The most important question you'll answer
Before choosing a platform, hiring staff, or designing a questionnaire, you must answer: What is this registry for?
Patient registries are not one-size-fits-all. A registry designed for disease characterization collects different data, operates under different governance, and uses different technology than one designed for clinical trial recruitment. Building without a clear purpose produces a dataset that serves no purpose well.
The six primary registry types
1. Natural History / Disease Characterization Registry
Purpose: Document the course of a disease over time, who gets it, when symptoms begin, how it progresses, what complications arise, and how participants are currently managed.
Typical data collected: Demographics, symptom onset and severity, diagnostic journey, comorbidities, current treatments, functional outcomes, quality of life measures, biosamples.
Value: Natural history studies are foundational to the entire drug development pipeline. Regulators (FDA, EMA) increasingly use patient registry natural history data as a comparator arm for single arm trials in rare diseases.
Key resource: FDA Guidance: Natural History Studies for Drug Development
2. Clinical Trial Recruitment Registry
Purpose: Identify and pre-qualify participants who may be eligible for clinical trials.
Typical data collected: Diagnosis confirmation, key inclusion/exclusion criteria (genotype, age at onset, current medications, prior treatments), willingness to participate, geographic location, treating physician contact.
Value: Rare disease trials fail most often due to insufficient enrollment. A recruitment registry can cut enrollment timelines dramatically and reduce screen failure rates.
Important distinction. A recruitment registry is not a clinical trial. It does not require IND or IDE. However, it does require IRB review if it collects identifiable information, and consent language must clearly distinguish registry participation from trial enrollment.
Key resource: NORD Rare Disease Registry Program
3. Contact / Participant Community Registry
Purpose: Maintain a list of participants and caregivers who want to stay connected to research updates, participate in surveys, or be contacted about future studies.
Typical data collected: Name, email, diagnosis, age, geographic region, expressed interests.
Value: Builds community, enables rapid surveys, and serves as a pipeline for more detailed data collection over time.
Privacy consideration: Even a simple contact registry requires privacy policy, consent, and a plan for data security. HIPAA may or may not apply depending on whether the registry operates within a covered entity.
4. EHR-Linked Registry
Purpose: Systematically extract structured clinical data from electronic health records, at scale, with minimal participant burden.
Data sources: Problem lists, medication records, lab results, imaging reports, clinical notes (via NLP), encounter data.
Value: Rich longitudinal clinical data without requiring participants to self report. Particularly powerful when linked to biobanks.
Complexity: High. Requires technical infrastructure (FHIR APIs, data harmonization), site agreements, and careful attention to the 21st Century Cures Act framework. See Module 9 and Module 10.
5. Biosample Registry
Purpose: Link participant clinical data to stored biological samples, DNA, plasma, tissue, CSF, available for future research.
Typical data collected: Sample type, storage location, collection date, associated clinical phenotype, consent scope for sample use.
Value: Enables biomarker discovery, genetic studies, and drug target identification. Samples with linked phenotype data are extraordinarily valuable to industry.
Key resource: ISBER Best Practices for Biobanking
6. Quality of Care / Outcomes Registry
Purpose: Track real world outcomes, treatment patterns, complications, health resource utilization, to improve standards of care.
Typical data collected: Treatments received, dosing, adverse events, hospitalizations, functional status over time.
Value: Informs clinical guidelines, supports comparative effectiveness research, and provides postmarket surveillance data valuable to industry partners.
Registries are not mutually exclusive
Most successful advocacy organization registries start with one purpose and expand over time. A common pathway:
Contact registry → Disease characterization → Recruitment registry → EHR-linked biosample registry
Design for expansion from the start. Even if you're starting with a simple contact registry, use standardized identifiers and data elements from the beginning. Retrofitting a contact list to become a natural history database is much harder than building in that capability from day one.
The purpose statement
Before proceeding to the next module, write a one-paragraph purpose statement for your registry. It should answer:
- Who are we collecting data about?
- What data will we collect?
- What scientific or clinical questions will this answer?
- Who will use this data and how?
- What does success look like in 5 years?
This statement will guide your IRB application, your platform selection, your data governance policy, and your partnership conversations.
Checklist
- [ ] Identified primary registry type(s)
- [ ] Written registry purpose statement
- [ ] Confirmed purpose with scientific advisory board (or identified need to form one, see Module 3)
- [ ] Identified 3 to 5 key scientific questions the registry will answer
- [ ] Mapped purpose to likely data elements (rough draft)
- [ ] Identified whether biosamples will be collected
- [ ] Determined geographic scope (single site, national, international)