Understanding natural variation in milk lipids: why parity and lactation stage matter for nutritional value

BACKGROUND: Bovine milk is a major dietary source of lipids whose fatty acid (FA) profile shapes its nutritional and technological quality, yet its natural variation is commonly described using fixed days-in-milk windows that ignore individual peak-yield timing and parity-specific metabolism. We used routine milk-recording data and Fourier transform mid-infrared spectroscopy (FT-MIR) to characterize parity- and stage-dependent dynamics in milk composition, defining lactation stages relative to each cow's model-based peak yield. RESULTS: A total of 10 209 test-day records from 677 Holstein cows (160 primiparous; 517 multiparous) from a commercial herd in Galicia (Spain) were analyzed for milk components, FA groups and individual FAs, β-hydroxybutyrate (BHB) and udder-health traits using linear mixed models. Multiparous cows had higher milk fat, BHB and somatic cell count and higher concentrations of most FA fractions, whereas primiparous cows had higher lactose, protein and urea (P < 0.05); peak-referenced stage effects were larger and more coherent in multiparous cows. Disaggregating multiparous cows by lactation number revealed consistent gradients with advancing parity (declining de novo synthesized FAs and rising BHB and somatic cell count) superimposed on, but not overturning, the primiparous-multiparous contrast. CONCLUSION: Parity structure and peak-referenced lactation staging are practical, FT-MIR-measurable determinants of raw-milk fat quality; accounting for them supports more precise milk characterization and stage- and parity-informed standardization of dairy products for human nutrition. © 2026 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.