Dairy and Eggs

TL;DR. Most adults on Earth cannot digest lactose, and the ability to do so as an adult is the mutation, not the norm. Dairy is the clearest case in nutrition where the food matrix matters more than the fatty acid profile: cheese and yoghurt carry substantial saturated fat and are not associated with the cardiovascular risk that saturated fat alone would predict, while butter behaves as expected. Eggs were restricted for decades because of their cholesterol and the restriction was dropped, because dietary cholesterol turned out to matter much less than saturated fat for most people.

Dairy

1. Lactase persistence: the mutation

Lactase, the brush border enzyme that splits lactose into glucose and galactose, is normally switched off after weaning in mammals. Continuing to produce it into adulthood is lactase persistence, and it is a relatively recent genetic adaptation.

It arose independently at least four times, in Northern Europe, East Africa, the Arabian peninsula, and elsewhere, in populations that herded animals. It is one of the strongest signals of recent natural selection in the human genome, and the selective advantage was large enough to sweep through populations in a few thousand years.

Approximate adult lactase persistence:

Globally, roughly two thirds of adults are lactose non-persistent. The Northern European norm is the global exception, which is worth knowing because most nutrition guidance is written from inside it.

Lactose intolerance is a dose problem, not a binary. Most non-persistent people tolerate around 12 g of lactose (a glass of milk) without symptoms, especially with a meal. Tolerance also improves with regular small exposure, because colonic bacteria adapt.

Cheese and yoghurt are naturally low in lactose. Hard aged cheeses (cheddar, parmesan, gouda) contain almost none, because lactose goes into the whey and what remains is fermented. Yoghurt contains live bacteria that supply their own lactase, so it is far better tolerated than milk.

Don't be confused: lactose intolerance and milk allergy are completely different. Lactose intolerance is an enzyme shortage causing bloating, wind, and diarrhoea, and is uncomfortable rather than dangerous. Cow's milk protein allergy is an immune reaction to casein or whey proteins, is common in infants (2 to 3 percent, usually outgrown), and can cause anaphylaxis. Lactose-free milk contains the same proteins and is not safe for someone with milk allergy.

2. What is in it

Per 100 g:

Two under-recognised points. Dairy is a major iodine source in several countries, particularly the UK, where it supplies a large share of intake via iodophor teat disinfectants and fortified feed. Switching from cow's milk to unfortified plant milk removes it silently (Chapter 16). And dairy calcium is well absorbed at around 32 percent, which is good though not the best: kale and bok choy are around 50 percent.

3. The saturated fat paradox

Dairy fat is roughly 65 percent saturated, which should, on the fatty acid argument, make it cardiovascularly harmful. The observational data do not show that consistently, and the pattern differs by product:

• Fermented dairy (yoghurt, cheese): neutral or inversely associated with cardiovascular disease and type 2 diabetes across large cohorts. Yoghurt in particular is repeatedly associated with lower diabetes risk.
• Milk: broadly neutral.
• Butter: modestly associated with higher risk, and it raises LDL in controlled feeding studies more than the same fat delivered as cheese.
• Full-fat versus low-fat: the expected advantage of low-fat dairy has not appeared clearly in cohort data, and the large PURE study found dairy intake, including full-fat, associated with lower mortality and cardiovascular events.

Proposed explanations, none conclusive:

• The food matrix. Cheese delivers its fat inside a protein-calcium matrix. Calcium binds fatty acids in the gut, forming soaps that are excreted, and controlled studies show cheese raises LDL less than butter with identical fat content and increases faecal fat excretion.
• Fermentation products, including specific peptides and vitamin K2 (menaquinone).
• What dairy displaces in the diet.
• Residual confounding, which cannot be excluded.

Where this leaves guidance: most bodies still recommend lower-fat dairy on the grounds that the LDL mechanism is causal and well established, while acknowledging the cohort data are inconsistent. A defensible reading is that yoghurt and cheese are fine within a good overall pattern, butter is the one to moderate, and the choice between full-fat and low-fat milk matters less than the rest of the diet.

Milk and cancer deserves a note in both directions: higher dairy intake is associated with lower colorectal cancer risk (fairly consistent, plausibly via calcium) and higher prostate cancer risk (also fairly consistent, mechanism unclear, possibly IGF-1). Both effects are modest.

4. Plant milks

Not nutritionally equivalent to dairy or to each other:

Two practical rules. Buy fortified versions (calcium, B12, vitamin D, and iodine if available; organic plant milks are frequently unfortified in the EU because of regulation, which is a genuine trap). And shake the carton: added calcium carbonate settles, and studies have found a large fraction of the calcium remains in the bottom of an unshaken container.

Children under five switching from dairy to plant milk need particular care over protein, calcium, iodine, and energy density.

Eggs

5. What is in one

A medium egg (about 50 g, 44 g edible):

Everything except the protein is in the yolk. Egg white is essentially protein and water; discarding the yolk discards the choline, the vitamins A, D, E, and K, the lutein, the selenium, and the flavour.

Choline is genuinely important and widely under-consumed: it is needed for cell membranes, acetylcholine, and, critically, fetal brain development. Eggs are one of very few rich sources alongside liver.

Lutein from eggs is absorbed several times better than from spinach, because it arrives dissolved in fat within a lipid matrix (Chapter 17).

6. The cholesterol reversal

For decades, guidance limited dietary cholesterol to 300 mg a day, which is roughly one and a half eggs, on the reasoning that dietary cholesterol raises blood cholesterol.

What changed: the body regulates cholesterol production. When intake rises, endogenous synthesis falls. Your liver makes around 1 to 1.5 g a day, several times any dietary intake. In controlled studies, dietary cholesterol raises LDL far less than saturated fat does, and in most people the effect is small.

The 2015 US Dietary Guidelines removed the numerical limit, stating cholesterol is "not a nutrient of concern for overconsumption." Several other national bodies followed.

Where nuance remains:

• Roughly a quarter to a third of people are hyper-responders whose LDL does move noticeably with dietary cholesterol.
• Cohort findings in people with type 2 diabetes are less reassuring and less consistent, and some analyses find associations with cardiovascular risk at higher egg intakes in that group.
• A 2019 JAMA pooled analysis found each additional half egg a day associated with a small increase in cardiovascular events, which sits against several other analyses finding no association. The evidence is genuinely mixed at high intakes.

A reasonable position: an egg a day is fine for most people. If you have high LDL, diabetes, or familial hypercholesterolaemia, treat your own response as the question worth answering and test it rather than assuming.

7. Egg safety

• Salmonella. In the UK, eggs bearing the British Lion mark come from vaccinated flocks, and since 2017 the FSA has considered them safe to eat raw or lightly cooked, including for pregnant women, infants, and older people. This is a UK-specific position and does not transfer. In the US and many other countries, guidance remains to cook eggs thoroughly for vulnerable groups, and pasteurised eggs are available for raw preparations.
• Refrigeration differs by country for a real reason. US eggs are washed at packing, which removes the natural protective cuticle, so they must be refrigerated. European eggs are not washed, the cuticle stays intact, and they are stored at room temperature. Neither is wrong; they are different systems, and moving an egg between them mid-life is the mistake: once refrigerated, keep it refrigerated, because condensation on a cold egg brought into a warm room draws bacteria through the shell.
• Do not wash eggs in a European system.
• The float test works: as an egg ages, moisture leaves and the air cell grows, so a fresh egg sinks and lies flat, an older one stands on end, and a floating egg should be discarded.
• Raw egg white contains avidin, which binds biotin; only relevant at extreme intakes.

8. Labels

Egg labelling in the EU and UK is coded on the shell with a number for the system: 0 = organic, 1 = free range, 2 = barn, 3 = caged. That single digit is the most informative thing on the box.

"Farm fresh," "natural," and similar terms mean nothing. Nutritional differences between systems are small; omega-3 enriched eggs are genuinely different, from hens fed flaxseed or algae, and deliver a modest amount of ALA and some DHA. Shell colour is determined by the breed of hen and has no nutritional meaning whatever. Yolk colour reflects the hen's carotenoid intake and is routinely manipulated with feed additives, so a deep orange yolk indicates diet, not quality.

The bottom line

• Most adults worldwide cannot digest lactose, and lactase persistence is a recent mutation, not the default. It is a dose problem: cheese and yoghurt are largely lactose-free and far better tolerated than milk.
• Lactose intolerance and cow's milk protein allergy are different conditions; lactose-free milk is not safe for the second.
• Dairy is the clearest case where the food matrix beats the fatty acid profile: cheese and yoghurt do not behave like their saturated fat content predicts, and butter does.
• Dairy is a major and easily overlooked iodine source. Fortified soy or pea milk is the closest plant substitute; almond, oat, and rice milks are not protein foods, and rice milk is unsuitable for young children.
• An egg is one of the highest-quality proteins available plus most of the day's choline and very well-absorbed lutein, all in the yolk. The old cholesterol limit was dropped; the evidence at high intakes and in diabetes remains mixed.
• The number stamped on a European egg tells you the farming system: 0 organic, 1 free range, 2 barn, 3 caged.

Sources and notes

Composition figures are from USDA FoodData Central. Lactase persistence genetics, its independent origins, and population frequencies follow Tishkoff et al., Nature Genetics, 2007, and Itan et al.'s modelling. Lactose tolerance thresholds and adaptation follow Suarez, Savaiano, and Levitt's controlled feeding work. Dairy and cardiovascular outcomes follow the PURE study (Dehghan et al., The Lancet, 2018) and Guo et al.'s meta-analyses; the cheese-versus-butter LDL comparison and faecal fat excretion follow Hjerpsted, Leedo, and Tholstrup, American Journal of Clinical Nutrition, 2011. The food matrix argument follows Thorning et al., Food and Nutrition Research, 2017. Dairy and colorectal versus prostate cancer follow the World Cancer Research Fund Continuous Update Project. Iodine in UK milk follows Bath and Rayman's surveys. Egg composition, choline content, and lutein bioavailability follow USDA data and Chung et al.'s absorption work. The dropped cholesterol limit follows the 2015 US Dietary Guidelines Advisory Committee report; the contested pooled analysis is Zhong et al., JAMA, 2019. British Lion egg safety follows the FSA's 2017 revised advice. US egg washing and refrigeration practice follows USDA and FDA rules; EU non-washing follows Regulation 589/2008. Egg shell coding follows EU marketing standards.

Open questions. Whether the neutrality of fermented dairy in cohort studies is a genuine matrix effect or residual confounding is unresolved. Egg intake at higher levels and in people with diabetes remains genuinely equivocal across large cohorts.

👉 Next: meat and fish.