The public conversation about food has been narrowed for decades into arguments about macronutrients. The arguments are useful at the edges and largely beside the point at the centre. The signal that survives is simpler: how much real food a person eats, how heavily it has been processed, whether protein is adequate, and the context of the meal around it.
The pattern beats the part
No nutrient in isolation reliably moves the endpoints that matter. Patterns do. The Mediterranean pattern — vegetables, legumes, fruit, whole grains, nuts, fish, olive oil, modest dairy, modest meat, very little ultra-processed food — has the largest randomised-trial evidence base for hard cardiovascular endpoints. PREDIMED and the earlier Lyon Diet Heart Study both showed roughly a 30% relative reduction in major events. DASH has the strongest evidence for blood pressure. Both share a structure more than a list of forbidden foods.
The repeated failure of single-nutrient megadose trials — beta-carotene, vitamin E, folate, selenium — is part of the same story. Food works through matrix, displacement, and adherence. A pattern captures all three. A capsule captures none of them.
Ultra-processed food is the variable to act on first
The most consequential dietary change for most patients is not switching macros. It is cutting back on products formulated to be eaten faster than the body can register satiety. In Hall's metabolic-ward crossover, at matched macros and calories on offer, an ultra-processed diet drove about 500 extra kilocalories a day and almost a kilogram of weight gain in two weeks. The 2024 BMJ umbrella review found convincing associations with cardiovascular mortality, common mental disorders, and type 2 diabetes across nearly ten million participants.
The practical implication is unromantic. A weekly shop dominated by single-ingredient or minimally processed items handles most of the work. Branding is irrelevant. Wholemeal industrial bread is not the same as a soft drink, but the overall direction is consistent enough to act on.
Protein, said plainly
The dietary advice many adults grew up on understated protein for the second half of life. Responsiveness to a single protein dose declines with age, and the fix is more protein, more often — typically 1.2–1.6 g/kg/day, divided across three or four meals. Morton's meta-analysis of 49 trials found additional lean-mass benefit plateaued near 1.62 g/kg/day, which is a reasonable ceiling. Per meal, around 0.4 g/kg — including 2.5–3 g of leucine — is the dose to aim for to trigger muscle-building. Plant-dominant patterns are not penalised on mortality endpoints; well-chosen plant blends approach animal-protein quality. The training context that makes this protein useful is in the note on strength and muscle.
Supplements, kept honest
A short list of supplements survives the evidence. Vitamin D where clinically deficient. B12 in older adults, strict vegans, and long-term metformin users. Creatine monohydrate at 3–5 g/day for anyone doing strength work. Icosapent ethyl at 4 g/day in statin-treated hypertriglyceridaemia — the only positive omega-3 trial of its kind; mixed EPA/DHA preparations have missed repeatedly. A multivitamin showed a small cognitive signal in older adults in COSMOS. Magnesium has a defensible niche in migraine prophylaxis.
The list of supplements with evidence of harm or no benefit is longer. High-dose vitamin E increased prostate cancer in SELECT. Beta-carotene increased lung cancer in smokers in ATBC and CARET. Most "longevity stacks" — NMN and NR included — raise biochemical markers without yet showing morbidity, mortality, or robust functional benefit in adequately powered trials. Supplements refine a dietary pattern that is already in place; they do not replace one.
Sourcing matters, modestly
Where food comes from is not a precious detail, but its magnitude should not be overstated. Organic produce shows lower pesticide residue detection and no meaningful nutrient advantage. Grass-fed beef has a better omega-6 to omega-3 ratio but small absolute n-3 amounts per serve — fatty fish remains the dominant source. The differences are real and worth pursuing where they are practical, but they sit downstream of the larger levers: cutting ultra-processed food, anchoring the pattern, getting protein adequate. The same logic of chronic, low-grade dose runs through the note on air, water, and soil.
Fasting, alcohol, and a few worth naming
Time-restricted eating delivers weight loss largely by cutting total calories. Head-to-head against simple calorie restriction at twelve months, an eight-hour eating window adds nothing on weight, fat, or metabolic markers. The old J-curve on alcohol — that light drinking is protective — has been eroded by genetic-causation studies; the current Australian guidance of no more than ten standard drinks a week and no more than four on any day reflects a no-safe-level position on net health gain. Coffee at three to four cups a day is associated with lower all-cause mortality across large cohorts. Sugary drinks are associated with higher mortality at two or more serves a day; artificial sweeteners have moved, on recent evidence, from neutral to mildly unfavourable, and the WHO has issued a conditional recommendation against using them for weight control. Fermented foods are the most defensible gut-targeted intervention currently in the literature.
What the work looks like
For most patients, the changes are not ascetic. A weekly shop reorganised around real food. One or two staple meals refined. A protein target made specific rather than vague, and spread across the day. A short list of items to stop buying altogether. A defensible supplement list of two to four items, not twenty. Sustained over a decade, this is a different physiology.
§ For professionals — mechanisms & evidence+
Patterns over nutrients
PREDIMED (Estruch 2018, NEJM, n=7,447 high-risk Spaniards) remains the strongest randomised evidence for a dietary pattern reducing major cardiovascular events: HR 0.69 (95% CI 0.53–0.91) for the EVOO-supplemented arm and HR 0.72 (0.54–0.95) for the nuts arm versus low-fat advice 1. The Lyon Diet Heart Study (de Lorgeril 1999) found HR 0.28 (0.15–0.52) for cardiac death plus non-fatal MI in secondary prevention 2. PREDIMED-Plus extends the signal to weight, glycaemia, and visceral fat with energy restriction and activity layered on 3. DASH-sodium remains the dominant blood-pressure pattern: SBP –11.5 mmHg in hypertensives on the combined low-sodium DASH arm 4.
Ultra-processed food
Hall's 2019 metabolic-ward crossover (n=20, two weeks each arm, matched for presented calories, macronutrients, sugar, sodium, fibre, and energy density) demonstrated +508 ± 106 kcal/day ad libitum intake on the ultra-processed arm and +0.9 kg weight gain versus –0.9 kg on unprocessed 5. The Lane 2024 BMJ umbrella review (14 meta-analyses, 45 pooled analyses, n=9.9 million) graded evidence as "convincing" for cardiovascular mortality (~+50%), common mental disorders (+48–53%), and type 2 diabetes (+12%); "highly suggestive" for all-cause mortality (+21%), obesity, and depression 6. NOVA heterogeneity remains a genuine limitation — wholemeal industrial bread and sweetened soft drinks are not equivalent exposures — but the population signal is no longer in doubt.
Protein, distribution, and ageing
The 0.8 g/kg/day RDA is a deficiency floor. PROT-AGE (Bauer 2013) and ESPEN (Deutz 2014) converge on 1.0–1.2 g/kg/day for healthy older adults and 1.2–1.5 g/kg/day with acute or chronic illness 7,8. Morton's 2018 BJSM meta-analysis (49 RCTs, n=1,863) found FFM gain of 0.30 kg (95% CI 0.09–0.52) and 1RM gain of 2.49 kg with supplemental protein in resistance-trained adults, plateauing at approximately 1.62 g/kg/day (95% CI 1.03–2.20) 9. Per-meal anabolic dose is approximately 0.4 g/kg, with a leucine bolus of 2.5–3 g. Naghshi 2020 BMJ (32 cohorts, n=715,128) reported plant-protein substitution at 3% of energy associated with all-cause mortality RR 0.95 (0.91–0.98) and CVD mortality RR 0.88 (0.80–0.96); animal protein was not associated with overall mortality 10. The training context that makes this protein useful is treated in the note on strength and muscle.
Supplements — the defensible list
VITAL (Manson 2019, n=25,871) was null for vitamin D 2,000 IU/day on invasive cancer (HR 0.96) and major CV events (HR 0.97), and for marine n-3 1 g/day on major CV events (HR 0.92) 11. REDUCE-IT (Bhatt 2019, icosapent ethyl 4 g/day in statin-treated hypertriglyceridaemia) reported a primary composite HR of 0.75 (0.68–0.83), with an atrial-fibrillation signal of HR 1.69 and ongoing controversy about the mineral-oil placebo 12. STRENGTH (Nicholls 2020, mixed EPA/DHA 4 g/day) was stopped for futility at HR 0.99 and showed a 69% increase in atrial fibrillation 13. SELECT (Klein 2011, n=35,533) found vitamin E 400 IU/day increased prostate cancer (HR 1.17) 14. COSMOS (Sesso 2022) found no significant CV or cancer benefit for a multivitamin overall, but the cognition substudy (Vyas 2024 AJCN meta-analysis of COSMOS-Mind, COSMOS-Web, and COSMOS-Clinic) found significant improvements in global cognition and episodic memory in adults aged 60 and over 15. Creatine monohydrate (Kreider 2017 ISSN position stand) retains the strongest sports-nutrition evidence base and is increasingly supported for older-adult strength work and cognition under sleep deprivation or stress 16. NMN and NR raise NAD+ without yet demonstrating morbidity, mortality, or robust functional benefit.
Time-restricted eating
TREAT (Lowe 2020, JAMA Intern Med, n=116) found no significant weight benefit of 16:8 TRE versus three meals a day, with lean-mass loss in the TRE arm 17. Liu 2022 NEJM (n=139, 12 months) found no additional benefit of an eight-hour window over caloric restriction alone for weight, fat, or metabolic markers 18. Effects of TRE on weight are largely mediated by caloric reduction; biologically plausible effects of early-window TRE on glycaemia and blood pressure (Sutton 2018) have not been confirmed on hard endpoints.
Sourcing — honest magnitudes
Dangour 2009 (AJCN, 162 studies over 50 years) found no meaningful nutrient difference between organic and conventional produce 19. Smith-Spangler 2012 (Ann Intern Med) found ~30% lower pesticide residue detection in organic produce without a clinically meaningful nutrient advantage 20. Daley 2010 reported a grass-fed beef omega-6:omega-3 ratio of 1.53:1 versus 7.65:1 for grain-fed, but absolute n-3 quantities per serve remain small — wild fatty fish dominates EPA and DHA intake 21. The signal here is real but modest. The same logic of chronic, low-grade dose is the throughline of the note on air, water, and soil.
Caloric restriction and ageing
CALERIE-2 (Kraus 2019 Lancet Diabetes Endocrinol, n=218, 11.9% caloric restriction achieved over two years) produced significant reductions in blood pressure, LDL, total cholesterol, insulin resistance, and metabolic syndrome score 22. Waziry 2023 (Nature Aging) reported that the same intervention slowed DunedinPACE epigenetic ageing by ~2–3%; PhenoAge and GrimAge were unchanged.
Other endpoints worth noting
Sugar-sweetened beverages: Malik 2019 (Circulation, n=118,363) reported all-cause mortality HR 1.21 at ≥2 serves/day 23. Non-sugar sweeteners: WHO 2023 conditionally recommended against use for weight control or NCD risk reduction; Debras 2022 (BMJ, NutriNet) reported CVD HR 1.09 in the highest intake group. Coffee: Poole 2017 (BMJ) found 3–4 cups/day associated with all-cause mortality RR 0.83. Fermented foods (Wastyk 2021 Cell) increased gut microbial alpha-diversity and lowered 19 inflammatory cytokines including IL-6 over ten weeks.
What does not work
Detoxes and cleanses, alkaline water, IgG food-sensitivity panels (explicitly advised against by allergy societies), DNA-personalised diets without phenotypic anchoring, and the broad class of "fat-burner" stimulant supplement blends. Most single-nutrient megadoses (vitamin E, beta-carotene in smokers) have failed or harmed.
Australian regulatory context
The TGA distinguishes AUST L (listed, sponsor self-certifies, not individually assessed for efficacy), AUST L(A) (assessed for efficacy), and AUST R (registered, fully assessed). Most vitamins and minerals are listed. Third-party verification (USP, NSF Certified for Sport, Informed-Sport) reduces contamination and banned-substance risk. FSANZ administers the bi-national Food Standards Code, including mandatory iodine and folate fortification of bread. The NHMRC's revision of the Australian Dietary Guidelines remains in its evidence phase; the 2013 edition is current.
What is the best diet for longevity?+
The Mediterranean pattern has the largest randomised-trial evidence base for hard cardiovascular endpoints. PREDIMED reported roughly a 30% relative reduction in major cardiovascular events over a median of 4.8 years; the earlier Lyon Diet Heart Study showed a similar magnitude in secondary prevention. DASH is the strongest pattern for blood pressure. No single nutrient does what a pattern does.
What counts as ultra-processed food, and why does it matter?+
NOVA category 4 captures industrial formulations using ingredients and processes not used in domestic kitchens: emulsifiers, isolates, non-nutritive sweeteners, hydrogenated oils, formulated flavour systems. In Hall's metabolic-ward crossover, matched for presented calories and macronutrients, an ultra-processed diet drove around 500 kcal/day greater intake and meaningful weight gain over two weeks. The BMJ 2024 umbrella review found convincing evidence for associations with cardiovascular mortality, type 2 diabetes, and common mental disorders.
How much protein should adults actually eat?+
The 0.8 g/kg/day RDA is a deficiency floor. The defensible target for most adults sits at 1.2 to 1.6 grams per kilogram per day, distributed across three or four meals to overcome anabolic resistance. Morton's 2018 meta-analysis of 49 trials found additional lean-mass benefit plateaued at approximately 1.62 g/kg/day. Plant-protein-dominant patterns are associated with lower mortality; blends can approach animal-protein quality on the DIAAS scale.
Is intermittent fasting worth doing?+
Time-restricted eating delivers modest weight loss, largely through caloric reduction. The 12-month NEJM trial from Liu in 2022 found no additional benefit of an eight-hour window over caloric restriction alone for weight, fat, or metabolic markers. It is a reasonable tool for some patients, but not uniquely powerful compared with the underlying dietary pattern.
Which supplements actually have evidence?+
A short list. Vitamin D replacement where clinically deficient. B12 in older adults, strict vegans, and long-term metformin users. Creatine monohydrate at 3 to 5 g/day for strength work and emerging cognitive signals. Icosapent ethyl at 4 g/day in statin-treated hypertriglyceridaemia (REDUCE-IT). A multivitamin showed cognitive benefit in older adults in COSMOS. Most others — including mixed EPA/DHA at standard doses (VITAL, STRENGTH), vitamin E (SELECT showed harm), beta-carotene in smokers (ATBC, CARET), and most longevity stacks — are neutral or worse.
Does organic and grass-fed food actually matter?+
The differences are real but modest. Organic produce shows lower pesticide residue detection but no clinically meaningful nutrient advantage. Grass-fed beef has a more favourable omega-6 to omega-3 ratio but the absolute n-3 amounts per serve are small — fatty fish remains the dominant EPA/DHA source. On the evidence, the magnitude of ultra-processed food reduction, dietary pattern shift, and protein adequacy dwarfs sourcing effects on hard endpoints. Sourcing is a refinement, not the foundation.
What about alcohol?+
The Global Burden of Disease analyses and large Mendelian-randomisation work (Millwood 2019) have eroded the old J-curve. Current Australian NHMRC guidance is no more than ten standard drinks per week and no more than four on any one day, with zero in pregnancy, breastfeeding, and under eighteen. There is no level of intake that improves health on net.
- 1. Estruch R, et al. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts (PREDIMED). NEJM. 2018.
- 2. de Lorgeril M, et al. Mediterranean diet, traditional risk factors, and the rate of cardiovascular complications after myocardial infarction: final report of the Lyon Diet Heart Study. Circulation. 1999.
- 3. Salas-Salvadó J, et al. Effect of a Lifestyle Intervention Program With Energy-Restricted Mediterranean Diet and Exercise on Weight Loss and Cardiovascular Risk Factors (PREDIMED-Plus). Diabetes Care. 2019.
- 4. Sacks FM, et al. Effects on Blood Pressure of Reduced Dietary Sodium and the DASH Diet. NEJM. 2001.
- 5. Hall KD, et al. Ultra-Processed Diets Cause Excess Calorie Intake and Weight Gain: An Inpatient Randomized Controlled Trial. Cell Metabolism. 2019.
- 6. Lane MM, et al. Ultra-processed food exposure and adverse health outcomes: umbrella review of epidemiological meta-analyses. BMJ. 2024.
- 7. Bauer J, et al. Evidence-based recommendations for optimal dietary protein intake in older people: PROT-AGE Study Group. JAMDA. 2013.
- 8. Deutz NEP, et al. Protein intake and exercise for optimal muscle function with aging: Recommendations from the ESPEN Expert Group. Clinical Nutrition. 2014.
- 9. Morton RW, et al. A systematic review, meta-analysis and meta-regression of the effect of protein supplementation on resistance training-induced gains in muscle mass and strength in healthy adults. BJSM. 2018.
- 10. Naghshi S, et al. Dietary intake of total, animal, and plant proteins and risk of all cause, cardiovascular, and cancer mortality: systematic review and dose-response meta-analysis. BMJ. 2020.
- 11. Manson JE, et al. Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease (VITAL). NEJM. 2019.
- 12. Bhatt DL, et al. Cardiovascular Risk Reduction with Icosapent Ethyl for Hypertriglyceridemia (REDUCE-IT). NEJM. 2019.
- 13. Nicholls SJ, et al. Effect of High-Dose Omega-3 Fatty Acids vs Corn Oil on Major Adverse Cardiovascular Events in Patients at High Cardiovascular Risk (STRENGTH). JAMA. 2020.
- 14. Klein EA, et al. Vitamin E and the risk of prostate cancer: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA. 2011.
- 15. Vyas CM, et al. Effect of multivitamin-mineral supplementation versus placebo on cognitive function: results from the COSMOS-Web ancillary study. AJCN. 2024.
- 16. Kreider RB, et al. International Society of Sports Nutrition position stand: safety and efficacy of creatine supplementation in exercise, sport, and medicine. JISSN. 2017.
- 17. Lowe DA, et al. Effects of Time-Restricted Eating on Weight Loss and Other Metabolic Parameters in Women and Men With Overweight and Obesity (TREAT). JAMA Internal Medicine. 2020.
- 18. Liu D, et al. Calorie Restriction with or without Time-Restricted Eating in Weight Loss. NEJM. 2022.
- 19. Dangour AD, et al. Nutritional quality of organic foods: a systematic review. AJCN. 2009.
- 20. Smith-Spangler C, et al. Are organic foods safer or healthier than conventional alternatives? A systematic review. Annals of Internal Medicine. 2012.
- 21. Daley CA, et al. A review of fatty acid profiles and antioxidant content in grass-fed and grain-fed beef. Nutrition Journal. 2010.
- 22. Kraus WE, et al. 2 years of calorie restriction and cardiometabolic risk (CALERIE): exploratory outcomes of a multicentre, phase 2, randomised controlled trial. Lancet Diabetes & Endocrinology. 2019.
- 23. Malik VS, et al. Long-Term Consumption of Sugar-Sweetened and Artificially Sweetened Beverages and Risk of Mortality in US Adults. Circulation. 2019.
- 24. Wastyk HC, et al. Gut-microbiota-targeted diets modulate human immune status. Cell. 2021.
- 25. National Health and Medical Research Council. Australian Dietary Guidelines.
- 26. National Health and Medical Research Council. Australian Guidelines to Reduce Health Risks from Drinking Alcohol. 2020.