Is saturated fat really bad for your heart?

Saturated fat is the fat that stays solid at room temperature, found in butter, cheese, fatty meat, and coconut and palm oils. Its best-established effect is on LDL cholesterol. LDL is a proven cause of atherosclerosis, based on genetic, cohort and trial data covering more than two million people⁷. Cutting saturated fat modestly lowers LDL¹.

The trials. Fifteen trials lasting at least two years (56,675 people) found that cutting saturated fat reduced cardiovascular events by 17% (risk ratio 0.83). Bigger cuts in saturated fat gave bigger benefits, and about 56 people needed to cut it for four years to prevent one event. Total deaths did not change clearly¹. In eight trials, replacing saturated fat with polyunsaturated vegetable oil lowered coronary events by 19%⁸.

Why cohorts look neutral. Pooled cohort analyses found saturated fat intake by itself was not linked to heart disease or death^9,10, and in 18 countries (PURE) higher intake went with lower stroke and death rates¹¹. The catch is "compared with what?" In two large US cohorts, replacing 5% of calories from saturated fat with polyunsaturated fat went with 25% lower coronary risk, with monounsaturated fat 15% and with whole grains 9%. Replacing it with refined starch and sugar showed no benefit⁵.

Food matters. One review argues whole-fat dairy, unprocessed meat and dark chocolate are not linked to higher cardiovascular risk³. Coconut oil, by contrast, raised LDL by 10.5 mg/dL in 16 trials compared with nontropical oils⁶. Our articles on dairy, red meat and coconut oil go further.

The counter-case. When the old Minnesota Coronary Experiment data were recovered, a corn-oil diet lowered cholesterol but did not reduce deaths¹². That is part of why polyunsaturated oils are debated, including in our seed oils articles.

Randomized trials. The Cochrane review (15 trials, 56,675 participants, at least 24 months) found a risk ratio of 0.83 (95% CI 0.70 to 0.98) for combined cardiovascular events with moderate-quality evidence, and greater cuts in saturated fat gave greater benefits. It found little or no effect on all-cause mortality (0.96, 0.90 to 1.03) or cardiovascular mortality (0.95, 0.80 to 1.12). The number needed to treat was 56 in primary prevention. Subgroups showed no clear difference between replacing saturated fat with polyunsaturated fat or carbohydrate, though data on monounsaturated fat were sparse. The first 2020 version reported a 21% reduction; the revised August 2020 version cited here reports 17%¹. Mozaffarian and colleagues pooled eight trials (13,614 people, 1,042 coronary events) that increased polyunsaturated fat to 14.9% of energy versus 5.0%, and found a 19% reduction (risk ratio 0.81, 0.70 to 0.95)⁸. The American Heart Association advisory judged the evidence strong enough to conclude lowering saturated fat in favor of unsaturated fats will lower cardiovascular disease, citing a reduction of about 30% in trials².

Cohorts and pooled analyses. Siri-Tarino and colleagues pooled 21 cohorts (347,747 people) and found relative risks of 1.07 for coronary disease (0.96 to 1.19), 0.81 for stroke and 1.00 for cardiovascular disease, none significant⁹. De Souza's meta-analysis gave 0.99 for all-cause mortality, 0.97 for cardiovascular mortality and 1.06 for total coronary disease, with certainty rated very low¹⁰. PURE followed 135,335 adults for 7.4 years: the top fifth of saturated fat intake had a hazard ratio of 0.86 for death and 0.79 for stroke compared with the bottom fifth, and no difference for heart attack, while high carbohydrate intake went with higher mortality (1.28). Diets differ greatly across those 18 countries, and the authors argued guidelines be reconsidered¹¹. Replacement analyses help reconcile this: a pooled analysis of 11 cohorts (344,696 people) found that replacing 5% of energy from saturated fat with polyunsaturated fat gave a hazard ratio of 0.87 for coronary events, while replacing it with carbohydrate gave 1.07⁴, and the Nurses' Health and Health Professionals cohorts (127,536 people, 24 to 30 years, 7,667 events) gave 0.75 for polyunsaturated fat and no benefit from refined carbohydrates⁵.

The contradicting trials. In the Minnesota Coronary Experiment (9,423 residents of nursing and mental hospitals), a diet swapping saturated fat for corn oil lowered cholesterol by 13.8% against 1.0% but did not reduce mortality, and mortality rose with greater cholesterol reduction (hazard ratio 1.22 per 30 mg/dL). A meta-analysis of five such trials (10,808 people) found no benefit for coronary death (1.13, 0.83 to 1.54) or all-cause mortality (1.07, 0.90 to 1.27)¹². The authors argue incomplete publication inflated earlier estimates; participants lived in institutions, which may limit how far the results generalize.

LDL and food sources. The European Atherosclerosis Society consensus concluded that LDL causes atherosclerotic disease, and that benefit scales with the size and duration of the LDL reduction⁷. Astrup and colleagues argue that saturated fat raises mostly large, less harmful LDL particles and that whole foods differ from nutrients, so dairy, unprocessed meat and dark chocolate need not be limited; their conclusion contradicts the AHA advisory³. A coconut oil meta-analysis (16 trials) found LDL up 10.47 mg/dL (95% CI 3.01 to 17.94) and HDL up 4.00 mg/dL, with no effect on glycemia, inflammation or adiposity⁶.

Bottom line (confidence 3 of 5). Lowering saturated fat probably gives a modest cut in heart events when unsaturated fat or whole grains replace it, but not in deaths, and not when refined carbs replace it. Cohort data are neutral and food source matters, so treat the nutrient as one input among many.