27/08/2026
Should you swap dairy for plant foods?
A recent narrative review argues that replacing dairy with nuts, legumes and wholegrains adds benefits for CVD, T2D, cancer and mortality. Appealing idea. It doesn't hold up.
Recent meta-analyses show dairy is protective up to around 250-300g per day, and dairy fat biomarkers aren't linked to increased mortality risk. Dairy isn't harmful.
So where do substitution models fall down?
🥛 They simulate outcomes mathematically. They don't track real dietary change in real people.
🥛 Calcium, iodine and B12 are the exact nutrients of concern in plant-based diets, and dairy is a primary source.
🥛 It's not like for like. Nobody replaces milk with lentils. Only soy milk compares favourably - other plant milks are 4-fold lower in protein, under 25% of the calcium, zero iodine.
🥛 In Song et al.'s analysis of 130,000+ people, the association between dairy protein and mortality vanished entirely in those with healthy lifestyles.
🥛 One model estimated halving dairy costs around five days of life. That's a null result in practical terms.
The review also omitted the PURE study (130,000 people, 21 countries) where whole-fat dairy was linked to lower mortality, ran no risk-of-bias assessment, and was published in an unindexed journal.
The real issue is the framing. A healthy diet can include both. Building a hierarchy between otherwise healthy foods isn't useful in clinical practice.
Full referenced living review available to Nutripaedia members. https://bit.ly/3SNg753