An audit of nearly 2,800 products on toddler-food shelves across Austin, Texas found that 81 percent were ultra-processed and 49 percent failed at least one World Health Organization nutrient benchmark for children — a finding presented as a conference abstract at NUTRITION 2026, the American Society for Nutrition’s annual meeting, and not yet a peer-reviewed published study.
What the audit found
Researchers from the University of Texas at Austin surveyed 2,783 packaged food products marketed for children aged 6 to 36 months across 21 grocery stores in Austin. The survey excluded infant formula, beverages, and first-stage purees, focusing instead on the snacks, pouches, and toddler-labeled meals that fill the “baby and toddler food” aisle.
The headline numbers, as presented:
- 81% of products were classified as ultra-processed.
- 49% failed at least one World Health Organization nutrient benchmark.
- Among the specific benchmark failures: 17% exceeded sugar limits, 25% exceeded sodium limits, 11% fell short on required fat content, and 12% missed energy-density targets.
“Many parents may assume products sold on baby/toddler food aisles must meet certain nutrient standards,” said Erin A. Hudson, JD, an American Society for Nutrition Science Policy Fellow and PhD candidate who presented the work, alongside co-author Marissa Burgermaster, PhD, an assistant professor at UT Austin. The research was presented July 26, 2026, at the Gaylord National Resort & Convention Center in National Harbor, Maryland.
Why the methodology is the real story
What makes this audit useful to research administrators and methods-minded readers isn’t just the topline percentage — it’s that the classification wasn’t left to the researchers’ own judgment call. The team scored every product against two existing, publicly documented standards rather than an in-house rubric:
- The Nova food classification system, a widely used four-group framework (unprocessed/minimally processed, processed culinary ingredients, processed foods, and ultra-processed foods) originally developed by researchers at the University of São Paulo and now used broadly in nutrition epidemiology to categorize foods by the extent and purpose of industrial processing rather than by nutrient content alone.
- The WHO Nutrient and Promotion Profile Model, the World Health Organization’s published benchmark set for the nutrient composition and marketing of foods aimed at infants and young children.
Because both instruments are named, publicly available standards rather than a proprietary or ad hoc scoring method, another team with access to the same 2,783 product labels and the same store list could, in principle, reclassify the same shelf audit and check whether they land on the same 81 percent and 49 percent figures. That’s a materially different evidentiary posture than a study that reports only its own summary statistics without naming the classification instrument behind them — it’s the same logic that underpins why CASRAI’s own guidance on research data management stresses using named, versioned, citable standards rather than bespoke internal ones: the audit trail is only as good as whether someone else can walk it.
What this is, and what it isn’t
This is a retail shelf audit of packaged products already on store shelves — not a clinical trial, not a study involving children as human subjects, and not a claim that any single named brand or product is unsafe. No IRB or human-subjects review question arises here, because no children or parents were enrolled as research participants; the “subjects” of the audit were product labels and packaging.
It’s also important to be precise about evidentiary status: this work was presented as a conference abstract at NUTRITION 2026. Abstracts presented at scientific meetings are typically screened for relevance and basic soundness by the meeting’s program committee, but — as the American Society for Nutrition itself notes of research presented at its meetings — they “generally have not completed the same peer review process required for research published in a scientific journal.” That means the specific percentages above should be read as preliminary findings pending full peer review and eventual publication, not as a finalized, journal-vetted result. A single-market audit (21 stores in one Texas metro area) also cannot be assumed to generalize nationally without further work — a point implicit in the “4 in 5” framing some coverage has used, which extrapolates from this one-city sample.
The regulatory gap behind the finding
The reason the audit’s authors frame the result as notable is a genuine gap in U.S. food regulation: unlike infant formula, which the FDA regulates under specific nutrient-content and labeling requirements, packaged foods marketed for toddlers past the formula stage are not subject to a dedicated nutrient-adequacy standard before they can carry “baby” or “toddler” branding on the package. A product can use toddler-oriented marketing without being formulated, or independently verified, against benchmarks like the WHO model applied here.
Sources
Primary source: American Society for Nutrition, coverage of the NUTRITION 2026 presentation, via ScienceDaily, “4 in 5 toddler foods are ultra-processed” (July 28, 2026), reporting on research presented July 26, 2026 at NUTRITION 2026, Gaylord National Resort & Convention Center, National Harbor, Maryland.
