Case Study

Iatrogenic Medications — Causal Query Explorer

How does prescribing weight-gaining or lipid-altering medications affect BMI, cholesterol, hospitalisation, and mortality? Toggle the query mode to move from association (Rung 1) to causation (Rung 2) to individual counterfactual (Rung 3).

Why this query explorer matters

Association tells you what tends to go together. Intervention tells you what would happen if you forced a change. Counterfactual tells you what would have happened for this specific patient under different circumstances. These are three distinct questions — and they have three distinct answers, even on the same data and the same model.

Rung 1 (obs) leaves all back-door paths open: conditioning on this patient's age and SES mixes causal signal with selection bias. Rung 2 (do) severs those incoming edges, isolating the downstream causal effect of the medication level. Rung 3 (counterfactual) anchors the patient's individual background — abducting their U node values from actual observations — then asks what would have happened for that specific person under a different prescription. Use the tabs above to decompose which paths carry the effect, how large the confounding gap is, and what the U nodes reveal about this patient's individual causal fingerprint.

Rung 2 — Intervention (do)
OutcomeMeanSD