{"entity":{"id":"idea-data-gcp-for-real-world-data","kind":"idea","name":"Good practice standards and inspection for real-world data sources","aka":[],"tldr":"Trials are inspected to check the data are real and traceable. Do the same for the hospital databases used to make regulatory decisions.","summary":"Clinical trial data are governed by Good Clinical Practice and inspected; real-world data sources used in regulatory submissions are not. The proposal is a 'GCP for real-world data': standards for provenance, traceability to source records, curation documentation, fitness-for-purpose assessment and inspection rights, with accreditation of data sources. The FDA's RWE framework and EMA's DARWIN EU data quality framework are starting points.","asOf":"2026-09-08","links":[{"label":"EMA DARWIN EU","url":"https://www.darwin-eu.org/"},{"label":"FDA Real-World Evidence framework","url":"https://www.fda.gov/science-research/science-and-research-special-topics/real-world-evidence"}],"tags":[],"related":[],"cancers":[],"sections":["ai-computation"],"technologies":[],"targets":[],"drugs":[],"companies":[],"institutions":[],"pathways":[],"terms":["real-world-evidence"],"trials":[],"people":[],"bottlenecks":["b-real-world-evidence","b-regulatory-fragmentation"],"keyPapers":[],"journals":[],"dependsOn":[],"notes":[],"hypothesis":"Accredited data sources will produce RWE that regulators accept as supportive evidence in a substantially higher share of submissions, and audit will reveal material provenance errors in a minority of current sources.","rationale":"Regulators discount RWE partly because they cannot verify it; accreditation with inspection is how the same problem was solved for trials and for laboratories (CLIA, ISO 15189).","test":"Pilot accreditation with five data sources (two commercial, two academic, one registry); publish audit findings and regulator uptake of RWE from accredited versus non-accredited sources over two years.","maturity":"early-clinical","actor":"regulator","cost":"medium","horizonYears":3},"route":"/ideas/idea-data-gcp-for-real-world-data/","neighbours":{"section":[{"id":"ai-computation","kind":"section","name":"AI & Computation","route":"/fronts/ai-computation/"}],"term":[{"id":"real-world-evidence","kind":"term","name":"Real-world evidence","route":"/terms/real-world-evidence/"}],"bottleneck":[{"id":"b-regulatory-fragmentation","kind":"bottleneck","name":"Regulatory divergence between regions","route":"/bottlenecks/b-regulatory-fragmentation/"},{"id":"b-real-world-evidence","kind":"bottleneck","name":"Weak real-world evidence and registries","route":"/bottlenecks/b-real-world-evidence/"}]}}