Paper
Condition families do not all hide the same way. Some dominate on adjusted stock, while others stand out because they are especially ghosted or black-boxed.
Which condition families remain worst once excess hiddenness is measured inside broad therapeutic portfolios rather than single sponsor tables? We analysed 249,507 eligible older closed interventional studies from the March 29, 2026 full-registry snapshot using one condition-family label per study. We ranked condition families by adjusted no-results excess, adjusted ghost excess, black-box stock, and strict-core carryover using the same study-mix adjustment as wave eight. Oncology carried the largest adjusted excess no-results stock at 543 studies, followed by cardiovascular at 373 and metabolic at 251. Healthy volunteers were different: near expected on no-results, yet 1,032 studies above expectation on ghost protocols and a 33.9 percent black-box rate. Condition families therefore split into stock-heavy disease backlogs and a separate healthy-volunteer silence pattern that is much more ghosted than merely overdue inside the same older-study registry universe overall. Condition families are keyword-derived registry groupings, so they approximate therapeutic portfolios rather than adjudicated disease ontologies or mutually exclusive clinical domains.