Mechanistic interpretability of reinforcement learning in Medicaid care coordination.

Abstract

To expose reasoning pathways of a reinforcement learning policy for Medicaid care coordination, develop an error taxonomy and implement fairness-aware guardrails. Retrospective interpretability audit using attention analysis, Shapley explanations, sparse autoencoder feature discovery and blinded clinician adjudication. Medicaid care coordination programmes in Washington, Virginia and Ohio (July 2023-June 2025). 250 000 intervention decisions; 200 divergent cases reviewed by five clinicians. Calibrated harm prediction; algorithmic clearance and residual harm rates; error taxonomy frequencies; subgroup fairness metrics. The conformal model achieved area under the receiver operating characteristic curve of 0.80 (95% CI 0.78 to 0.82), clearing 89.5% (95% CI 88.9% to 90.1%) of decisions with 1.22% (95% CI 1.14% to 1.30%) residual harm versus 6.67% (95% CI 6.02% to 7.32%) for flagged decisions. Sparse autoencoders identified seven reasoning motifs linking social determinants to clinical cascades. The error taxonomy revealed premise errors (48%, 95% CI 41% to 55%), calibration failures (27%, 95% CI 21% to 33%) and contextual blind spots (25%, 95% CI 19% to 31%). Divergence was higher for telehealth visits (11.2%) and behavioural health patients (10.7% vs 6.9%, p<0.001). Fairness optimisation reduced race-group disparity by 37% (95% CI 22% to 48%) and sex-group disparity by 28% (95% CI 14% to 39%). Reviewers rated 23% (95% CI 17% to 29%) of overridden recommendations as well-matched, confirming appropriate human oversight. Mechanistic interpretability transforms opaque algorithmic assistance into auditable decision support, providing a governance scaffold for clinical artificial intelligence deployment.