The peer-reviewed case for raw milk has expanded substantially over the last twenty years, much of it from large European birth-cohort studies that aren't widely cited in the US conversation.
What follows is a synthesis across five domains, with the underlying papers behind each claim. Where the evidence is strong, that's noted. Where it's thin or contested, that's noted too.
None of this is medical advice, and The Melk Stand makes no health claims about its products. The material below is a literature review, provided so customers can read the source papers themselves.
This is the strongest single area of raw-milk research. Five large European birth cohorts — PARSIFAL, GABRIELA, PASTURE, ALEX,and the Agricultural Lung Health Study in the US — converge on the same finding: children who drink raw farm milk in early life develop substantially less asthma, hay fever, eczema, and atopic sensitization than children who don't.
The mechanism is reasonably well-characterized. The heat-sensitive components implicated are native whey proteins (BSA, α-lactalbumin, β-lactoglobulin), omega-3 fatty acids, bioactive microRNAs (raw milk contains roughly three times the level present in pasteurized milk; UHT eliminates them), and alkaline phosphatase. Boiling the same farm milk eliminates the protective effect. In mouse models, raw milk prevents airway inflammation; heated raw milk does not.
The protection extends beyond childhood. The Wyss study followed 3,061 American adults and found those who drank raw milk as children had measurably better lung function decades later— FEV1 up about 50 mL, FVC up about 66 mL, holding into adulthood.
One illustrative comparison: the Amish, who drink raw milk and live around dairy cattle, have a 7% allergic-sensitization rate. The genetically similar Hutterites, who drink industrially-processed milk, have four to six times more asthma— despite shared Northern European ancestry.
Raw and pasteurized milk affect the immune system differently. The 2020 Butler study — the first published human investigation of raw milk and the gut microbiome — found unpasteurized dairy intake associated with significantly increased Lactobacillus abundance in the gut. The dataset is small, but it is currently the only direct human read on this question, and the direction of effect is consistent with the bench science.
The components pasteurization knocks out are well-catalogued. Native lactoferrin denatures roughly 80 times faster at 85°C than at 65°C; UHT essentially eliminates it. Bovine immunoglobulins (IgG, IgA, IgM), milk fat globule membrane proteins, lactoperoxidase, and TGF-β-bearing extracellular vesicles are all heat-sensitive. The 2017 Brick processing-intensity study quantified 23 immune-related proteins reduced by pasteurization.
Bovine IgG also appears to actively train human innate immunity. The van Splunter study showed bovine IgG from milk induces trained immunityin human monocytes — the same kind of immune memory associated with BCG vaccines. That is a measurable biological signal, not just a correlational observation.
Pasteurization is effective at one job: killing pathogens. The trade-off is that heat is non-selective — it also denatures or destroys a meaningful list of bioactive components. The data on which components are lost is mature and well-quantified. Recent 2025 work is more pointed still: raw milk preserves 100% of native α-lactalbumin and β-lactoglobulin, while UHT and 95°C treatment completely denature lactoferrin and IgG.
Enzyme losses are total. Alkaline phosphatase is destroyed by definition— it is the universal pasteurization marker, the test the FDA uses to confirm milk has been heated. Bile salt-stimulated lipase, lactoperoxidase, and the native lipases are all destroyed. These are digestion-relevant enzymes endogenous to the milk that do not survive heat treatment.
On vitamins, the 2011 Macdonald meta-analysis of 40 studies found significant decreases in B1, B2, vitamin C, and folateafter pasteurization, with B12 and E qualitatively reduced. Most reviewers describe the absolute nutritional impact as “modest” on the basis that milk is not most people's primary source of these vitamins — a reasonable framing, though it does depend on dietary context.
On fat: pasture-grazed cows produce milk with roughly five times the conjugated linoleic acid (CLA), more omega-3s, vaccenic acid, and β-carotene than confined-feedlot cows. The Williamson 1978 BMJ study found preterm infants grew faster on raw mother's milk than pasteurized, and Andersson 2007 attributed the gap to reduced fat absorption after heat treatment.
A common report from raw-milk drinkers: “I can't drink milk — but I tried raw and I was fine.” Survey data echoes the anecdote at scale. A 2014 Maryland survey of raw-milk drinkers found 99% reported no discomfort with raw milk, including many who had been told they were lactose intolerant for years. A 2,500-person Michigan survey of self-identified lactose-intolerant adults found roughly three out of four had no symptoms drinking raw.
The single best peer-reviewed test of “does raw milk help with lactose intolerance” is a Stanford crossover trial led by Christopher Gardner. It found no benefit. Raw milk produced the same digestive symptoms as pasteurized in confirmed lactose-maldigesters over an 8-day arm.
The most credible reconciliation in the literature is that the underlying issue often isn't lactose at all. It is A1 vs A2 β-casein— a protein difference between cow breeds that produces lactose-intolerance-like symptoms in some drinkers regardless of lactose load.
Many small Utah dairies and the breeds The Melk Stand sources from (Jerseys, Guernseys, heritage stock) are A2-rich or A2-only by genetics. When customers report tolerating this milk but not commodity grocery milk, the most defensible mechanism is not preserved lactase — it is the absence of A1 β-casein.
Sometime around 5,000 to 10,000 years ago, a single point mutation in cow DNA flipped a proline to a histidine at position 67 of the β-casein protein — one amino acid. The pre-mutation version is A2. The post-mutation version is A1. Most ancient cattle — African breeds, Asian breeds, all of Bos indicus— never picked up the mutation and still produce A2-only milk. The dominant European dairy breed, the Holstein-Friesian black-and-white that produces roughly 90% of American grocery-store milk, is heavily A1.
The functional difference: when A1 milk encounters digestive enzymes, gut proteases cleave a 7-amino-acid opioid peptide called BCM-7 (beta-casomorphin-7)from the protein. A2 milk does not release BCM-7 — the proline at position 67 makes it enzymatically resistant. BCM-7 binds to opioid receptors in the gut, slows transit time, increases inflammation, and produces a symptom profile that closely resembles lactose intolerance.
Beyond gut symptoms, more speculative work has connected dietary A1 to ecological correlations with Type 1 diabetesincidence (r = +0.726 across 0–14-year-olds in international datasets), to atherosclerotic plaque formation in rabbit models, and to elevated urinary BCM-7 in autistic children. The 2009 EFSA review concluded these non-GI claims “cannot be established” causally — a fair characterization. The GI evidence is robust; the cardiovascular and T1D evidence is suggestive but ecological.
The practical implication: most “milk doesn't agree with me” reports in the US describe Holstein milk specifically. The Jerseys, Guernseys, and heritage breeds dominant on small Utah dairies are A2-rich or A2-only.Tolerance differences customers report between this milk and commodity grocery milk are most likely explained by β-casein composition, independent of pasteurization status.
The research case for raw, locally-sourced, pasture-grazed milk has strengthened substantially over the last twenty years, particularly around childhood asthma, allergy, and digestive comfort for people who have been told they can't drink milk.
It is not a complete case. The lactose-intolerance picture is more nuanced than commonly told, and the gut-microbiome literature still needs more direct human studies. Those gaps are noted in the relevant sections above rather than glossed.
The research summarized on this page is provided for educational interest only and is not medical advice. The Melk Stand makes no health claims about its products. Raw milk products, no matter how carefully produced, may be unsafe. Risks are greater for infants, young children, elderly individuals, pregnant women, and those with compromised immune systems. Consume at your own discretion.