The product review as outcomes data

A parent buying a PKU medical formula in 2026 has access to roughly the same product information that a parent buying a PKU formula in 1996 had. Brand websites describe the formula's nutritional profile and prescribing indications. The metabolic dietitian provides clinical guidance on amino acid composition and dosing. Other parents share their experience through Facebook groups, conference hallway conversations, and the small number of advocacy nonprofits that maintain product databases.

What the parent does not have is structured outcomes data. Whether Formula A is associated with better dietary adherence than Formula B in a population matched for age, severity, and metabolic control. Whether one formula's gastrointestinal tolerability is consistently better than another's in toddlers versus school-age children. Whether the higher-cost formula's clinical advantage persists when the lower-cost generic is available. The questions are concrete. The answers do not exist in a form anyone can query.

The structural reason is that medical food and rare disease product markets have not had the data infrastructure to produce real-world outcomes evidence. Manufacturers have sales figures. Insurers have claims data. Clinicians have anecdote. The community has Facebook posts. None of these sources, taken alone or combined, produces the comparative outcomes data that the question requires.

What structured community reviews could be

The technical specification of a structured review is straightforward. A patient or family member who uses a product reports the product, the diagnosis the product is being used for, the clinical context (severity, comorbidities, concurrent treatments), the duration of use, the outcome the product was used to influence, the side effects experienced, and the overall assessment. The structured fields produce a row in a database that supports query and aggregation. The free-text portion of the review captures the qualitative dimension that the structured fields cannot.

The aggregate dataset that emerges from enough structured reviews is a real-world evidence dataset for the product category. Comparative effectiveness questions that the manufacturer has never been asked to answer become answerable from the data the customer base produces in the course of using the product. The dataset has properties that traditional market research and clinical trials do not match.

The properties that matter for clinical inference are sample size, longitudinal duration, real-world variability, and patient-attributed authorship. Sample size is achievable in a community-driven dataset because the community is the population. Longitudinal duration emerges naturally because patients use these products for years. Real-world variability is the substrate, not a confounder, because the dataset is meant to capture how the product performs in actual use rather than in idealized trial conditions. Patient-attributed authorship is the property that traditional clinical trials cannot replicate, because the patient is the data source rather than a study participant.

What the dataset enables

Three categories of question become tractable when structured reviews accumulate at sufficient scale.

The first is comparative effectiveness within a product class. If 200 PKU patients reviewing three different medical food products report adherence, palatability, blood phenylalanine impact, gastrointestinal side effects, and cost, the aggregate data supports comparison across the products in a way that no single trial has been designed to support. The comparison is a pragmatic comparative effectiveness analysis conducted by patients in their own homes. The data is generated as a byproduct of using the product, not as a research output.

The second is regulatory consideration. Real-world evidence frameworks at the FDA, EMA, and other regulatory agencies have evolved to accept patient-reported outcome data and post-market real-world evidence under defined conditions. A structured review dataset that meets the data quality standards for real-world evidence becomes citable in regulatory submissions, coverage decisions, and clinical guideline development. The line between consumer review and outcomes evidence is the line of structure and provenance, both of which are engineering choices in the dataset's design.

The third is competitive market pressure. The medical food market for inborn errors of metabolism has been described, accurately, as a stagnation market for forty years. Formulas have not changed substantially since the 1980s in part because there is no data-driven competitive pressure to improve them. A community-controlled dataset where the manufacturer of the worst-performing formula in objective metrics faces the same data the manufacturer of the best-performing formula faces creates the market dynamic that has been absent. The forty-year stagnation has structural causes, and the data infrastructure is the missing piece.

What the data trust adds

The data trust governance model adds three properties that distinguish the structured review dataset from the existing alternatives.

Patient-controlled access means the contributor decides who queries their data. Manufacturers can pay for aggregate analysis. Researchers can request specific cohort access under terms the community establishes. The patient retains the right to withdraw the data, restrict its use, and review the analyses produced from it.

Aggregation rights with reidentification protection means the dataset supports population-level analysis without exposing individual patient identity. The anonymization is technical (differential privacy, secure multiparty computation, federated analysis) and contractual (the access terms forbid reidentification attempts). The combination produces analytic value at the population level without privacy cost at the individual level.

Persistent custody means the dataset survives sponsor transitions, manufacturer acquisitions, and institutional changes. The data is not held by the manufacturer whose product is being reviewed. The data is held by a fiduciary structure that the affected community establishes. The dataset accumulates over decades regardless of which companies are in the market at any given time.

The product review as outcomes data is the structural shift the rare disease product market has needed for forty years and has not been built to produce. The infrastructure question is what kind of consumer review system, with what fields, under what governance, generates the dataset. The technical answer is achievable. The governance answer is the work.