbra protocol,” but that protocol itself uses tens of thousands of IU per day to treat active autoimmune disease under close medical supervision — that’s disease treatment, not a general wellness strategy. Don’t conflate the two. And to be fair to the evidence: no mega-trial has directly tested whether hitting 40–60 beats simply being adequate.

A-Tier

High-polyphenol olive oil — the hardest endpoints on this list

20–30 ml per day, taken straight. What matters is the polyphenol content, not the volume.

Olive oil is unusual here for one reason: the kind of evidence behind it. Almost everything else on this list rests on surrogate markers: a biomarker moving the right way, a plausible mechanism. Olive oil has hard clinical endpoints from a large randomized trial. In PREDIMED, roughly 7,400 people at high cardiovascular risk were randomized, and the extra virgin olive oil group had about 30% fewer heart attacks, strokes and cardiovascular deaths. Those aren’t lab values — those are events you actually want to avoid.

On top of that it carries something almost no supplement has: an approved EFSA health claim. The EU permits the statement that olive oil polyphenols protect blood lipids from oxidative stress, at 5 mg or more of hydroxytyrosol and derivatives per 20 g of oil. EFSA rejects this kind of claim routinely; the fact that this one passed means a panel judged there to be an established cause-and-effect relationship from dose-dependent human evidence.

So why A-tier and not S? Two reasons, and they’re the rubric doing its job. PREDIMED was retracted and republished in 2018 after randomisation irregularities at some sites — the reanalysis held, but it isn’t untouchable. And the benefit was shown in a high-risk population eating a Mediterranean pattern, not in the general population as a standalone habit. Strong trial evidence, narrower group: that’s the definition of A.

The therapeutic dose, precisely: the EFSA threshold is 5 mg hydroxytyrosol and derivatives per 20 g of oil (20 g ≈ 22 ml). That’s the number that matters, and it means volume alone tells you nothing — a supermarket oil at 100 mg/kg won’t reach it at any sensible dose, while a strong early-harvest oil clears it several times over in a single tablespoon. Practical version:

  • Aim for ≥250 mg/kg polyphenols — the level at which an oil can legally carry the EU claim.
  • 400+ mg/kg is the threshold popularised by Bryan Johnson’s protocol, who takes ~30 ml daily of an oil testing around 425 mg/kg — roughly 12 mg polyphenols per dose.
  • 1,000+ mg/kg exists in early-harvest single-variety oils (Coratina from Puglia is the classic example, sometimes exceeding 1,200 mg/kg). At that strength, 20 ml delivers around 25 mg — several times the EFSA threshold, at lower calorie cost.

One trap worth knowing: different labs measure polyphenols differently. The same oil can read ~400 mg/kg by HPLC and ~1,000 mg/kg by NMR. So compare numbers only when you know the method — check the certificate of analysis rather than the marketing line. Also, polyphenols degrade with light, heat, oxygen and time, so the harvest-date number isn’t what’s in the bottle a year later. Buy dark glass or bag-in-box, store it cool and dark, and prefer recent harvests.

📐 The studies behind it

PREDIMED randomized ~7,400 participants at high cardiovascular risk to a Mediterranean diet supplemented with extra virgin olive oil, one with nuts, or a control low-fat diet, reporting roughly a 30% reduction in major cardiovascular events in the EVOO arm[63]. Important caveat: PREDIMED was retracted and republished in 2018 after irregularities in randomisation were identified at some study sites; the reanalysis upheld the direction and broad magnitude of the findings, but the evidence is somewhat less bulletproof than the headline figure implies. The EFSA claim (EFSA Journal 2011;9(4):2033, implemented via EC Regulation 432/2012) permits “olive oil polyphenols contribute to the protection of blood lipids from oxidative stress” for oils containing ≥5 mg hydroxytyrosol and its derivatives per 20 g[64][65]. EFSA explicitly noted that many olive oils are too low in phenolics for a normal diet to reach that 5 mg intake[64]. High-phenolic EVOO is conventionally defined as ≥250 mg polyphenols/kg[65]. Mechanistically, hydroxytyrosol and tyrosol activate Nrf-2 signalling and suppress NF-κB activation — the two pathways most implicated in ageing and chronic disease[66]. A 2023 systematic review and meta-analysis of RCTs found olive oil polyphenols improved HDL cholesterol and lipid metabolism markers[67]. Coratina’s exceptional phenolic content is a varietal characteristic, amplified by early harvest[68]. Context caveat: the trial evidence comes from a Mediterranean dietary pattern in a high-risk population — this is not established as a “shot on top of an otherwise poor diet.”

Omega-3 (EPA) — for high triglycerides and heart risk

Context-dependent. The big trial used 4 g/day of purified EPA — that’s prescription strength.

Omega-3 is a good example of why “does it work?” is the wrong question. In people with high triglycerides and high cardiovascular risk, a high dose of purified EPA cut major heart events by about a quarter. In the general, lower-risk population, standard fish oil doses showed no meaningful benefit in two large trials. So it helps if you’re in the high-risk group or eat very little oily fish — it’s not a universal heart upgrade. The ratio side of this matters too, which I covered when I looked at what the research actually says about seed oils and omega-6 intake. If you’re on blood thinners, talk to your doctor.

📐 The studies behind it

REDUCE-IT (8,179 high-risk patients) found icosapent ethyl, a purified EPA ester at 4 g/day, reduced major cardiovascular events by roughly 25%[9], with the benefit tracking achieved blood EPA levels[10] and holding regardless of baseline LDL control[11]. Against that, the general-population VITAL trial found no significant reduction in its primary cardiovascular composite from standard-dose fish oil[12], and ASCEND was similarly null. The pattern: benefit concentrates in high-risk or EPA-deficient populations.

Magnesium bisglycinate — sleep quality

250 mg elemental magnesium per day. Bisglycinate form.

A real effect, but a modest one — and it works best if you were running low to begin with. In a trial of 155 poor sleepers, magnesium bisglycinate improved insomnia scores more than placebo, with most of the improvement showing up in the first two weeks. The people who benefited most were those with low magnesium in their diet. If you track recovery, better sleep tends to show up there as well — I wrote up what actually moved my own numbers in this HRV experiment. Note the form: bisglycinate is well absorbed and gentle on the stomach, which is exactly what you want here (see the gut section below for why oxide is a different tool entirely).

📐 The studies behind it

A 2025 randomized, double-blind, placebo-controlled trial (155 adults, 4 weeks) found magnesium bisglycinate reduced Insomnia Severity Index scores significantly more than placebo (−3.9 vs −2.3, p=0.049), though with a small effect size (Cohen’s d=0.2), and exploratory analysis pointed to bigger gains in those with lower baseline dietary magnesium[13][14]. A separate RCT using magnesium-L-threonate with objective Oura-ring measurement also found significant improvements in sleep and next-day function[15].

B-Tier

Creatine — for the brain

Same 3–5 g/day. But the honest answer is: unproven, except possibly in older adults.

This one is a good lesson in how science corrects itself. A 2024 meta-analysis reported that creatine improved memory, attention and processing speed — and it got a lot of coverage. Then reviewers found a statistical problem: several studies had counted multiple test results from the same participants as if they were separate people, which inflates the apparent evidence. When the numbers were re-run properly, the memory benefit disappeared, except specifically in older adults. Europe’s food safety authority reached the same conclusion. So: promising in older people, not established in general.

📐 The studies behind it

Xu et al. (2024, Front Nutr) concluded creatine improved memory, attention time and processing speed[16]. A later commentary flagged non-independent outcome pooling — multiple cognitive tests from the same participants inflating effective sample size[17]. On re-analysis with appropriate methods, the memory effect in Prokopidis et al.’s related 2023 meta-analysis was no longer significant except in older adults[18], and EFSA’s 2024 scientific opinion concluded no firm conclusions could be drawn from the pooled data[19].

Creatine — artery stiffness

Same dose. Early evidence, and it seems to depend on who you are.

Stiffer arteries are a real marker of cardiovascular ageing. A small trial in older men found a week of creatine improved stiffness measures, and another found it blunted the temporary stiffness spike after heavy exercise. But a study in young healthy adults found nothing at rest. Read that as: possibly useful for older or training populations, not a general “creatine is good for your arteries” claim.

📐 The studies behind it

A 2024 RCT in older men found 7 days of creatine improved arterial stiffness and atherosclerosis parameters measured by cardio-ankle vascular index[20]. An earlier trial found creatine suppressed the post-exercise rise in brachial-ankle pulse wave velocity after isokinetic exercise[21]. A study in young healthy adults found no independent effect on resting vascular measures[22].

Vitamin K2 (MK-7) with D3 — artery calcification

360–720 mcg/day MK-7 in the trials. Mixed results, and one hard rule.

The theory is elegant: K2 activates a protein that keeps calcium in your bones and out of your artery walls. The trials are less tidy. The largest one, in 389 older men over two years, found no overall effect on calcium buildup in arteries — though a subgroup that already had a lot of calcification did see slower progression. Bone markers respond fairly consistently; hard clinical outcomes haven’t been demonstrated. Worth knowing, not worth overselling.

⚠️ Important: K2 works directly against warfarin — doctors literally use vitamin K to reverse warfarin when someone is bleeding. If you take warfarin or another vitamin-K antagonist, do not add K2 without talking to your prescriber.

📐 The studies behind it

AVADEC (389 older men, 720 mcg MK-7 + 25 mcg D3/day, 24 months) found no significant overall effect on aortic valve or coronary calcium progression[23], though a subgroup with baseline CAC >400 showed significantly slowed progression[24]. A separate trial in type 2 diabetes found 6 months of K2 alone did not affect arterial calcification or bone mineral density[25]. A 2026 review concluded the matrix-Gla-protein mechanism is sound and bone biomarker effects reasonably consistent, but that no adequately powered trial has shown the combination beating its individual components on a hard clinical endpoint[26]. On the interaction: menaquinone antagonises vitamin-K-antagonist anticoagulants directly[27].

Collagen peptides — joints, bone and connective tissue

Dose depends entirely on the target: 40 mg UC-II or 10 g for joints, 5 g for bone, 2.5 g for nails. Always with vitamin C. Not a whey substitute.

Collagen is one of the most over-marketed supplements on the shelf, and there is still a real evidence base underneath the noise — but it’s target-specific, and the dose that works for one thing isn’t the dose for another. The short version:

Goal Dose Evidence strength
Joints / cartilage 10 g/day hydrolysed, or 40 mg/day UC-II Strongest case
Bone (postmenopausal) 5 g/day Strong, but one ingredient maker
Tendons (around training) 15 g, 60 min before loading Synthesis markers only
Nails 2.5 g/day Well-supported, underrated
Skin 2.5–10 g/day Modest, funding-biased
Muscle Doesn’t work; use whey

The confusion worth clearing up first, because it costs people money: whey and collagen are not interchangeable. Whey is a complete protein — all essential amino acids, plenty of leucine — which is what triggers muscle building. Collagen is an incomplete protein: no tryptophan, and it supplies only around a third of the leucine you’d get from an equivalent amount of whey protein. It will not build muscle the way a complete protein does. What it does have is an unusual amino acid profile (heavy on glycine, proline and hydroxyproline) that supplies building blocks for connective tissue. Different tool, different job — take them together, not one instead of the other.

Joints are collagen’s best use. Two quite different approaches both hold up: around 10 g/day of hydrolysed collagen works nutritionally, with absorbed peptides accumulating in cartilage; or just 40 mg/day of undenatured type II collagen (UC-II), which works by an immune mechanism and beat glucosamine plus chondroitin head-to-head for knee osteoarthritis.

Bone density after menopause is the finding that makes collagen worth a second look. Most nutrients at best slow bone loss; actually increasing density is rare. In a 12-month double-blind trial, 131 postmenopausal women taking 5 g/day of specific collagen peptides gained bone density in the spine (+3.0%) and femoral neck (+6.7%) while the placebo group lost it. Bone formation markers moved the right way, and a four-year follow-up showed the gains held.

Nails are the underrated one: 2.5 g/day for 24 weeks increased nail growth by 12% and cut breakage by 42%. Skin is the weakest case — and the most heavily marketed. A 2025 meta-analysis separated trials by funding source and found that in studies not funded by supplement makers, the effect disappeared.

⚠️ Quality matters more here than for most supplements. One analysis of 28 top-selling collagen products found measurable arsenic in 64%, lead in 37% and mercury in 34%. Marine collagen carries the higher heavy-metal risk because it bioaccumulates. Buy third-party-tested bovine collagen with a published heavy-metal certificate of analysis.

The full breakdown — skin, bones, joints, muscle, nails and hair, the funding-bias problem, sourcing, and my own protocol — is in the dedicated collagen article.

📐 The studies behind it

Bone: König et al. (2018, Nutrients) randomized 131 postmenopausal women with age-related BMD reduction to 5 g specific collagen peptides or placebo for 12 months; spine BMD +3.0% vs −1.3% placebo, femoral neck +6.7% vs −1.0% (full text, Figure 3; the abstract reports the equivalent T-score shifts: spine +0.1 vs −0.03, p=0.030; femoral neck +0.09 vs −0.01, p=0.003), with P1NP (formation) rising in the collagen group and CTX-1 (degradation) rising in controls[56]. A four-year open-label follow-up showed continued spine gains[57]. Caveat: one manufacturer’s specific ingredient, industry-supported, not independently replicated, and no fracture-reduction data. Joints: Lugo et al. (Nutr J 2016) found 40 mg/day UC-II outperformed glucosamine plus chondroitin for knee osteoarthritis; separate trials support ~10 g/day hydrolysed collagen in athletes with joint pain over 24 weeks. Muscle: Oikawa et al. (Am J Clin Nutr 2020) confirmed whey but not collagen drives muscle protein synthesis; Zdzieblik et al. (Br J Nutr 2015) found collagen plus resistance training improved fat-free mass in sarcopenic older men, likely via connective tissue rather than contractile fibres. Nails: Hexsel et al. (J Cosmet Dermatol 2017), 2.5 g/day, 24 weeks, +12% growth rate and −42% breakage. Skin: Myung & Park (Am J Med 2025) found the pooled effect significant overall but absent in non-industry-funded and high-quality trials.

Vitamin C — mostly as collagen’s required partner

Deficiency is rare in Europe. But if you take collagen, take vitamin C with it — 50 mg is enough, and more does no harm.

Vitamin C and collagen belong together, and it’s stronger than a “works nicely with” claim: your body physically cannot finish building collagen without vitamin C. Collagen isn’t assembled like a normal protein — after the chains are made, specific building blocks have to be chemically modified before the whole thing can fold into its stable rope-like structure. The enzymes that do that modification need vitamin C to function. No vitamin C, no stable collagen. That’s literally what scurvy is: connective tissue falling apart.

The practical consequence: taking collagen peptides without vitamin C is giving your body the raw materials while withholding a tool it needs to use them. The studied amount is small — around 50 mg alongside the collagen is plenty to keep those enzymes running. Higher doses are common and harmless within sensible limits, they just aren’t doing more for collagen specifically.

On its own, though, vitamin C earns a modest ranking. Frank deficiency is uncommon on a normal European diet, and it doesn’t do much extra once you’re topped up — the same “corrects a shortfall, doesn’t upgrade a sufficient person” logic as vitamin D. It’s cheap and safe, so pairing it with collagen is sensible; taking grams of it daily as a general health measure isn’t supported.

📐 The studies behind it

Ascorbic acid is an obligatory cofactor for prolyl hydroxylase and lysyl hydroxylase, the enzymes that hydroxylate proline and lysine residues in procollagen — the post-translational step required for correct folding and cross-linking of the collagen triple helix[52][53]. Collagen makes up roughly 85–95% of tendon dry weight, so this matters structurally[52]. A systematic review of vitamin C after musculoskeletal injury found supportive basic-science and animal evidence for enhanced collagen synthesis and cross-linking, with human clinical evidence more limited[53]. The pre-loading protocol used 48–50 mg vitamin C alongside gelatine[54]. A separate RCT in Achilles tendinopathy compared a collagen peptide, chondroitin sulphate, hyaluronate and vitamin C combination against diclofenac[55] — promising, but combination products make it hard to isolate vitamin C’s contribution.

Magnesium and your gut — two different things

For constipation: magnesium oxide. For everything else: bisglycinate. These are not interchangeable.

This trips people up constantly, so it’s worth being clear. Magnesium oxide works as a laxative because it’s poorly absorbed — it stays in your intestine and pulls water in with it. That’s great if you’re constipated and useless if you’re trying to raise your body’s magnesium levels. Bisglycinate is the opposite: well absorbed, gentle, gets into your system. Pick the form based on what you actually want.

On colorectal cancer: there’s a consistent finding that people who eat more magnesium have somewhat lower rates of colorectal cancer — roughly 10–12% lower in the highest-intake groups, strongest for colon cancer specifically. The mechanism makes sense (magnesium is involved in DNA repair and controlling inflammation). But be careful how much weight you put on it: this comes from observational studies tracking diet, not from trials where people were given magnesium supplements and followed for cancer outcomes. Real signal, modest size, not proof.

📐 The studies behind it

Magnesium oxide’s osmotic laxative action follows from its low bioavailability, roughly 4–15%[28][41]; chelated bisglycinate shows better absorption and tolerability, particularly in people with impaired magnesium uptake[42]. On cancer risk: a 2012 meta-analysis of 8 prospective studies (~339,000 participants) found the highest-vs-lowest intake group had an 11% lower colorectal cancer risk (RR 0.89, 95% CI 0.79–1.00), with colon cancer specifically at RR 0.81[43]; a separate meta-analysis found each 100 mg/day increment associated with 12% lower CRC risk and 13% lower adenoma risk[44]. Proposed mechanisms include DNA synthesis and repair, regulation of proliferation and apoptosis, and inflammatory modulation[44]. All of this is epidemiological. Microbiome claims remain largely mechanistic and animal-based[29][30].

Glycine + NAC (GlyNAC) — cellular energy

2.4–7.2 g/day combined, glycine and NAC in a 1:1 ratio.

This is the most interesting finding in the whole list, and also the one most in need of confirmation. As you age, your mitochondria — the parts of your cells that make energy — get worse at burning fat for fuel and start leaning on glucose instead. In a trial of older adults, GlyNAC reversed that: fat-burning capacity improved by 78% over 16 weeks, back to levels seen in young adults. What makes it unusual is that the researchers looked inside muscle tissue and found the actual energy-production machinery being rebuilt, including the enzyme that physically manufactures ATP. The placebo group showed none of it. The catch: it’s 24 people, from one lab. Impressive, not yet confirmed.

📐 The studies behind it

In a randomized trial (24 older adults, 16 weeks vs isonitrogenous alanine placebo, J Gerontol A 2023), older participants started with 33–42% lower mitochondrial fatty-acid oxidation and elevated glucose oxidation. GlyNAC improved fatty-acid oxidation 29% at 2 weeks and 78% at 16 weeks, normalising values to those of young adults[31][32]. Muscle biopsy showed significantly improved expression of PGC1α (mitochondrial biogenesis), CPT1b (fatty-acid transport), HADHA (β-oxidation), electron transport chain complexes I, II, III and V, and ATP5A — a subunit of ATP synthase itself[33]. No equivalent changes in placebo. Limitation: modest sample, and the work comes largely from Sekhar’s group at Baylor; independent replication would move this to A-tier.

C-Tier

Boron — free testosterone in men

3–10 mg/day in the studies.

Boron is cheap, found in fruit and nuts, and has a wide safety margin — but the evidence is genuinely thin. The most-quoted study gave 10 mg a day to eight men for one week and found free testosterone went up and estradiol went down. Eight men, one week, no placebo group. That’s a hint, not a finding. It’s a reasonable cheap add-on if you’re curious; it’s not something to build a plan around.

🧪 My own take: I take 6 mg a day and noticed a subjective effect within one to two weeks — which sits right in the range those small studies used. One person’s experience isn’t evidence, and the same caveats about tiny samples and short timeframes apply to me as to the published work.

📐 The studies behind it

Naghii et al. (2011, J Trace Elem Med Biol) supplemented 8 healthy men with 10 mg boron daily for one week, reporting increased plasma free testosterone and decreased estradiol, with SHBG reduction as the proposed mechanism[34]. Nielsen et al. (1987) found 3 mg/day over 7 weeks raised both estradiol and testosterone in postmenopausal women[35]. Small, short, and mostly non-blinded.

GlyNAC — the broader anti-ageing claims

Same 2.4–7.2 g/day, but a much weaker claim than the energy one above.

Separate from the specific mitochondrial finding, GlyNAC is often sold as reversing multiple “hallmarks of ageing” at once — inflammation, insulin resistance, cognition, strength. That broader claim leans on the same small trial, and a bigger follow-up found the benefit concentrated in people who started out with high oxidative stress and low glutathione. In other words: helps most if you actually need it, less impressive across the general population.

📐 The studies behind it

The multi-system claim rests on the same 24-person, 16-week trial reporting improvements in glutathione, oxidative stress, inflammation, insulin resistance, endothelial function, strength and some cognitive measures[36]. A larger dose-ranging RCT (114–117 healthy older adults, 2 weeks) found significant glutathione increases specifically in participants above median oxidative stress and below median baseline glutathione[37].

Zinc — and why more is not better

10–15 mg/day. Keep your total from food plus supplements under 40 mg.

Fixing an actual zinc deficiency clearly helps immune function. The problem with zinc isn’t whether it works — it’s people taking too much for too long. Zinc and copper compete for the same absorption route in your gut, so sustained high doses can quietly cause a copper deficiency, which shows up as anaemia and low white blood cells. This has landed people in hospital. Modest dose, don’t stack multiple products containing it, and don’t run high doses for months.

📐 The studies behind it

Zinc and copper share intestinal transport pathways; chronic intake around 50–60 mg/day has produced signs of copper deficiency within 10 weeks[38]. Case reports document zinc-induced copper deficiency presenting as anaemia and neutropenia after prolonged high-dose use[39]. A 2023 Br J Clin Pharmacol study argued that prescribing guidance permitting up to 135 mg/day is too permissive and should be revised downward[40]. The adult tolerable upper intake level is 40 mg/day.

NAC on its own

Strong evidence — but not for what people take it for.

NAC is genuinely important medicine: it’s the hospital treatment for paracetamol overdose and works as a mucolytic. As a general “antioxidant for longevity” supplement, the evidence is much thinner than its reputation suggests. Most of NAC’s recent positive ageing data comes from the GlyNAC combination above, paired with glycine — not from NAC alone.

Timing: where it matters, and where it doesn’t

Most supplement timing advice is noise. Two cases here are worth getting right, and they point in opposite directions — one where timing genuinely matters, and one where the received wisdom is largely a marketing story.

Collagen + vitamin C: 60 minutes before training — timing really matters

Take 15 g collagen or gelatine with ~50 mg vitamin C, roughly an hour before you load the tendon.

This one has a specific, well-explained mechanism. After you swallow collagen, the relevant amino acids — glycine, proline, hydroxyproline — take about 60 minutes to peak in your blood. So the hour isn’t arbitrary: you’re timing that peak to land exactly when you start training. The second half of the reason is anatomical: tendons and ligaments have poor blood supply at rest, and exercise is what opens up circulation to them. Deliver the building blocks while the tap is open, and they actually reach the tissue.

In the trial that established this, 15 g of vitamin-C-enriched gelatine an hour before six minutes of rope skipping doubled the blood marker for collagen synthesis compared to placebo. Dose mattered too — 15 g worked, 5 g didn’t produce the same effect. That’s why this number is higher than the 5 g daily bone protocol above: different goal, different study, different amount. The 5 g is a steady daily intake for bone; the 15 g is a training-day dose timed to the session. This is one of the few timing protocols in nutrition with a clear reason behind it rather than a vague “for best results.”

Whey after training: much less critical than you’ve been told

Total daily protein is what matters. The 30-minute window is largely a myth.

The idea that you must drink a shake within half an hour of your last set or waste the session is probably the most successful piece of marketing in supplement history — and the evidence doesn’t support it. Here’s the twist that unravelled it: a meta-analysis of 23 controlled trials found that the studies apparently showing a post-workout timing benefit had a hidden flaw. The groups told to eat right after training were also eating 25% more protein per day overall. Once total daily intake was accounted for, the timing advantage vanished completely — for muscle growth and for strength. What looked like a window was a dosing difference in disguise.

A direct test made the same point: ten weeks of training with whey taken before versus after the session produced no meaningful difference in muscle or strength. Muscle protein synthesis stays elevated for many hours after training, so the practical window is more like 4–6 hours around your session than 30 minutes after it. If you ate protein an hour or two before training, those amino acids are still circulating anyway.

None of that makes a post-workout shake wrong — it’s a convenient, easy-to-remember habit. It just isn’t load-bearing. If you’re going to spend effort optimising something, spend it on hitting your daily protein target and spreading it across meals, not on watching the clock.

📐 The studies behind it

Collagen timing: Shaw et al. (2017, Am J Clin Nutr 105:136-143) ran a randomized, double-blinded crossover in 8 healthy men consuming placebo, 5 g or 15 g vitamin-C-enriched gelatine (48 mg vitamin C) one hour before 6 minutes of rope skipping, repeated 3×/day over 3 days. Circulating glycine, proline, hydroxyproline and hydroxylysine peaked one hour after ingestion; the 15 g group showed double the amino-terminal propeptide of collagen I (PINP), indicating roughly doubled collagen synthesis[54][58]. Engineered ligaments treated with participants’ post-supplement serum showed increased collagen content and improved mechanics[58]. The 60-minute interval derives directly from the amino acid peak timing[59]. Caveat: 8 participants, all young men, with synthesis markers rather than injury outcomes as the endpoint.

Protein timing: Schoenfeld, Aragon & Krieger (2013, J Int Soc Sports Nutr 10:53) meta-analysed 23 RCTs (525 subjects). Studies showing a timing benefit had timing groups consuming 1.66 vs 1.33 g/kg/day — 25% more protein. Controlling for total protein intake as a covariate, the timing effect was non-significant for both hypertrophy (p=0.18) and strength (p=0.49)[60]. A 10-week RCT in resistance-trained men found pre- versus post-exercise protein produced similar muscular adaptations[61]. Reviews place the practical window at 4–6 hours around training rather than 30–60 minutes post-exercise, with daily intake around 1.6–2.2 g/kg the dominant variable[62].

Frequently asked questions

If I only took two things from this list, what would they be?

On evidence strength alone: creatine monohydrate, and vitamin D3 after testing your level. Those two have the biggest, most repeated evidence bases here, and creatine has an unusually clean safety record after decades of study. Hitting your daily protein target isn’t a supplement, but it outranks both.

Why is vitamin D3 S-tier if VITAL found no benefit?

Because the S-tier claim is narrow: correcting an actual deficiency. VITAL tested D3 in people who mostly weren’t deficient and found no bonus on top. That’s a different question from “does fixing a deficiency help,” which it does.

Why does GlyNAC show up twice, in two different tiers?

Because it’s two claims wearing one name. The narrow one — that it rebuilds mitochondrial energy machinery, measured directly in muscle tissue — has decent trial evidence and sits at B-tier. The broad one — that it reverses ageing across many systems at once — leans on the same small study and needs independent replication, so it stays C-tier.

Can collagen replace my protein powder?

No. Collagen is missing tryptophan and low in leucine, so it doesn’t trigger muscle protein synthesis the way whey does. It counts toward your total grams of protein but does a different job. Take them together if you want both muscle and connective tissue support.

Can I take several of these together?

Most have no known problem interaction. Two real exceptions: K2 with warfarin or any vitamin-K antagonist, and long-term high-dose zinc, which can crowd out copper. Beyond that, watch for the same ingredient appearing in several products and pushing your total dose higher than you think.

Does S-tier mean I should take it?

No. The tier reflects how strong the evidence is, not whether it fits you. D3 is S-tier for people who are deficient — if you’re already fine, more does little. Test where you can, and food before supplements.

Sources

  1. Devries MC, Phillips SM. Creatine Supplementation during Resistance Training in Older Adults—A Meta-analysis. Med Sci Sports Exerc. 2014;46(6):1194-1203.
  2. Efficacy of Creatine Supplementation Combined with Resistance Training on Muscle Strength and Muscle Mass in Older Females: A Systematic Review and Meta-Analysis. PMC. 2021.
  3. Impact of creatine supplementation and exercise training in older adults: a systematic review and meta-analysis. Eur Rev Aging Phys Act. 2025.
  4. von Haehling S, Morley JE, Anker SD. An overview of sarcopenia: facts and numbers on prevalence and clinical impact. J Cachexia Sarcopenia Muscle. 2010;1(2):129-133.
  5. A short review of the most common safety concerns regarding creatine ingestion. Front Nutr / PMC. 2025.
  6. Is It Time for a Requiem for Creatine Supplementation-Induced Kidney Failure? A Narrative Review. PMC. 2023.
  7. Manson JE, et al. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. N Engl J Med. 2019;380:33-44.
  8. Effect of vitamin D on cognitive decline: results from two ancillary studies of the VITAL randomized trial. PMC. 2021.
  9. Bhatt DL, et al. REDUCE-IT: Profound reductions in first and total cardiovascular events with icosapent ethyl. PMC. 2020.
  10. REDUCE-IT EPA Trial Shows Association Between Higher EPA Levels, Reduced CV Events. American College of Cardiology. 2020.
  11. Cardiovascular Outcomes With Icosapent Ethyl by Baseline LDL-C: A Secondary Analysis of REDUCE-IT. J Am Heart Assoc. 2024.
  12. Marine n-3 Fatty Acids and Prevention of Cardiovascular Disease: A Novel Analysis of the VITAL Trial. PMC. 2023.
  13. Schuster J, et al. Magnesium Bisglycinate Supplementation in Healthy Adults Reporting Poor Sleep: A Randomized, Placebo-Controlled Trial. Nat Sci Sleep. 2025.
  14. ibid.
  15. Magnesium-L-threonate improves sleep quality and daytime functioning in adults with self-reported sleep problems. ScienceDirect. 2024.
  16. Xu C, et al. The effects of creatine supplementation on cognitive function in adults: a systematic review and meta-analysis. Front Nutr. 2024.
  17. Commentary: The effects of creatine supplementation on cognitive function in adults. Frontiers. 2026.
  18. Eckert I, Pascher E. Letter to the editor, re: Prokopidis et al. creatine and memory meta-analysis. Nutr Rev. 2023.
  19. European Food Safety Authority. 2024 scientific opinion on creatine and cognition.
  20. Effects of acute creatine supplementation on cardiac and vascular responses in older men; a randomized controlled trial. ScienceDirect. 2024.
  21. Creatine supplementation attenuates hemodynamic and arterial stiffness responses following an acute bout of isokinetic exercise. Eur J Appl Physiol. 2011.
  22. The effect of creatine supplementation on the response of central and peripheral pulse wave velocity to high-intensity resistance exercise. Taylor & Francis. 2018.
  23. Vitamin K2 and D in Patients With Aortic Valve Calcification: A Randomized Double-Blinded Clinical Trial (AVADEC). Circulation. 2022.
  24. Targeting Coronary Artery Calcium: Vitamins K2 and D in Cardiovascular Health. Balchem. 2026.
  25. Six months vitamin K treatment does not affect systemic arterial calcification or bone mineral density in diabetes mellitus 2. Eur J Nutr / PMC. 2021.
  26. Beyond Bone Health: Exploring the “Heart-Brain-Bone” Axis Modulated by Lipid-Soluble Nutrients. PubMed. 2026.
  27. Vitamin K2 and Warfarin drug interaction data. Drugs.com.
  28. Cymbiotika. Does Magnesium Help with Gut Health? Exploring the Connection.
  29. The essential role of magnesium in immunity and gut health: Impacts of dietary magnesium restriction. PubMed. 2025.
  30. Magnesium Orotate and the Microbiome-Gut-Brain Axis Modulation. MDPI Nutrients. 2022.
  31. Sekhar RV, et al. Supplementing Glycine and N-Acetylcysteine (GlyNAC) in Older Adults Improves Glutathione Deficiency, Oxidative Stress, Mitochondrial Dysfunction, Inflammation, Physical Function, and Aging Hallmarks. J Gerontol A Biol Sci Med Sci. 2023;78(1):75-89.
  32. ibid., mitochondrial fatty-acid oxidation results.
  33. ibid., muscle expression of PGC1α, CPT1b, HADHA, electron transport complexes I/II/III/V, and ATP5A.
  34. Naghii MR, et al. Comparative effects of daily and weekly boron supplementation on plasma steroid hormones and proinflammatory cytokines. J Trace Elem Med Biol. 2011.
  35. Nielsen FH, et al. Effect of dietary boron on mineral, estrogen, and testosterone metabolism in postmenopausal women. 1987.
  36. Kumar P, et al. Glycine and N-acetylcysteine (GlyNAC) supplementation in older adults improves glutathione deficiency, oxidative stress, mitochondrial dysfunction. Clin Transl Med. 2021.
  37. A Randomized Controlled Clinical Trial in Healthy Older Adults to Determine Efficacy of Glycine and N-Acetylcysteine Supplementation on Glutathione Redox Status and Oxidative Damage. PMC. 2022.
  38. Zinc Deficiency. StatPearls / NCBI Bookshelf. 2025.
  39. Anemia Due to Unexpected Zinc-Induced Copper Deficiency. PMC. 2025.
  40. Duncan A, et al. Iatrogenic copper deficiency: Risks and cautions with zinc prescribing. Br J Clin Pharmacol. 2023.
  41. Lindberg JS, et al. Magnesium bioavailability from magnesium citrate and magnesium oxide. J Am Coll Nutr. 1990;9(1):48-55.
  42. Bioavailability of magnesium diglycinate vs magnesium oxide in patients with ileal resection. JPEN. 1994.
  43. Chen GC, et al. Magnesium intake and risk of colorectal cancer: a meta-analysis of prospective studies. Eur J Clin Nutr. 2012;66(11):1182-1186.
  44. Dai Q, et al. Magnesium intake and colorectal tumor risk: a case-control study and meta-analysis. Am J Clin Nutr. 2012;96(3):622-631.
  45. Safety Data in Patients with Autoimmune Diseases during Treatment with High Doses of Vitamin D3 According to the “Coimbra Protocol.” Nutrients. 2022;14(8):1575.
  46. Physiology of Vitamin D — Focusing on Disease Prevention. PMC. 2024.
  47. Deutz NEP, et al. Protein intake and exercise for optimal muscle function with aging: Recommendations from the ESPEN Expert Group. Clin Nutr. 2014;33(6):929-936.
  48. Bauer J, et al. PROT-AGE Study Group recommendations on protein intake and per-meal leucine distribution for older adults.
  49. Dietary protein and amino acid intakes for mitigating sarcopenia in humans. Crit Rev Food Sci Nutr. 2024.
  50. Nutritional Interventions: Dietary Protein Needs and Influences on Skeletal Muscle of Older Adults. J Gerontol A Biol Sci Med Sci. 2023;78(Suppl 1):67.
  51. Dietary protein requirements of older adults with sarcopenia determined by the indicator amino acid oxidation technology. PMC. 2025.
  52. Effect of Vitamin C Injection on Flexor Tendon Healing in Zone II: A Randomized Controlled Trial. Cureus. 2026.
  53. DePhillipo NN, et al. Efficacy of Vitamin C Supplementation on Collagen Synthesis and Oxidative Stress After Musculoskeletal Injuries: A Systematic Review. Orthop J Sports Med. 2018.
  54. Effect of Collagen/Vitamin C in Jumper’s Knee: a randomized controlled trial (JUMPFOOD study). ClinicalTrials.gov NCT05407194.
  55. Comparing Effectiveness of Combination of Collagen Peptide Type-1, Chondroitin Sulphate, Sodium Hyaluronate and Vitamin C Versus Oral Diclofenac in Achilles Tendinopathy. PMC. 2021.
  56. König D, Oesser S, Scharla S, Zdzieblik D, Gollhofer A. Specific Collagen Peptides Improve Bone Mineral Density and Bone Markers in Postmenopausal Women—A Randomized Controlled Study. Nutrients. 2018;10(1):97.
  57. Specific Bioactive Collagen Peptides in Osteopenia and Osteoporosis: Long-Term Observation in Postmenopausal Women. J Bone Metab. 2021;28(3):207.
  58. Shaw G, Lee-Barthel A, Ross MLR, Wang B, Baar K. Vitamin C-enriched gelatin supplementation before intermittent activity augments collagen synthesis. Am J Clin Nutr. 2017;105(1):136-143.
  59. Gelatin, Collagen, Gummy and PINP. ClinicalTrials.gov NCT03357432.
  60. Schoenfeld BJ, Aragon AA, Krieger JW. The effect of protein timing on muscle strength and hypertrophy: a meta-analysis. J Int Soc Sports Nutr. 2013;10:53.
  61. Schoenfeld BJ, et al. Pre- versus post-exercise protein intake has similar effects on muscular adaptations. PeerJ / PMC. 2017.
  62. Aragon AA, Schoenfeld BJ. Is There a Postworkout Anabolic Window of Opportunity for Nutrient Consumption? Clearing up Controversies. J Orthop Sports Phys Ther. 2018;48(12):911-914.
  63. Estruch R, et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet (PREDIMED). N Engl J Med. Originally 2013; retracted and republished 2018;378:e34.
  64. EFSA Panel on Dietetic Products, Nutrition and Allergies. Scientific Opinion on the substantiation of health claims related to polyphenols in olive oil. EFSA Journal. 2011;9(4):2033.
  65. Exploring the Benefits of Extra Virgin Olive Oil on Cardiovascular Health Enhancement and Disease Prevention: A Systematic Review. PMC. 2025.
  66. Serreli G, Deiana M. Extra Virgin Olive Oil Polyphenols: Modulation of Cellular Pathways Related to Oxidant Species and Inflammation in Aging. Cells. 2020;9(2):478.
  67. Zupo R, et al. Olive Oil Polyphenols Improve HDL Cholesterol and Promote Maintenance of Lipid Metabolism: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. Metabolites. 2023;13(12):1187.
  68. Bellumori M, et al. The EFSA Health Claim on Olive Oil Polyphenols: Acid Hydrolysis Validation and Total Hydroxytyrosol and Tyrosol Determination in Italian Virgin Olive Oils. Molecules. 2019;24(11):2179.

This is general science journalism, not personal medical advice. None of the above is a recommendation to start, stop, or combine supplements for your specific situation — dosing, interactions, and what’s appropriate depend on your health history, medications, and lab values. If you’re on any prescription medication (especially blood thinners), pregnant, or managing a health condition, talk to a doctor or pharmacist before adding any of these. This article reflects the evidence as of the trials cited above; supplement science moves, and some of these gradings will shift as bigger trials report.

Pablo Vicento
Written by

Pablo Vicento

Liked this? There's more where it came from.

Get our digest — articles worth your time, no spam, unsubscribe in one click.

Subscribe to Factnetize →

Leave a comment

Your email address will not be published. Required fields are marked *

Weekly · No spam

Get smarter,
one Sunday at a time.

Join our weekly digest — the articles worth your time, plus one thing that made us think differently.