Clinical Data Registry

A clinical data registry sits between two more familiar research-administration concepts: it collects clinical information at the scale and structural discipline of a research database, but organizes that collection around a defined patient population rather than around a single tested hypothesis. That distinction — population-defined and standardized versus protocol-defined and hypothesis-testing — is what separates a registry from a traditional clinical trial, even though registries and trials frequently interact (a trial may recruit from a registry, or a registry may serve as an external comparator for a trial).

How registries differ from clinical trials

A clinical trial is organized around an investigational question: a protocol specifies eligibility criteria, assigns participants to an intervention (or comparator/placebo), follows a fixed visit schedule, and closes enrollment once a pre-calculated sample size is reached. A registry, by contrast, is organized around a population defined by a shared disease, procedure, device, or exposure. It does not assign an intervention — it documents what happens to patients under ordinary or specialty-society-recommended care — and it typically stays open on an ongoing or periodically renewed basis rather than closing once a target N is hit. The two are frequently linked in practice: a device or drug registry required as a condition of regulatory approval is a form of mandated post-market surveillance, and registries are a common source of external control arms and real-world comparators for trials.

What standardization means in this context

The defining operational feature of a registry, beyond its population, is a shared data dictionary: a pre-specified, versioned set of variables, definitions, and coding conventions that every contributing site applies the same way. This is what makes pooled, risk-adjusted, cross-site benchmarking possible — without it, a multi-site collection effort is just a set of incompatible local databases. Specialty-society registries typically publish and periodically revise this data dictionary, and participating sites are expected to map their local data collection to it, often through dedicated abstraction or registry-coordinator roles.

Purposes registries serve

• Quality improvement and benchmarking. Risk-adjusted outcomes reporting lets a participating site compare its own performance against a specialty or national benchmark, which is the original and still dominant purpose of most society-run registries.
• Observational research. Large, standardized, longitudinal registry populations support epidemiological and comparative-effectiveness research that would be impractical or unethical to study via a randomized trial.
• Public health surveillance. Disease- or exposure-specific registries (e.g., cancer registries, rare disease registries) track incidence, prevalence, and trends at a population level.
• Regulatory reporting and post-market surveillance. Some registries exist specifically to satisfy a condition of device or drug approval, an accreditation requirement, or a payer coverage-with-evidence-development requirement.

The regulatory role — a bounded and evolving one

Registries are one of the source types FDA recognizes as Real-World Data (RWD) — data on patient health status or care delivery collected outside a conventional randomized controlled trial — under the Real-World Evidence framework FDA established following the 21st Century Cures Act (2016) and its December 2018 RWE Framework. FDA’s own position is that a registry’s data can, in principle, contribute to Real-World Evidence supporting a regulatory decision (such as a new indication for an already-approved product, or satisfying a post-approval study commitment), but only where the underlying data are assessed as fit for that specific regulatory use: relevant to the question being asked, and reliable in terms of data quality, provenance, and completeness. Not every registry is built to that standard, and being an RWD source does not automatically make a given registry’s data adequate for a given regulatory submission — that determination is made case by case, in the context of a specific study design and question, generally in consultation with FDA. Sponsors and institutions relying on registry data for a regulatory purpose should treat this as a case-specific regulatory-strategy question rather than a general property of registries as a category, and should consult current FDA guidance directly rather than assuming any particular registry automatically qualifies.

Governance and data considerations

Registries raise many of the same data-governance questions as other clinical research data: participating sites need clarity on data-sharing agreements, whether registry data is collected under a HIPAA authorization, waiver, or a public-health/quality-improvement exception, and how long-term identifiers and linked data are protected. Because registries are often sponsored by a third party (a specialty society, a manufacturer, a public agency) rather than by the enrolling institution itself, the governing agreement between sponsor and site — covering data ownership, publication rights, and permitted secondary uses — is a standard part of standing up institutional participation.