Domain A: Data Coverage & Federation¶
Using the domain reference
Domain A includes the five maturity descriptors and the source-verified minimum evidence for L3–L5. Expand the evidence section beneath each indicator and retain an auditable assessment record. See How to Apply HDRL.
Focus: Availability, linkage, and flow of health data for research
Indicators: 10 (7 Core, 3 Enhancement)
The business question
Do we actually have the data researchers want? Without this, the shop shelves are empty.
Data coverage is the foundation of any health data research service. If core datasets — primary care, hospital episodes, mortality, prescribing — are unavailable or incomplete, no amount of governance or infrastructure investment can compensate. This domain also assesses whether data can be linked across sources and federated across organisational boundaries, which is critical for the UK's multi-nation architecture.
Indicator Summary¶
| ID | Indicator | Type | Class | Unit |
|---|---|---|---|---|
| A.1.1 | Core Dataset Availability | Core | B0 | System |
| A.1.2 | Data Currency & Timeliness | Core | B0 | System |
| A.1.3 | Data Equity & Representativeness | Core | B0 | System |
| A.2.1 | Patient Identifier Infrastructure | Core | B0 | System |
| A.2.2 | Linkage Services | Core | B0 | Both |
| A.3.1 | Federated Query Capability | Enhancement | O | Both |
| A.3.2 | UK Gateway Connectivity | Core | B0 | Both |
| A.3.3 | Federation Operating Model & Assurance | Core | B0 | Both |
| A.4.1 | Consented Cohort Integration | Enhancement | C2 | System |
| A.4.2 | Multi-Modal Data Access | Enhancement | C3 | System |
A.1 — Secondary Use Data Flows¶
A.1.1 Core Dataset Availability¶
CORE · B0 · System
| Level | Description |
|---|---|
| L1 | No systematic inventory. Data flows ad-hoc. Core datasets (primary care, secondary care, prescribing, mortality) have unknown or highly restricted availability. |
| L2 | Inventory initiated. Feasibility assessed. Strategy documented. Pilot agreements in discussion. |
| L3 | Core datasets partially available: secondary care and mortality accessible; primary care <50% coverage or not refreshed; prescribing not linked. |
| L4 | Core datasets available with >= 70% population coverage. Refreshed at least quarterly. Data sharing agreements with major providers. |
| L5 | Core datasets >= 90% coverage with monthly refresh. Automated flows. Proactive provider engagement. Coverage gaps systematically addressed. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Documented dataset inventory showing which core datasets are available and under what conditions
- Evidence of access for at least some core datasets (agreements/approvals; delivered extracts/workspaces)
L4 minimum evidence
- Published data inventory/catalogue for core datasets (incl. coverage method)
- Signed data sharing agreements with major providers (or controller decisions) for those datasets
- Coverage and refresh evidence meeting Level 4 threshold (report/dashboard + extract dates)
L5 minimum evidence
- Automated ingestion/refresh logs showing monthly (or better) updates
- Coverage improvement plan + evidence of actions taken (e.g., onboarding new providers) and resulting change
- Proactive monitoring/reporting demonstrating sustained coverage and refresh performance
A.1.2 Data Currency & Timeliness¶
CORE · B0 · System
| Level | Description |
|---|---|
| L1 | Currency unknown or variable. Some datasets years out of date. No monitoring. |
| L2 | Requirements defined. Baseline measured. Targets established but not achieved. |
| L3 | Core datasets refreshed within 6 months. Latency monitored. Some achieve monthly; others delayed. |
| L4 | Refreshed quarterly, median latency <= 120 days. SLAs met >= 80%. |
| L5 | Median <= 60-day latency. Near-real-time for priority use cases. SLAs met >= 95%. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Defined timeliness requirements/targets + baseline measurement
- Refresh/latency evidence for core datasets consistent with Level 3 claim (extract dates; monitoring output)
L4 minimum evidence
- Latency definition + monitoring dashboard showing quarterly refresh and median latency claim
- SLA/targets for refresh and delivery + evidence of >=80% compliance
- Exception log showing how delays are detected and addressed
L5 minimum evidence
- Monitoring dashboard showing median latency <=60 days (and near-real-time where claimed)
- SLA report showing >=95% compliance for priority datasets/use cases
- Evidence of continuous improvement (trend over time; change log)
A.1.3 Data Equity & Representativeness¶
CORE · B0 · System
| Level | Description |
|---|---|
| L1 | No assessment of coverage by demographic/socioeconomic characteristics. Representativeness unknown. |
| L2 | Equity dimensions identified (deprivation, ethnicity, geography, age, sex). Baseline assessment initiated. |
| L3 | Coverage monitored by key dimensions. Known gaps documented. Improvement actions identified but not systematic. |
| L4 | Routine monitoring by deprivation, ethnicity, geography, protected characteristics. Annual equity reporting. Active programmes to address gaps. |
| L5 | Comprehensive equity framework with published reports. Demonstrated improvement. Equity embedded in acquisition priorities. Contributing to national guidance. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Baseline representativeness assessment using agreed equity dimensions
- Documented gap register and initial actions (plans, provider engagement)
L4 minimum evidence
- Equity monitoring report with agreed dimensions (deprivation, ethnicity, geography, protected characteristics)
- Documented gap register + prioritised improvement actions
- Published (or internally approved) annual equity report for data coverage/quality
L5 minimum evidence
- Published equity framework + repeated reporting over time
- Evidence of measurable improvement in identified gaps (before/after metrics)
- Evidence equity is embedded in acquisition priorities (e.g., business case decisions, procurement criteria)
A.2 — Identity & Linkage¶
A.2.1 Patient Identifier Infrastructure¶
CORE · B0 · System
| Level | Description |
|---|---|
| L1 | No consistent identifier. Probabilistic matching with significant errors. |
| L2 | National identifier exists but incomplete adoption. Strategy documented. Interim approaches defined. |
| L3 | Identifier (CHI, NHS number) used in majority of datasets. Deterministic linkage for most core datasets. |
| L4 | Identifier consistently applied across core datasets. Linkage accuracy >= 99%. Validation in place. |
| L5 | Universal identifier across all datasets including cohorts and multi-modal. Externally validated. Supports cross-UK linkage. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Evidence identifier is present in the majority of relevant datasets (audit/sample)
- Linkage validation approach documented with initial results
L4 minimum evidence
- Documentation showing identifier coverage across core datasets
- Linkage accuracy validation report (method + results) supporting >=99% claim
- Ongoing control for identifier quality (audit/checks and remediation process)
L5 minimum evidence
- External/independent validation of linkage accuracy and process (audit or peer review)
- Evidence identifier applies across extended datasets (cohorts/multi-modal where applicable)
- Evidence of cross-UK linkage capability (documented approach + successful linkage case)
A.2.2 Linkage Services¶
CORE · B0 · Both
| Level | Description |
|---|---|
| L1 | No dedicated service. Linkage ad-hoc with inconsistent methodology. |
| L2 | Function identified. Methodology documented. Available for selected projects. |
| L3 | Operational service. Standard process. Turnaround variable (weeks to months). Limited combinations. |
| L4 | Routinely available with SLAs. Turnaround <= 4 weeks. Flexible linkage. Quality metrics reported. |
| L5 | Turnaround <= 2 weeks. Automated workflows. Advanced capabilities (fuzzy matching, privacy-preserving). |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Linkage service SOP/process documented and operational for projects
- Evidence of delivered linkages + basic quality reporting (match rate/error rate)
L4 minimum evidence
- Standard linkage service SOP + published SLA
- Turnaround statistics (median + tail) showing <=4 weeks for routine requests
- Linkage quality metrics report (match rate, error rate) and QC process
L5 minimum evidence
- Workflow automation evidence (tooling/pipeline docs) + stats showing <=2 weeks
- Evidence of advanced methods where used (e.g., privacy-preserving/fuzzy matching) with evaluation
- Independent review or benchmarking of linkage service performance/quality
A.3 — Federation & Interoperability¶
A.3.1 Federated Query Capability¶
ENHANCEMENT · O · Both
| Level | Description |
|---|---|
| L1 | No federated capability. All analysis requires data transfer. |
| L2 | Concepts understood. Options assessed. Pilot in planning. |
| L3 | Federated possible for selected datasets. Bespoke setup. Limited tools. |
| L4 | Routinely supported. Standard APIs. Compatible with DataSHIELD, OHDSI. |
| L5 | Default for appropriate uses. Rich API ecosystem. Active in UK/international networks. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Pilot or limited operational federated query implementation (architecture/API evidence)
- Evidence of at least one federated use case delivered (project log)
L4 minimum evidence
- Operational federated query implementation documentation (APIs/standards supported)
- Evidence of routine use (project list/logs) and supported toolchain (e.g., DataSHIELD/OHDSI)
- Service guidance/SOP for researchers and node operators
L5 minimum evidence
- Policy/architecture showing federated access is default where appropriate
- Participation evidence in federated networks/standards groups + interoperability test results
- Performance and reliability KPIs for federated services (availability, query latency)
A.3.2 UK Gateway Connectivity¶
CORE · B0 · Both
| Level | Description |
|---|---|
| L1 | No awareness of UK Gateway specs. Architecture does not consider UK-wide interoperability. |
| L2 | Specs reviewed. Gap analysis completed. Roadmap defined. |
| L3 | Architecture aligned. Connectivity in development/testing. Manual workarounds required. |
| L4 | Operational connectivity. Metadata discoverable. Standard requests flow through Gateway. |
| L5 | Full integration with automated sync. Complex UK-wide queries supported. Contributing to standards. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Gateway connectivity roadmap + technical alignment evidence (architecture mapping)
- Evidence of connectivity in development/testing (test results; manual workarounds documented)
L4 minimum evidence
- Gateway integration test evidence (metadata discovery + request routing) and operational sign-off
- Operational logs/screenshots showing metadata is discoverable and requests flow via gateway
- SOP for maintaining gateway connectivity (change control + incident handling)
L5 minimum evidence
- Automated synchronisation evidence (scheduled jobs, APIs) with monitoring/alerts
- Demonstration of complex UK-wide requests/queries supported (test cases)
- Evidence of contribution to gateway standards/spec evolution (change proposals, working group outputs)
A.3.3 Federation Operating Model & Assurance¶
CORE · B0 · Both
| Level | Description |
|---|---|
| L1 | No federation operating model. Roles and cross-node processes undefined. |
| L2 | Draft operating model. Limited bilateral agreements; inconsistent use. |
| L3 | Model used for priority pathways. Issues logged; assurance limited. |
| L4 | Standard model in routine use. Cross-node SOPs and performance reporting. |
| L5 | Optimised federation. Joint improvement cycle and independent assurance/benchmarking. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Draft operating model (roles, escalation, change control) with initial adoption evidence
- Evidence of operating model used for at least one cross-node pathway (issues/decisions logged)
L4 minimum evidence
- Signed federation operating model (roles, RACI, escalation, change control) + SOP set
- Joint governance minutes/decision logs showing routine use
- Cross-node KPIs/service reporting and evidence of interoperability testing
L5 minimum evidence
- Evidence of joint improvement cycle (backlog, retrospectives, release notes) and tracked KPI improvement
- Independent assurance/assessment of federation operations (audit/peer review)
- Evidence model has been shared/adopted beyond the node (templates, guidance, community contribution)
A.4 — Multi-Modal & Cohort Data¶
A.4.1 Consented Cohort Integration¶
ENHANCEMENT · C2 · System
Capability module
Within HDRL v1.0's original six-capability mapping, this indicator is included only when assessing Capability 2 (Consented cohort data). The mapping preserves the initial 2025 UK Health Data Research Service (HDRS) proposition; it should not be treated here as a current programme requirement.
| Level | Description |
|---|---|
| L1 | No systematic linkage to cohorts/biobanks. |
| L2 | Key cohorts identified. Consent reviewed. Pilot planned. |
| L3 | Selected cohorts linkable. Bespoke arrangements. Coverage incomplete. |
| L4 | Major cohorts routinely linkable. Standard processes. Catalogue maintained. |
| L5 | Comprehensive linkage. Recall-by-genotype. UK-wide cohort integration. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- List of priority cohorts/biobanks and consent/permissions review summary
- Evidence of at least one linkage arrangement delivered or in active operation
L4 minimum evidence
- Catalogue of linkable cohorts/biobanks + consent/permissions summary
- Standard linkage process + template agreements
- Evidence of routine linkage for major cohorts (delivery logs/metrics)
L5 minimum evidence
- Evidence of comprehensive linkage coverage across major cohorts/biobanks
- Demonstrated advanced use case (e.g., recall-by-genotype or equivalent) with governance approval
- Evidence of UK-wide cohort integration arrangements (cross-node agreements + delivered project)
A.4.2 Multi-Modal Data Access¶
ENHANCEMENT · C3 · System
Capability module
Within HDRL v1.0's original six-capability mapping, this indicator is included only when assessing Capability 3 (Multi-modal data). The mapping preserves the initial 2025 HDRS proposition; it should not be treated here as a current programme requirement.
| Level | Description |
|---|---|
| L1 | Multi-modal data (imaging, genomics, pathology, clinical letters) not available. |
| L2 | Strategy defined. Priority types identified. Pilot planned. |
| L3 | Selected multi-modal available. Coverage incomplete; linkage partial. |
| L4 | Routine access to >= 2 types with >= 25% coverage each. Linked to core datasets. |
| L5 | Comprehensive access: imaging, genomics, pathology, clinical letters. Population-scale. NLP-processed text. |
Minimum evidence for L3–L5
Use this source-defined minimum evidence with the maturity descriptors above. Record the artefact, date, scope and the claim it supports.
L3 minimum evidence
- Multi-modal strategy/inventory with at least one modality accessible for research
- Evidence of linkage for at least one modality to core datasets (technical + governance sign-off)
L4 minimum evidence
- Inventory of multi-modal datasets with coverage estimates for >=2 modalities
- Evidence of linkage to core datasets (technical + governance sign-off)
- Access process and documentation (data dictionaries, quality notes) for those modalities
L5 minimum evidence
- Evidence of population-scale multi-modal availability (coverage/refresh metrics per modality)
- Evidence of advanced handling where relevant (e.g., NLP-processed text pipeline + QC)
- Demonstration of routine multi-modal research delivery (project examples, throughput metrics)