Skip to content

Latest commit

 

History

History
59 lines (48 loc) · 2.71 KB

File metadata and controls

59 lines (48 loc) · 2.71 KB

SHIELD Product Plan

Thesis

Score care-delivery networks at a declared scale, identify measurable access, capacity, workforce, continuity, affordability, and resilience gaps, and report where infrastructure methods do not transfer cleanly.

Implemented product shape

  • Seven-crate Rust workspace: six shared analysis crates plus the feature-specific shield-cms-access public-data result.
  • DIM-01..13 scale-aware evidence contracts.
  • Explicit transfer-strain and null-result posture.
  • A deterministic CMS-derived aggregate fixture; no patient records.
  • A fourteen-section held HLT handoff that cannot authorize savings, allocation, facility or clinical action, public release, or rate changes.
  • A deterministic CMS–USDA county rurality join covering 5,360 of 5,432 facilities while preserving 72 unmatched rows as an explicit residual.
  • A current HRSA primary-care HPSA registry census that distinguishes 79,150 component rows, designation status, 7,682 currently designated IDs, and component/rural-status multiplicity.
  • A separately preserved June 30, 2026 HRSA quarterly total of 9,003 primary-care designations; different-vintage totals are not forced together.
  • A same-vintage HRSA geography bridge separating 2,838 area designations from 4,844 facility designations, preserving 762 multi-component and 155 multi-county area designations, and exposing 18 invalid county-key residuals.
  • A same-vintage HRSA designation-formula capacity baseline covering 2,838 area and 550 correctional designations, with exact FTE/shortage arithmetic and 4,294 policy-excluded facility designations kept distinct from zero.

Next public work

The facility, rurality, and formal shortage-registry baselines are complete. Next:

  1. Build a compatible CMS/HRSA facility-identity bridge only if explicit source identifiers support it; never infer identity from county co-location.
  2. Add an official provider/site identity or staffed service-line source only where a compatible geographic denominator and baseline/surge basis exist; do not relabel designation-formula physician FTE as hospital staffing.
  3. Keep county class, formal shortage designation, facility presence, staffed service availability, and patient-relevant travel access distinct.
  4. Test a specific HLT candidate against access, quality/safety, equity, adequacy/resilience, cost, and delivery evidence before any fiscal admission.

Non-goals

No medical advice, patient-level analysis, licensing, accreditation, Certificate-of-Need, payer, coverage, or treatment determination.

Validation

cargo fmt --all -- --check
cargo clippy --workspace --all-targets -- -D warnings
cargo test --workspace --locked
cargo run -p shield-cli -- --help