Catalog

Every open problem in the collection, with its verification verdict

Use the search box to filter across ID, title, category, priority, verdict, and maturity; the column headers to sort; and the category list in the margin to narrow to one topic. Each row links to the full entry: what the problem is, why it is open, what has already been tried, and the probable solution or research direction.

Read the verdict column first. It records what happened when the claim was checked against the literature and against two independent external auditors, not what the source document asserted. A verdict is a judgment about the claim, not about the problem’s importance.

Verdict Meaning
Confirmed open Genuinely open. Auditors agreed and no counterevidence was found.
Partially addressed Real, but existing work covers part of it. The entry names which part.
Overstated The underlying difficulty is real, but a claim about its severity, or about no method existing, is stronger than the evidence supports.
Resolved since report Closed in the literature or in code, either already or since the source was written. Kept because knowing something was recently closed is as useful as knowing it is open.
Not supported No accessible source substantiates the claim as stated. Retained, with the reasons shown.
Unverifiable Auditors disagreed with evidence on both sides, or the evidence was insufficient to decide. The disagreement is shown on the entry.

Full protocol: How this was verified.

ID Problem Category Priority Verdict Maturity Priority tractability
ADJ-01 No transportability-first PAIC workflow matches MAIC’s usability Adjacent-field methods Very high Overstated Promising 1  
ADJ-02 No general-purpose transport library plugs into the NMA stack Adjacent-field methods Very high Overstated Promising 1  
ADJ-03 Efficient influence-function transport estimators for ITC networks Adjacent-field methods Very high Overstated Promising 1  
ADJ-06 Distributionally robust bounds over compatible target laws Adjacent-field methods Very high Confirmed open Promising 1  
ADJ-07 Partial identification bounds under drift or omitted modifiers Adjacent-field methods Very high Partially addressed Established 1  
ADJ-12 Routine NMA output omits graph diagnostics and value of information Adjacent-field methods Very high Overstated Promising 1  
ADJ-14 Federated ML-NMR and federated model checking do not exist Adjacent-field methods Very high Partially addressed Promising 1  
CMP-01 Cross-subnetwork component constancy is the untestable bridge Component methods (CNMA and component PAIC) Very high Confirmed open Emerging 1  
CMP-02 Adjusted component bridges need interaction constancy across the gap Component methods (CNMA and component PAIC) Very high Confirmed open Emerging 1  
CMP-03 Strict component additivity may be clinically false Component methods (CNMA and component PAIC) Very high Partially addressed Emerging 1  
CMP-05 No versioned intervention ontology maps onto CNMA component matrices Component methods (CNMA and component PAIC) Very high Overstated Emerging 1  
CMP-08 Identification beyond row-space screens in nonlinear component ML-NMR Component methods (CNMA and component PAIC) Very high Partially addressed Emerging 1  
CMP-09 cMAIC combines marginal effects through an additive component model Component methods (CNMA and component PAIC) Very high Confirmed open Promising 1  
CMP-10 Component PAIC lacks a marginal target-standardized estimand path Component methods (CNMA and component PAIC) Very high Partially addressed Promising 1  
CMP-11 Two-stage component PAIC adjusts only the IPD edges Component methods (CNMA and component PAIC) Very high Partially addressed Promising 1  
CMP-13 Shared Gamma conflates within- and between-study modification Component methods (CNMA and component PAIC) Very high Overstated Promising 1  
CMP-14 Aggregate-data-only interactions may be prior-driven Component methods (CNMA and component PAIC) Very high Partially addressed Promising 1  
CMP-17 Survival pseudo-IPD reconstruction uncertainty is not propagated Component methods (CNMA and component PAIC) Very high Partially addressed Emerging 1  
CMP-18 Proportional or time-constant component effects may fail Component methods (CNMA and component PAIC) Very high Partially addressed Promising 1  
CMP-21 Missing data and MNAR sensitivity are unaddressed in component PAIC Component methods (CNMA and component PAIC) Very high Overstated Emerging 1  
CMP-26 Unreleased cpaic has no tracked simulation benchmark or replication Component methods (CNMA and component PAIC) Very high Partially addressed Established 1  
CMP-30 No diagnostic for anchor choice in additive component models Component methods (CNMA and component PAIC) Very high Confirmed open Emerging 1 3
CMU-02 Priors, not data, can drive posteriors in weakly identified models Computation & reproducibility Very high Partially addressed Promising 1  
CMU-05 Failed fits are deleted rather than classified and reported Computation & reproducibility Very high Partially addressed Established 1  
CMU-08 No cross-package benchmark or reference pipeline for ITC software Computation & reproducibility Very high Overstated Emerging 1  
COV-01 Deciding which covariates are true effect modifiers Covariates: selection, target moments, measurement Very high Partially addressed Promising 1  
COV-07 Unknown target joint covariate distribution Covariates: selection, target moments, measurement Very high Partially addressed Promising 1 4
DEC-01 Method evaluations score estimator error, not decision error Decision analysis, regulation & reporting Very high Partially addressed Emerging 1  
DEC-02 No factorial study ranks which choice most often flips a decision Decision analysis, regulation & reporting Very high Overstated Emerging 1  
DEC-04 No PAIC-specific reporting standard or credibility appraisal tool Decision analysis, regulation & reporting Very high Partially addressed Promising 1  
DEC-05 Committees get no one-page statement of what an estimate rests on Decision analysis, regulation & reporting Very high Partially addressed Promising 1  
DEC-06 Estimand-first guidance exists but is not enforced in PAIC practice Decision analysis, regulation & reporting Very high Unverifiable Promising 1  
DEC-07 Estimand, effect scale and absolute target outcomes go undeclared Decision analysis, regulation & reporting Very high Partially addressed Promising 1  
DEC-08 No provenance label separates bridged from prior-driven contrasts Decision analysis, regulation & reporting Very high Overstated Promising 1  
DEC-09 Modifier selection and target moments reported without provenance Decision analysis, regulation & reporting Very high Partially addressed Promising 1  
DEC-10 Trimming redefines the target population without naming it Decision analysis, regulation & reporting Very high Unverifiable Promising 1  
DEC-11 Intervals omit selection, reconstruction and target-summary error Decision analysis, regulation & reporting Very high Partially addressed Promising 1  
DEC-12 Nothing requires showing how far a conclusion is assumption-driven Decision analysis, regulation & reporting Very high Overstated Promising 1  
DEC-20 No rule says what proves similarity from an indirect comparison Decision analysis, regulation & reporting Very high Confirmed open Emerging 1 4
DEC-25 No agreed rule turns an NMA posterior into a recommendation Decision analysis, regulation & reporting Very high Partially addressed Promising 1 4
DIA-01 No diagnostic can falsify unmeasured effect modification in PAIC Diagnostics & empirical validation Very high Overstated Emerging 1  
DIA-02 No agreed ESS definition and no support-geometry diagnostics Diagnostics & empirical validation Very high Partially addressed Promising 1  
DIA-03 PAIC diagnostics are not evaluated as classifiers of realized error Diagnostics & empirical validation Very high Overstated Promising 1  
DIA-04 ADEMP has no identification map for simulation estimands Diagnostics & empirical validation Very high Overstated Emerging 1  
DIA-05 No shared benchmark platform with common scenarios, seeds and truth Diagnostics & empirical validation Very high Confirmed open Promising 1 5
DIA-06 Failure signatures across PAIC estimator families are uncharacterized Diagnostics & empirical validation Very high Overstated Emerging 1  
DIA-08 PAIC simulations generate only shared, linear effect modification Diagnostics & empirical validation Very high Overstated Promising 1  
DIA-09 Narrow outcome families and hazard-ratio-only survival scoring Diagnostics & empirical validation Very high Confirmed open Promising 1  
DIA-11 No scenario library or performance measures for estimability failure Diagnostics & empirical validation Very high Partially addressed Emerging 1  
DIA-12 No study ranks competing bridges by worst-case misspecification Diagnostics & empirical validation Very high Overstated Emerging 1  
DIA-13 QBA bias mechanisms are not crossed with the primary factors Diagnostics & empirical validation Very high Confirmed open Emerging 1  
DIA-14 QBA is judged on oracle recovery, not honest classification Diagnostics & empirical validation Very high Partially addressed Emerging 1  
DIA-16 No curated public suite for bridge-deletion validation with IPD Diagnostics & empirical validation Very high Overstated Promising 1 5
DIA-17 Held-out IPD is not used to benchmark transport against truth Diagnostics & empirical validation Very high Confirmed open Promising 1  
DIS-01 Cross-gap contrasts are absent from the randomized likelihood Disconnected networks & bridges Very high Confirmed open Established 1  
DIS-02 No data-driven criterion can rank candidate bridge families Disconnected networks & bridges Very high Overstated Emerging 1  
DIS-11 Predictive criteria and population comparison cannot validate a bridge Disconnected networks & bridges Very high Partially addressed Established 1  
DIS-12 No standard ledger names the parameter carrying the gap Disconnected networks & bridges Very high Partially addressed Promising 1  
DIS-13 Bridge-ensemble weights cannot be learned for cross-gap accuracy Disconnected networks & bridges Very high Confirmed open Speculative 1 2
DIS-14 Reported intervals condition on the bridge being exactly true Disconnected networks & bridges Very high Partially addressed Promising 1  
DIS-15 Cross-gap bridge bounds are not propagated to net benefit Disconnected networks & bridges Very high Overstated Emerging 1  
DIS-16 No design method chooses which edge to add across a disconnected gap Disconnected networks & bridges Very high Overstated Emerging 1 3
EST-01 Target population is unnamed and set by data availability Estimands & target populations Very high Overstated Emerging 1  
EST-02 Anchored PAIC needs a second transport to the decision population Estimands & target populations Very high Partially addressed Emerging 1  
EST-03 Non-collapsible marginal effects are not portable across populations Estimands & target populations Very high Partially addressed Promising 1  
EST-04 Methods return incompatible marginal and conditional estimands Estimands & target populations Very high Partially addressed Promising 1  
EST-05 Nothing requires declaring whether a result is marginal or conditional Estimands & target populations Very high Overstated Promising 1  
EST-06 The MAIC paradox: swapping the IPD trial can reverse the comparison Estimands & target populations Very high Partially addressed Emerging 1  
EST-07 Reported target moments are treated as known constants Estimands & target populations Very high Partially addressed Promising 1  
EST-08 Effect scales chosen by convention rather than transportability Estimands & target populations Very high Overstated Promising 1  
EST-10 Bucher ITC and aggregate-data NMA name no target population Estimands & target populations Very high Partially addressed Established 1  
EST-15 No bound for effect-measure non-transitivity in anchored survival comparisons Estimands & target populations Very high Confirmed open Promising 1 3
EVB-02 Evidence-set assembly can favor the sponsor before modeling begins Evidence-base construction & data access Very high Overstated Emerging 1  
EVB-03 Trials combined under one outcome label are not one estimand Evidence-base construction & data access Very high Partially addressed Promising 1  
EVB-04 No decision-value method chooses which IPD or bridging trial to get Evidence-base construction & data access Very high Partially addressed Speculative 1  
HET-01 Design and conduct differences survive baseline covariate balancing Heterogeneity, consistency & multi-arm structure Very high Overstated Promising 1  
IDN-01 Transitivity and conditional constancy cannot be tested in-sample Identification, transitivity & conditional constancy Very high Partially addressed Promising 1  
IDN-02 No assumption map spans identification, transport and consistency Identification, transitivity & conditional constancy Very high Overstated Emerging 1  
IDN-03 No procedure shows an unanchored adjustment set is complete Identification, transitivity & conditional constancy Very high Partially addressed Established 1  
IDN-04 Recorded-covariate adjustment cannot remove unmeasured modifier bias Identification, transitivity & conditional constancy Very high Confirmed open Emerging 1  
IDN-05 The shared effect modifier check has low and uncalibrated power Identification, transitivity & conditional constancy Very high Overstated Promising 1  
IDN-06 ML-NMR interactions can rest solely on aggregate-data variation Identification, transitivity & conditional constancy Very high Partially addressed Promising 1  
IDN-07 No agreed criterion for when a held-out-trial discrepancy falsifies Identification, transitivity & conditional constancy Very high Unverifiable Promising 1  
IDN-08 No curated benchmark validates bridged cross-gap comparisons Identification, transitivity & conditional constancy Very high Overstated Promising 1 5
IDN-09 Negative controls for ITC lack catalogues, power and calibration Identification, transitivity & conditional constancy Very high Overstated Promising 1  
IDN-10 Transported control-arm outcomes are uncalibrated where it matters Identification, transitivity & conditional constancy Very high Partially addressed Promising 1  
IDN-14 Identification, estimation, computation and decision are conflated Identification, transitivity & conditional constancy Very high Partially addressed Established 1  
IDN-17 No criterion for discarding a present but non-exchangeable anchor Identification, transitivity & conditional constancy Very high Confirmed open Emerging 1 3
MIS-01 Missing aggregate summaries and unavailable IPD have no remedy Missing data & uncertainty propagation Very high Partially addressed Promising 1  
MIS-03 Nothing propagates sampling error in reported target moments Missing data & uncertainty propagation Very high Overstated Promising 1  
MOD-01 Conventional STC reports a conditional effect at target means Model specification, double robustness & targeted learning Very high Partially addressed Established 1  
MOD-02 Convenience functional forms and untested flexible alternatives Model specification, double robustness & targeted learning Very high Confirmed open Promising 1  
MOD-03 Ecological versus within-study effect modification Model specification, double robustness & targeted learning Very high Partially addressed Promising 1  
MOD-05 No doubly robust PAIC estimator is indexed by bias parameters Model specification, double robustness & targeted learning Very high Partially addressed Established 1  
MOD-06 No TMLE or RMST-targeted estimator for population-adjusted transport Model specification, double robustness & targeted learning Very high Overstated Promising 1  
MOD-07 No influence-function theory for mixed-granularity multi-arm networks Model specification, double robustness & targeted learning Very high Confirmed open Promising 1 2
MOD-09 No criterion says when ML-NMR should be preferred to MAIC or STC Model specification, double robustness & targeted learning Very high Partially addressed Established 1  
MOD-13 No reporting convention exposes prior-dominated contrasts in PAIC Model specification, double robustness & targeted learning Very high Partially addressed Emerging 1  
MOD-19 No valid selection criterion for cut points and powers in time-varying survival NMA Model specification, double robustness & targeted learning Very high Confirmed open Promising 1 3
OUT-01 Target-standardized marginal risks are not the HTA reporting default Outcome-specific problems Very high Overstated Established 1  
OUT-07 Informative censoring and outcome-dependent missingness Outcome-specific problems Very high Partially addressed Promising 1  
OUT-11 Non-proportional hazards and the transported hazard ratio Outcome-specific problems Very high Overstated Promising 1 3
OVL-01 Weak overlap leaves target regions the source data cannot support Positivity, overlap & extrapolation Very high Partially addressed Established 1  
OVL-02 ESS is an insufficient overlap diagnostic and no battery is agreed Positivity, overlap & extrapolation Very high Partially addressed Promising 1  
OVL-04 Weight trimming silently changes the estimand Positivity, overlap & extrapolation Very high Overstated Emerging 1  
QBA-01 General estimand-specific QBA theory for unanchored PAIC Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1 4
QBA-02 No prospectively blinded calibration program for PAIC and QBA Quantitative bias analysis for unanchored PAIC Very high Overstated Emerging 1 5
QBA-03 QBA modes and interval semantics are routinely conflated Quantitative bias analysis for unanchored PAIC Very high Partially addressed Established 1  
QBA-04 Generic unmeasured confounding collapses distinct bias mechanisms Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-06 Measurement and misclassification QBA under population adjustment Quantitative bias analysis for unanchored PAIC Very high Partially addressed Promising 1  
QBA-10 Target population and joint-distribution uncertainty Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1 4
QBA-11 Outcome-model misspecification and doubly robust QBA Quantitative bias analysis for unanchored PAIC Very high Confirmed open Promising 1 2
QBA-13 Simulated-covariate QBA is an STC route, not a general solution Quantitative bias analysis for unanchored PAIC Very high Confirmed open Promising 1  
QBA-14 Sensitivity parameters carry their own identification problem Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-15 Multiple correlated omitted variables and their dependence structure Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-16 Probabilistic QBA needs elicited and validated bias distributions Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-17 Partial identification and robust decision regions for HTA Quantitative bias analysis for unanchored PAIC Very high Confirmed open Promising 1  
QBA-18 MAIC sensitivity grids: infeasible calibration and failed fits Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-19 Latent-U Bayesian QBA hides nonidentification behind a proper prior Quantitative bias analysis for unanchored PAIC Very high Unverifiable Emerging 1  
QBA-20 Component PAIC needs joint population-and-bridge bias analysis Quantitative bias analysis for unanchored PAIC Very high Confirmed open Emerging 1 4
QBA-22 Total bias is non-additive, so one-at-a-time analyses mislead Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1 4
QBA-23 Propagating every uncertainty layer without prohibitive cost Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-24 Survival QBA beyond the hazard ratio with limited aggregate data Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1 3
QBA-25 Decision QBA: net benefit, tipping surfaces and extrapolation Quantitative bias analysis for unanchored PAIC Very high Partially addressed Emerging 1  
QBA-27 Prespecification and outcome-driven choice of sensitivity ranges Quantitative bias analysis for unanchored PAIC Very high Partially addressed Promising 1  
QBA-28 Reference QBA software and a minimum reporting standard Quantitative bias analysis for unanchored PAIC Very high Partially addressed Promising 1  
QBA-29 No agreed parameterization or format for unanchored PAIC sensitivity Quantitative bias analysis for unanchored PAIC Very high Confirmed open Promising 1  
QBA-31 Shifted-null margins are not calibrated to residual adjustment bias Quantitative bias analysis for unanchored PAIC Very high Confirmed open Emerging 1 4
SFW-01 No package unifies PAIC, transport, estimands and reporting Software, benchmarks & interface standards Very high Confirmed open Emerging 1  
SFW-02 No machine-readable specification standard for ITC analyses Software, benchmarks & interface standards Very high Confirmed open Emerging 1  
SFW-03 No minimum validation standard for ITC and PAIC packages Software, benchmarks & interface standards Very high Partially addressed Emerging 1  
SFW-04 No neutral cross-package benchmark suite for PAIC implementations Software, benchmarks & interface standards Very high Partially addressed Emerging 1 5
SFW-05 No coherent QBA software object for population-adjusted comparison Software, benchmarks & interface standards Very high Confirmed open Emerging 1  
SFW-07 multinma lacks partial IPD, missing data and a data-informed bridge Software, benchmarks & interface standards Very high Overstated Established 1  
SFW-10 maicplus variance, trimming and estimand disclosure gaps Software, benchmarks & interface standards Very high Partially addressed Established 1  
ADJ-04 Flexible outcome surfaces and support diagnostics in standardization Adjacent-field methods High Confirmed open Promising 2  
ADJ-05 Optimal transport for distributional rather than moment balance Adjacent-field methods High Partially addressed Emerging 2  
ADJ-08 Nonparametric and commensurate priors in disconnected evidence Adjacent-field methods High Overstated Promising 2  
ADJ-09 Cross-environment invariance as a screen for bridges and modifiers Adjacent-field methods High Confirmed open Promising 2  
ADJ-15 Privacy guarantees and governance for shared trial-level data Adjacent-field methods High Partially addressed Emerging 2  
ADJ-16 No ITC component vocabulary or datum-level extraction provenance Adjacent-field methods High Confirmed open Emerging 2  
CMP-04 Additivity model selection can be unstable Component methods (CNMA and component PAIC) High Partially addressed Promising 2  
CMP-06 Measurement error and component misclassification are ignored Component methods (CNMA and component PAIC) High Partially addressed Emerging 2  
CMP-07 Causal meaning of a component effect is ambiguous Component methods (CNMA and component PAIC) High Unverifiable Emerging 2  
CMP-12 No two-stage adjusted estimator carries multi-arm covariance Component methods (CNMA and component PAIC) High Partially addressed Promising 2  
CMP-15 Target joint covariate distribution is only approximately known Component methods (CNMA and component PAIC) High Unverifiable Emerging 2  
CMP-16 Unreleased cpaic shares one tau and one IPD-row residual variance Component methods (CNMA and component PAIC) High Unverifiable Promising 2  
CMP-19 Competing risks and recurrent events are absent from component PAIC Component methods (CNMA and component PAIC) High Confirmed open Emerging 2  
CMP-20 Poisson aggregate likelihood needs zero exposure-rate covariance Component methods (CNMA and component PAIC) High Unverifiable Emerging 2  
CMP-22 IPD availability may be informative rather than random Component methods (CNMA and component PAIC) High Partially addressed Emerging 2  
CMP-23 Bridge fragility ignores interaction and correlated component drift Component methods (CNMA and component PAIC) High Overstated Emerging 2  
CMP-25 Component ML-NMR integration has no parallel or adaptive scheme Component methods (CNMA and component PAIC) High Partially addressed Promising 2  
CMP-28 Component NMA has no dose-response dimension Component methods (CNMA and component PAIC) High Partially addressed Emerging 2 3
CMP-29 No component-level transitivity and coherence assessment Component methods (CNMA and component PAIC) High Confirmed open Emerging 2 3
CMP-31 No practical-identification diagnostic for component effects Component methods (CNMA and component PAIC) High Confirmed open Promising 2 4
CMP-33 No screen separates randomized from between-study component identification Component methods (CNMA and component PAIC) High Confirmed open Promising 2 5
CMP-34 No sensitivity or reporting standard for discretionary component coding Component methods (CNMA and component PAIC) High Confirmed open Emerging 2 4
CMU-03 Bayesian component ML-NMR and bridge models lack SBC validation Computation & reproducibility High Partially addressed Established 2  
CMU-04 Computational diagnostics and Monte Carlo error go unreported Computation & reproducibility High Partially addressed Established 2  
CMU-06 Simulation studies publish no registry, seeds, or environment locks Computation & reproducibility High Partially addressed Established 2  
CMU-07 Published PAIC analyses do not record the choices that drive them Computation & reproducibility High Partially addressed Emerging 2  
COV-02 No evidence on how the modifier set changes across effect scales Covariates: selection, target moments, measurement High Partially addressed Emerging 2  
COV-03 Design-dependent roles of prognostic covariates Covariates: selection, target moments, measurement High Overstated Established 2  
COV-04 Selection uncertainty in PAIC interaction shrinkage is uncalibrated Covariates: selection, target moments, measurement High Partially addressed Promising 2  
COV-05 Legitimate versus dangerous roles for flexible learners Covariates: selection, target moments, measurement High Partially addressed Promising 2  
COV-08 Entropy balancing has no network extension or variance theory Covariates: selection, target moments, measurement High Overstated Promising 2  
COV-09 Measurement non-equivalence of covariates across trials Covariates: selection, target moments, measurement High Partially addressed Promising 2  
COV-10 Effect-modifier selection does not check interaction consistency Covariates: selection, target moments, measurement High Confirmed open Promising 2 3
COV-11 Landmark external controls lack covariate inputs for population adjustment Covariates: selection, target moments, measurement High Confirmed open Emerging 2 2
COV-12 No aggregate integration method for median-based covariate summaries Covariates: selection, target moments, measurement High Confirmed open Promising 2 4
COV-13 No review method for assembling transferable effect-modifier sets Covariates: selection, target moments, measurement High Confirmed open Emerging 2 3
COV-14 Per-comparator adjustment sets yield non-comparable PAIC contrasts Covariates: selection, target moments, measurement High Confirmed open Emerging 2 4
DEC-03 Value of information is not extended to IPD or bridge acquisition Decision analysis, regulation & reporting High Overstated Emerging 2 3
DEC-13 Published PAIC analyses cannot be reconstructed from the record Decision analysis, regulation & reporting High Partially addressed Promising 2  
DEC-14 No error control for repeated looks at a living network Decision analysis, regulation & reporting High Confirmed open Promising 2 3
DEC-15 No certainty-of-evidence framework rates component effects Decision analysis, regulation & reporting High Confirmed open Emerging 2 3
DEC-16 No joint error rate for the many claims a network asserts Decision analysis, regulation & reporting High Partially addressed Promising 2 3
DEC-18 NMA rank output lacks a decision-theoretic reconciliation with net benefit Decision analysis, regulation & reporting High Partially addressed Promising 2 4
DEC-19 Evidence contribution has no definition inside a Bayesian fit Decision analysis, regulation & reporting High Confirmed open Promising 2 3
DEC-21 The effect-preservation fraction is not calibrated to indirect bias Decision analysis, regulation & reporting High Confirmed open Emerging 2 4
DEC-22 No confidence rating covers a network that mixes designs Decision analysis, regulation & reporting High Confirmed open Promising 2 4
DEC-23 The treatment hierarchy is published without a certainty rating Decision analysis, regulation & reporting High Partially addressed Promising 2 4
DEC-24 Borrowing intensity has no scale that spans sharing mechanisms Decision analysis, regulation & reporting High Partially addressed Promising 2 4
DEC-27 No rule says which of three conflicting estimates to use Decision analysis, regulation & reporting High Confirmed open Established 2 4
DEC-28 No inference is reported for an estimated optimal-treatment boundary Decision analysis, regulation & reporting High Confirmed open Emerging 2 4
DEC-29 Ranking metrics reward imprecision and no correction exists Decision analysis, regulation & reporting High Partially addressed Promising 2 4
DEC-32 Threshold analysis can perturb an estimate’s location but not its precision Decision analysis, regulation & reporting High Confirmed open Established 2 3
DEC-33 Value-of-information output is conditional on the chosen synthesis structure Decision analysis, regulation & reporting High Confirmed open Promising 2 3
DIA-07 Scenario grids omit network topology and information placement Diagnostics & empirical validation High Partially addressed Promising 2  
DIA-10 Data access is simulated as benign rather than informative Diagnostics & empirical validation High Partially addressed Promising 2  
DIA-15 Coding-error bias, rank reversal and drift calibration go unscored Diagnostics & empirical validation High Overstated Emerging 2  
DIA-18 Constancy assumptions are never tested across calendar time Diagnostics & empirical validation High Partially addressed Promising 2  
DIA-24 No reference distribution says when inconsistency is large Diagnostics & empirical validation High Partially addressed Promising 2 4
DIA-25 Rival study-contribution definitions feed the same certainty rating Diagnostics & empirical validation High Partially addressed Established 2 5
DIA-26 No validated diagnostic localizes inconsistency to a design Diagnostics & empirical validation High Partially addressed Promising 2 4
DIS-03 Baseline risk is not a neutral anchor for disconnected evidence Disconnected networks & bridges High Overstated Promising 2  
DIS-04 Dose-response bridges assume a curve that may not transport Disconnected networks & bridges High Confirmed open Promising 2  
DIS-05 Class-effect bridges extrapolate to treatments with no randomized link Disconnected networks & bridges High Confirmed open Promising 2  
DIS-06 Component additivity across a gap cannot be checked from the data Disconnected networks & bridges High Unverifiable Promising 2  
DIS-07 External-control bridges become cross-design causal synthesis Disconnected networks & bridges High Partially addressed Promising 2  
DIS-09 Structured elicitation is not applied to the cross-gap contrast Disconnected networks & bridges High Partially addressed Promising 2  
DIS-10 multinma’s baseline-synthesis bridge is an unmerged pull request Disconnected networks & bridges High Confirmed open Emerging 2  
DIS-17 Cross-outcome borrowing is a bridge with no diagnostic Disconnected networks & bridges High Confirmed open Promising 2 3
DIS-19 No population-adjusted likelihood mixes a randomized core with single-arm evidence Disconnected networks & bridges High Partially addressed Emerging 2 3
DIS-20 No aggregate-data bridge exists for disconnected survival networks Disconnected networks & bridges High Confirmed open Emerging 2 3
DIS-21 Match acceptance thresholds are uncalibrated against bridge bias Disconnected networks & bridges High Confirmed open Emerging 2 4
EST-09 No convention for reading divergence between estimand rungs Estimands & target populations High Partially addressed Emerging 2  
EST-11 No simulation crosses target choice with second-stage transport error Estimands & target populations High Overstated Emerging 2  
EST-12 Treatment hierarchies lack a declared target-population referent Estimands & target populations High Confirmed open Promising 2 4
EST-16 Synthetic-patient indirect comparisons lack a validation standard Estimands & target populations High Confirmed open Emerging 2 4
EVB-01 Selective IPD availability threatens transport across the network Evidence-base construction & data access High Partially addressed Promising 2  
EVB-05 No reporting standard covers what population adjustment needs Evidence-base construction & data access High Unverifiable Emerging 2  
EVB-06 No usable correction for overlapping participant populations across studies Evidence-base construction & data access High Confirmed open Emerging 2 3
EVB-07 NMA does not model informative trial-design selection Evidence-base construction & data access High Confirmed open Emerging 2 3
EVB-09 No common estimand or aggregate-data reconstruction for enriched and withdrawal trials Evidence-base construction & data access High Confirmed open Emerging 2 3
EVB-10 No general NMA framework for non-parallel randomized trial designs Evidence-base construction & data access High Confirmed open Emerging 2 3
EVB-11 No missing-evidence appraisal for bridged cross-gap contrasts Evidence-base construction & data access High Confirmed open Emerging 2 3
EVB-13 Platform trials lack stage-level reporting for network synthesis Evidence-base construction & data access High Partially addressed Emerging 2 4
EVB-14 Reporting bias and inconsistency cannot be disentangled in NMA Evidence-base construction & data access High Confirmed open Promising 2 3
HET-02 Multi-arm covariance is lost when contrasts are adjusted pairwise Heterogeneity, consistency & multi-arm structure High Overstated Promising 2  
HET-03 Single shared heterogeneity variance across classes and populations Heterogeneity, consistency & multi-arm structure High Partially addressed Promising 2  
HET-04 Node splitting compares paths without standardizing to a target Heterogeneity, consistency & multi-arm structure High Partially addressed Emerging 2  
HET-06 Diagnostic-driven study deletion in NMA lacks a calibrated rule Heterogeneity, consistency & multi-arm structure High Partially addressed Promising 2 4
HET-07 Heterogeneity structure is undefined for multi-parameter survival NMA Heterogeneity, consistency & multi-arm structure High Partially addressed Promising 2 3
HET-08 Inconsistency conclusions depend on an uncalibrated heterogeneity error specification Heterogeneity, consistency & multi-arm structure High Confirmed open Established 2 5
HET-10 No calibrated small-study interval method for population-adjusted NMA Heterogeneity, consistency & multi-arm structure High Partially addressed Promising 2 3
HET-11 No covariance correction for overlapping reported subgroups Heterogeneity, consistency & multi-arm structure High Confirmed open Emerging 2 4
HET-15 Composite comparator nodes are collapsed rather than modeled as reported mixtures Heterogeneity, consistency & multi-arm structure High Confirmed open Emerging 2 4
HET-17 Heterogeneity has no principled place in a sequential or living NMA boundary Heterogeneity, consistency & multi-arm structure High Confirmed open Emerging 2 3
HET-20 The design-by-treatment inconsistency test has no small-sample null Heterogeneity, consistency & multi-arm structure High Confirmed open Established 2 4
HET-21 Multivariate synthesis needs within-study correlations that publications do not report Heterogeneity, consistency & multi-arm structure High Partially addressed Promising 2 3
IDN-11 Trial-engagement effects are mistaken for prognostic imbalance Identification, transitivity & conditional constancy High Partially addressed Emerging 2  
IDN-12 Five mechanisms of cross-study effect variation are lumped together Identification, transitivity & conditional constancy High Partially addressed Emerging 2  
IDN-13 Time, geography and setting are confounded in non-crossed evidence Identification, transitivity & conditional constancy High Confirmed open Emerging 2  
IDN-15 Global inconsistency diagnostics lack calibrated operating characteristics Identification, transitivity & conditional constancy High Partially addressed Promising 2 4
IDN-18 No estimand or correction for randomized-withdrawal anchors Identification, transitivity & conditional constancy High Confirmed open Emerging 2 4
IDN-19 No principled basis for shared effect-modifier treatment classes Identification, transitivity & conditional constancy High Confirmed open Promising 2 4
MIS-02 No robustness theory for censoring-weighted survival transport Missing data & uncertainty propagation High Unverifiable Emerging 2  
MIS-04 Model and bridge selection uncertainty never enters the interval Missing data & uncertainty propagation High Partially addressed Promising 2  
MOD-04 No post-selection inference for model choice in population adjustment Model specification, double robustness & targeted learning High Confirmed open Promising 2  
MOD-08 Cross-fitting folds and rate conditions undefined for PAIC Model specification, double robustness & targeted learning High Partially addressed Promising 2 2
MOD-12 Mixture-weight calibration is unvalidated for subnetwork bridges Model specification, double robustness & targeted learning High Partially addressed Promising 2 2
MOD-14 Network meta-interpolation has no stated estimand and no implementation Model specification, double robustness & targeted learning High Partially addressed Emerging 2 4
MOD-15 No calibrated rule says when within- and across-trial interactions may be pooled Model specification, double robustness & targeted learning High Confirmed open Promising 2 4
MOD-16 No criterion chooses pooled or arm-separate matching in anchored MAIC Model specification, double robustness & targeted learning High Confirmed open Established 2 5
MOD-17 No selection-adjusted interval for the treatment a network ranks best Model specification, double robustness & targeted learning High Confirmed open Emerging 2 3
MOD-20 No criterion or error-based evidence selects arm-synthesis over contrast-synthesis Model specification, double robustness & targeted learning High Confirmed open Established 2 4
OUT-02 Rare events, separation, and zero-event studies in weighted synthesis Outcome-specific problems High Partially addressed Promising 2  
OUT-03 Outcome misclassification in binary indirect comparisons Outcome-specific problems High Partially addressed Promising 2  
OUT-04 Different instruments and standardized mean difference transport Outcome-specific problems High Partially addressed Promising 2  
OUT-08 Overdispersed counts and recurrent events as total disease burden Outcome-specific problems High Unverifiable Emerging 2 2
OUT-09 Endogenous exposure time and target-standardized rate contrasts Outcome-specific problems High Partially addressed Emerging 2  
OUT-10 Proportional odds, category probabilities, and ordinal utilities Outcome-specific problems High Overstated Emerging 2  
OUT-12 Survival extrapolation is unidentified beyond the observed follow-up Outcome-specific problems High Partially addressed Promising 2 3
OUT-13 Reconstruction uncertainty from digitized Kaplan-Meier curves Outcome-specific problems High Partially addressed Promising 2 3
OUT-14 Pseudo-IPD cannot express delayed entry or assessment windows Outcome-specific problems High Overstated Emerging 2  
OUT-15 Treatment switching and subsequent therapies in survival PAIC Outcome-specific problems High Confirmed open Emerging 2  
OUT-16 Competing risks and cumulative incidence transport Outcome-specific problems High Confirmed open Emerging 2 2
OUT-17 Transition-specific component effects in multistate PAIC Outcome-specific problems High Confirmed open Speculative 2 2
OUT-18 Component-specific time-varying treatment effects Outcome-specific problems High Partially addressed Emerging 2 3
OUT-19 No multivariate population-adjusted model for correlated outcomes Outcome-specific problems High Overstated Promising 2  
OUT-20 No PAIC workflow carries dependence through transport to net benefit Outcome-specific problems High Partially addressed Emerging 2  
OUT-22 No consistency test for vector-valued effects in survival NMA Outcome-specific problems High Partially addressed Promising 2 3
OUT-23 No estimand aligns outcomes reported on different assessment schedules Outcome-specific problems High Partially addressed Emerging 2 3
OUT-24 No rule selects the common RMST horizon across unequal follow-up Outcome-specific problems High Partially addressed Promising 2 4
OUT-25 No treatment-invariance check for summed rating-scale outcomes Outcome-specific problems High Confirmed open Emerging 2 3
OUT-26 No criterion says when a network supports surrogate-based prediction Outcome-specific problems High Partially addressed Promising 2 3
OVL-03 No support diagnostic aggregates over a PAIC target distribution Positivity, overlap & extrapolation High Partially addressed Emerging 2  
OVL-05 Optimal transport for distributional balancing in PAIC Positivity, overlap & extrapolation High Confirmed open Promising 2  
OVL-06 No standardized sample-size planning guidance or software for PAIC Positivity, overlap & extrapolation High Confirmed open Emerging 2  
QBA-05 Transport bias functions are not adapted to aggregate-data PAIC Quantitative bias analysis for unanchored PAIC High Confirmed open Promising 2  
QBA-07 Outcome ascertainment and the observation process in external data Quantitative bias analysis for unanchored PAIC High Unverifiable Promising 2  
QBA-08 Index-date, immortal time and delayed entry in external comparators Quantitative bias analysis for unanchored PAIC High Partially addressed Promising 2  
QBA-09 No sensitivity parameters index departures from censoring at random Quantitative bias analysis for unanchored PAIC High Overstated Emerging 2  
QBA-12 Treatment version, care context and subsequent-therapy drift Quantitative bias analysis for unanchored PAIC High Confirmed open Emerging 2  
QBA-21 Unreleased cpaic has no QBA interface beyond box-bounded drift Quantitative bias analysis for unanchored PAIC High Partially addressed Promising 2  
QBA-26 Benchmarking and negative controls as data-based calibration Quantitative bias analysis for unanchored PAIC High Partially addressed Promising 2  
QBA-30 Added bias uncertainty can narrow rather than widen a pooled interval Quantitative bias analysis for unanchored PAIC High Confirmed open Promising 2 4
QBA-32 No criterion selects among design-aware models for a mixed-design network Quantitative bias analysis for unanchored PAIC High Partially addressed Promising 2 4
SFW-06 ML-NMR runtime can constrain repeated sensitivity analyses at scale Software, benchmarks & interface standards High Unverifiable Established 2  
SFW-08 No agreed leave-one-out unit for mixed IPD and aggregate likelihoods Software, benchmarks & interface standards High Partially addressed Established 2  
SFW-09 maicplus supports only binary and time-to-event endpoints Software, benchmarks & interface standards High Confirmed open Established 2  
SFW-12 outstandR is pairwise only with no network or disconnected validation Software, benchmarks & interface standards High Confirmed open Promising 2  
SFW-13 cpaic and mlumr are unreleased and independently unvalidated Software, benchmarks & interface standards High Partially addressed Emerging 2  
SFW-14 drMAIC has no methods paper and no independent evaluation Software, benchmarks & interface standards High Confirmed open Emerging 2  
SFW-17 No end-to-end error propagation study for automated synthesis pipelines Software, benchmarks & interface standards High Confirmed open Emerging 2 4
ADJ-10 Data-driven modifier and causal-structure discovery with few trials Adjacent-field methods Medium-high Partially addressed Emerging 3  
ADJ-18 Simulation-based inference for mechanistic and pharmacometric bridges Adjacent-field methods Medium-high Confirmed open Emerging 3  
CMP-27 Component models cannot represent delivery order or sequencing Component methods (CNMA and component PAIC) Medium-high Confirmed open Emerging 3 3
CMP-32 No sample-size or allocation method for component NMA Component methods (CNMA and component PAIC) Medium-high Confirmed open Emerging 3 4
CMP-35 No unique definition of evidence contribution for component estimates Component methods (CNMA and component PAIC) Medium-high Confirmed open Emerging 3 4
CMU-01 ML-NMR integration cost is unaddressed for flexible survival models Computation & reproducibility Medium-high Overstated Promising 3  
COV-06 No causal forest aggregates to a target known only by marginals Covariates: selection, target moments, measurement Medium-high Partially addressed Emerging 3  
DEC-17 Evidence-provenance measures do not propagate heterogeneity uncertainty Decision analysis, regulation & reporting Medium-high Confirmed open Promising 3 4
DEC-26 Nothing predicts whether a network will beat separate comparisons Decision analysis, regulation & reporting Medium-high Confirmed open Emerging 3 4
DEC-30 ROB-MEN reliability is unmeasured and CINeMA agreement collapses on large networks Decision analysis, regulation & reporting Medium-high Partially addressed Established 3 4
DEC-31 Evaluating a treatment strategy can disconnect the network it is evaluated on Decision analysis, regulation & reporting Medium-high Confirmed open Emerging 3 3
DEC-34 Network meta-analysis displays are tested for acceptability, not for accuracy Decision analysis, regulation & reporting Medium-high Partially addressed Emerging 3 4
DIA-19 NMA outlier and influence screens have no calibrated null Diagnostics & empirical validation Medium-high Confirmed open Promising 3 4
DIA-20 Transitivity dissimilarity screens have no calibrated threshold Diagnostics & empirical validation Medium-high Confirmed open Emerging 3 4
DIA-21 Competing contribution measures disagree and none is agreed correct Diagnostics & empirical validation Medium-high Confirmed open Established 3 3
DIA-22 Empirical method comparisons exclude the networks where methods diverge Diagnostics & empirical validation Medium-high Confirmed open Established 3 4
DIA-23 No convention requires network simulations to score basic parameters Diagnostics & empirical validation Medium-high Confirmed open Established 3 5
DIS-08 Mechanistic bridges import structural model error Disconnected networks & bridges Medium-high Confirmed open Speculative 3  
DIS-18 Unequal dose coverage moves a dose-response bridge one way Disconnected networks & bridges Medium-high Confirmed open Promising 3 4
EST-13 Network intervals condition on plugged-in variance components Estimands & target populations Medium-high Partially addressed Promising 3 4
EST-14 Model-performance NMA lacks a target case-mix estimand Estimands & target populations Medium-high Confirmed open Emerging 3 4
EVB-08 Network small-study-effects diagnostics lack an assumption-light default Evidence-base construction & data access Medium-high Confirmed open Established 3 4
EVB-12 No selective-inference correction for outcome-triggered new trials pooled into NMA Evidence-base construction & data access Medium-high Confirmed open Emerging 3 3
EVB-15 Selective non-publication of commissioned indirect comparisons is unmeasured Evidence-base construction & data access Medium-high Confirmed open Emerging 3 3
HET-05 Additive and multiplicative heterogeneity models have not been compared in NMA Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Promising 3 5
HET-09 Multivariate NMA inconsistency covariance is unidentifiable in sparse networks Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Emerging 3 3
HET-12 No empirical prior exists for residual heterogeneity after covariate adjustment Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Promising 3 4
HET-13 No corpus-derived prior distributions exist for the NMA inconsistency variance Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Promising 3 3
HET-14 No evidence base or reporting rule governs the heterogeneity versus inconsistency split Heterogeneity, consistency & multi-arm structure Medium-high Partially addressed Established 3 4
HET-16 Exchangeable and comparison-specific heterogeneity priors are blocked by the covariance constraint Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Promising 3 3
HET-18 Arm-based between-treatment correlations are imposed as exchangeable with no way to test or vary the structure Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Promising 3 3
HET-19 Node splitting reports one arbitrary member of a family of multi-arm parameterizations Heterogeneity, consistency & multi-arm structure Medium-high Confirmed open Established 3 4
IDN-16 Covariate-indexed consistency has no calibrated licensed region Identification, transitivity & conditional constancy Medium-high Confirmed open Emerging 3 3
IDN-20 No synthesis framework for treatment-specific eligibility patterns Identification, transitivity & conditional constancy Medium-high Confirmed open Emerging 3 3
MIS-05 Missing-outcome methods do not cover NMA inconsistency and rankings Missing data & uncertainty propagation Medium-high Partially addressed Promising 3 4
MOD-10 Flexible interaction surfaces inside network evidence synthesis Model specification, double robustness & targeted learning Medium-high Confirmed open Promising 3  
MOD-11 No mixture-over-populations model transports to a declared target Model specification, double robustness & targeted learning Medium-high Partially addressed Emerging 3  
MOD-18 Risk-modeling network meta-regression treats a fitted risk score as observed Model specification, double robustness & targeted learning Medium-high Confirmed open Emerging 3 3
OUT-05 Unequal variances, skew, and bounded scales in continuous outcomes Outcome-specific problems Medium-high Confirmed open Promising 3  
OUT-06 Change scores, trajectories, and responder-threshold estimands Outcome-specific problems Medium-high Overstated Promising 3  
OUT-21 Absolute effects across a network share no baseline risk or time horizon Outcome-specific problems Medium-high Partially addressed Established 3 5
SFW-15 netmeta and gemtc carry no target-population representation Software, benchmarks & interface standards Medium-high Confirmed open Established 3  
SFW-16 MBNMAdose and crossnma bridges are unidentified across a disconnection Software, benchmarks & interface standards Medium-high Overstated Established 3  
ADJ-11 No proximal ITC method works with an aggregate-only comparator Adjacent-field methods Medium Partially addressed Speculative 4  
ADJ-13 Topology-only graph neural network output cannot identify effects Adjacent-field methods Medium Confirmed open Speculative 4  
ADJ-17 Representation learning for low-dimensional population encodings Adjacent-field methods Medium Confirmed open Speculative 4  
ADJ-19 Shapley attribution of regimen effects to treatment components Adjacent-field methods Medium Confirmed open Emerging 4  
ADJ-20 Reinforcement learning and dynamic regimes for adaptive sequences Adjacent-field methods Medium Partially addressed Speculative 4  
CMP-24 Edge influence is approximate for multi-arm and random effects Component methods (CNMA and component PAIC) Medium Partially addressed Promising 4  
SFW-11 Archived MAIC packages leave no maintained CRAN installation path Software, benchmarks & interface standards Medium Overstated Established 4  
No matching items