Is milk bad for cholesterol?

For blood cholesterol, the trials say no measurable harm. A meta-analysis of randomized trials of dairy foods found no significant change in total or LDL cholesterol. When 18 healthy adults drank half a liter a day of whole milk or skim milk for 3 weeks each, whole milk did not raise total or LDL cholesterol and raised HDL. Genetic data from up to 1.9 million people link drinking more milk to lower LDL. The part that has not gone away is the fat. Milk fat is saturated fat, and butter, its concentrated form, raises LDL against plant oils.

Myth. The idea that a glass of milk pushes up your cholesterol has been tested and did not hold up in randomized trials. The pooled analysis that says so covers dairy foods in general rather than milk alone, and its abstract gives no pooled numbers or trial count, so the size of any small change is unknown. The smaller trials of milk itself point the same way.

Unsettled. Whether whole or low-fat milk is the better daily choice over years is not settled. The trials lasted 3 to 12 weeks, mostly in people with normal cholesterol, and none measured heart attacks. Cohort data link high-fat milk to more cardiovascular deaths, and US and heart association advice now point in opposite directions.

What the trials found

Milk and dairy against no dairy or low-fat dairy

In the crossover trial, healthy adults with normal cholesterol added 17 fl oz (0.5 liters) a day of whole milk, at 3.5 percent fat, or skim milk, at 0.1 percent, to their usual diet for 3 weeks each. Of 18 who started, 17 finished. Total and LDL cholesterol did not differ between the two milks. HDL was higher after whole milk (P below 0.05). It is a very small and short trial.

A 12-week trial in 72 adults with metabolic syndrome compared a diet limited in dairy with 3.3 servings a day of either low-fat or full-fat milk, yogurt and cheese. Full-fat dairy had no effect on total, LDL or HDL cholesterol or triglycerides against either diet (P above 0.1). Lipids were exploratory outcomes in this trial, and the dairy was a mix rather than milk alone.

Genetic data

A Mendelian randomization study used a gene variant that keeps lactose digestion going into adulthood, and so raises milk drinking, in up to 1,904,220 people. Genetically higher milk intake went with lower LDL, total and HDL cholesterol (for LDL, P = 3.6e-6), and with higher body mass index. This method assumes the variant acts on cholesterol only through milk, which cannot be fully checked.

Milk fat compared with other fats

Whole milk in the US averages 3.2 g of fat and 1.86 g of saturated fat per 3.5 oz (100 g), from lab samples. That fat is not neutral when it is concentrated. A network meta-analysis of 54 randomized trials found that safflower, sunflower, rapeseed, flaxseed, corn, olive, soybean, palm and coconut oil, and beef fat, all lowered LDL by 8.9 to 16 mg/dL (0.23 to 0.42 mmol/L) more than butter. That result is about butter, not a glass of milk.

A 2026 network meta-analysis of 14 randomized trials with 543 adults found that replacing 17 fl oz (500 ml) a day of cow milk with soy drink may lower LDL by 18 mg/dL (95% CI 3.9 to 33; metric: 0.47 mmol/L, 0.10 to 0.85), on low-certainty evidence. HDL did not change. Soy protein lowers LDL on its own, so this gap may reflect what soy does rather than harm from milk.

Not all of milk fat behaves the same. A 2026 meta-analysis of 13 randomized trials in adults without disease found that concentrated milk fat globule membrane, the coat around milk fat droplets, lowered LDL by 4.6 mg/dL (1.2 to 8.5; metric: 0.12 mmol/L, 0.03 to 0.22) over 2 to 16 weeks, on moderate-certainty evidence. These were products made from the membrane at 19.8 to 5000 mg a day, not ordinary milk. A meta-analysis of cheese trials found cheese lowered LDL by 7.3 mg/dL (4.6 to 10; metric: 0.19 mmol/L, 0.12 to 0.27) against the same fat eaten as butter. Swapping cheese for milk showed no benefit to blood lipids, but only 2 trials tested that.

Heart outcomes, from cohort studies

Observational data from a meta-analysis of cohort studies link high-fat milk to a 9 percent higher risk of cardiovascular death (1.02 to 1.16) and a 15 percent higher risk of death from any cause (1.09 to 1.20). The review gives these as relative risks only, and confounding is likely because high-fat milk drinkers differ in diet and lifestyle. A separate review of 55 cohort studies found milk, cheese and yogurt showed inconsistent links with heart disease, stroke and high blood pressure, on low to moderate quality evidence.

Who should be careful

The firmest evidence on this page about heart events is not about milk at all. A Cochrane review found that cutting saturated fat for at least two years lowered combined cardiovascular events by 21 percent (risk ratio 0.79, 0.66 to 0.93), across 11 trials and 53,300 people, of whom 8 percent had an event, on moderate-quality evidence. The trials cut saturated fat from every source, and whole milk is one of them.

Not for everyone. If your LDL is already high, or you have familial hypercholesterolemia, the reassuring milk trials above do not cover you. They were run in people with normal cholesterol or metabolic syndrome, and we found no trial in familial hypercholesterolemia. In that case whole milk, cream and butter count toward your saturated fat like any other source, and the choice of milk is worth taking to the person treating you.

What expert bodies say

They disagree, and the disagreement is recent. The American Heart Association's 2021 dietary guidance advises choosing low-fat or fat-free dairy products. The Dietary Guidelines for Americans 2025-2030 say that when consuming dairy, people should include full-fat dairy with no added sugars, at 3 dairy servings a day for a 2,000 calories diet, which reverses the low-fat dairy advice of the previous edition. The same US guidelines also say saturated fat should in general stay at or below 10 percent of daily calories. WHO makes the same 10 per cent recommendation for adults and children, rated strong.

So both governments cap saturated fat at the same level. What changed in the US is where whole milk sits inside that cap. The short trials on this page found no difference in LDL between full-fat and low-fat dairy in either direction, which is why neither side can claim them.

How we searched

What would change this answer

• A large randomized trial of full-fat against low-fat milk lasting a year, with LDL and apoB as outcomes. The longest trial we found ran 12 weeks.
• Outcome data. No trial has tested whole against skim milk on heart attacks or strokes.
• Trials in people with high LDL or familial hypercholesterolemia, who are the readers most likely to ask this question and the ones the existing trials left out.
• The pooled numbers from the dairy trial meta-analysis, which would show how small the non-significant changes were.

What milk contains, glass by glass, is on the milk page. How fats in general move cholesterol is on its own evidence page.

Sources

1. ev-mlkc-01 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs · n = not stated in the abstract
   Dairy foods were associated with a non-significant reduction in triacylglycerol level, and a non-significant increase in total cholesterol, low density lipoprotein cholesterol, and high-density lipoprotein cholesterol level.
   
   Who: adults in randomized controlled trials of dairy food intake
   Effect: non-significant: small rise in total, LDL and HDL cholesterol, small fall in triglycerides
   Certainty: Abstract gives no pooled numbers, trial count or N; no_pmc. Dairy foods pooled, not milk alone.
   Diabetes Metab Syndr, 2021 · checked 2026-10-07 · we read the abstract
2. ev-mlkc-02 · Meta-analysis or systematic review · systematic review and network meta-analysis of RCTs, GRADE · n = 543
   Replacing 500 mL/d of cow's milk with soy drink may reduce low-density lipoprotein (LDL) cholesterol [MD: -0.47 mmol/L (-0.85; -0.10); low certainty], but showed no effect for high-density lipoprotein (HDL) cholesterol [MD: 0.01 mmol/L (-0.03; 0.05); moderate certainty].
   
   Who: adults in RCTs of at least 3 weeks comparing soy, rice or oat drinks with cow milk
   Effect: soy drink vs cow milk: LDL MD -18 mg/dL (-0.47 mmol/L) (95% CI -0.85 to -0.10; low certainty); HDL MD 0.01 (-0.03 to 0.05)
   Certainty: Low certainty; soy protein itself lowers LDL, so the gap may reflect soy rather than harm from milk.
   Adv Nutr, 2026 · checked 2026-10-07 · we read the abstract
3. ev-mlkc-03 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs, GRADE · n = 13 studies
   Pooled analyses demonstrated that 2 to 16 wk of MFGM/MPL intake (19.8-5000 mg/d) reduced fasting total cholesterol [weighted mean difference (WMD): -0.18 mmol/L; 95% confidence interval (CI): -0.30, -0.07], low-density lipoprotein cholesterol (WMD: -0.12 mmol/L; 95% CI: -0.22, -0.03), apolipoprotein B concentrations (WMD: -0.05 g/L; 95% CI: -0.09, -0.01), and the total cholesterol:high-density lipoprotein (HDL) cholesterol ratio (WMD: -0.17; 95% CI: -0.33, -0.01), with low heterogeneity (I2 = 0%-9%) and moderate certainty of evidence.
   
   Who: disease-free adults; MFGM or milk polar lipids 19.8 to 5000 mg/day vs isoenergetic dairy comparators
   Effect: total cholesterol WMD -7 mg/dL (-12 to -2.7; metric: -0.18 mmol/L, -0.30 to -0.07); LDL -0.12 (-0.22 to -0.03); apoB -0.05 g/L; moderate certainty
   Certainty: Concentrated MFGM products, not ordinary milk; shows that milk fat is not uniform in its effect.
   Adv Nutr, 2026 · checked 2026-10-07 · we read the abstract
4. ev-mlkc-04 · Meta-analysis or systematic review · systematic review and network meta-analysis of RCTs · n = 54 trials
   Safflower, sunflower, rapeseed, flaxseed, corn, olive, soybean, palm, and coconut oil as well beef fat were more effective in reducing LDL-C (-0.42 to -0.23 mmol/l) as compared with butter.
   
   Who: adults in randomized trials of at least 3 weeks comparing oils and solid fats, per 10% isocaloric exchange
   Effect: LDL -16 to -8.9 mg/dL (-0.42 to -0.23 mmol/L) for safflower, sunflower, rapeseed, flaxseed, corn, olive, soybean, palm, coconut oil and beef fat vs butter
   Certainty: About butter, not liquid milk; relevant to cream and whole-milk fat in large amounts.
   J Lipid Res, 2018 · checked 2026-10-07 · we read the abstract
5. ev-mlkc-05 · Meta-analysis or systematic review · systematic review and meta-analysis of RCTs, RoB 2.0 · n = 10 articles
   We found no evidence of a benefit from replacing cheese for ≥14 d with milk on fasting blood lipid markers (n = 2).
   
   Who: adults in RCTs of isoenergetic substitution of hard or semi-hard cheese with other dairy products for at least 14 days
   Effect: cheese vs butter: LDL WMD -7.3 mg/dL (-10 to -4.6; metric: -0.19 mmol/L, -0.27 to -0.12); cheese replaced by milk: no fasting lipid benefit (2 studies)
   Certainty: Only 2 trials compare cheese and milk; all trials some concerns for bias.
   Adv Nutr, 2023 · checked 2026-10-07 · we read the abstract
6. ev-mlkc-06 · Meta-analysis or systematic review · Cochrane systematic review and meta-analysis of RCTs · n = 53,300
   The included long-term trials suggested that reducing dietary saturated fat reduced the risk of combined cardiovascular events by 21% (risk ratio (RR) 0.79; 95% confidence interval (CI) 0.66 to 0.93, 11 trials, 53,300 participants of whom 8% had a cardiovascular event, I² = 65%, GRADE moderate-quality evidence).
   
   Who: adults with or without cardiovascular disease in trials of at least 24 months reducing saturated fat
   Effect: combined cardiovascular events RR 0.79 (95% CI 0.66 to 0.93); 11 trials; moderate-quality evidence; small reductions in LDL cholesterol
   Certainty: About saturated fat from all sources, not milk; whole milk is one source of it.
   Cochrane Database Syst Rev, 2020 · checked 2026-10-07 · we read the abstract
7. ev-mlkc-07 · Randomized controlled trial(s) · randomized controlled crossover trial · n = 18
   There were no significant differences between whole milk and skimmed milk in effects on total and LDL cholesterol, triacylglycerol, insulin, and glucose concentrations.
   
   Who: healthy adults; 0.5 L/day of whole milk (3.5% fat) or skim milk (0.1% fat) added to habitual diet, 2 x 3 weeks; 17 completed
   Effect: no significant difference in total or LDL cholesterol; HDL higher on whole milk (P < 0.05)
   Certainty: Very small and short; normocholesterolemic adults only.
   Eur J Clin Nutr, 2018 · checked 2026-10-07 · we read the abstract
8. ev-mlkc-08 · Randomized controlled trial(s) · randomized controlled trial, exploratory endpoints (NCT02663544) · n = 72
   In men and women with metabolic syndrome, a diet rich in full-fat dairy had no effects on fasting lipid profile or blood pressure compared with diets limited in dairy or rich in low-fat dairy.
   
   Who: men and women with metabolic syndrome; limited dairy, or 3.3 servings/day of low-fat or full-fat milk, yogurt and cheese for 12 weeks
   Effect: no intervention effect on total, LDL or HDL cholesterol or triglycerides (P > 0.1); per-protocol n = 66
   Certainty: Lipids were exploratory outcomes; dairy mix, not milk alone.
   Am J Clin Nutr, 2021 · checked 2026-10-07 · we read the abstract
9. ev-mlkc-09 · Observational data · two-sample Mendelian randomization with meta-analysis · n = up to 1,904,220
   Our large-scale MR study provides genetic evidence for the association of milk consumption with higher BMI but lower serum cholesterol levels.
   
   Who: UK Biobank, 1958 British Birth Cohort, Health and Retirement Study and consortia; lactase persistence variant LCT-13910 C>T as instrument
   Effect: genetically higher milk intake: lower LDL-C (P = 3.6e-6), total and HDL cholesterol; higher BMI; lower coronary artery disease risk OR 0.86 (0.75 to 0.99) for the T allele
   Certainty: Genetic association; assumes the variant acts only through milk intake.
   Int J Obes (Lond), 2021 · checked 2026-10-07 · we read the abstract
10. ev-mlkc-10 · Observational data · systematic review and dose-response meta-analysis of prospective cohort studies · n = not stated in the abstract
    High-fat milk consumption was significantly associated with a greater risk of all-cause (Pooled ES: 1.15; 95% CI: 1.09-1.20, I2=24.5%, p = 0.22), CVD (Pooled ES: 1.09; 95% CI: 1.02-1.16, I2=4.5%, p = 0.38) and cancer mortality (Pooled ES: 1.17; 95% CI: 1.08-1.28, I2=30.1%, p = 0.19).
    
    Who: adults in prospective cohort studies of milk and dairy intake and mortality
    Effect: high-fat milk: CVD mortality ES 1.09 (1.02 to 1.16), all-cause 1.15 (1.09 to 1.20); total dairy: CVD mortality 0.93 (0.88 to 0.98)
    Certainty: Observational; residual confounding likely (high-fat milk drinkers differ in diet and lifestyle).
    Crit Rev Food Sci Nutr, 2022 · checked 2026-10-07 · we read the abstract
11. ev-mlkc-11 · Observational data · systematic review and meta-analysis of prospective cohort studies, NutriGrade · n = 55 studies
    Milk, cheese, or yogurt consumption showed inconsistent associations with the cardiovascular outcomes in high compared with low intake and dose-response meta-analyses.
    
    Who: adults in prospective cohort studies of dairy intake and hypertension, CHD and stroke
    Effect: total dairy, highest vs lowest: CHD RR 0.96 (0.92 to 1.00); stroke 0.90 (0.85 to 0.96); milk inconsistent
    Certainty: Low to moderate quality, considerable heterogeneity.
    Adv Nutr, 2022 · checked 2026-10-07 · we read the abstract
12. ev-mlkc-12 · Position of an expert body · scientific statement of the American Heart Association · n = —
    choose healthy sources of protein (mostly plants; regular intake of fish and seafood; low-fat or fat-free dairy products; and if meat or poultry is desired, choose lean cuts and unprocessed forms)
    
    Who: general population, cardiovascular prevention
    Effect: low-fat or fat-free dairy recommended
    Certainty: Disagrees with DGA 2025-2030 (full-fat dairy). Not in corpus; quote read twice via Europe PMC REST (EXT_ID and DOI queries); ahajournals.org returned 403.
    Circulation, 2021, 2021 Dietary Guidance to Improve Cardiovascular Health: A Scientific Statement From the American Heart Association · checked 2026-10-07 · we read the abstract
13. ev-mlkc-13 · Position of an expert body · WHO guideline · n = —
    WHO recommends that adults and children reduce saturated fatty acid intake to 10% of total energy intake (strong recommendation).
    
    Who: adults and children
    Effect: SFA at most 10% of total energy (strong recommendation); replacement with PUFA (strong)
    Certainty: General limit on saturated fat; whole milk counts toward it.
    WHO, 2023, Saturated fatty acid and trans-fatty acid intake for adults and children: WHO guideline · checked 2026-10-07 · we read the section
14. ev-mlkc-14 · Position of an expert body · national dietary guideline · n = —
    When consuming dairy, include full-fat dairy with no added sugars.
    
    Who: general US population
    Effect: recommends including full-fat dairy without added sugars; 3 dairy servings/day at 2,000 calories
    Certainty: Reverses DGA 2020-2025 (low-fat dairy) and disagrees with AHA 2021.
    Dietary Guidelines for Americans, 2025-2030 (HHS, USDA), January 2026 · checked 2026-10-07 · we read the section
15. ev-mlkc-15 · Position of an expert body · national dietary guideline · n = —
    In general, saturated fat consumption should not exceed 10% of total daily calories.
    
    Who: general US population
    Effect: saturated fat no more than 10% of total daily calories
    Dietary Guidelines for Americans, 2025-2030 (HHS, USDA), January 2026 · checked 2026-10-07 · we read the section
16. milk-b-14 · Position of an expert body · government fact sheet announcing the 2025-2030 edition of the Dietary Guidelines for Americans · n = —
    The guidance calls for receiving the bulk of fat from whole food sources, such as meats, poultry, eggs, omega 3–rich seafood, nuts, seeds, full-fat dairy, olives, and avocados.
    
    Who: US general population aged 2 and over
    Effect: whole milk and full-fat dairy named among whole-food fat sources the guidance directs people toward, reversing the previous low-fat dairy direction
    Certainty: this is a position of an authority, not evidence, and it is in open conflict with the 2023 WHO saturated fat guideline held in our own T2 layer. Two governments reading broadly the same literature now say different things about the same glass of milk, and the honest page reports the disagreement rather than picking a side. Verified by two independent fetches: the first returned a paraphrase, the second the verbatim sentence
    Dietary Guidelines for Americans 2025-2030, USDA/HHS fact sheet · checked 2026-09-21 · we read the page
17. wmlk-daily-r4-01 · Position of an expert body · reference dataset, not a study · n = —
    FDC 746782 Milk, whole, 3.25% milkfat, with added vitamin D: Total lipid (fat) 3.2 g (range 2.59-3.41, 24 samples) · Fatty acids, total saturated 1.86 g · Protein 3.27 g · Calcium 123.0 mg (range 116.0-130.0, 11 samples) · Vitamin B-12 0.54 ug per 100 g
    
    Who: US retail whole milk, 3.25% milkfat, Foundation samples
    Effect: fat 3.2 g (2.59 to 3.41, 24 samples), saturated fat 1.86 g, protein 3.27 g, calcium 123 mg (116 to 130, 11 samples), B12 0.54 ug per 3.5 oz (100 g)
    Certainty: US samples; vitamin D added.
    USDA FoodData Central, Foundation Foods, release 2026-04-30, FDC 746782 (Milk, whole, 3.25% milkfat, with added vitamin D) · checked 2026-09-29 · we read the dataset