Universal Transplant Causal Mortality Analysis

Joint Causal Effect — All-Cause Mortality Heart · 65M · 5yr

1.000

Causal HR

P(Y|do(X))

0.0%

Risk Reduction

1 − joint HR

—%

5-yr Mortality

Baseline × HR

0.0%

% of Max RR

Pareto efficiency

95% CI (HR):

——

Select interventions

SD: — 95% CI RR: — Compliance adj: 100% Active N: 0

Residual risk vs baseline 100%

Evidence Table — Active Organ Protocol HR · CI · E-value · Relevance

Acronyms: HR=Hazard Ratio · CI=Confidence Interval · E-val=E-value · CAV=Cardiac Allograft Vasculopathy · CNI=Calcineurin Inhibitor · MMF=Mycophenolate Mofetil · mTOR=Mammalian Target of Rapamycin · CMV=Cytomegalovirus · DSA=Donor-Specific Antibody · SGLT2=Sodium-Glucose Cotransporter-2 · PCP=Pneumocystis jirovecii Pneumonia · NODAT=New-Onset Diabetes After Transplant · BOS=Bronchiolitis Obliterans Syndrome · HCC=Hepatocellular Carcinoma · AMR=Antibody-Mediated Rejection

#	Abbrev	Intervention	Cat.	HR (95%CI)	E-val	RR% (95%CI)	Relevance	Pareto	Active

Pareto Analysis — Cumulative Risk Reduction Greedy Forward Selection · Organ-Adjusted

Greedy algorithm adds the highest marginal-gain intervention at each step using organ-relevance-adjusted HRs. Max achievable HR = — (RR = —). N80 = — interventions · N99 = — interventions.

Rank	Abbrev	Eff. HR	Marginal	Cumul. HR	Cumul. RR%	% of Max	5-yr Mort.	Zone

Curedo / Protocol Adherence Analysis Dose-Response · Compliance Curve

The compliance slider scales all active intervention dose intensities proportionally, modelling partial adherence. Curedo analysis identifies the adherence threshold above which additional compliance gains diminish — analogous to a pharmacological "cure dose" (minimum effective dose for maximum attainable benefit). Move the slider to explore how outcomes change from 0% (no effective intervention) to 100% (full protocol adherence).

Compliance%	Effective HR	Risk Reduction%	5-yr Mortality	% of Max RR

Curedo threshold = adherence level at which the marginal benefit per additional 1% compliance drops below 0.3% RR. Below this threshold: high returns per unit of adherence. Above: diminishing returns (plateau zone). The curve shape is specific to the current intervention selection and organ profile.

What-If Scenario Analysis Organ-Adapted Protocols

Click any card to load that scenario. All HRs adjusted for current organ type.

Scenario	N	Interventions	Adj. HR	RR%	5-yr Mort.

Counterfactual — If-Not-For Analysis Pearl Twin Network · PN

Compares the factual world (interventions applied) to the counterfactual world (no interventions). PN = Probability of Necessity: probability the intervention was causally necessary for survival (Pearl 2000).

Counterfactual (If-Not-For)

1.000

No interventions

—%

→

Actual World (With Selection)

1.000

Interventions applied

—%

If-Not-For Causal Effect (PN)

Select interventions

Sensitivity Analysis — E-Values VanderWeele & Ding 2017

E-value = minimum strength of unmeasured confounder to explain away the causal effect. Active interventions shown at full opacity.

Baseline Mortality Sensitivity (current organ / age / sex / years)

Structural Causal Model — Causal DAG Active pathways highlighted

Organ-relevant pathways shown. Active intervention categories and their mediator contributions highlighted in teal.

AI Clinical Narrative Anthropic Claude API

Generates a STEM graduate-level clinical synthesis tailored to the current organ type, patient profile, Pareto position, and uncertainty analysis.