Portfolio pools — every R-validated meta-analysis in this corpus

The endpoint this number measures is not recorded anywhere we hold, so a value above 1 cannot be read as benefit or as harm. The trials have endpoints; we did not record them. This is our recording gap, not a property of the evidence. -- outputs/r_validation/AGYW_HIV_PREP.json

The store refused to pool this and recorded why: Nothing is pooled: one trial. This is not a withheld estimate -- the value stands and is CLEAR Outcomes' own. -- ssot/bempedoic-acid-review/bempedoic-acid-review.json

The store refused to pool this and recorded why: Nothing is pooled: one trial. This is not a withheld estimate -- the value stands and is CLEAR Outcomes' own. -- ssot/bempedoic-acid-review/bempedoic-acid-review.json

The endpoint this number measures is not recorded anywhere we hold, so a value above 1 cannot be read as benefit or as harm. The trials have endpoints; we did not record them. This is our recording gap, not a property of the evidence. -- outputs/r_validation/CAB_PREP_HIV.json

The store refused to pool this and recorded why: THE NUMERATORS AND THE DENOMINATORS ON THIS PAGE COME FROM DIFFERENT OUTCOMES, ON ALL THREE TRIALS. Each row carried the primary composite's participant denominators -- 3,889 and 3,865; 2,654 and 2,641; 5,470 and 5,469, all exactly right -- against event counts of 8 and 5, 6 and 18, and 18 and 18. Those counts are NOT the composite's. They are ALL-CAUSE MORTALITY, a SECONDARY outcome on every one of the three: CHAMPION-PCI's 'Individual Incidence of All-cause Mortality' is 8 against 5, CHAMPION-PLATFORM's 'Incidence of All-cause Mortality' is 6 against 18, and CHAMPION-PHOENIX's 18 against 18 is the DEATH row of its stent-thrombosis/death/MI breakdown. THE REGISTRY'S OWN PRIMARY-COMPOSITE COUNTS ARE 290 AND 276, 185 AND 210, AND 257 AND 322 -- between one and two ORDERS OF MAGNITUDE larger than what the page used. A mortality numerator over a composite denominator is not a small error in a rate; it is a quantity that corresponds to nothing. AND THE CORRECTION REVERSES THE CONCLUSION, WHICH IS WHY THIS MATTERS MORE THAN A TIDINESS COMPLAINT. Pooling the registry's own primary counts by DerSimonian-Laird gives OR 0.8955 (0.7526 to 1.0656), I-squared 64 per cent -- AN INTERVAL THAT CROSSES NO EFFECT. The card published OR 0.81 (0.71 to 0.91), which EXCLUDES it. The page reported a significant benefit where the trials' own primary outcomes do not establish one. SEPARATELY AND SUFFICIENT ON ITS OWN: THE COMPOSITES ARE NOT THE SAME. CHAMPION-PHOENIX's registered primary is 'The Composite Incidence of All-cause Mortality, Myocardial Infarction (MI), Ischemia-driven Revascularization (IDR) AND STENT THROMBOSIS (ST)' -- four components. CHAMPION-PCI's and CHAMPION-PLATFORM's are mortality, MI and IDR: three. One whole component is present in one trial and absent from the other two, so even with correct counts these three would not share one estimand. -- ssot/cangrelor-pci-review/cangrelor-pci-review.json

The store refused to pool this and recorded why: THE NUMERATORS AND THE DENOMINATORS ON THIS PAGE COME FROM DIFFERENT OUTCOMES, ON ALL THREE TRIALS. Each row carried the primary composite's participant denominators -- 3,889 and 3,865; 2,654 and 2,641; 5,470 and 5,469, all exactly right -- against event counts of 8 and 5, 6 and 18, and 18 and 18. Those counts are NOT the composite's. They are ALL-CAUSE MORTALITY, a SECONDARY outcome on every one of the three: CHAMPION-PCI's 'Individual Incidence of All-cause Mortality' is 8 against 5, CHAMPION-PLATFORM's 'Incidence of All-cause Mortality' is 6 against 18, and CHAMPION-PHOENIX's 18 against 18 is the DEATH row of its stent-thrombosis/death/MI breakdown. THE REGISTRY'S OWN PRIMARY-COMPOSITE COUNTS ARE 290 AND 276, 185 AND 210, AND 257 AND 322 -- between one and two ORDERS OF MAGNITUDE larger than what the page used. A mortality numerator over a composite denominator is not a small error in a rate; it is a quantity that corresponds to nothing. AND THE CORRECTION REVERSES THE CONCLUSION, WHICH IS WHY THIS MATTERS MORE THAN A TIDINESS COMPLAINT. Pooling the registry's own primary counts by DerSimonian-Laird gives OR 0.8955 (0.7526 to 1.0656), I-squared 64 per cent -- AN INTERVAL THAT CROSSES NO EFFECT. The card published OR 0.81 (0.71 to 0.91), which EXCLUDES it. The page reported a significant benefit where the trials' own primary outcomes do not establish one. SEPARATELY AND SUFFICIENT ON ITS OWN: THE COMPOSITES ARE NOT THE SAME. CHAMPION-PHOENIX's registered primary is 'The Composite Incidence of All-cause Mortality, Myocardial Infarction (MI), Ischemia-driven Revascularization (IDR) AND STENT THROMBOSIS (ST)' -- four components. CHAMPION-PCI's and CHAMPION-PLATFORM's are mortality, MI and IDR: three. One whole component is present in one trial and absent from the other two, so even with correct counts these three would not share one estimand. -- ssot/cangrelor-pci-review/cangrelor-pci-review.json

The store refused to pool this and recorded why: TWO OF THE THREE TRIALS ON THIS PAGE REGISTER ONLY CONTINUOUS PRIMARY OUTCOMES, AND THIS PAGE POOLED EVENT COUNTS. Verbatim: STEP-HFpEF and STEP-HFpEF DM each register two co-primaries, 'Change in Kansas City Cardiomyopathy Questionnaire Clinical Summary Score' and 'Change in Body Weight'. Both are CONTINUOUS CHANGE SCORES. Neither is an event count and neither can produce a 2x2 table. The counts stored for STEP-HFpEF DM, 7 of 310 against 18 of 306, CORRESPOND TO NO REGISTERED PRIMARY OUTCOME OF THAT TRIAL. SUMMIT is different and is the only row here with a defensible source: it registers a continuous KCCQ primary AND 'First Occurrence of the Composite Endpoint of Heart Failure (HF) Outcomes', and its stored 29 of 364 against 52 of 367 do correspond to that event co-primary. So the pool joins one trial's registered event composite to another trial's unregistered counts. SEPARATELY: THE OUTCOME NAME AND THE MEASURE DESCRIBE DIFFERENT QUANTITIES. The outcome is named 'Change in Kansas City Cardiomyopathy Questionnaire Clinical Summary Score' and the stored measure is an ODDS RATIO. A change score has no odds. SEPARATELY AGAIN, AND IT IS WHY THIS TOPIC WAS INVISIBLE: THIS PAGE'S CARD WAS INVISIBLE TO OUR OWN TOOLING. The card existed and published 'HR 0.41 (0.22 to 0.79), k=3' -- a measure, a value and a trial count that all differ from the object's OR 0.4846 (0.3178 to 0.7389) at k=2. It was reported as having NO CARD because two scripts matched card links with the pattern [A-Z0-9_]+\.html, which cannot match the lowercase 'p' in HFpEF. A page invisible to the tool that counts pages is worse than an uncounted one: it is counted as a different thing, and this card escaped card-alignment checking entirely for as long as those tools have run. -- ssot/incretin-hfpef-review/incretin-hfpef-review.json

This review's published estimate is RATE_RATIO 0.8066 (0.6856-0.9488) k=2. The odds ratio previously shown here was computed over a different trial set and is not a re-expression of it, so two surfaces were reporting different quantities with neither naming the other. The odds ratio is withheld rather than deleted: it is kept in the object as withheld_pool. -- withheld OR was OR 0.657 k=4

This review's published estimate is HR 0.7636 (0.7062-0.8258) k=3. The odds ratio previously shown here was computed over a different trial set and is not a re-expression of it, so two surfaces were reporting different quantities with neither naming the other. The odds ratio is withheld rather than deleted: it is kept in the object as withheld_pool. -- withheld OR was OR 0.709 k=5

This review's published estimate is HR 0.7171 (0.6246-0.8234) k=2. The odds ratio previously shown here was computed over a different trial set and is not a re-expression of it, so two surfaces were reporting different quantities with neither naming the other. The odds ratio is withheld rather than deleted: it is kept in the object as withheld_pool. -- withheld OR was OR 0.602 k=2

This review's published estimate is HR 0.7171 (0.6246-0.8234) k=2. The odds ratio previously shown here was computed over a different trial set and is not a re-expression of it, so two surfaces were reporting different quantities with neither naming the other. The odds ratio is withheld rather than deleted: it is kept in the object as withheld_pool. -- withheld OR was OR 0.602 k=2