LRH · EMERGENCY DEPARTMENT

Guideline sources

Every upstream body this manual takes numbers from: which edition each tool is written against, when a human last checked it, and exactly which tools and cards move if it changes.

⚠ THIS PAGE IS PROVENANCE, NOT CLINICAL CONTENT. A green chip means somebody confirmed the manual still matches the current edition on that date — it is not an assurance that the content is correct, and it is never a substitute for reading the guideline. NEVER VERIFIED means exactly that: nobody has checked yet through this process. Nothing on this page is stored or sent anywhere.

FRESHNESS AT A GLANCE

BY SOURCE — WORST FIRST

Sorted by how much attention it needs, then by how much of the manual moves if it changes. Filter by body, tool or topic.

BY TOOL — WHAT EACH ONE OWES

Before changing a clinical value in a tool, this is the list of guidelines that tool answers to. Adding a value from a source not listed here means adding a row to guidelines.js first.

HOW THIS IS KEPT UP TO DATE — AND WHAT IT CANNOT CATCH

A scheduled job (check_guidelines.mjs, run monthly by GitHub) checks each body two ways: a PubMed search for newly indexed guidelines, and a fingerprint of the body’s own guideline index page. When either moves, it opens a GitHub issue naming the tools and cards affected.

Neither check is complete, and the design assumes that. A body that swaps a revised PDF in at the same address, or reorganizes its website, can change a guideline without either probe noticing. So every source also carries a review interval — 180 days for the ones that drive the most numbers, 365 for the rest — and going past it puts the source on the list whether or not anything was detected. The crawler shortens the time to notice; the calendar is what guarantees the look happens.

A quiet month therefore reads as no change detected, never as nothing changed. Process: GUIDELINE-WATCH.md.

— personal/unofficial aid, not part of the hospital IT system/EHR, not a substitute for clinical judgment. Editions and review dates are recorded by hand and can lag reality; thresholds are assay- and site-specific — verify locally against the guideline itself.