Whole milk: the short trials are calm, the long worries are observational

In a randomized crossover trial, 17 healthy adults with normal cholesterol drank 17 fl oz (0.5 L) a day of whole milk (3.5% fat) or skim milk for 3 weeks each. Whole milk did not worsen fasting blood lipids, glucose or insulin compared with skimmed. The trial was small and short. A cup is 244 g in US survey coding, and 3.5 oz (100 g) of US whole milk averages 3.2 g of fat (2.59 to 3.41 across 24 samples), 1.86 g of saturated fat, 3.27 g of protein and 123 mg of calcium. The long-term evidence is observational and points both ways. Among 73,860 Norwegian adults followed for a median 33 years, high whole milk intake compared with none was linked to 15 percent higher all-cause mortality (HR 1.15, 95% CI 1.10 to 1.21). The cohort was recruited in 1974 to 1988, drank a great deal of milk, and other differences between drinkers may explain part of the link. In 14 studies of 20,897 children aged 1 to 18, whole milk against reduced-fat milk was linked to lower odds of overweight or obesity (OR 0.61, 95% CI 0.52 to 0.72). Those studies were mostly cross-sectional with no randomized trial, so heavier children may simply have been switched to lower-fat milk. For adults, a 2024 USDA review found no trial at all of milk fat level and body composition, and the 14 cohorts it did find disagreed.

The numbers, per 3.5 oz (100 g)

As people eat it, per 3.5 oz (100 g)

How much is actually in it

• US whole milk averages 3.2 g fat, 1.86 g saturated fat, 3.27 g protein and 123 mg calcium per 3.5 oz (100 g) across lab samples. — US retail whole milk, 3.25% milkfat, Foundation samples Evidence E source
  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)
• Iodine is the most variable nutrient in whole milk: 24 US samples ranged from 17.6 to 83.8 micrograms per 3.5 oz (100 g). — US retail whole milk, Foundation samples Evidence E source
  Iodine 37.9 ug per 3.5 oz (100 g), range 17.6 to 83.8, 24 samples
• UK tables list pasteurized whole milk at 63 calories, 3.6 g fat and 120 mg calcium per 3.5 oz (100 g), a little fattier than the US sample average. — UK pasteurized whole milk, average Evidence E source
  63 calories, fat 3.6 g, saturated fat 2.29 g, protein 3.4 g, lactose 4.6 g, calcium 120 mg, iodine 31 ug per 3.5 oz (100 g)

What a real portion looks like

• In US dietary survey coding a cup or a school carton of whole milk is 244 g. — US portion measures for whole milk Evidence E source
  1 cup 244 g, 1 individual school container 244 g, 1 fl oz 30 g

What your body absorbs

• In a crossover trial in 19 adults drinking 15 fl oz (430 ml), the share of vitamin A absorbed was similar from whole, fortified whole and fortified skim milk. — 19 healthy volunteers, 10 women and 9 men, single 15 fl oz (430 ml) doses on separate days (n = 19) Evidence B source
  Relative vitamin A absorption did not differ among the three milks
• After leg exercise, 8.4 oz (237 g) of whole milk led to 21% uptake of ingested threonine by muscle against 11% for fat-free milk. — Three groups of volunteers drinking milk 1 h after leg resistance exercise Evidence B source
  Threonine uptake 21 +/- 6% (whole) vs 11 +/- 5% (fat-free), 12 +/- 3% (isocaloric fat-free)

What cooking and storage change

• Laboratory data from 40 studies suggest pasteurization lowers some milk vitamins, including B2 (SMD -0.41), with minimal effect on overall nutritive value. — 40 studies measuring vitamins in raw and pasteurized milk (n = 40 studies) Evidence D source
  B1 SMD -1.77, B2 -0.41, C -2.13, folate -11.99, B6 not significant
• In a crossover RCT in 20 women drinking 17 fl oz (500 mL), UHT milk raised blood essential amino acids more than pasteurized milk (AUC 55,324 vs 36,598). — 20 healthy female habitual dairy consumers, 17 fl oz (500 mL) of UHT or pasteurized milk (n = 20) Evidence B source
  Essential amino acid AUC 55,324 vs 36,598 umol.min/L, P = 0.006

What it does to your health

• Observational data from 73,860 Norwegians followed 33 years link high whole milk intake to 15% higher all-cause mortality than none. — 73,860 adults in 3 Norwegian counties, median follow-up 33 years (n = 73860) Evidence C source
  All-cause mortality HR 1.15 (95% CI 1.10 to 1.21), IHD HR 1.13 (1.01 to 1.27)
• Observational data link higher whole milk intake in women to higher ovarian cancer risk (RR 1.35), while low-fat milk was linked to lower risk. — Women in 29 case-control and cohort studies (n = 29 studies) Evidence C source
  Whole milk RR 1.35 (1.15 to 1.59), low-fat milk RR 0.84 (0.73 to 0.96)

Safety, limits and interactions

• In an RCT of 60 infants with cow's milk allergy, one supervised low dose of whole milk at diagnosis meant 65% reached drinking plain milk by 12 months against 35%. — 60 infants under 12 months with suspected IgE-mediated cow's milk allergy, 12 months follow-up (n = 60) Evidence B source
  24/37 (65%) vs 7/20 (35%) at step 12 of the milk ladder, p = 0.03

What people believe that the data does not support

• Observational data link whole milk in children to lower odds of overweight or obesity than reduced-fat milk (OR 0.61). — Children aged 1 to 18 in 14 studies (n = 20897) Evidence C source
  OR 0.61 (95% CI 0.52 to 0.72), I2 73.8%
• In a 3 week crossover RCT in 17 healthy adults, 17 fl oz (0.5 L) a day of whole milk did not worsen fasting blood lipids, glucose or insulin compared with skim milk. — 18 healthy adults randomized, 17 completed, 0.5 L/d whole (3.5% fat) or skim milk for 3 weeks each (n = 17) Evidence B source
  No adverse effect on fasting blood lipids, glucose or insulin vs skim milk

What is still unknown

• A 2024 USDA systematic review found no trials testing whether milk fat level changes body composition or obesity risk in adults. — Adults and older adults, 14 prospective cohort studies (n = 14 studies) Evidence C source
  No trials, cohort findings inconsistent

The bottom line

Composition shifts between countries and samples. UK tables list pasteurized whole milk at 63 calories, 3.6 g of fat and 120 mg of calcium per 3.5 oz (100 g), a little fattier than the US average. Iodine varies most: 24 US samples ranged from 17.6 to 83.8 micrograms per 3.5 oz (100 g). Fat in the milk did not change how much vitamin A people absorbed. In a crossover trial of 19 adults drinking single 15 fl oz (430 ml) doses, the share absorbed was similar from whole, fortified whole and fortified skim milk over 6.5 hours of measurement. After leg resistance exercise, muscle took up 21 percent of ingested threonine from whole milk against 11 percent from fat-free milk, in small groups. Fat-free milk matched for calories gave 12 percent, and whole milk was not significantly ahead of it. Processing matters a little. In a crossover RCT of 20 women who habitually eat dairy, one 17 fl oz (500 mL) serving of UHT milk raised blood essential amino acids more than pasteurized milk (AUC 55,324 against 36,598 umol.min/L). Laboratory data from 40 studies suggest pasteurization lowers some vitamins, and the authors judged the effect on nutritive value minimal because milk holds little of those vitamins to begin with. Observational data from 29 case-control and cohort studies link the highest whole milk intake in women to higher ovarian cancer risk (RR 1.35, 95% CI 1.15 to 1.59) and low-fat milk to lower risk (RR 0.84, 0.73 to 0.96). Case-control designs are prone to recall bias, so this is an association. The caution with firm data is cow's milk allergy in babies. An open randomized trial enrolled 60 infants under 12 months with suspected IgE-mediated allergy. There 24 of 37 (65 percent) given one low dose of milk under medical supervision at diagnosis had reached step 12 of the milk ladder at 12 months, against 7 of 20 (35 percent) without it. The control group was small. The dose was given in clinic under medical supervision, never at home, so the trial says nothing about trying it without a doctor. We have no card on lactose intolerance, pregnancy or medicines for whole milk.

Sources

1. wmlk-daily-r4-01 · reference dataset
   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
   USDA FoodData Central, Foundation Foods, release 2026-04-30, FDC 746782 (Milk, whole, 3.25% milkfat, with added vitamin D) · checked 2026-09-29
2. wmlk-daily-r4-02 · reference dataset
   FDC 746782 Milk, whole, 3.25% milkfat, with added vitamin D: Iodine, I 37.9 ug per 100 g (range 17.6-83.8, median 36.6, 24 samples)
   USDA FoodData Central, Foundation Foods, release 2026-04-30, FDC 746782 (Milk, whole, 3.25% milkfat, with added vitamin D) · checked 2026-09-29
3. wmlk-daily-r4-03 · reference dataset
   CoFID 12-596 Milk, whole, pasteurised, average: Energy (kcal) 63 · Fat 3.6 g · Satd FA 2.29 g · Protein 3.4 g · Lactose 4.6 g · Calcium 120 mg · Iodine 31 ug per 100 g
   CoFID 2021, Public Health England, code 12-596 (Milk, whole, pasteurised, average) · checked 2026-09-29
4. wmlk-daily-r4-04 · reference dataset
   FNDDS 2705385 Milk, whole (food code 11111000): 1 cup 244 g · 1 individual school container 244 g · 1 fl oz 30 g
   USDA FNDDS 2021-2023, food code 11111000, fdc_id 2705385 (Milk, whole) · checked 2026-09-29
5. wmlk-daily-r4-05 · RCT
   The percent relative absorption of vitamin A did not differ among the three types of milk.
   Eur J Nutr, 2006 · checked 2026-09-29
6. wmlk-daily-r4-06 · RCT
   Threonine uptake relative to ingested was significantly (P < 0.05) higher for WM (21 +/- 6%) than FM (11 +/- 5%), but not IM (12 +/- 3%).
   Med Sci Sports Exerc, 2006 · checked 2026-09-29
7. wmlk-daily-r4-07 · meta-analysis of laboratory studies
   The effect of pasteurization on milk's nutritive value was minimal because many of these vitamins are naturally found in relatively low levels.
   J Food Prot, 2011 · checked 2026-09-29
8. wmlk-daily-r4-08 · RCT
   Although GCV remained larger for a longer duration following UHT-M (treatment × time interaction, P = 0.002), plasma essential amino acid AUC was greater following UHT-M than PAST-M (55,324 ± 3809 compared with 36,598 ± 5673 μmol·min·L-1, P = 0.006).
   Am J Clin Nutr, 2024 · checked 2026-09-29
9. wmlk-daily-r4-11 · cohort study
   For whole milk, high compared with no intake was positively associated with CVD (HR: 1.07; 95% CI: 0.99, 1.16), IHD (HR: 1.13; 95% CI: 1.01, 1.27), and all-cause mortality (HR: 1.15; 95% CI: 1.10, 1.21).
   Am J Clin Nutr, 2025 · checked 2026-09-29
10. wmlk-daily-r4-12 · meta-analysis of observational studies
    For comparisons of the highest v. lowest intakes, higher whole milk intake was associated with increased ovarian cancer risk (RR 1·35; 95 % CI 1·15, 1·59), whereas decreased risks were observed for higher intakes of low-fat milk (RR 0·84; 95 % CI 0·73, 0·96), dietary Ca (RR 0·71; 95 % CI 0·60, 0·84) and dietary vitamin D (RR 0·80; 95 % CI 0·67, 0·95).
    Br J Nutr, 2020 · checked 2026-09-29
11. wmlk-daily-r4-13 · RCT
    However, 24/37 (65%) of the intervention group were at step 12 compared to 7/20 (35%) of the control group (p = .03).
    Allergy, 2022 · checked 2026-09-29
12. wmlk-daily-r4-09 · meta-analysis of observational studies
    Among children who consumed whole (3.25% fat) compared with reduced-fat (0.1-2%) milk, the OR of overweight or obesity was 0.61 (95% CI: 0.52, 0.72; P < 0.0001), but heterogeneity between studies was high (I2 = 73.8%).
    Am J Clin Nutr, 2020 · checked 2026-09-29
13. wmlk-daily-r4-10 · RCT
    Intake of 0.5 L/d of whole milk did not adversely affect fasting blood lipids, glucose, or insulin compared to skimmed milk.
    Eur J Clin Nutr, 2018 · checked 2026-09-29
14. wmlk-daily-r4-14 · systematic review
    There were no trials directly assessing the effect of different fat content in milk on the outcomes of body composition and risk of obesity in adults.
    USDA Nutrition Evidence Systematic Review, Dairy Milk and Milk Alternatives and Growth, Body Composition, and Risk of Obesity, 2024 · checked 2026-09-29