[
  {
    "from": "prereg:f476c2c1bb6f1121fbc97b1b248bfd6a6d91edd7049fe775be1ef3c3632e7207",
    "kind": "unstated",
    "claims": [
      "C1"
    ],
    "detail": "After source tracing identified two confirmed endpoint substitutions, the controlled main sensitivity replaced only those rows. The Evolut observed-event discrepancy was held fixed there and examined by omission or observed-count substitution only as separate arithmetic sensitivities; incomplete follow-up prevents a corrected survival effect from simple observed counts."
  },
  {
    "from": "prereg:f476c2c1bb6f1121fbc97b1b248bfd6a6d91edd7049fe775be1ef3c3632e7207",
    "kind": "changed",
    "claims": [
      "C1"
    ],
    "detail": "The plan sought source verification across the complete trial scope. All seven programs were inspected, but some exact current-database one-year PARTNER counts and the later NOTION union count could not be resolved from accessible sources, and NOTION-2 prints one-year risks rather than exact event counts. These inputs are retained and explicitly unresolved, rather than claiming a fully corrected clinical meta-analysis."
  },
  {
    "from": "prereg:f476c2c1bb6f1121fbc97b1b248bfd6a6d91edd7049fe775be1ef3c3632e7207",
    "kind": "unstated",
    "claims": [
      "C1"
    ],
    "detail": "R 4.5.1/metafor 5.2-1 was used after the Python implementation for an independent-software author check. Its result was archived with the input digest and is compared in default reproduction; a separate pinned R environment and script regenerate the reference. This is not independent external review."
  }
]
