The paradox of safer donor milk
Donor human milk banking is the subject here: a system that collects, checks, processes and distributes human milk for infants who may need it urgently. The central question is not whether banking makes milk “good” or “bad,” but what is gained and what may change when a biological material is made safer for vulnerable recipients. Understanding that distinction lets us read claims about pasteurization, preservation and new processing technologies without mistaking a laboratory difference for a clinical verdict.
That question matters because the recipients are often preterm or low-birthweight infants, whose nutritional and immune needs are especially demanding. When a mother’s own milk is unavailable or insufficient, the World Health Organization (WHO) identifies donor human milk as the next best alternative, provided safe and affordable milk-banking facilities are available. Donor milk is also described as a bridge: it can help stabilize feeding while a mother has time to initiate or restore lactation. It is not a reason to treat breastfeeding as optional, and it is not evidence that milk from a bank is biologically identical to milk directly from the infant’s mother.
The apparent paradox follows from those two facts. Banking has to reduce hazards that could harm a recipient. Yet human milk is valuable partly because it contains structurally delicate biological components—immune proteins, enzymes, lipids and complex carbohydrates—not merely calories and bulk nutrients. A process that improves one dimension of the product can alter another. The task is to understand that trade-off precisely, not to choose between romanticizing “raw” milk and celebrating safety as if it had no biological cost.
Why donor milk exists—and why it does not replace breastfeeding
The starting point is a clinical gap, not a technology contest. A preterm infant may need human milk before the mother can produce enough, or a mother may temporarily be unable to provide it. In that situation, donor human milk can supply nutrition and some immunological protection while the feeding relationship develops. WHO’s standards place this use inside a broader priority order: mother’s own milk comes first whenever it is available and appropriate; donor milk follows when it is not enough or cannot be provided.
That order protects two relationships at once. The recipient needs a safe product, but the donor’s own infant must continue to receive their mother’s milk. A milk bank is therefore not simply a place where surplus liquid is acquired. It is an institution that has to decide whether a donation is genuinely surplus, support the donor’s lactation, and avoid turning the act of donation into pressure to produce more milk than is healthy. The quality of the final bottle cannot be separated entirely from the conditions under which it was obtained.
The same point applies to the recipient. Donor human milk is not a universal promise of identical outcomes for every infant. The WHO standards describe it as a critical intervention for preterm and low-birthweight infants, but they also acknowledge that much of the evidence behind the detailed recommendations is low or very low certainty. The fact that donor milk is preferred to formula in a specific care context does not prove that every processing method preserves every benefit, or that a difference in a molecule will change growth, infection risk or development.
A useful way to see the subject is to follow the milk as a product and as a relationship. It begins with a donor and the donor’s infant. It passes through eligibility decisions, expression and storage. At a bank, contributions may be pooled, treated and tested before they reach a vulnerable recipient. At every transition, the system is trying to preserve enough of the milk’s value while reducing hazards and keeping the supply available.
This pathway also explains why the word “safer” needs a careful object. Safer for whom, against which hazard, at which point in the chain, and with what consequence for the rest of the product? Those questions are more informative than a general contrast between natural and processed milk.
What a milk bank changes before the milk is heated
Pasteurization is visible because it has a name, a temperature and a time. Much of the bank’s work is less conspicuous, but it changes the safety problem before treatment begins. Screening can identify health conditions, exposures or behaviors that could put a recipient at risk. WHO recommends a risk questionnaire and serological screening, including exclusion of donors who test positive for HIV or HTLV-1. Education and support address hand hygiene, equipment cleaning, safe expression and changes in health or medication during the donation period.
Cold-chain control manages a different risk. Expressed milk should be refrigerated or frozen promptly, and the bank must maintain monitored storage and transport conditions. Time at room temperature can allow bacterial load to rise; inadequate freezing, thawing or temperature monitoring can increase the chance that a batch fails later testing. Treatment cannot make every earlier handling decision irrelevant. A product that has been exposed to uncontrolled growth is not equivalent to one that arrived under reliable cold-chain conditions simply because both pass through the same heater.
Containers and batch decisions create further interfaces. Food-grade containers need to close effectively, resist breakage and be easy to clean, disinfect and label. Milk from several donors may be pooled to reduce variation in protein and fat and to make the nutritional product more consistent. But pooling also means that a contamination event can affect a larger batch, and it can make it harder to preserve meaningful information about individual contributions. That is why traceability matters: a bank should be able to track milk from donor to recipient and trace it back in the event of an adverse event.
Even the milk entering this chain is not uniform. Its composition varies with factors such as lactation stage, gestational age and maternal diet. Protein, some vitamins and certain minerals may change across lactation, while other components follow different patterns. Milk collected early after birth is not simply “better” than mature milk; it is different, and the relevance of that difference depends on the recipient and the evidence available. Pooling can smooth some variation, but it cannot turn a changing biological material into a perfectly standardized one.
This is why WHO’s standards include donor welfare and equitable access alongside microbiological safeguards. Financial incentives are discouraged, while actual costs such as transport or expression supplies may be reimbursed. The concern is not that every paid donation is automatically unsafe. It is that financial pressure can reward volume over quality, encourage donors to withhold information, exploit people in difficult circumstances, or increase the volume entering a system faster than screening and processing capacity can grow. Practical support can reduce the burden of donation without creating the same incentive.
In this wider view, safety is a chain of partially independent controls. Donor screening addresses hazards associated with the source. Hygiene and containers address contamination at collection. Refrigeration and freezing limit growth during storage. Treatment reduces many microbial hazards. Post-pasteurization testing checks whether the batch remains within a defined safety limit. Traceability makes a failure investigable. None of these controls is a complete substitute for the others, and none promises that the milk remains untouched by the process.
Safer is not the same as unchanged
The most important distinction is between three questions that are often compressed into one. First: did the process reduce a microbial hazard? Second: did it preserve the composition or activity that makes human milk biologically valuable? Third: did the processed product improve an outcome that matters to an infant? The answers can differ.
A treatment may reduce microbial load while lowering the activity of a heat-sensitive immune protein. A total nutrient concentration may remain stable while a particular molecular profile changes. A protein may look preserved in a chemical measurement but behave differently during digestion. And even a demonstrated change in activity may not translate into a measurable difference in infection, feeding tolerance, growth or development. Each step requires a different kind of evidence.
This is the boundary around the title’s paradox. Saying that banked milk is not biologically unchanged is not the same as saying it is unsafe or unworthy. It means that safety interventions have consequences that should be measured rather than hidden behind a broad label. Conversely, saying that an alternative process preserves more of one component is not enough to establish that it is the best choice for a milk bank. The method must also control relevant hazards, work reliably at the required scale, fit available testing and storage systems, and remain accessible to the infants who need donor milk.
The current evidence supports this cautious posture. WHO’s standards generally recommend heat treatment by Holder pasteurization or high-temperature short-time processing for milk from screened healthy donors, while allowing that untreated milk might be appropriate in settings with stronger screening and testing. The standards do not recommend retort treatment and describe high-pressure processing and other non-thermal methods as promising but not yet supported well enough for broad recommendation. A recent multi-omics review likewise finds that processing effects vary by technique and by molecular domain.
Those findings invite a better question than “Does pasteurization destroy the benefits of human milk?” The question is: which properties change under which conditions, how certain is the evidence, and what does the change mean for the recipient and the system that must deliver the milk? The answer begins with the standard treatment itself. Holder pasteurization is widely used because it offers an established safety layer, but its temperature–time exposure also gives us a clear case for examining how microbial protection and biological change occur together.