Desk live·
ForensicPost
Breaches/Availability/File 24-1231

Four 2024 Incidents Did Their Damage by Stopping Things, Not Taking Them

Four 2024 incidents did most of their damage by stopping things rather than by taking them. One of them reached a patient. None of them produced an availability figure anybody had to file.

Constructed geometry · not a chart of case data
S. Rosler13 min readConfidence: medium4 sources reviewed

This desk files availability as the half of security nobody counts: breach law is built around confidentiality, an incident where nothing is stolen but a service stops generates no notification and no statistic, and the harm to the affected person is immediate and concrete.

Four incidents from 2024 make that case better than the argument ever did.

What stopped, and what it costThis corpus; see the linked files
TimeEventEvidence
Change HealthcareClaims and payment routing severed across much of US healthcare24-0221 · providers unpaid, no availability figure filed
Ascension≈140 hospitals to paper charting, ambulances diverted, ≈1 month24-0508 · delayed care counted nowhere
Synnovis10,000+ appointments and 1,710 operations postponed24-0603 · and a contributed death at 24-0604
CDK Global≈15,000 dealerships to pen and paper for ≈2 weeks24-0711 · >$1bn modelled, borne by the dealerships

Every One Of Them Filed A Confidentiality Number Instead

192.7 million at Change Healthcare. 5.6 million at Ascension. Around a million potentially affected at Synnovis. Those figures exist because a law requires them.

No equivalent figure exists for a month of paper charting or two weeks of manual sales, because no law requires one. The corpus has been reading the record through the only aperture the record has.

What Changed, And What Did Not

What changed is that at 24-0604 an NHS trust’s own patient safety investigation connected an outage to a death. That is the mechanism the corpus filed as missing at 25-1219 — the one connecting an incident to a person — appearing once.

What did not change is the counting. One finding, arrived at because a health service investigates deaths as a matter of course, is not a measurement regime. It is an accident of a sector that already had one.

The Limits Of This File

Four incidents is not a year and not a sample. They are here because they were large, Western, and covered — which is the same selection this corpus documented at 25-0502 and measured against incident-response data at 25-1225b.

Graded medium. The individual facts are established in the linked files; the claim that 2024 was distinctive is this desk’s reading, and a reading built on the four cases the record happened to preserve.

This is an analysis file

It draws on the files linked in the timeline rather than adding new material; each carries its own sources. The claim that these four are representative of anything is explicitly not made — they were selected by coverage, which is the bias this corpus documents. Graded medium. Corrections: corrections@forensicpost.com.

Sources
  1. Ransomware attack continues to disrupt healthcare in LondonThe Record
  2. The Change Healthcare cyberattackCongressional Research Service
  3. The CDK Global outage: explaining how it happenedTechTarget
  4. Ascension ransomware attack affects 5.6 million patientsHIPAA Journal
S. Rosler
Covers extortion groups and leak-site economics. Verifies our sample sets.
// the chain of custody — tuesdays

Get the next file first.

One incident a week, taken apart properly. Logs, timelines, and what the filing left out.

PGP-signed edition · no tracking pixels · one-click unsubscribe
© 2026 ForensicPost Media · the desk · newsletter · searchGlossary