🥗 Nutrition & Supplements · 12 min read · Topic 8 of 10

Longevity Supplement Candidates: The Second Tier, Audited

Past the Big Five, the supplement shelf gets murkier and the marketing louder. This page audits the six names that dominate longevity podcasts: glycine, berberine, CoQ10, ashwagandha, astaxanthin, and resveratrol — the evidence for each, the doses trials actually used, and, because every active compound has a cost, the side effects and interactions nobody prints on the bottle.

🔎 Evidence Snapshot ★★★☆☆ Tiered — the best of this tier is moderate; the worst is preliminary

What the evidence supports

  • Glycine (3 g pre-bed) improved subjective sleep quality in small Japanese crossover RCTs — real trials, small, industry-funded.
  • Berberine lowers fasting glucose and HbA1c in meta-analysis — metformin-territory numbers built on mostly low-quality trials.
  • CoQ10 shows a modest systolic blood-pressure reduction (~−4.8 mmHg in meta-analysis) and signals in heart failure (Q-SYMBIO, n=420).
  • Ashwagandha reduces perceived stress and cortisol in multiple small RCTs — with a rare-but-real liver-injury flag.

What remains uncertain

  • Astaxanthin's skin evidence rests on small, short, largely industry-funded trials.
  • Resveratrol hits a bioavailability wall — null blood-pressure meta-analyses and a null large metabolic trial.
  • Long-term safety data are thin across the whole tier.

Evidence last reviewed: September 16, 2026. Conclusions may change as new research is published.

the second tier, audited

Product picks are generic categories, not brands. We may earn a commission on Amazon or iHerb purchases at no cost to you — this never changes our evidence conclusions. Full disclosure

Glycine powder

The sleep-trial molecule: 3 g before bed in the small crossover studies.

⚠️ Sweet-tasting; industry-funded evidence base; not an insomnia treatment.

Check price on Amazon →

Why a Second Tier Exists

The Big Five earned their slot by clearing a bar: hundreds of trials, or correction of a measurable deficiency, or both. The supplements below don't clear it. They have plausible mechanisms, trials that number in the dozens, and modest effect sizes even when the trials agree. That's what "second tier" means here — not "worthless," and not "the next creatine." Set expectations before the ledger: everything on this page sits at two to three stars of evidence. The best of the tier is moderate; none of it approaches the first tier's support. (NAD+ precursors like NMN, the seventh name on the influencer list, are deliberately deferred — the weakest human evidence of all, a biomarker-only story so far.)

6
Candidates audited — each with a watch-items column
2
Evidence tiers represented: moderate (★★★☆☆) and preliminary (★★☆☆☆)
0
Promises made — stars rate evidence quality, never certainty of results

The Ledger: Six Candidates at a Glance

This table is the page. Read it the way an auditor would: what the trials showed, what it seemed to do, and what it cost — on this site, no benefit is ever presented without its cost.

SupplementHuman evidenceBest-supported useMain watch-items (side effects / interactions)Verdict
🌙 Glycine Small crossover RCTs (n≈11–58, Japan, Ajinomoto-funded) on subjective sleep quality (Yamadera 2007; Inagawa 2006) Pre-bed sleep quality — subjective, not dramatic Sweet taste; GI upset at high doses; industry-funded, small evidence base ★★★☆☆ Moderate
🌿 Berberine 37-trial glucose meta-analysis: FPG −0.82 mmol/L, HbA1c −0.63% (Xie 2022) — pooled from mostly low-quality trials Glucose-lowering adjunct territory — conversations, not substitutions CYP3A4/2D6/2C9 interactions (cyclosporine, statins, many drugs); GI upset and constipation; contraindicated in pregnancy; hypoglycemia risk with diabetes medications ★★★☆☆ Moderate*
⚡ CoQ10 Blood-pressure meta-analysis (~−4.8 mmHg systolic; Moradi 2022); Q-SYMBIO (n=420) in heart failure (Mortensen 2014) Blood-pressure and heart-failure adjunct discussions Late-day dosing can disturb sleep; mild GI upset; may add to blood-pressure medication effects; warfarin caution ★★★☆☆ Moderate
🌰 Ashwagandha Multiple small stress/cortisol RCTs, generally consistent direction Stress and sleep support Rare but real liver injury (26-case review including one transplant; TGA 2024 safety alert); thyroid stimulation; adds to sedatives; avoid in pregnancy ★★★☆☆ Moderate ⚠️
🦐 Astaxanthin Small skin trials (4–12 mg, 8–16 weeks), largely industry-funded; meta-analysis exists (Nutrients 2021) Skin endpoints — elasticity, wrinkles, moisture Orange skin discoloration at high doses; insufficient pregnancy data ★★☆☆☆ Preliminary
🍇 Resveratrol Null blood-pressure meta-analysis (Liu 2014); the bioavailability wall (Walle 2004); STEP null on metabolic endpoints Mostly a lesson in hype versus evidence High-dose GI effects; possible kidney signal at 2.5–3 g/day (STEP); CYP interactions theoretical ★★☆☆☆ Preliminary

How to read this table: the stars rate the quality of the human evidence, never the size of the promise. The watch-items column is not a footnote; it exists because every biologically active compound has a cost, and marketing prints the benefit in large type and the cost in none. *Berberine's asterisk is the whole lesson of this page: its headline meta-analysis (Xie et al., 2022, PMID 36467075) pools 37 mostly low-quality trials, and Dong's earlier one (2012) reads the same way. Volume is not quality — see the chart below.

What "Moderate" Actually Means

This page is where the middle of the site's star system lives. Five stars is creatine territory — hundreds of trials, consistent results, clean safety data; four is strong cohorts with fewer trials. Three stars — "moderate" — means consistent small randomized trials: real blinding and placebo arms, but samples you can count without a calculator, durations of weeks, and endpoints that are often subjective. Two stars — "preliminary" — means early, mixed, or industry-shaped evidence where the next well-run trial could flip the sign.

Human RCT Count by Candidate
Approximate pooled randomized trials supporting each headline use. Counts include small, short, and low-quality trials — volume is not strength. CoQ10's bar includes Q-SYMBIO (n=420), the tier's largest single trial.
Berberine (glucose) ~40 CoQ10 (BP / heart) ~30 Resveratrol (BP / meta) ~20 Ashwagandha (stress) ~15 Astaxanthin (skin) ~12 Glycine (sleep) ~5 Colors match the verdict tiers: blue = moderate (★★★☆☆), amber = preliminary (★★☆☆☆).

The chart's quiet joke is glycine: five small trials carry the same podcast weight as berberine's forty — when you hear "well-studied," ask studied how well.

The Pattern in the Watch-Items Column

Read down the fourth column of the ledger and a pattern appears: the safety data are thinner than the efficacy data, for every row. Small trials are powered to detect whether a compound does something; they cannot catch uncommon harms. A trial of 40 people can show a supplement lowers cortisol; it would need tens of thousands to detect a one-in-ten-thousand liver injury. Ashwagandha's liver flag came not from trials but from case reports — 26 cases across 19 studies in one systematic review (Egyptian Journal of Internal Medicine, 2026), including one patient who needed a transplant, plus a parallel review of 25 patients (Cureus, 2025). Australia's Therapeutic Goods Administration issued a formal safety alert in February 2024. None of that was visible in the stress RCTs that made the supplement famous.

The Doses Trials Actually Used

Doses below describe the literature, not recommendations — they're here so any conversation with a clinician starts from what was actually studied, not a marketer's scoop size.

SupplementDoses used in trialsContext worth knowing
🌙 Glycine3 g before bedA single pre-bed dose; effects on subjective quality and next-day alertness
🌿 Berberine~1–1.5 g/day, divided 2–3 timesDivided dosing limits GI upset; the meta-analysis effect (−0.63% HbA1c) sat here
⚡ CoQ10100–300 mg/dayFat-soluble — take with a meal; the ~−4.8 mmHg average came from this range
🌰 Ashwagandha300–600 mg/day of standardized root extractThe stress/cortisol RCT range; more is not the studied direction
🦐 Astaxanthin4–12 mg/dayThe skin-trial range (8–16 weeks); discoloration reports sit at the high end and above
🍇 Resveratrol150–500 mg/day in most trialsFar above dietary intake; STEP's kidney signal appeared at 2.5–3 g/day

⚠️ Doses are descriptions, not prescriptions

Nothing on this page tells you to take anything. If you take prescription medications, are pregnant or planning to be, or manage a chronic condition, the dosing conversation belongs with a clinician who can see your full list. Never start, stop, or change a medication to accommodate a supplement.

Who Should Be Extra Careful

Product picks are generic categories, not brands. We may earn a commission on Amazon or iHerb purchases at no cost to you — this never changes our evidence conclusions. Full disclosure

CoQ10

The most medically grounded of the tier — 100–300 mg/day in BP and heart-failure research.

⚠️ Late-day doses can disturb sleep; clinician check on BP or blood-thinning medication.

Check price on Amazon →

Product picks are generic categories, not brands. We may earn a commission on Amazon or iHerb purchases at no cost to you — this never changes our evidence conclusions. Full disclosure

Ashwagandha root extract

300–600 mg/day of root extract in the stress RCTs — with a real liver caution label.

⚠️ Rare liver-injury reports; avoid in pregnancy and liver disease.

Check price on Amazon →

The Bottom Line

  1. This is a 2–3 star tier — the best of it (glycine, berberine, CoQ10, ashwagandha) is "moderate," and the rest is preliminary; nothing here is first-tier.
  2. Read the watch-items column first — berberine's enzyme interactions and ashwagandha's liver flag are real costs, not fine print.
  3. Trial volume is not trial quality — forty low-quality trials can out-count five good ones and still mean less.
  4. The fundamentals outrank the whole tier — nothing here outperforms the Big Five, protein, and fiber; on any medication, involve a clinician first.

Go Deeper: Longevity Supplement Candidates

Each candidate gets its own full audit — trial detail, dosing, and the interactions that matter on prescription drugs.

Related Topics

Sources & further reading