Structural limits of single-barrier reform in algorithmic recourse: a formal series-system model (analysis code and data)

Analysis code, derived data, and a one-command reproducibility pipeline for the manuscript 'Structural limits of single-barrier reform in algorithmic recourse: a formal series-system model with implications for digital health' (PLOS Digital Health, PDIG-D-26-00342). A formal 11-stage multiplicative series-system model of algorithmic recourse across three layers (data integration, data accuracy, institutional access), parameterized from federal datasets (CFPB, FTC, LSC) and one healthcare algorithmic audit. Includes: closed-form single-barrier bound and higher-order-interaction propositions; factorial decomposition; Shapley attribution; Sobol and Morris sensitivity; alternative-topology robustness (Gaussian-copula and repeated-attempt); systemic-bias placement sensitivity; and a full parameter-derivation audit trail. All stochastic analyses are seeded (seed 42) and reproducible via `make all` / `make verify`, with regression tests pinning the deterministic headline values (baseline 0.0018%, maximum single-barrier gain 0.0054%, three-way factorial share 87.6%). v2.0.0 corresponds to the PLOS Digital Health major-revision package.