Decision analysis, regulation & reporting

34 open problems

Bias and RMSE do not measure wrong reimbursement decisions. Linking statistical performance to the decision it feeds.

ID Problem Priority Verdict Maturity prank
DEC-01 Method evaluations score estimator error, not decision error Very high Partially addressed Emerging 1
DEC-02 No factorial study ranks which choice most often flips a decision Very high Overstated Emerging 1
DEC-04 No PAIC-specific reporting standard or credibility appraisal tool Very high Partially addressed Promising 1
DEC-05 Committees get no one-page statement of what an estimate rests on Very high Partially addressed Promising 1
DEC-06 Estimand-first guidance exists but is not enforced in PAIC practice Very high Unverifiable Promising 1
DEC-07 Estimand, effect scale and absolute target outcomes go undeclared Very high Partially addressed Promising 1
DEC-08 No provenance label separates bridged from prior-driven contrasts Very high Overstated Promising 1
DEC-09 Modifier selection and target moments reported without provenance Very high Partially addressed Promising 1
DEC-10 Trimming redefines the target population without naming it Very high Unverifiable Promising 1
DEC-11 Intervals omit selection, reconstruction and target-summary error Very high Partially addressed Promising 1
DEC-12 Nothing requires showing how far a conclusion is assumption-driven Very high Overstated Promising 1
DEC-20 No rule says what proves similarity from an indirect comparison Very high Confirmed open Emerging 1
DEC-25 No agreed rule turns an NMA posterior into a recommendation Very high Partially addressed Promising 1
DEC-03 Value of information is not extended to IPD or bridge acquisition High Overstated Emerging 2
DEC-13 Published PAIC analyses cannot be reconstructed from the record High Partially addressed Promising 2
DEC-14 No error control for repeated looks at a living network High Confirmed open Promising 2
DEC-15 No certainty-of-evidence framework rates component effects High Confirmed open Emerging 2
DEC-16 No joint error rate for the many claims a network asserts High Partially addressed Promising 2
DEC-18 NMA rank output lacks a decision-theoretic reconciliation with net benefit High Partially addressed Promising 2
DEC-19 Evidence contribution has no definition inside a Bayesian fit High Confirmed open Promising 2
DEC-21 The effect-preservation fraction is not calibrated to indirect bias High Confirmed open Emerging 2
DEC-22 No confidence rating covers a network that mixes designs High Confirmed open Promising 2
DEC-23 The treatment hierarchy is published without a certainty rating High Partially addressed Promising 2
DEC-24 Borrowing intensity has no scale that spans sharing mechanisms High Partially addressed Promising 2
DEC-27 No rule says which of three conflicting estimates to use High Confirmed open Established 2
DEC-28 No inference is reported for an estimated optimal-treatment boundary High Confirmed open Emerging 2
DEC-29 Ranking metrics reward imprecision and no correction exists High Partially addressed Promising 2
DEC-32 Threshold analysis can perturb an estimate’s location but not its precision High Confirmed open Established 2
DEC-33 Value-of-information output is conditional on the chosen synthesis structure High Confirmed open Promising 2
DEC-17 Evidence-provenance measures do not propagate heterogeneity uncertainty Medium-high Confirmed open Promising 3
DEC-26 Nothing predicts whether a network will beat separate comparisons Medium-high Confirmed open Emerging 3
DEC-30 ROB-MEN reliability is unmeasured and CINeMA agreement collapses on large networks Medium-high Partially addressed Established 3
DEC-31 Evaluating a treatment strategy can disconnect the network it is evaluated on Medium-high Confirmed open Emerging 3
DEC-34 Network meta-analysis displays are tested for acceptability, not for accuracy Medium-high Partially addressed Emerging 3
No matching items

Back to the full catalog