Skip to content

Interoperability, identity and continuity

Codes: INT — interoperability and continuity; IDPOP — identity, registration and population management.

Scope

This finding concerns whether sufficient, correctly attributed information follows the person and the accountable care team across technical, organisational, geographic and temporal boundaries. INT covers exchange and continuity; IDPOP covers the identity, registration, cohort and source-selection rules that determine which information and people are visible.

Interoperability is therefore more than message transport. It includes positive identity, semantic meaning, provenance, source coverage, arrival in the receiving workflow and reconciliation of conflicting or missing states.

What recurs in the current series

  • Information exists elsewhere but is not available in the care view. Stephen Cassidy (representative case) concerns a known allergy held in a national record but not routinely accessible or integrated into hospital care.
  • Separate service records can create several reasonable but incomplete views. David Buttriss concerns GP, out-of-hours and mental-health records that were not reliably accessible across services; Adrian Jennings adds separate systems and structured fields that did not readily carry the required clinical context. The former was framed as a future-death risk without a separate finding that the systems caused the death; the latter formed part of a wider communication and support pathway.
  • The transfer artefact is separate from the receiving record. Morris Reddington (representative case) concerns an ambulance ePRF behind separate credentials and unfamiliar navigation, encouraging reliance on verbal handover.
  • Registration can select the wrong longitudinal source. In Samuel Jordan (representative case), the receiving service obtained the permanent-practice record rather than the more recent temporary-practice record.
  • Cohort status can be absent from the computable feed. Rachelle Ross (representative case) concerns missing screening non-responder status and a resulting dependency on local manual entry.
  • Information can attach to the wrong subject or context. Hannah Booth concerns maternal risk information placed only in the baby's record, requiring both correct attribution and cross-record relevance.
  • Exchange can fail at transaction state. Oliver Billings concerns prescription cancellation after one pharmacy had downloaded the prescription, leaving duplicate active supply across endpoints.
  • Manual bridges can delay or obscure transfer state. Steven Welch concerns print-and-fax transfer of radiology images and reports across a Wales–England boundary; Paul Price concerns incompatible GP and mental-health systems, fax, post, outage and a lost callback. The Welch report expressly said the identified errors were unlikely to have caused the death; both reports nevertheless identify prospective risk in the transfer mechanism.
  • A shared database can still omit the recent operational history. Jacob Sulaiman concerns response officers who could see only information entered in the shared database, not a readily retrievable account of recent calls, visits and outcomes. The report said fuller information may have affected assessment and management, not that the database failure caused the death.
  • Shared access can still be incomplete. Lily Jahany, Akhona Moyo, Edward Muwanga and Tammy Milward concern different combinations of unavailable source records, partial summaries, multiple platforms and separately recorded risk messages.

The transferable mechanism is not simply “systems do not talk.” It is failure to establish a sufficiently complete, attributable and actionable clinical state at the point responsibility changes.

Variation and limits

  • The series includes incompatible systems, absent interfaces, permission or consent dependencies, partial summaries, delayed records, separate logins, hybrid paper/electronic handover and semantic mismatch. These require different controls.
  • Local configuration matters. Evidence in the Emma Sanders case described variation in Summary Care Record configuration by ICB; one local state should not be generalised nationally.
  • A single combined record can itself misattribute information. Hannah Booth cautions against assuming that co-location alone solves identity and relevance.
  • In Lily Jahany and several later reports, information-system concerns were not found causal or more than minimally contributory. The prospective safety concern remains distinct from the conclusion on the death.

The current review counts show recurrence of INT and IDPOP within the focused series. The sample was targeted, codes overlap and reports differ in causal status; the counts are not prevalence estimates and do not rank vendors or architectures.

Defence implications — analytical translation

The following are review-derived implications, not coronial findings about Defence systems:

  • separate current care responsibility, physical location, administrative registration and authoritative record sources;
  • show which expected sources and periods are present, unavailable or not yet reconciled;
  • require a minimum transfer dataset, including allergies, medicines, active risks, current plans and outstanding work, to enter the receiving workflow;
  • maintain cohort visibility for screening, vaccination, readiness and vulnerable-person searches despite posting, attachment, embarkation or temporary registration; and
  • test transfers between Defence, NHS, allied and deployed environments, including duplicate, delayed and conflicting states.

See the DPHC and Programme CORTISONE analysis.

Evidence basis: included-case dataset, coding framework, and current review counts.