Food composition and dietary-fat arithmetic

• Nearly all natural foods supply protein; pure sugar and pure fats are the exceptions, while butter still has some protein.
• Natural foods largely separate carbohydrate-and-protein foods—fruit, vegetables, grains and pulses—from fat-and-protein foods—meat, fish, eggs and dairy; nuts, seeds and avocado are uncommon mixed examples.
• Meat and fish are generally carbohydrate-free, eggs have traces, and fluid dairy can contain 5–10 g carbohydrate per 100 g; processed foods combine fat and carbohydrate in highly palatable products.
• With real-food protein typically around 15% of calories, a 30% fat ceiling leaves about 55% for carbohydrate; a 25% fat target leaves 60%, while low-carbohydrate eating requires more fat.
• Every fat-containing food includes saturated, monounsaturated and polyunsaturated fats; among ordinary food groups, dairy has more saturated than unsaturated fat.
• USDA food data put a typical sirloin steak at about 2% saturated fat, while oily fish has twice its total fat and 1.5 times its saturated fat; olive oil has 14 times the total fat and seven times the saturated fat of that steak. [1]
• One tablespoon of olive oil contains more saturated fat than a 100 g pork chop, and lard is predominantly monounsaturated; simplified warnings about meat, eggs and saturated fat do not match these food-composition comparisons.

How the guidelines emerged

• The 1977 US McGovern committee introduced dietary-fat goals, incorporated into the 1980 Dietary Guidelines for Americans; Britain followed with NACNE proposals in 1983 and the COMA cardiovascular document in 1984. [2][3][4]
• Earlier medical advice identified floury, vegetable and sugary foods as fattening, illustrated by an 1869 Tanner quotation reproduced in Gary Taubes’s Good Calories, Bad Calories. [5]
• The new recommendations shifted the dietary focus to ceilings of 30% of calories from total fat and 10% from saturated fat; those limits necessarily raised the carbohydrate share without evidence establishing that level as healthy or safe.
• Coronary heart disease motivated the policy. In the US in 1950, annual all-cause mortality was about 1.45%, and about 589 people per 100,000 died from heart disease: roughly 0.59% of the population but about 40% of deaths.
• Those measures convey different risks; heart disease accounting for 40% of deaths does not mean 40% of living people die from it each year.
• Early Russian pathologists fed cholesterol-rich animal material to rabbits and observed arterial damage, although rabbits do not naturally eat animal foods; this model differs substantially from human diets.
• Ancel Keys tested dietary cholesterol in people in the early 1950s; in a 1954 symposium he concluded that dietary cholesterol did not affect human blood cholesterol or atherosclerosis. These findings clear animal foods of raising blood cholesterol or causing atherosclerosis, while fat remained the focus. [6]

First review: randomized trials available when policy began

• The February 2015 analysis included six historical randomized trials involving 2,467 men, nearly all already ill; none tested both the 30% total-fat and 10% saturated-fat targets. [7]
• Headlines vindicating butter or crediting fat with longer life went beyond the review: it tested evidence for the guidelines, not butter or fat’s effects on lifespan.
• They included the Rose Corn Oil trial, LA Veterans trial, Oslo Diet-Heart Study and Sydney Diet-Heart Study; one trial was released after the 1977 committee met. [8][9][10][11]
• Pooled all-cause deaths were 370 versus 369, with no significant difference; coronary-heart-disease mortality likewise did not improve, and the forest-plot estimates lay on the line of no effect.
• The Sydney intervention reached the 10% saturated-fat target alone, but had significantly more deaths in its intervention arm; its original authors also warned about possible toxicity of the vegetable oil.
• Blood cholesterol fell in both trial arms and more in the diet-intervention arms, yet this did not coincide with fewer heart deaths; cholesterol reduction alone did not establish improved mortality in these trials.
• Plant sterols in vegetable oils can compete with cholesterol absorption and lower blood cholesterol without improving heart-disease outcomes; a 2014 paper identified a possible adverse coronary effect, leading to advice against large vegetable-oil intakes. [12]
• The 1965 Research Committee trial concluded that a low-fat diet had no role after myocardial infarction, and the Rose Corn Oil trial cautioned that corn oil might cause harm. [13][8]
• Dietary advice was nevertheless applied to about 276 million Britons and Americans on the basis of trials involving fewer than 2,500 mostly ill men; that population offered no direct evidence for healthy people or women.

Second review: observational evidence before the guidelines

• The review of six prospective cohorts available before the UK recommendations included only men, mostly healthy, followed for roughly four to 20 years. [14]
• The Western Electric, London Banks and Busmen, F. Heart, Honolulu and Seven Countries cohorts found no association between total fat and coronary-heart-disease deaths. [15][16][17][18]
• Only the Seven Countries Study linked saturated fat and heart-disease deaths, through a between-country correlation, not a within-population relationship; other cohorts identified smoking and paternal age.
• Some cohorts found inverse associations between alcohol and heart disease, potentially reflecting affluence and its benefits, not a benefit from drinking; Keys found no association with smoking, activity or body weight in his study.
• The Seven Countries Study enrolled about 13,000 men aged 40–59 from 1956 and linked saturated fat to cholesterol and cholesterol to coronary disease; its original conclusions concerned associations, not demonstrated causation. [18]
• In a 25-year follow-up, Menotti and Keys obtained a cholesterol–heart-disease correlation of 0.72 (R²≈0.52); a reanalysis yielded a stronger latitude correlation of 0.96 (R²≈0.92), with sunlight/vitamin D as a possible explanation across 16 cohorts. [19]
• The stronger latitude relationship points toward sunlight and vitamin D, with lower exposure in Finland and higher exposure nearer the equator; country comparisons also differ in climate, history and wealth.
• Across the historical cohorts, all-cause mortality was just under 5% and coronary mortality around 1%; in Seven Countries, five-year coronary mortality was 21% with pre-existing coronary disease versus 1% without it.
• Neither the six observational cohorts nor the randomized evidence supported those population-wide fat limits.

Third and fourth reviews: more recent trials and cohorts

• Updating the randomized evidence to ten trials and about 62,000 participants added women, healthy people, and studies lasting one to eight years, including Minnesota, DART and the Women’s Health Initiative. [20][21][22][23]
• The pooled results remained null for all-cause and coronary-heart-disease mortality despite larger cholesterol reductions in nine of the ten intervention comparisons.
• Only the Minnesota Coronary Survey included healthy men and women together, but they were institutionalized psychiatric patients who might have been taking cardiovascularly relevant medication, limiting applicability to the general population.
• Even these later trials never directly tested both official fat limits; STARS came closest at 27% total fat and 8–10% saturated fat, but it enrolled only 55 people and its initial positive result lost significance with Fisher’s exact test. [24]
• The updated prospective-cohort review identified seven cohorts, six usable in meta-analysis, with the Kushi cohort excluded; its pooled results again found no significant association between total or saturated fat and coronary deaths. [25]
• Four isolated positive cohort findings failed Bradford Hill causal criteria, and the combined results of these four reviews gave no support for the fat guidelines.

Other research and the Cochrane controversy

• A 20-year vegetarian history preceded the research; its findings favored animal foods despite an initial inclination away from meat.
• Skeaff and Miller’s 2009 review combined 28 studies and about 280,000 people without a significant effect; Siri-Tarino’s 2010 analysis of about 350,000 people likewise found no saturated-fat association with coronary disease. [26][27]
• Mozaffarian’s 2010 review found a benefit when saturated fat was exchanged for polyunsaturated fat, while Hooper’s 2011 Cochrane review contained one positive result amid eleven nonsignificant results. [28][29]
• Chowdhury’s 2014 review found a significant association for trans fats but not saturated, monounsaturated or polyunsaturated fats; Schwingshackl and Hoffmann’s review also found no significant effects. [30][31]
• Hooper’s 2015 saturated-fat Cochrane update was a 150-page review with a single significant signal: cardiovascular events after lowering saturated fat; seven other outcomes, including all-cause, cardiovascular and coronary mortality, heart attacks and strokes, were nonsignificant. [32]
• Ravnskov and colleagues criticized the Mozaffarian analysis for excluding the Rose Corn Oil and Sydney trials while including a Finnish Mental Hospital study that was neither randomized nor adequately controlled. [33][8][11]
• The Cochrane cardiovascular-event estimate pooled eleven trials although only one documented an actual exchange of saturated for polyunsaturated fat; the others intended dietary changes without verifying them.
• The positive signal depended on adding unpublished heart-event data from four studies originally about skin cancer, high cholesterol, insulin resistance or diabetes, involving only 646 additional people.
• Those extra data were not peer reviewed; study heterogeneity was high (I²=65%), and the positive cardiovascular-event signal lost statistical significance in sensitivity analysis.
• No trial contributing to that Cochrane result studied healthy men and women together, limiting its applicability to the general population.
• Malhotra, Redberg and Meier’s 2017 editorial challenged the saturated-fat artery-clogging theory, while the Hooper 2015 cardiovascular-event result remained central to criticism of that view. [34]

After the guidelines: obesity and the sources of dietary fat

• Mark Hegsted and legislators acknowledged in 1977 that the evidence lacked proof and scientific consensus, yet the committee asserted that the recommended dietary changes could cause no physical or mental harm. [2]
• UK obesity rose from about 2.7% in 1972 to 25% by the end of the century; US obesity accelerated around the 1976–1980 guideline transition. The timing raises a question about a possible connection.
• The 2015 US dietary-guideline charts attribute much saturated fat intake to pizza, desserts, candy, chips, tortillas, sauces, mixed ready meals and processed meat, before reaching cheese, milk, butter, nuts and seeds. [35]
• The UK Family Food figures likewise place pastries, cakes, biscuits and sweets prominently among fat sources; real meat, fish and eggs account for only 7%, with substantial intake also coming from dairy, added fats and processed products. [36]
• Processed food, not saturated fat itself, is the main dietary target; the central public-health guidance is to eat real food.

References

1. [06:13] USDA FoodData Central — https://fdc.nal.usda.gov/
2. [07:44] Dietary Goals for the United States (1977) — https://www.govinfo.gov/app/details/CPRT-95SPRT23395O
3. [07:50] Nutrition and Your Health: Dietary Guidelines for Americans (1980) — https://www.dietaryguidelines.gov/about-dietary-guidelines/previous-editions/1980-dietary-guidelines-americans
4. [08:07] Diet and Cardiovascular Disease: COMA Report (1984) — https://assets.publishing.service.gov.uk/media/5bab9af640f0b60633988006/Diet_and_Cardiovascular_Disease__1984_.pdf
5. [08:20] Good Calories, Bad Calories — Gary Taubes — https://www.penguinrandomhouse.com/books/176680/good-calories-bad-calories-by-gary-taubes/9781400033461/
6. [13:04] Symposium on Atherosclerosis (1954), including Ancel Keys — https://doi.org/10.17226/20269
7. [14:11] Evidence from randomised controlled trials did not support the introduction of dietary fat guidelines in 1977 and 1983 — Harcombe et al. (2015) — https://doi.org/10.1136/openhrt-2014-000196
8. [15:42] Corn Oil in Treatment of Ischaemic Heart Disease — Rose et al. (1965) — https://doi.org/10.1136/bmj.1.5449.1531
9. [15:43] A Controlled Clinical Trial of a Diet High in Unsaturated Fat in Preventing Complications of Atherosclerosis — Dayton et al. (1968) — https://doi.org/10.1016/S0140-6736(68)91531-6
10. [15:44] Effect of Diet on Five-Year Mortality in the Oslo Diet-Heart Study (1970) — https://doi.org/10.1161/01.cir.42.5.935
11. [15:47] Dietary Prevention of Coronary Heart Disease: The Sydney Diet Heart Study — Woodhill et al. (1978) — https://doi.org/10.1007/978-1-4684-0967-3_18
12. [18:35] Plant sterols lower cholesterol, but increase risk for coronary heart disease — Harcombe and Baker (2014) — https://doi.org/10.3844/ojbsci.2014.167.169
13. [19:51] Low-fat Diet in Myocardial Infarction: A Controlled Trial — Research Committee (1965) — https://doi.org/10.1016/S0140-6736(65)91469-8
14. [20:19] Evidence from prospective cohort studies did not support the introduction of dietary fat guidelines in 1977 and 1983 — Harcombe et al. (2016) — https://doi.org/10.1136/bjsports-2016-096409
15. [20:50] Diet and Coronary Death: The Western Electric Study — Shekelle et al. (1981) — https://doi.org/10.1056/NEJM198101083040201
16. [20:52] Diet and Heart: A Postscript — Morris et al. (1977), London banks and busmen — https://doi.org/10.1136/bmj.2.6098.1307
17. [20:54] Framingham Heart Study — Official research milestones — https://www.framinghamheartstudy.org/fhs-about/research-milestones/
18. [20:56] Seven Countries Study — Official publications archive — https://www.sevencountriesstudy.com/study-findings/publications/
19. [23:41] Inter-cohort Differences in Coronary Heart Disease Mortality in the 25-Year Follow-up of the Seven Countries Study — Menotti et al. (1993) — https://doi.org/10.1007/BF00209531
20. [27:15] Evidence from Randomised Controlled Trials Does Not Support Current Dietary Fat Guidelines — Harcombe et al. (2016) — https://doi.org/10.1136/openhrt-2016-000409
21. [27:54] The Minnesota Coronary Survey (1989) — https://doi.org/10.1161/01.atv.9.1.129
22. [27:57] Diet and Reinfarction Trial (DART) (1989) — https://doi.org/10.1016/S0140-6736(89)90828-3
23. [27:59] Low-Fat Dietary Pattern and Risk of Cardiovascular Disease: Women’s Health Initiative (2006) — https://doi.org/10.1001/jama.295.6.655
24. [29:48] St Thomas’ Atherosclerosis Regression Study (STARS) (1992) — https://doi.org/10.1016/0140-6736(92)90863-x
25. [30:35] Evidence from Prospective Cohort Studies Does Not Support Current Dietary Fat Guidelines — Harcombe et al. (2016) — https://doi.org/10.1136/bjsports-2016-096550
26. [32:18] Dietary Fat and Coronary Heart Disease: Summary of Evidence from Prospective Cohort and Randomised Controlled Trials — Skeaff and Miller (2009) — https://doi.org/10.1159/000229002
27. [32:41] Meta-analysis of Prospective Cohort Studies Evaluating the Association of Saturated Fat with Cardiovascular Disease — Siri-Tarino et al. (2010) — https://doi.org/10.3945/ajcn.2009.27725
28. [32:56] Effects on Coronary Heart Disease of Increasing Polyunsaturated Fat in Place of Saturated Fat — Mozaffarian et al. (2010) — https://doi.org/10.1371/journal.pmed.1000252
29. [32:59] Reduced or Modified Dietary Fat for Preventing Cardiovascular Disease — Hooper et al., Cochrane (2011) — https://doi.org/10.1002/14651858.CD002137.pub2
30. [33:11] Association of Dietary, Circulating, and Supplement Fatty Acids with Coronary Risk — Chowdhury et al. (2014) — https://doi.org/10.7326/M13-1788
31. [33:24] Dietary Fat Intake and Risk of Cardiovascular Disease and All-Cause Mortality — Schwingshackl and Hoffmann (2014) — https://doi.org/10.1136/bmjopen-2013-004487
32. [33:44] Reduction in Saturated Fat Intake for Cardiovascular Disease — Hooper et al., Cochrane (2015) — https://doi.org/10.1002/14651858.CD011737
33. [34:07] The Questionable Benefits of Exchanging Saturated Fat with Polyunsaturated Fat — Ravnskov et al. (2014) — https://doi.org/10.1016/j.mayocp.2013.11.006
34. [37:30] Saturated Fat Does Not Clog the Arteries: Coronary Heart Disease Is a Chronic Inflammatory Condition — Malhotra, Redberg and Meier (2017) — https://doi.org/10.1136/bjsports-2016-097285
35. [39:20] Dietary Guidelines for Americans, 2015–2020 — https://www.dietaryguidelines.gov/about-dietary-guidelines/previous-editions/2015-dietary-guidelines
36. [40:08] Family Food Statistics — UK Government — https://www.gov.uk/government/collections/family-food-statistics