The causal map LIVING · continuously updated

A knowledge graph of UK health data: domain nodes you can drag, grey edges are measured join links, coloured arrows are causal claims with counterfactual structure - green tested-supported, red tested-refuted (falsifications are a feature), amber registered hypotheses. Click any node or edge. Maintained from a versioned registry the science team extends through its pipeline.

Counterfactual cards

Click a causal edge (coloured arrow) to read its claim: mechanism, the counterfactual (population / exposure / outcome / time), identification design, evidence and verdict. Click a node for the domain's live footprint.

How to read it

supported claim · refuted claim · hypothesis · grey = join plumbing (opt-in) · drag nodes, scroll to zoom.

Registry: causal-edges.json in the science repo - every edge is a PR, every verdict traces to a paper.

Copilotdeep · LLM + SQL