Supplement science: what the evidence actually says about the supplements in your protocol

Supplement research quality varies enormously — from rigorous, multi-centre randomised controlled trials to single in-vitro studies on isolated cells. Before accepting any supplement claim, the relevant questions are: (1) Has this been studied in humans or only in cell cultures and animal models? (2) Was the study population similar to you (healthy adults vs elderly patients vs athletes)? (3) Was the dose used in the study achievable from the product you are considering? (4) Who funded the study? (5) Has the finding been independently replicated? ProtocolHub uses a four-level evidence classification: Strong (multiple independent RCTs in relevant human populations), Moderate (at least one quality RCT plus consistent mechanistic data), Emerging (mechanistically plausible with limited human trial data), and Anecdotal (primarily user reports, minimal controlled data). This framework is applied consistently across all supplement and peptide profiles on the site.

Three supplements have accumulated evidence bases that are, by supplement industry standards, genuinely strong. Creatine monohydrate is the most studied sports nutrition supplement in existence — over 500 published studies consistently demonstrating improvements in strength, power output, and lean mass. It also has emerging strong data for cognitive protection during ageing. Omega-3 EPA/DHA has strong evidence for cardiovascular risk reduction, anti-inflammatory activity, and mood support. The dose matters: 2–4g combined EPA/DHA daily is the level used in cardiovascular trials; lower doses produce smaller effects. Vitamin D deficiency is prevalent in Australia despite the climate (due to sun avoidance behaviours) and has been associated in observational studies with increased risk of over 20 chronic conditions — though intervention trials do not uniformly show disease prevention benefit from supplementation. The evidence supports supplementation to correct deficiency; the evidence for prevention in replete individuals is less clear.

Magnesium deficiency is common in the Australian population (estimated 50–60% of adults consuming below the RDI) and has documented effects on sleep quality, blood pressure, insulin sensitivity, and muscle function. The form matters significantly: magnesium glycinate and magnesium L-threonate have superior bioavailability and tolerability compared to magnesium oxide (the cheapest and most common form). NMN's evidence base for raising blood NAD+ levels in humans is now well-established from multiple RCTs — what remains uncertain is the degree to which this translates to the longevity and healthspan benefits seen in pre-clinical models. Collagen peptides have moderate evidence for improving skin elasticity and joint pain when taken with vitamin C — the cofactor for collagen synthesis. The benefit for tendons specifically has mechanistic support and is consistent with the recovery protocol rationale.

Ashwagandha (KSM-66 extract) has multiple human trials showing statistically significant reductions in cortisol, improvements in self-reported stress and anxiety, and modest but consistent improvements in testosterone levels in men with low-normal baseline testosterone. The evidence is emerging-to-moderate and the effect sizes are real but not dramatic. Berberine has attracted significant attention as an AMPK activator with mechanisms overlapping with metformin — clinical trials show meaningful reductions in fasting glucose and HbA1c in type 2 diabetes populations, and more modest effects in metabolically healthy individuals. It is a genuinely useful adjunct to GLP-1 fat loss protocols. Lion's Mane has moderate evidence for NGF upregulation and emerging evidence from small human trials for improvements in mild cognitive impairment — the mechanism is credible and the evidence trajectory is positive.

The supplement industry is poorly regulated for bioavailability claims, meaning that two products labelled identically can have dramatically different biological effects depending on the form of the active compound they contain. Magnesium oxide has approximately 4% bioavailability compared to 80%+ for magnesium glycinate — yet both appear on product labels as 'magnesium.' Turmeric supplements vary enormously in curcumin content and bioavailability — curcumin requires either fat or piperine for absorption, and most cheap turmeric capsules deliver negligible bioavailable curcumin. Fish oil quality varies by EPA/DHA concentration per gram (cheap fish oils may contain only 30% omega-3 per gram, requiring 6–8 capsules to reach the therapeutic dose), freshness (oxidised fish oil may be counterproductive), and molecular form (ethyl ester vs triglyceride form). ProtocolHub's supplement profiles include a 'What to look for' section for every supplement that addresses the form-specific quality markers most relevant for that compound.

Supplement underdosing is arguably as common as supplement fraud in the industry — products are formulated at the lowest dose that technically allows a mechanism claim while remaining cost-competitive. The clinical dose of omega-3 for cardiovascular benefit is 2–4g combined EPA/DHA daily. Most fish oil capsules contain 250–500mg EPA/DHA per capsule — meaning 4–16 capsules are required to reach the clinical dose. The clinical dose of ashwagandha in the cortisol-reduction trials was 600mg KSM-66 extract daily; most products contain 300mg per capsule. Magnesium L-threonate clinical trials used 144mg elemental magnesium as L-threonate daily — most products provide this in 3 capsules. When assessing whether a supplement 'worked' or 'didn't work' for you, the first question to ask is whether you were taking the dose that was actually studied. ProtocolHub's supplement profiles list the clinically studied dose for every supplement alongside the recommended form.