Paper
The regulated-looking subset is cleaner than the raw registry, but it is still far from clean.
How large is the likely U.S.-nexus reporting debt once older CT.gov studies are filtered through strict and broad ACT-style sensitivity layers? We analysed 249,507 eligible older closed interventional studies from the March 29, 2026 full-registry snapshot. We created three proxy layers using recorded U.S. locations, intervention families, and phase exclusions within older closed interventional records: broad U.S. nexus, drug/bio non-phase1, and strict drug/bio/device U.S. nexus. The strict proxy contains 58,598 older studies, with 18,475 missing-results studies and a 31.5 percent no-results rate. Debt is still old inside that supposedly regulated core: mean unresolved time is 10.92 years beyond the two-year mark, and Pre-FDAAA strict-proxy cohorts remain 87.0 percent unresolved even after U.S.-nexus filtering. The regulatory backlog therefore sits far below the raw all-study rate but remains large, old, and institutionally distributed across OTHER, INDUSTRY, and government-linked slices. These layers are conservative proxies built from registry-visible fields, not formal ACT or FDAAA legal determinations or enforceability judgments.