Cohort Identification

Sources and review notes

These links support definition-level research and do not establish the regulatory status, safety, or suitability of any product.

The NIH All of Us Researcher Workbench's Data Explorer uses explicit inclusion and exclusion criteria to build cohorts and supports reviewing a subset to validate those criteria. FDA's final guidance on EHR and medical claims real-world data describes study-population definitions, data relevance and reliability, and quality assurance considerations for a specific regulatory research context. HHS's HIPAA research guidance explains that PHI use or disclosure for research, preparatory review, and recruitment depends on authorization or limited regulatory pathways and safeguards. These sources support transparent criteria, fit-for-purpose data review, validation, and privacy governance but do not validate a cohort-identification product, guarantee complete or unbiased membership, or authorize clinical action, outreach, marketing, reporting, or research use. Teams must define the purpose, population, time window, data sources, terminology and code versions, missingness, patient matching, fairness checks, manual review, consent or legal basis, IRB or Privacy Board needs, and correction process for each cohort.

• Source 1: support.researchallofus.org/hc/en-us/articles/29767527455124-Cohort-Builder-and-Dataset-Builder
• Source 2: www.fda.gov/regulatory-information/search-fda-guidance-documents/real-world-data-assessing-electronic-health-records-and-medical-claims-data-support-regulatory
• Source 3: www.hhs.gov/hipaa/for-professionals/special-topics/research/index.html