Skip to content

Regulation 28 case crosswalk for the article

Evidence boundary: This dossier explains the evidence and reasoning behind an Analysis article. It does not establish that DMICP caused harm, contains the civilian defects described, is unsafe, or fails DCB0129/DCB0160. Evidence cut-off: 15 July 2026.

The manuscript uses seven Prevention of Future Deaths reports as illustrative hazard-discovery signals. This is not the complete targeted corpus and is not a frequency analysis. Each transfer to Defence is a prospective question to test, not a finding about DMICP.

Seven illustrative cases

Case and article ref Report-specific digital mechanism Defence assurance question Controlled trace Essential boundary
Stephen Rhodes, ref 19 A markedly abnormal result and recommended action were filed without sufficient salience. Can a critical result remain salient, owned, escalated and closed when workload is high or the recipient is absent? SRC-001; CLM-001; REQ-004, REQ-005, REQ-009 Interface, display, workload and workflow contributions cannot be isolated. SRC-018 is a respondent account, not the coroner's finding.
Stephen Stringer, ref 20 A digital enquiry entered an administrative route, did not enter the clinical record and was not seen by a doctor. Does every patient message enter accountable clinical triage, record incorporation, escalation and closure? SRC-014; CLM-005; REQ-005; AST-005 The official report does not identify the software supplier. SRC-023 is response evidence.
Morris Reddington, ref 21 An ambulance electronic record required a separate retrieval path and was routinely not reviewed during hospital handover. Does a transfer record arrive inside the receiving workflow, become visible and acquire an accountable owner? SRC-004; CLM-004; REQ-006; AST-006 The proposed FHIR remedy in SRC-015 is a respondent proposal, not a coronial finding.
Samuel Jordan, ref 22 Administrative registration selected a record source that omitted recent care. Can registration, care responsibility, location or record-source selection hide recent care or remove someone from a safety cohort? SRC-021; CLM-010; REQ-001, REQ-002, REQ-003 The civilian mechanism is direct evidence for the case; Defence applicability remains an analytical hypothesis.
Stephen Cassidy, ref 23 A known drug allergy was stranded outside the receiving hospital workflow. Is safety-critical information available, visibly complete and reconciled across organisational boundaries? SRC-020; CLM-009; REQ-003, REQ-006 The report is not evidence of Defence-platform availability or behaviour.
Rachelle Ross, ref 24 Safety depended on manual creation of a specific cervical-screening non-responder warning. Are screening response states computable, visible and owned, including exclusions and failed attendance? SRC-002, SRC-016, SRC-076; CLM-002; REQ-002, REQ-009 Generic smear-due alerts existed; the missing element was the specific non-responder status. Recipient explanations remain separate from the coroner's concern.
Alexander Braund, ref 25 Amendment provenance was present in audit evidence but not apparent in the exported summary examined at inquest. Do live, printed, exported and investigation views preserve author, event time, entry time, prior content and amendment status? SRC-003, SRC-017; CLM-003; REQ-008; AST-008 The live audit trail retained prior content; the concern involved what the exported summary exposed.

How the cases are used

The seven cases make the article's six seams concrete:

  • Visibility: Rhodes.
  • Work ownership: Rhodes, Stringer and Reddington.
  • Interoperability and identity: Reddington, Jordan and Cassidy.
  • Integrity and provenance: Braund.
  • Population safety and decision support: Ross and Jordan.
  • Governance: responsibility and evidence boundaries across every case.

Resilience is supported more strongly by the wider targeted corpus and Defence inspection evidence than by these seven illustrative reports. The thematic analysis and Findings preserve that broader context.

Interpretation rules

  1. A PFD report is authoritative for the coroner's stated concern, not for prevalence or product-wide performance.
  2. A recipient response is the respondent's account and is never silently converted into a coronial finding.
  3. A future-risk concern does not necessarily establish that the digital mechanism caused the death.
  4. A similar mechanism in a civilian system makes a Defence test foreseeable; it does not prove transferability.
  5. Candidate REQ- requirements remain Defence-unvalidated and AST- scenarios remain unexecuted.