# Build an ITC feasibility workbook

Create a decision-support workbook using the exact stable headers in `../assets/itc-feasibility-workbook.md`. Populate every available field from cited sources; write `unavailable` rather than infer missing information. Include network, endpoint/estimand, directed contrast, modifier, risk, scenario, and human decision tables.

For each contrast, state numerator, denominator, endpoint definition, timepoint, effect measure, target population, and whether the target is conditional or marginal. Separate prognostic variables from plausible effect modifiers. State whether the network is anchored or unanchored and the assumptions that follow. For MAIC, require covariate support/overlap and effective-sample-size diagnostics; for STC, require a stated outcome model and target standardization. Do not generate a `proceed` decision from a score. A named accountable owner must choose `proceed`, `restrict`, or `stop` and record the rationale.

Flag endpoint mismatch, absent common comparator, unavailable material covariates, incompatible timepoints, unsupported transportability, and unknown effect scale as unresolved risks. Do not calculate an effect estimate unless the supplied evidence and protocol authorize it.