Calcium

The worry is straightforward. People who cannot digest lactose drink less milk, milk is a major source of dietary calcium, and calcium builds bone. So do lactase non-persistent people end up with weaker bones?

This question has an unusually good answer, because the variant controlling lactase persistence — rs4988235 — is assigned at conception and does not shift with income, diet fashion or anything else. It works as a natural experiment.

A study of 97,811 Danish adults did exactly that. Each copy of the persistence allele was worth 0.58 more glasses of milk per week (95% CI 0.49–0.68) — so the genotype does change behaviour, measurably, across a lifetime. The hazard ratio for hip fracture per allele was 1.01 (0.94–1.09). Nothing. Pooled with other Northern European studies, the odds ratio for any fracture was 0.86 (0.61–1.21) — an interval wide enough to hold a benefit, a harm and no effect at all.

Femoral neck bone density was very slightly higher in persistent homozygotes (0.10 g/cm², 0.02–0.18). Lumbar spine and total hip showed no difference. A small density difference at one site, and no difference in the thing that actually breaks.

None of this means calcium does not matter, and none of it is advice for someone already diagnosed with osteoporosis. It means that if you avoid dairy because of lactose, the fracture argument for a calcium supplement is weaker than it sounds. See also lactose intolerance.