Fats: which one replaces which

The fat question changed shape in the 1990s, and the Nurses' Health Study is where it happened. Until then the advice was to eat less fat. Then the cohorts started separating the types of fat and asking what took the fat's place in the diet, and the answer stopped being about quantity.

In the Nurses' Health Study, with 80,082 women followed for 14 years and 939 cases of nonfatal myocardial infarction or death from coronary heart disease, total fat intake was not significantly related to coronary risk (relative risk 1.02 per 5% more energy from fat, 95% confidence interval 0.97 to 1.07). [1]

Within the same cohort, the types of fat behaved quite differently from one another and from the total.

In the same cohort, a 5% increase in energy from saturated fat, compared with the same energy from carbohydrate, was associated with a 17% higher risk of coronary disease (relative risk 1.17, 95% confidence interval 0.97 to 1.41, P = 0.10), while a 5% increase from polyunsaturated fat was associated with a relative risk of 0.62 (0.46 to 0.85) and from monounsaturated fat 0.81 (0.65 to 1.00). [1] [1]

The same study estimated that replacing 5% of energy from saturated fat with energy from unsaturated fats would reduce coronary risk by 42% (95% confidence interval 23 to 56), and that replacing 2% of energy from trans fat with unhydrogenated unsaturated fats would reduce it by 53% (34 to 67). [1] [1]

The Nurses' Health Study and the Health Professionals Follow-up Study were then followed for far longer, which is where the substitution numbers get big enough to argue about.

In 84,628 women and 42,908 men followed for 24 to 30 years, with 7,667 incident cases of coronary heart disease, replacing 5% of energy from saturated fat with polyunsaturated fat was associated with a 25% lower risk (hazard ratio 0.75, 95% confidence interval 0.67 to 0.84), with monounsaturated fat 15% lower (0.85, 0.74 to 0.97) and with carbohydrate from whole grains 9% lower (0.91, 0.85 to 0.98). [2]

And the thing that replaced the saturated fat mattered as much as the fat that left the diet.

In the same two cohorts, carbohydrate from refined starches and added sugars was positively associated with coronary heart disease (hazard ratio 1.10, 95% confidence interval 1.00 to 1.21), and replacing saturated fat with those carbohydrates was not significantly associated with coronary risk. [2] [2]

The same cohorts were later used to take saturated fat apart into its individual fatty acids, on the theory that they might not behave alike. They mostly did, with one instructive exception.

In 73,147 women and 42,635 men with 7,035 cases of coronary heart disease, individual saturated fatty acids were not alike: the hazard ratio comparing highest with lowest intake was 1.07 (95% confidence interval 0.99 to 1.15) for lauric acid, 1.13 (1.05 to 1.22) for myristic acid and 1.18 (1.09 to 1.27) for palmitic acid. [3]

In the same cohorts, replacing 1% of energy from the four main saturated fatty acids combined was associated with a hazard ratio for coronary heart disease of 0.92 (95% confidence interval 0.89 to 0.96) when replaced by polyunsaturated fat, 0.95 (0.90 to 1.01) when replaced by monounsaturated fat, and 0.93 (0.89 to 0.97) when replaced by plant protein. [3]

All of that is about heart disease. The same cohorts were also followed to death from any cause, which is a different and blunter question.

Across 3,439,954 person-years in the Nurses' Health Study and the Health Professionals Follow-up Study, with 33,304 deaths, the hazard ratio for total mortality comparing extreme quintiles was 1.08 (95% confidence interval 1.03 to 1.14) for saturated fat, 0.81 (0.78 to 0.84) for polyunsaturated fat, 0.89 (0.84 to 0.94) for monounsaturated fat and 1.13 (1.07 to 1.18) for trans fat. [4]

In the same cohorts, replacing 5% of energy from saturated fat with polyunsaturated fat was associated with an estimated 27% lower total mortality (hazard ratio 0.73, 95% confidence interval 0.70 to 0.77) and with monounsaturated fat 13% lower (0.87, 0.82 to 0.93). [4]

Cohorts can only adjust for what they thought to measure. The trials asked the same question with randomisation instead, and they are more modest.

A Cochrane review of 15 randomised trials with 56,675 participants found that reducing saturated fat intake for at least two years reduced combined cardiovascular events by 17% (risk ratio 0.83, 95% confidence interval 0.70 to 0.98, moderate-quality evidence), with 56 people needing to reduce their intake for about four years for one person to avoid an event. [5] [5]

The same review found little or no effect of reducing saturated fat on all-cause mortality (risk ratio 0.96, 95% confidence interval 0.90 to 1.03) or cardiovascular mortality (0.95, 0.80 to 1.12), and reported that subgrouping did not suggest significant differences between replacing saturated fat with polyunsaturated fat or with carbohydrate. [5] [5]

A meta-analysis of the older trials, which used harder endpoints than cholesterol, found a larger effect than the Cochrane review did.

A meta-analysis of eight randomised trials with 13,614 participants and 1,042 coronary events found that increasing polyunsaturated fat in place of saturated fat reduced coronary heart disease events by 19% (risk ratio 0.81, 95% confidence interval 0.70 to 0.95). [6]

Against all of that sits a systematic review of the observational evidence that reached the opposite conclusion about saturated fat, and rated its own certainty as very low.

A systematic review of observational studies found saturated fat intake was not associated with all-cause mortality (relative risk 0.99, 95% confidence interval 0.91 to 1.09), cardiovascular mortality (0.97, 0.84 to 1.12) or total coronary heart disease (1.06, 0.95 to 1.17), and rated the certainty of those associations very low. [7] [7]

The same review found trans fat intake was associated with all-cause mortality (relative risk 1.34, 95% confidence interval 1.16 to 1.56), coronary heart disease mortality (1.28, 1.09 to 1.50) and total coronary heart disease (1.21, 1.10 to 1.33), with moderate certainty for the coronary outcomes. [7] [7]

The largest cardiology body in the United States read the same literature and stated a position rather than a doubt.

The American Heart Association's presidential advisory reports that randomised trials which lowered saturated fat and replaced it with polyunsaturated vegetable oil reduced cardiovascular disease by about 30%, and that replacing saturated fat with mostly refined carbohydrates and sugars was not associated with lower rates of cardiovascular disease. [8] [8]

One more trial belongs here, because it is the only large randomised test of a whole dietary pattern in this entry, and because its history is part of its evidence.

In a trial of 7,447 people at high cardiovascular risk in Spain, the incidence of major cardiovascular events was lower in the groups assigned a Mediterranean diet with extra-virgin olive oil (hazard ratio 0.69, 95% confidence interval 0.53 to 0.91) or with nuts (0.72, 0.54 to 0.95) than in the group advised to reduce dietary fat. [9]

That trial is a corrected and republished article: the authors withdrew their earlier report after identifying protocol deviations, including enrolment of household members without randomisation and assignment to a study group without randomisation at one of the eleven sites, and the results above come from analyses that do not rely exclusively on assuming all participants were randomly assigned. [9] [9]

The honest summary of the fat evidence is narrower than either side's slogan. Replacing saturated fat with polyunsaturated fat lowers heart-disease events; replacing it with refined carbohydrate does not; and the effect on dying of anything at all is small in the trials and larger in the cohorts.

What this section does not show

Substitution, not reduction

Almost every number here describes what happened when one nutrient replaced another at the same calorie level. None of it is a statement about eating less fat.

Cohorts and trials disagree

The cohorts find larger effects than the randomised trials do, and the observational review rates its own certainty as very low. Both are on the page rather than the flattering one.

The trial history is part of the evidence

PREDIMED is a corrected and republished article. Its authors withdrew the first report after identifying protocol deviations, and the numbers quoted come from the revised analysis.

Nothing here tells you what to eat

A substitution finding is a result about groups. This site does not know what any reader eats now, and it gives no advice about butter, oil or anything else.

All twelve elements

The entry's full limits block covers the dates, the study designs, who was and was not studied, the comparators, the adjustment sets, the uncertainty, the counter-signals and what we did not check. Read it on the entry's first page.

This page is general information about published research. It is not medical advice, and it is not a substitute for advice from a qualified health professional. No clinician-patient relationship exists, and we do not know your diet, your medications, your test results, your medical conditions or your history. Talk to a health professional before changing your diet if you are pregnant or breastfeeding, feeding an infant or child, over 65, taking medication, living with a medical condition, or if you have been diagnosed with a nutrient deficiency or have a history of disordered eating. The evidence described is about groups and diets, not about you, and not about any product listed on this site. If you are unwell, contact a health professional; in an emergency, call 911. No clinician reviewed this entry. It is a compilation of published sources by this site.

Sources for this section

Every numbered sentence above was checked against the source with that number on the date shown. The number links to the entry in this list, and the entry's first page lists every source it cites.

1. Dietary fat intake and the risk of coronary heart disease in women. The New England Journal of Medicine. prospective cohort study. Open the source ↗
   Checked 19 September 2026. Backs 3 sentences in this section.
2. Saturated Fats Compared With Unsaturated Fats and Sources of Carbohydrates in Relation to Risk of Coronary Heart Disease: A Prospective Cohort Study. Journal of the American College of Cardiology. prospective cohort study. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.
3. Intake of individual saturated fatty acids and risk of coronary heart disease in US men and women: two prospective longitudinal cohort studies. The BMJ. prospective cohort study. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.
4. Association of Specific Dietary Fats With Total and Cause-Specific Mortality. JAMA Internal Medicine. prospective cohort study. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.
5. Reduction in saturated fat intake for cardiovascular disease. Cochrane Database of Systematic Reviews. systematic review. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.
6. Effects on coronary heart disease of increasing polyunsaturated fat in place of saturated fat: a systematic review and meta-analysis of randomized controlled trials. PLOS Medicine. systematic review and meta-analysis of randomised trials. Open the source ↗
   Checked 19 September 2026. Backs 1 sentence in this section.
7. Intake of saturated and trans unsaturated fatty acids and risk of all cause mortality, cardiovascular disease, and type 2 diabetes: systematic review and meta-analysis of observational studies. The BMJ. systematic review and meta-analysis. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.
8. Dietary Fats and Cardiovascular Disease: A Presidential Advisory From the American Heart Association. Circulation. position paper. Open the source ↗
   Checked 19 September 2026. Backs 1 sentence in this section.
9. Primary Prevention of Cardiovascular Disease with a Mediterranean Diet Supplemented with Extra-Virgin Olive Oil or Nuts. The New England Journal of Medicine. randomised trial. Open the source ↗
   Checked 19 September 2026. Backs 2 sentences in this section.