Paper
A CT.gov record with no linked paper is not always an unpublished study. Often the paper trail exists, but the registry page does not show it.
How much of CT.gov publication-link missingness survives when no-link studies are re-audited with broader exact-ID search beyond the earlier PubMed-only pass? We re-audited the existing 1,050-study sponsor-class-stratified sample of older no-link studies drawn from 140,363 eligible no-link records in the March 29, 2026 registry snapshot. We compared PubMed exact-ID matches with Europe PMC exact-ID matches, separating total rescue, non-MED rescue, and weighted publication-only visibility. Weighted PubMed exact-ID matching captured only 1.2 percent of no-link records, but Europe PMC exact-ID rescue added 39.6 points, lifting weighted any-match visibility to 40.8 percent. Non-MED rescue was much smaller at 2.8 points, while weighted external-publication-only visibility still reached 29.1 percent. Publication-link missingness therefore combines real silence with a much larger indexing and linkage gap than the PubMed-only audit suggested, especially in NIH-linked and network-linked no-link portfolios. The audit is sample-based, exact-ID only, and cannot adjudicate whether every retrieved paper fully reports the registered study or whether links were added later.