Dietary fat

INTRODUCTION — 

Historically, the chief health concern about dietary fats has been their role in coronary heart disease and obesity and possible roles in the genesis of diabetes, brain health, and cancers. Following decades of research, the relationship between fat intake and these health outcomes is becoming clearer. Starting in the 1980s, several authoritative national and international organizations recommended decreasing the total amount of fat in the diet. However, over the last two decades, scientific investigations have clarified that the quality or type of fat consumed is far more important than amount (proportion of calories) of total fat consumed. Based on this, the 2015 Dietary Guidelines Advisory Committee (DGAC) recommended no upper limit on the percentage of calories from total fat, and the 2020 DGAC did not re-review this evidence [1,2].

This topic reviews the relationship between dietary fat and health outcomes. The management of patients with hypercholesterolemia and the role of diet and dietary supplements in lipid lowering are discussed separately. (See "Low-density lipoprotein cholesterol-lowering therapy in the primary prevention of cardiovascular disease" and "Lipid management with diet or dietary supplements".)

RECOMMENDATIONS FOR PATIENTS — 

For patients who do not require special diets for existing diseases (eg, chronic kidney disease, celiac disease, eating disorder), we provide the following dietary advice (table 1):

●Increase consumption of plant-, seafood-, and yogurt-based fats from minimally processed, bioactive-rich foods like fruits, nuts, seeds, vegetables, legumes, whole-grain products, plant oils, yogurt, and seafood, which are linked to lower risks of cardiovascular diseases (CVDs), diabetes, and obesity.

●Reduce consumption of processed meats and carbohydrate-rich foods high in refined starch, added sugars, trans fat, or sodium.

●Consume minimally processed animal foods like cheese, milk, eggs, poultry, and unprocessed red meats in moderation.

●Do not assume that low-fat or "fat-free" varieties of packaged and processed foods (eg, snack foods, salad dressings, deli meats) are more healthful or better for CVD, weight control, or other health outcomes [3]. These foods are often rich in refined starches and added sugars, which increase levels of triglyceride-rich lipoproteins, lower high-density lipoprotein cholesterol levels [4], and are associated with a higher incidence of type 2 diabetes [5,6] and obesity [7] and failure to gain the benefits of more healthful alternatives.

●Reduce the intake of industrially produced trans fatty acids by avoiding foods containing partially hydrogenated vegetable oils (eg, vanaspati or stick margarine, commercially baked goods, deep-fried fast foods, and street vendor foods). Notably, in the United States and several other high-income nations, the US Food and Drug Administration or a similar body has determined that partially hydrogenated vegetable oils are no longer "generally regarded as safe." This has largely removed industrial trans fats from the food supply in these nations [8].

●Use plant oils from fruits, seeds, and nuts with high amounts of monounsaturated and/or polyunsaturated fatty acids, especially n-3 polyunsaturated fats, for cooking, spreads, sauces, and dressings. Examples of healthful plant oils include extra virgin olive oil, canola oil, and soybean oil (figure 1).

●Choose foods rich in healthy fats including nuts, seeds, avocadoes, and fish.

These suggestions are generally consistent with the 2020 Dietary Guidelines for Americans [1]. (See 'Dietary fat advice' below.)

As discussed below, health effects of dairy fats remain uncertain and controversial. Most guidelines continue to recommend low-fat dairy products.

TOTAL FAT INTAKE — 

Healthy dietary patterns depend on types of foods, rather than macronutrient targets (such as the percentage of energy from total fat or carbohydrate) [2]. Emphasizing total fat intake has obscured this focus. In 2002, the Dietary Reference Intake recommended a daily fat intake of 20 to 35 percent of total energy. Subsequently, multiple prospective cohort studies and randomized trials have shown that lowering total fat intake has no meaningful effects on heart disease, stroke, cancers, diabetes, or long-term weight control [9]. This is likely because total dietary fat includes different fatty acids and comes from different food sources with divergent effects on health. Additionally, reductions in dietary fat are typically accompanied by increased carbohydrate intake, particularly refined starches and sugars.

Other issues with regard to the recommendation for a low-fat diet include:

●The role of dietary fat in the etiology of obesity remains controversial. Some scientists and organizations continue to recommend low-fat diets. Others recommend only calorie counting without regard to specific macronutrient targets; others recommend food-based healthy diet patterns; and still others recommend high-fat, low-carbohydrate diets. (See "Obesity in adults: Dietary therapy".)

●Compared with making direct recommendations about specific foods to consume or avoid, it is difficult to communicate to patients how to achieve a target level of total fat intake.

●The message to reduce dietary fat has been translated by food manufacturers and consumers into a potentially harmful set of food choices, particularly ultra-processed foods rich in starch and sugar, which appear to have contributed to the obesity and type 2 diabetes epidemics.

Effects of total dietary fat intake — The evidence is strong that the level of intake of total fat per se has no meaningful effects on risk of heart disease, stroke, cancer, or obesity.

Heart disease — Multiple randomized trials have examined the effect of jointly reducing total and saturated fat intake upon cardiovascular endpoints, and none have shown significant effects [10-15]. As an example, in a trial of 48,835 females who were predominantly without coronary heart disease at baseline, randomization to an intensive behavior-modification group reduced total fat intake and increased the intake of vegetables, fruits, and grains but did not reduce rates of cardiovascular disease (CVD), stroke, or coronary heart disease after a mean follow-up of 8.1 years, compared with the control group [11]. In a prespecified subgroup analysis, the low-fat intervention increased the risk of recurrent coronary heart disease among women with prevalent CVD at baseline. Three additional large, randomized trials with a total of 63,879 male participants did not show significant reductions in cardiovascular endpoints from a low total fat, low saturated fat diet [10,12,14,15]. In all trials, the intervention group reduced saturated and trans fats but also reduced monounsaturated and polyunsaturated fats while increasing refined carbohydrates. (See 'Type of fat' below.)

The results of prospective cohort studies are consistent with randomized trials, suggesting no meaningful associations of total fat consumption with the incidence of coronary heart disease. As an example, in an analysis from the Nurses' Health Study involving 80,000 women followed for 20 years, the multivariate adjusted relative risk (RR) of coronary heart disease for the highest quintile of dietary fat compared with the lowest was 0.92 (95% CI 0.77-1.09) [16]. A meta-analysis of cohort studies, including the Nurses' Health Study, revealed no association of total fat intake with either coronary events or death [17]. In the international Prospective Urban Rural Epidemiology (PURE) cohort study, 135,335 individuals from 18 countries were followed for total mortality and major cardiovascular events [18]. Intake of total fat was not significantly associated with the risk of myocardial infarction or CVD mortality.

Cancer — The relationship between fat intake and the development of cancer is even less well-documented than that of coronary heart disease. Some cross-cultural studies have suggested a role for high total fat intake in the genesis of breast, colon, and prostate cancers, but these studies, like those of heart disease, could be confounded by other factors.

●Case-control studies of diet and breast cancer have suggested a modest role for total fat intake. In one summary of results from 12 case-control studies, the RR of breast cancer was estimated to be 1.35 for each 100 g increment in daily intake of fat [19]. By contrast, cohort studies, which do not have the inherent recall bias of case-control studies, have generally not found this association. In a pooled analysis of seven cohort studies, for example, each 25 g increment of fat intake, adjusted for energy intake, was associated with a RR of breast cancer of only 1.05 (95% CI 0.94-1.11) [20]. Furthermore, in an analysis of the Nurses' Health Study, a prospective cohort study of 88,795 women, there was no evidence that either lower total fat intake or lower intake of specific types of fat decreased the risk of breast cancer [21].

●Associations between fat intake and colon cancer may be explained by red meat consumption rather than total fat intake [22].

●Systematic reviews from the WCRF/AICR continuous update project have concluded that there is no probable or convincing evidence that total dietary fat influences the risk of any cancers [23].

●Consistent with these observational studies, the Women's Health Initiative randomized clinical trial described above found that, at a mean follow-up of 8.1 years, the intervention to reduce total fat intake did not significantly affect the risk of colorectal cancer (hazard ratio [HR] 1.08, 95% CI 0.90-1.29) or breast cancer (HR 0.91, CI 0.83-1.01) [24,25].

Obesity — The prevalence of obesity in the United States has been rising for decades [26-28], prompting researchers to examine the role of dietary constituents in promoting an increase in body fat.

The comparative roles of different diet compositions, including low-fat diets, in reducing weight are discussed in detail separately. (See "Obesity in adults: Dietary therapy".)

Type 2 diabetes — In a meta-analysis of eight observational cohort studies, total dietary fat consumption was not associated with incidence of diabetes [29]. Further, in the Women's Health Initiative randomized trial, significant lowering of total fat intake over mean 8.1 years of follow-up had no significant effect on incidence of diabetes or on insulin resistance [30].

Total mortality — In the large, international PURE cohort study including a mix of high-, middle-, and low-income countries, higher intake of total fat was associated with significantly lower risk of death from all causes [18]. While the results may be confounded from poorer individuals consuming more starchy staples and having less access to (often more expensive) animal source foods, the findings provide little support for mortality benefits of lowering total fat consumption.

TYPE OF FAT — 

The type of fat consumed and its food sources appear to have some beneficial impacts on health outcomes (table 2) [17]. In general, industrial trans fats should be avoided. The Dietary Guidelines for Americans have concluded that substituting polyunsaturated for saturated fats appears to lower cardiovascular risk [31], whereas substituting carbohydrates for saturated fat does not reduce risk (and may cause excess weight gain) (see 'Polyunsaturated fatty acids' below). Evidence regarding the relationship between different types of dietary fat and clinical outcomes comes from large, long-term, randomized trials and observational studies of clinical events.

Trans fatty acids — Low levels of trans fatty acids occur naturally in some foods, especially dairy and meats from ruminants (cows, sheep, goats). However, much higher levels of trans fatty acid consumption can occur as a result of the industrial partial hydrogenation of unsaturated fatty acids. Trans fatty acids from this source appear to confer harmful cardiovascular effects [32-38].

Starting in 2006, the US Food and Drug Administration (FDA) required that Nutrition Facts labels include trans fat content. Several areas in the United States and Europe have passed legislation to restrict trans fats from foods served commercially [39,40]. In 2018, the FDA implemented its 2015 ruling that partially hydrogenated vegetable oils are no longer "generally regarded as safe," effectively removing industrial trans fatty acids from the United States' food supply [8]. In 2021, the European Union mandated that all foods contain less than 2 percent of trans fat (other than naturally occurring trans fat) as a percentage of all fat [41], effectively removing industrial trans fatty acids from their food supply.

Nonetheless, industrial trans fatty acids remain present in many nations. Globally, today's margarines contain less trans fatty acids than previously [42]. However, internationally, many deep-fried foods and commercial baked goods, such as cookies and cakes, still contain trans fatty acids. A clue to their presence is the words "partially hydrogenated" on the list of package ingredients. Restaurants and street vendors in low- and middle-income countries also commonly use partially hydrogenated oils for frying and cooking.

Trans fatty acids raise blood low-density lipoprotein (LDL) and lower high-density lipoprotein (HDL) cholesterol concentrations [33]. In a meta-analyses of randomized controlled feeding studies, each 1 percent energy consumption of trans fatty acids in place of saturated fats, monounsaturated fats, or polyunsaturated fats, respectively, increased the total cholesterol to HDL cholesterol ratio by 0.31, 0.54 and 0.67; increased the apolipoprotein (apo) B/apoA1 ratio by 0.007, 0.010 and 0.011; and increased lipoprotein(a) by 3.76, 1.39 and 1.11 mg/L (p<0.05 for each) [43].

By comparison, consumption of saturated fats, as compared with carbohydrate, also raises LDL cholesterol concentrations but concurrently lowers levels of triglyceride-rich lipoproteins and raises HDL cholesterol, with no net change in the total cholesterol to HDL ratio (see 'Saturated fatty acids' below). Thus, effects of saturated fats on the lipid profile are far less adverse than those of trans fatty acids.

Trans fatty acids may also interfere with the desaturation and elongation of n-3 (omega-3) fatty acids. These are important for the prevention of heart disease and complications of pregnancy. (See 'Heart disease' below.)

Several observational studies have linked the consumption of trans fatty acids, or foods that contain them, with adverse cardiovascular outcomes [32-37,44,45]. In an analysis from the Nurses' Health Study, for each increase of 2 percent of energy from trans fat, the relative risk (RR) for incident coronary heart disease was 1.93 (95% CI 1.43-2.61) [32]. Calculations suggest that replacement of trans fatty acids with other sources of fat, even saturated fatty acids, can reduce coronary risk [43].

Trans fat intake does not appear associated with diabetes risk. Based on a meta-analysis of seven cohorts, estimated trans fat consumption from dietary questionnaires was not significantly associated with incident type 2 diabetes [29]; and, based on individual-level analyses from an international consortium of 18 prospective studies with measured blood biomarkers, trans fatty acid blood levels were not associated with higher incidence of diabetes [46].

In contrast to the trans fatty acids discussed above, trans-palmitoleic acid is a trans fatty acid derived from naturally occurring dairy and other ruminant trans fats. Biomarkers of trans-palmitoleic acid intake are not associated with higher cardiovascular risk [33] and indeed have been associated with lower risk of diabetes in several cohort studies, perhaps conferring benefit by decreasing insulin resistance [47]. In a pooled de novo analysis among 63,682 participants from 16 prospective cohorts across 12 countries, higher blood biomarker levels of trans-palmitoleic acid were associated with lower incidence of type 2 diabetes, with 18 percent lower risk across the interquintile range [48].

More research is needed to determine the causality and mechanisms underlying a possible metabolic benefit of trans-palmitoleic acid or its major food sources, such as dairy fat. The evidence is strong to reduce consumption of trans fatty acids from partially hydrogenated vegetable oils and other industrial sources, given their adverse effects on cardiovascular risk factors discussed above.

Saturated fatty acids — While recommendations on saturated fat have conventionally grouped all saturated fatty acids together, increasing evidence indicates that different saturated fatty acids, and different food sources of saturated fat, have divergent effects on cardiovascular and metabolic health [49]. For example, saturated fatty acids with carbon chain lengths of 14 (myristic) and 16 (palmitic), chiefly found in dairy products and red meats, appear most potent in increasing both LDL and HDL cholesterol (figure 2) [50] and decreasing triglyceride-rich lipoproteins. Stearic acid (18 carbons), another component of beef and the chief fatty acid of cocoa butter, has far smaller effects on LDL and HDL cholesterol. Combining these effects on different lipid fractions, compared with isocaloric consumption of total carbohydrate, the consumption of myristic and palmitic acid have relatively little effect on the ratio of total cholesterol to HDL cholesterol, while stearic acid and especially lauric acid (12 carbons) significantly lower this ratio (figure 3).

Studies have shown a continuous, graded relationship between the total serum cholesterol concentration and coronary heart disease events and mortality (figure 4) [51,52]. However, concentrations of triglycerides and HDL cholesterol also independently predict risk. Thus, more discriminative predictors of risk appear to be ratios, such as the ratios of total to HDL cholesterol, non-HDL to HDL cholesterol, and apoB to apoA1 particles. The most common dietary saturated fats, palmitic and myristic, exert a neutral effect on these ratios. This is consistent with findings from systematic reviews and meta-analyses of prospective cohorts, which show no association between the overall intake of total saturated fat and risk for coronary heart disease [38,53] or incident type 2 diabetes [29].

By contrast, higher consumption of unsaturated fats improves blood cholesterol and lipoprotein ratios, compared with higher consumption of total carbohydrate or total saturated fat. Similarly, systematic reviews and meta-analyses of randomized clinical trials and prospective cohort studies have found that replacing saturated fats with polyunsaturated fats, rather than carbohydrates, appears to decrease the risk of coronary heart disease events [54,55]. This is consistent with the conclusions of the 2020 Dietary Guidelines Advisory Committee (DGAC) that there is strong evidence that replacing saturated fats with polyunsaturated fats reduces the risk of coronary heart disease events and cardiovascular disease (CVD) mortality, there is limited evidence available on whether replacing saturated fats with monounsaturated fats improves CVD endpoints, and there is strong evidence that replacing saturated fat with carbohydrate does not reduce risk [1,2].

Thus, a focus on increasing intake of unsaturated fats, in particular polyunsaturated fats, is based on good evidence. The relative intake of saturated fat, in relation to unsaturated fats, may be more important than its absolute intake. In a randomized feeding study, diets varying widely in saturated fat content (from 7 to 21 percent of total energy) had similar effects on LDL cholesterol concentrations when they were accompanied by similar proportional changes in unsaturated fats. Diets with the highest saturated and unsaturated fat contents, as replacements for carbohydrate, further reduced levels of lipoprotein(a) and atherogenic dyslipidemia [56].

Very low intakes of saturated fat, such as levels below 7 percent of total calories such as seen in some Asian countries, have been linked to higher risk of stroke, especially hemorrhagic stroke, in individual cohort studies as well as the large multi-country PURE cohort [18]. Specifically, across quintiles, higher intake of saturated fat was associated with lower risk of total mortality (cross-quintile hazard ratio [HR] 0.86, 95% CI 0.76-0.99) and stroke (HR 0.79, 0.64-0.98). While causality of these associations has not been established, hypothesized mechanisms include increased cerebral vascular fragility from low intakes of saturated fat and/or animal protein.

Less is known about health effects of other saturated fatty acids. Medium-chain triglycerides (including saturated fats with carbon lengths at or below 12 carbons) have been thought to be metabolically protective, while very long-chain saturated fats (lengths of 20, 22, or 24 carbons) have been linked to lower risk of coronary heart disease and heart failure [57,58]. Medium-chain saturated fats are derived from certain food sources, such as dairy fats, while very long-chain saturated fats are derived from nuts or endogenous synthesis.

Food sources of saturated fat exhibit even more divergent associations with cardiovascular and metabolic risk than individual saturated fatty acids [49]. In particular, higher intakes of processed meats, but less so unprocessed red meats and not dairy fat or plant sources of saturated fats, are consistently linked to higher cardiovascular risk [59]. Higher intakes of processed meats, and to a lesser extent unprocessed red meats, are also linked to higher risk of type 2 diabetes [59]. Intake of dairy foods, especially yogurt and cheese, are linked to lower risk of type 2 diabetes. The mechanisms for the latter protective associations have not been confirmed but may include benefits of medium-chain triglycerides, vitamin K2, milk fat globule membrane, and several other compounds in dairy fat [60]. Relatively little is known about long-term health effects of tropical fruit oils, such as palm oil and coconut oil, which are rich in different saturated fats.

In sum, the divergent associations of different individual saturated fatty acids with blood lipids and clinical endpoints and the divergent associations of different major food sources with clinical endpoints, have led some scientists to call into question the wisdom of maintaining a dietary recommendation that limits total saturated fat intake from all sources, as compared with food-specific recommendations [49,61,62]. Nonetheless, this remains controversial, and the 2018 draft World Health Organization (WHO) report on fatty acids maintains a recommendation to keep total saturated fat consumption below 10 percent of total calories. The 2020 DGAC report, while maintaining this general recommendation, adds that "the health effects of dietary saturated fat—or any other nutrient—depends not only on the total amount consumed, but also the specific type of saturated fatty acids inherent within the food matrix, sources and degree of processing, and the overall dietary pattern" [1]. Consistent with this, the 2020 DGAC partly translates the nutrient guidance to food-based guidance, "substituting some animal-source foods, especially processed meats and certain dairy products, with sources of polyunsaturated fats such as seafood, seeds, nuts, legumes, and appropriate vegetable oils," rather than focusing on avoiding all saturated fat.

Monounsaturated fatty acids — Evidence from randomized trials and observational studies supports increasing monounsaturated fatty acids from plant sources in place of saturated fatty acids. Plant sources of monounsaturated fats include olive oil, canola oil, tree nuts, and avocados.

In contrast to the diversity of individual saturated fatty acids, nearly all (>90 percent) monounsaturated fat in foods is oleic acid, an 18-carbon fatty acid with a single double bond. Food sources of oleic acid (and therefore monounsaturated fatty acids) remain diverse, however, with much coming from red meats and dairy fats (each containing approximately equal amounts of saturated fat and monounsaturated fat) and a smaller amount coming from plant oils like nuts, avocadoes, and olive oil. Olive oil is a major source in Mediterranean countries such as Spain and Italy.

Evidence from randomized trials supports the cardiovascular benefit of consuming plant foods rich in monounsaturated fat as part of a Mediterranean diet. The Mediterranean diet is discussed separately. (See "Healthy diet in adults", section on 'Mediterranean diet'.)

Data from the following randomized trials support replacing saturated with monounsaturated fats:

●In randomized trials, replacing saturated fats with monounsaturated fats decreases LDL cholesterol and triglycerides while maintaining HDL cholesterol [63-65]. Monounsaturated fatty acids can also decrease the oxidation of LDL cholesterol, a key step in atherosclerosis [66]. The in vivo implications of these findings are not yet clear, in part because much of the monounsaturated fats in the United States diet have come from animal rather than vegetable sources, and many studies have not been able to distinguish the two sources. (See "Lipoprotein classification, metabolism, and role in atherosclerosis".)

●A 2016 systematic review and meta-analysis including 102 randomized controlled feeding trials found that replacing carbohydrates with either monounsaturated fat or polyunsaturated fat improved several markers of glycemia, including HbA1c [67].

Polyunsaturated fatty acids — Polyunsaturated fatty acids can be categorized into two classes, the n-6 family of polyunsaturated fatty acids (eg, linoleic acid, arachidonic acid) and the n-3 family (eg, alpha-linolenic acid, eicosapentaenoic acid [EPA], docosahexaenoic acid [DHA]). The 18-carbon n-6 (linoleic) and n-3 (alpha-linolenic) fatty acids are essential nutrients, that is, they cannot be synthesized by humans, and their absence in the diet causes clinical deficiency. Thus, in contrast to saturated and monounsaturated fats, for which blood and tissue levels largely represent de novo synthesis in the liver from either dietary starch or sugars, blood and tissue levels of the major n-6 and n-3 fatty acids reflect dietary consumption.

Like monounsaturated fats, there is relatively little diversity in types of dietary n-6 fatty acids, which are predominantly (>90 percent) linoleic acid with small additional amounts of dietary arachidonic acid. Dietary n-3 fatty acids are either plant-derived (alpha linolenic acid, eg, from walnuts, canola, soybean) or fish/shellfish-derived (in particular EPA and DHA, present in all fish and shellfish but especially oily fish).

n-6 and n-3 fatty acids have been extensively studied in relation to cardiovascular, metabolic, and cancer risk factors and outcomes.

Heart disease — Intake of both n-6 and n-3 polyunsaturated fatty acids has been associated with decreased risk of coronary heart disease.

n-6 fatty acids — Based on randomized, controlled feeding studies, it is well established that n-6 fatty acids from plant oils (eg, soybean, safflower, sunflower, and corn oils) improve lipid and lipoprotein profiles, lowering serum LDL cholesterol, apoB levels, and triglycerides (very low-density lipoprotein cholesterol) while also increasing HDL cholesterol and apoA1 levels, leading to a significant overall reduction in the ratio of total to HDL cholesterol and ratio of apoB to apoA1 [65].

In randomized, controlled feeding studies, intake of n-6 fatty acids also improves long-term glycemic control, insulin resistance, and insulin production (see 'Diabetes mellitus' below). Some controlled trials have seen possible other benefits of n-6-rich plant oils, such as reduced liver fat accumulation and postprandial triglycerides [68,69]. Despite theorized proinflammatory effects of n-6 fatty acids or seed oils containing these fatty acids, such effects are not seen in human trials [68,70,71].

Meta-analyses of mostly older, often methodologically limited, controlled trials suggest that replacing butter and other animal fats with n-6-rich oils (predominantly soybean oil) is associated with a lower incidence of coronary heart disease.

Consistent with the beneficial effect on lipids and glycemia, meta-analyses of observational studies of both estimated dietary consumption or measured blood biomarkers of total n-6 fatty acids and linoleic acid have been associated with a lower risk of coronary heart disease. For example, in individual-level pooled analyses from 30 international prospective studies including nearly 70,000 participants, higher blood biomarker levels of linoleic acid were associated with lower risks of total CVD, cardiovascular mortality, and ischemic stroke, with HRs per interquintile range of 0.93 (95% CI 0.88-0.99), 0.78 (0.70-0.85), and 0.88 (0.79-0.98), respectively. The association between blood levels of linoleic acid and lower coronary heart disease risk did not reach statistical significance (0.94, 95% CI 0.88-1.00). Blood biomarker levels of arachidonic acid were not associated with higher risk of cardiovascular outcomes, but rather with lower risk of total CVD across quintiles (0.92, 95% CI 0.86-0.99).

The observed benefits of linoleic acid, the predominant dietary n-6 fatty acid, for coronary heart disease in cohort studies are seen whether linoleic acid is consumed in place of saturated fat or carbohydrate [72].

Studies evaluating the ratio of n-6 to n-3 fatty acids in the diet or blood sometimes show that higher ratios are linked to worse health outcomes. However, such associations in humans are generally driven by differences in levels of n-3 fatty acids rather than differences in levels of n-6 fatty acids [73].

Based on the lipid and other physiologic benefits, meta-analyses of observational cohort studies, and older clinical trials, the US Dietary Guidelines and many other guidelines recommend diets rich in n-6 fatty acids [74].

n-3 fatty acids — The n-3 family of polyunsaturated fatty acids includes EPA and DHA, which are found in fish oil, especially cold-water oily fish such as salmon, anchovies, mackerel, herring, sardines, and tuna. The third member of the family is alpha-linolenic acid, which is found in the oil of plants such as walnuts, soybeans, and canola.

Supplement or prescription n-3 fatty acids, including their effects on cardiovascular risk factors, are discussed in detail elsewhere. (See "Fish oil: Physiologic effects and administration".)

The evidence for cardiovascular benefits of plant-derived alpha-linolenic acid remains uncertain. A meta-analysis of 27 observational studies demonstrated that top versus bottom tertile of dietary intake of alpha-linolenic acid was associated with a borderline significant 10 percent lower risk of CVD (13 studies, pooled RR 0.90, 95% CI 0.81-0.99), with a similar but not statistically significant inverse association for biomarker studies of alpha-linolenic acid [75]. In general, Americans tend to consume relatively low amounts of alpha-linolenic acid, which is found in canola oil and some nuts and seeds (eg, flaxseed) [76].

Stroke — Due to conflicting results of studies of n-6 and n-3 polyunsaturated fats and risk of stroke, no firm recommendations are warranted regarding the effect of polyunsaturated fat intake on stroke risk [77].

●Consumption of polyunsaturated fats (combined n-3 and n-6 fatty acids) in the Framingham study was not associated with risk of stroke, whereas intakes of both saturated and monounsaturated fats were associated with lower risk of ischemic stroke [78].

●By contrast, a prospective study of 43,732 men participating in the Health Professionals Follow-up Study found no association between stroke and the intake of saturated fat, monounsaturated fat, polyunsaturated fat, trans unsaturated fat, or dietary cholesterol [79]. Intake of n-3 and n-6 polyunsaturated fatty acids were also not associated with ischemic or hemorrhagic stroke.

●In separate reports from the Nurses' Health (women) and Health Professionals (men) studies, participants with higher intake of fish and n-3 fatty acids had a reduced risk of ischemic stroke [80,81]. In the men, eating fish once per month or more was associated with a reduced risk of ischemic stroke compared with fish consumption less than once per month (RR 0.57, 95% CI 0.35-0.95) [81]. In the women, the effect was strongest among those who did not take aspirin regularly [80]. A meta-analysis that included these studies also found a reduced risk of stroke with increased fish consumption [82], as did a subsequent observational study in 4775 older adults [83]. It is a challenge to put these findings in context with the null findings among men for all types of fat discussed above [79]. It is possible that fish intake may provide benefits by some other mechanism than n-3 fatty acids or may be a marker for other healthy behaviors that reduce the risk of stroke.

Diabetes mellitus — Data are conflicting on the relationship between polyunsaturated fats and the development of diabetes. One cohort study of 11-year follow-up among 36,000 women showed inverse relations with incident type 2 diabetes for intake of plant fat and substitution of polyunsaturated fatty acids for saturated fatty acids and dietary cholesterol [84]. Some additional studies have also suggested a protective role for polyunsaturated fatty acids and a harmful role for trans fats, though other studies have not [85].

●In a systematic review and meta-analysis of 102 randomized controlled feeding trials, higher intakes of monounsaturated fat and polyunsaturated fat improved several markers of glycemia [67]. Specifically, consuming polyunsaturated fat in place of saturated fat reduced fasting glucose, HbA1c, and insulin resistance; and consuming polyunsaturated fat in place of carbohydrates, saturated fat, or monounsaturated fat also improved insulin secretion capacity. Replacing saturated fat with carbohydrates had no significant effect on fasting glucose, HAbA1c, or insulin resistance. These findings support a potential beneficial role of monounsaturated fatty acids and especially polyunsaturated fatty acids for glycemic control.

●In a meta-analysis of observational cohort studies, dietary polyunsaturated fats were not associated with incidence of type 2 diabetes when evaluated as a linear-dose response, but in nonlinear analysis, protective associations were seen for dietary vegetable fats and polyunsaturated fats up to modest levels of intake [29].

●In pooled analysis across 20 cohorts from 10 nations, blood biomarker levels of linoleic acid were inversely associated with incident diabetes, with 35 percent lower risk per interquintile range (RR 0.65, 95% CI 0.60-0.72). Biomarker levels of arachidonic acid were not significantly associated with type 2 diabetes risk overall (RR per interquintile range 0.96, 95% CI 0.88-1.05) [86].

Other diseases — Dietary fat intake may affect the risk of Alzheimer disease. Studies have suggested that intake of saturated fats and trans fats may increase the risk of Alzheimer disease and cognitive decline, while intake of monounsaturated and polyunsaturated fats may decrease this risk [87,88].

Plant versus animal fat — Plant sources of fat are generally high in mono- and polyunsaturated fatty acids (see 'Monounsaturated fatty acids' above and 'Polyunsaturated fatty acids' above), whereas animal sources are generally high in saturated fatty acids. In a prospective cohort study of approximately 407,531 United States adults, higher intakes of plant fat, especially fat from grains and vegetable oils, were associated with lower rates of all-cause and cardiovascular death (HR 0.91 and 0.86; adjusted risk differences -1.15 and 0.73 percent, respectively), comparing the lowest with highest quintiles [89]. Conversely, higher intakes of animal fat were associated with increased rates of all-cause and cardiovascular death (HR 1.16 and 1.14; adjusted risk differences 0.78 and 0.32 percent, respectively). However, different food sources of animal fat have divergent associations with mortality and cardiovascular risk. (See 'Saturated fatty acids' above.)

Dietary cholesterol — Dietary cholesterol raises the total serum cholesterol but is a less important contributor than saturated fat [90,91]. Eggs are a chief source of dietary cholesterol, but the association between regular egg consumption and risk of coronary heart disease and stroke is uncertain. As an example, in a 2013 meta-analysis of eight prospective cohort studies (more than 474,000 participants), there was no association between egg consumption and risk for coronary heart disease or stroke (RR 0.99 and 0.91 for each additional daily egg, 95% CI 0.85-1.15 and 0.81-1.02, respectively) [92].

For most individuals, only a minor emphasis on reducing dietary cholesterol alone is needed since it is a less important contributor to serum cholesterol and clinical endpoints than is saturated or trans fat [2,93]. However, some patients who either consume very large amounts of dietary cholesterol or whose serum LDL cholesterol response to moderate cholesterol intake is unfavorable may benefit from reduction in cholesterol intake.

Consistent with this evidence, the 2015 DGAC report de-emphasized the previous recommendation to limit consumption of dietary cholesterol to 300 mg per day, stating that dietary cholesterol is no longer a "nutrient of concern" for Americans [2]. The role of a low-cholesterol diet is discussed in detail elsewhere. (See "Healthy diet in adults", section on 'Low-cholesterol diet'.)

DIETARY FAT ADVICE — 

Based on advances in evidence, the 2015 Dietary Guidelines Advisory Committee (DGAC) report recommended no upper limit for total dietary fat, and the 2020 DGAC did not re-review this question [1,2]. By contrast, the earlier 2002 Institute of Medicine Dietary Reference Intakes, which guide the Nutrition Facts panel and other United States Department of Agriculture panels, have not been updated and recommended limiting total fat intake (for adults, between 20 and 35 percent of energy from fat [2,74]; for 4- to 18-year-olds, 25 to 35 percent; and for children one to three years of age, 30 to 40 percent). The 2020 DGAC report emphasized the importance of quality and types of fat and food sources, as reflected in the following advice:

●Consume less than 10 percent of calories from saturated fat by replacing them with polyunsaturated fat. Only limited evidence supported replacing saturated fat with monounsaturated fat, and replacing saturated fat with carbohydrate was not recommended.

●Keep trans fat consumption as low as possible by limiting foods that contain industrial sources of trans fats, such as partially hydrogenated oils.

●Reduce the intake of calories from solid fats (eg, from meats, butter). Use plant oils to replace these solid fats where possible.

●Replace protein foods that are higher in solid fats (eg, red and processed meats) with choices that are lower in solid fats and calories and/or are sources of oils. These include seafood, legumes (beans and peas), nuts, seeds, plant oils, and soy products, as well as lean meats, poultry, and eggs.

●Health effects of dairy fats remain uncertain and controversial. Most guidelines continue to recommend low-fat dairy products.

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Healthy diet in adults".)

INFORMATION FOR PATIENTS — 

UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topic (see "Patient education: Diet and health (The Basics)")

●Beyond the Basics topic (see "Patient education: Diet and health (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Heterogeneous effects of dietary fat – Given the heterogeneous effects of different types of fatty acids and food sources, the amount of total fat and perhaps total saturated fat and monounsaturated fat in the diet appear to be less important than the types of foods consumed. (See 'Total fat intake' above.)

●Fat intake and disease

•Heart disease – Industrial trans fats contribute to the genesis of heart disease, while n-3 polyunsaturated fats and n-6 polyunsaturated fats are protective. It is uncertain whether total saturated and monounsaturated fats have important effects on the risk of heart disease, likely due to substantial heterogeneity in subclasses of specific fatty acids, the diversity of food sources of these fats, and the various foods and nutrients they might replace in the diet. (See 'Type of fat' above.)

•Other diseases – Fewer data are available regarding the role of intake of fats in the incidence of stroke, cancer, type 2 diabetes, and other outcomes, although most of the evidence suggests results that are consistent with those concerning heart disease.

●Counseling patients about fat intake – Reasonable messages that are relatively easy to discuss with patients who do not require special diets for existing diseases (eg, chronic kidney disease, celiac disease, eating disorders) include the following, which are consistent with the 2020 Dietary Guidelines for Americans (see 'Recommendations for patients' above):

•Focus on components of a healthy diet – It can be counterproductive to recommend that patients restrict total fat intake. Instead, we focus on encouraging patients to eat more foods rich in healthful fats, vitamins, flavonols, and/or other bioactive compounds such as nuts, seeds, plant oils, fish, yogurt, fruits, nonstarchy vegetables, legumes, and minimally processed whole-grain products. Patients should limit intake of unprocessed red meats to one to two portions per week.

•Limit processed foods – Suggest that patients limit processed red meats and ultraprocessed foods that are rich in refined starch, added sugars, trans fat, and/or sodium.

•Low fat not necessarily healthier – Do not assume that low-fat or "fat-free" varieties of packaged and processed foods (eg, snack foods, salad dressings) are more healthful or better for weight control.

•Choose plant oils – When fats are needed for cooking, spreads, dressings, and other uses, recommend plant oils with high amounts of unsaturated fatty acids, including extra virgin olive oil and plant oils containing n-3 polyunsaturated fats such as soybean and canola oils.

ACKNOWLEDGMENT — 

The UpToDate editorial staff acknowledges Matthew W Gilman, MD, SM, who contributed to earlier versions of this topic review.