Cohort validation pack

OpenSAFELY

Generated 6 Oct 2026, 10:37 · ID 51

Unverified

Last verified: 16/04/2026

Instructions for reviewers

Review each field against your knowledge of this cohort. Mark Accurate if the value is right, Needs correction if it is wrong or inaccurate (add notes), Not sure if you cannot verify it, or Does not apply if the field is not relevant.
Evidence shown is from our source registry. Fields marked No source yet hold values we could not yet pin to a document — those were AI-inferred and need your input most. Fields marked Searched — not published are ones we looked for and could not find publicly; the note lists where we checked.
Your selections and notes are kept in this browser tab only (not sent to the database). Print or save as PDF when finished, then return it to the DIGIT team.

Overview

Total fields

Field-cited

Whole-record source only

AI-inferred, no source yet

Searched — not published

No value held

Needs review

Source coverage

Summary of how well fields in this pack are linked to evidence.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Field-level source coverage Optional	41 of 70 reviewable fields are cited to a source (59%) · 8 with no source yet · 21 searched — not published · 30 with no value held	Computed	No citation linked to this field yet.	Response for Field-level source coverageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Identity and ownership

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Short name Required	OpenSAFELY	Field-cited	OpenSAFELY — OpenSAFELY in brief https://www.opensafely.org/os-in-brief/ Section: title “OpenSAFELY in brief - OpenSAFELY” Page: https://www.opensafely.org/os-in-brief/ DIGIT name field uses the public product token OpenSAFELY from the page <title> element.	Response for Short nameAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Full name Required	OpenSAFELY	Field-cited	OpenSAFELY — OpenSAFELY in brief https://www.opensafely.org/os-in-brief/ Section: title “OpenSAFELY in brief - OpenSAFELY” Page: https://www.opensafely.org/os-in-brief/ full_name mirrors the canonical public product name OpenSAFELY (same <title> substring).	Response for Full nameAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Managing organisation Recommended	University of Oxford / London School of Hygiene & Tropical Medicine	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: OpenSAFELY is a public asset “OpenSAFELY is a public asset. It is publicly funded, built by researchers and software developers at the University of Oxford, all IP is shared openly, and the Data Controller is NHS England.” Page: https://www.opensafely.org/os-in-detail/ managing_organisation lists Oxford plus LSHTM EHR group as the academic/engineering partnership named alongside Oxford on the same page (LSHTM bullet in the collaboration list).	Response for Managing organisationAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Primary site Recommended	No value held	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: Data Controller “NHS England is the Data Controller for the service as a whole.” Page: https://www.opensafely.org/os-in-detail/ UNSOURCEABLE for a single DIGIT sites.id anchor: NHS England is the Data Controller and execution is vendor-hosted; Oxford/LSHTM are development partners rather than a discrete recruitment site row in DIGIT.	Response for Primary siteAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Website URL Recommended	https://www.opensafely.org/	Field-cited	OpenSAFELY — OpenSAFELY in brief https://www.opensafely.org/os-in-brief/ Section: Canonical site “https://www.opensafely.org/” Page: https://www.opensafely.org/os-in-brief/ Public programme entry point is opensafely.org (also linked from in-detail navigation).	Response for Website URLAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Catalogue of Mental Health Measures URL Recommended	No value held	No source yet	No citation linked to this field yet.	Response for Catalogue of Mental Health Measures URLAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Contact name Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no single public cohort contact name mandated for the whole platform on reviewed pages.	Response for Contact nameAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Contact email Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: platform-wide public contact email not extracted from reviewed pages (site has contact routes).	Response for Contact emailAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Summary and status

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Description Required	Secure, transparent software platform for analysis of NHS electronic health records. Research code runs against pseudonymised records inside vendor and NHS-hosted environments; aggregated outputs leave the secure perimeter. NHS England is the Data Controller for the service; development is led from the University of Oxford with LSHTM EHR science collaboration.	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “Working on behalf of NHS England, we created OpenSAFELY—a secure health analytics platform that covers 40% of all patients in England and holds patient data within the existing data centre of a major vendor of primary care electronic health records.” Page: https://doi.org/10.1038/s41586-020-2521-4 OpenSAFELY — OpenSAFELY in brief https://www.opensafely.org/os-in-brief/ Section: OpenSAFELY is a software platform “OpenSAFELY is a software platform for analysing large, sensitive datasets – without compromising any of the secrets inside.” Page: https://www.opensafely.org/os-in-brief/ Description synthesises in-brief platform framing with Nature 2020 secure analytics / pseudonymous linkage language and in-detail governance (NHS England controller; code-to-data model). OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: OpenSAFELY is a secure, transparent, open-source software platform “Rather than allowing researchers to make copies of the raw data, OpenSAFELY leaves the data where it's already stored. Instead, OpenSAFELY takes the research code to the data – then brings back aggregated results in the form of tables and graphs.” Page: https://www.opensafely.org/os-in-detail/	Response for DescriptionAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Total N Required	No value held	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “covers 40% of all patients in England” Page: https://doi.org/10.1038/s41586-020-2521-4 total_n is NULL because sources give a coverage fraction (40% in 2020 Nature description) and later multi-vendor expansion without a single audited headcount in the cited pages; removed legacy 58M figure as unsupported by captured files.	Response for Total NAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Total N qualifier Recommended	No value held	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “covers 40% of all patients in England” Page: https://doi.org/10.1038/s41586-020-2521-4 total_n_qualifier NULL alongside NULL total_n because the published evidence is a proportion, not a qualifier tag on an integer N.	Response for Total N qualifierAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Year established Recommended	2020	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: dc.source “Nature 2020 584:7821” Page: https://doi.org/10.1038/s41586-020-2521-4 year_established=2020 from Nature publication year in page metadata.	Response for Year establishedAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Active cohort Recommended	Yes	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: Platform News / Optum access “as of January 2026, research access to data held by Optum data is temporarily closed, pending agreed funding arrangements between Optum and NHS England.” Page: https://www.opensafely.org/os-in-detail/ is_active=TRUE because the platform and TPP-backed access continue while Optum/EMIS-side access is partially paused rather than the programme ending.	Response for Active cohortAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Ongoing recruitment Recommended	No	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: OpenSAFELY is used in various different ways “OpenSAFELY is used in various different ways, but the most important way is with data from the NHS – particularly GP records.” Page: https://www.opensafely.org/os-in-brief/ ongoing_recruitment=FALSE: entity models the platform/archive rather than a single recruiting study.	Response for Ongoing recruitmentAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Refreshed cohort Recommended	Yes	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: OpenSAFELY in brief (body) “Taken together, health records from millions of people are combined in huge datasets.” Page: https://www.opensafely.org/os-in-brief/ Live EHR backends are continually appended in routine care -> is_refreshed=TRUE for this administrative-analytics resource (analogous to other national EHR products).	Response for Refreshed cohortAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Recontact success rate Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: not applicable to non-recruited EHR platform.	Response for Recontact success rateAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Retention rate Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: not applicable to non-recruited EHR platform.	Response for Retention rateAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Published references Recommended	doi:10.1038/s41586-020-2521-4, doi:10.1136/bmjopen-2022-071261	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: citation_doi “10.1038/s41586-020-2521-4” Page: https://doi.org/10.1038/s41586-020-2521-4 ClinicalTrials.gov https://clinicaltrials.gov Section: referencesModule.references[0].citation “Herrett E, Tomlin K, Lin LY, Tomlinson LA, Jit M, Briggs A, Marks M, Sandmann F, Parry J, Bates C, Morley J, Bacon S, Butler-Cole B, Mahalingasivam V, Dennison A, Smith D, Gabriel E, Mehrkar A, Goldacre B, Smeeth L, Eggo RMM. Protocol for an OpenSAFELY cohort study collecting patient-reported outcome measures using the TPP Airmid smartphone application and linked big data to quantify the health and economic costs of long COVID (OpenPROMPT). BMJ Open. 2023 Feb 17;13(2):e071261. doi: 10.1136/bmjopen-2022-071261.” Page: https://clinicaltrials.gov/study/NCT05552612 Second DOI supports OpenSAFELY-linked OpenPROMPT programme referenced from ClinicalTrials.gov JSON.	Response for Published referencesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Notes Recommended	Primary care coverage spans TPP SystmOne and Optum (formerly EMIS) deployments in England; OpenSAFELY in detail notes a January 2026 pause on research access to Optum-held data pending funding arrangements with NHS England. Mental health phenotypes are typically derived from coded diagnoses and prescribing in routine records rather than research-only assessments.	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: Optum access “as of January 2026, research access to data held by Optum data is temporarily closed, pending agreed funding arrangements between Optum and NHS England.” Page: https://www.opensafely.org/os-in-detail/ Notes also summarise MH phenotyping via coded diagnoses/prescribing rather than research-only instruments.	Response for NotesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

3D cohort mapping

Consent, characterisation, and data availability axes used on the cohort map.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Consent level (axis) Required	2 — Section 251 / CAG (pseudonymised, no individual consent)	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “Primary care records of 17,278,392 adults were pseudonymously linked to 10,926 COVID-19-related deaths.” Page: https://doi.org/10.1038/s41586-020-2521-4 Step1 identifiable individuals within the controlled environment: YES (individual-level pseudonymous records). Step2 access without individual study consent under NHS England statutory/IG mechanisms (Data Provision Notice etc.): YES per in-detail governance text -> consent_level=2 per data-dictionary.md § Consent Level (Section 251 / CAG row). NOTE: data-population-plan.md Appendix examples still list OpenSAFELY under level-1 examples; dictionary decision tree takes precedence. OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: Data Provision Notice “Data Provision Notice to GP practices explaining the legal obligation they are under to share patient data to OpenSAFELY.” Page: https://www.opensafely.org/os-in-detail/ Supports Step2 (non-individual-consent lawful processing pathway) consistent with consent_level 2.	Response for Consent level (axis)AccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Characterisation depth (axis) Required	1 — Routine EHR	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 Characterisation_depth=1: reviewed sources describe routine primary-care EHR phenotyping without research-grade MH instrument batteries on the platform core (per data-dictionary.md decision tree step 1).	Response for Characterisation depth (axis)AccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Data availability (axis) Required	1 — Clinical records only	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 DERIVED per data-dictionary.md: modality flags on the cohort row are all FALSE (including has_linked_health_records per overlap rule for an NHS-is-the-dataset platform) -> data_availability=1.	Response for Data availability (axis)AccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Cohort map group (derived) Recommended	Group 4	Internal	Internal catalogue field	—	—
Population type Recommended	Data resource	Not published	DIGIT Cohort Portfolio Review — population type and engagement classification (Aug 2026) Classified data_resource in the DIGIT cohort portfolio review (Aug 2026). Federated analytics platform over NHS primary care records; not a recruited cohort.	Response for Population typeAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Industry engagement status Recommended	Unknown	Not published	Searched — not published UNSOURCEABLE: industry engagement status is an organisational property of the custodian and is not publicly documented; default 'unknown' pending a cohort-owner meeting.	Response for Industry engagement statusAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Parent cohort / programme Recommended	No value held	No source yet	No citation linked to this field yet.	Response for Parent cohort / programmeAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
TWiCs eligible Recommended	No	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: UK Policy Framework for Health and Social Care Research “Gaining appropriate research ethics approval is a prerequisite for using the OpenSAFELY platform.” Page: https://www.opensafely.org/os-in-detail/ Platform access is project-based with ethics; not a TWiCs-eligible consented intervention cohort -> twics_eligible=FALSE.	Response for TWiCs eligibleAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Data modalities

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Questionnaire / interview Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE for the platform archive: validated research questionnaires exist only in separate consented studies (e.g. OpenPROMPT) rather than as standing fields inside the core OpenSAFELY EHR corpus.	Response for Questionnaire / interviewAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Cognitive testing Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE: no named cognitive test battery is evidenced as part of the core platform dataset in reviewed Nature summary.	Response for Cognitive testingAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Physical measures Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE: structured research physical measurement programme not described in reviewed Nature summary (routine clinical measures may exist as EHR entries but are not evidenced as research-grade physical measures flags).	Response for Physical measuresAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Biosamples Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE.	Response for BiosamplesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Genomic data Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE.	Response for Genomic dataAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Omics Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE.	Response for OmicsAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Imaging Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE.	Response for ImagingAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Wearable / sensor Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE.	Response for Wearable / sensorAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Digital phenotyping Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE for the core platform dataset (passive device phenotyping streams are not described here).	Response for Digital phenotypingAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Linked health records Recommended	No	Field-cited	DIGIT — docs/data-dictionary.md (in-repo snapshot) https://raw.githubusercontent.com/avantgarda/digit-capabilities-db/main/docs/data-dictionary.md Section: Cohort Data Type Flags “Use the specific linkage sub-fields (`has_nhs_linkage`, `has_gp_linkage`, etc.) to record details.” Page: https://raw.githubusercontent.com/avantgarda/digit-capabilities-db/main/docs/data-dictionary.md Second excerpt from same table cell as 151002 row: documents where to record GP vs hospital vs death modalities instead of has_linked_health_records. OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: particularly GP records “OpenSAFELY is used in various different ways, but the most important way is with data from the NHS – particularly GP records.” Page: https://www.opensafely.org/os-in-brief/ FALSE for has_linked_health_records per data-dictionary.md overlap rule (FALSE-when: NHS dataset is health records, not linked-to-NHS). Corroboration local_file cohorts/opensafely/digit_data-dictionary_2026-04-16.md. Use has_gp_linkage, has_nhs_linkage, has_death_registry_linkage for modalities.	Response for Linked health recordsAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Biosample types Recommended	No value held	Not published	Searched — not published UNSOURCEABLE / FALSE implied: no biosample bank for the platform core.	Response for Biosample typesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Genotyping platform Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no genotyping platform for the platform core.	Response for Genotyping platformAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Imaging modalities Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no imaging archive for the platform core.	Response for Imaging modalitiesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Mental health measures Recommended	GP-recorded mental disorder diagnoses (Read v2 / CTV3 / SNOMED CT codelists), Primary care prescribing (e.g. BNF-classified antidepressants, antipsychotics, anxiolytics)	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: What using the service actually looks like “to extract and shape data from the available data sets.” Page: https://www.opensafely.org/os-in-detail/ mental_health_measures summarises modalities on the cohort row (coded diagnoses and prescribing): ehrQL extraction sentence plus routine primary-care framing from Nature 2020 (separate rows) — not named PROM batteries on the platform core.	Response for Mental health measuresAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Data subcategories

Structured modality and subcategory rows used alongside the high-level data-type flags.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Data subcategories Recommended List key data subcategories using distinct rows for category markers, named questionnaire/interview instruments, biospecimen types, omics/genomics data products, imaging modalities, sensor streams, and linked-record types.	No value held	No source yet	No citation linked to this field yet.	Response for Data subcategoriesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Field

Current value

Evidence

Source evidence

Response

Notes / correction

Data subcategories

Recommended

List key data subcategories using distinct rows for category markers, named questionnaire/interview instruments, biospecimen types, omics/genomics data products, imaging modalities, sensor streams, and linked-record types.

No value held

No source yet

No citation linked to this field yet.

AccurateNeeds correctionNot sureDoes not apply

Consent and use permissions

Specific use permissions separate from the 3D consent-level axis.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Consent and use permissions Recommended Confirm health/biomedical research use, industry/commercial consideration, non-commercial restrictions, governance modifiers, recontact, trial approach, biosample use, linkage, sharing, international transfer, and any Section 251/CAG support.	No value held	No source yet	No citation linked to this field yet.	Response for Consent and use permissionsAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Field

Current value

Evidence

Source evidence

Response

Notes / correction

Consent and use permissions

Recommended

Confirm health/biomedical research use, industry/commercial consideration, non-commercial restrictions, governance modifiers, recontact, trial approach, biosample use, linkage, sharing, international transfer, and any Section 251/CAG support.

No value held

No source yet

No citation linked to this field yet.

AccurateNeeds correctionNot sureDoes not apply

Access pathways

One or more routes by which researchers can request or analyse data.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Access pathways Recommended Confirm all data access pathways, including route, access level, TRE/SDE or download/API environment, DAC review, collaboration requirements, data sharing agreements, output checking, timelines, and fees.	No value held	No source yet	No citation linked to this field yet.	Response for Access pathwaysAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Field

Current value

Evidence

Source evidence

Response

Notes / correction

Access pathways

Recommended

Confirm all data access pathways, including route, access level, TRE/SDE or download/API environment, DAC review, collaboration requirements, data sharing agreements, output checking, timelines, and fees.

No value held

No source yet

No citation linked to this field yet.

AccurateNeeds correctionNot sureDoes not apply

Within-cohort coverage

Aggregate counts or percentages showing how broadly data, consent, or access characteristics apply within the cohort.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Within-cohort coverage metrics Recommended Provide numerator, denominator, unit type, numerator/denominator exactness qualifiers, denominator type, basis, date/release, original source wording, and whether each metric is marginal, joint, conditional, or a release total.	No value held	No source yet	No citation linked to this field yet.	Response for Within-cohort coverage metricsAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Field

Current value

Evidence

Source evidence

Response

Notes / correction

Within-cohort coverage metrics

Recommended

Provide numerator, denominator, unit type, numerator/denominator exactness qualifiers, denominator type, basis, date/release, original source wording, and whether each metric is marginal, joint, conditional, or a release total.

No value held

No source yet

No citation linked to this field yet.

AccurateNeeds correctionNot sureDoes not apply

Linkage and access

Field	Current value	Evidence	Source evidence	Response	Notes / correction
NHS linkage Recommended	Yes	Field-cited	OpenSAFELY documentation — Data sources overview https://docs.opensafely.org/data-sources/ Section: Data sources overview (toc) “Hospital admissions (hospital spells, outpatients and A&E)” Page: https://docs.opensafely.org/data-sources/ has_nhs_linkage=TRUE: documentation lists hospital admissions (APC) as an available backend dataset class, matching DIGIT has_nhs_linkage (HES / hospital records) intent.	Response for NHS linkageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
GP linkage Recommended	Yes	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 TRUE: platform is anchored on English primary care EHR vendors.	Response for GP linkageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Death registry linkage Recommended	Yes	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “Primary care records of 17,278,392 adults were pseudonymously linked to 10,926 COVID-19-related deaths.” Page: https://doi.org/10.1038/s41586-020-2521-4 TRUE: Nature 2020 OpenSAFELY analysis describes pseudonymous linkage from primary care records to mortality outcomes.	Response for Death registry linkageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Education linkage Recommended	No	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 FALSE: no education record linkage described in reviewed Nature summary.	Response for Education linkageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Data access mechanism Recommended	tre	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: OpenSAFELY in brief (body) “Researchers only get to see insights. They never get unconstrained access to the raw data.” Page: https://www.opensafely.org/os-in-brief/ Mapped to data_access_mechanism=tre (trusted research environment style execution) per controlled environment access.	Response for Data access mechanismAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
TRE platform Recommended	OpenSAFELY (vendor/NHS-hosted secure execution environments)	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: deployed within the secure data centres “The software is deployed within the secure data centres of the two largest electronic health record providers in the NHS: TPP and Optum (formerly Emis).” Page: https://www.opensafely.org/os-in-detail/ tre_platform string names OpenSAFELY execution model (vendor/NHS-hosted secure environments).	Response for TRE platformAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Data dictionary URL Recommended	https://docs.opensafely.org/	Field-cited	OpenSAFELY — OpenSAFELY in detail https://www.opensafely.org/os-in-detail/ Section: Documentation “All our documentation is published in the open.” Page: https://www.opensafely.org/os-in-detail/ data_dictionary_url points to docs.opensafely.org, linked explicitly from in-detail page body and navigation.	Response for Data dictionary URLAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Data access timeline (days) Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: NHS England project approval timelines not quantified as days on reviewed pages.	Response for Data access timeline (days)AccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Data access cost Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no public fee schedule on reviewed pages.	Response for Data access costAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Participant profile and dates

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Age distribution Recommended	No value held	Not published	Searched — not published UNSOURCEABLE in this pass: no Table 1-style age JSON extracted into downloaded files for the platform entity.	Response for Age distributionAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Sex distribution Recommended	No value held	Not published	Searched — not published UNSOURCEABLE in this pass: sex distribution JSON not transcribed from full-text tables.	Response for Sex distributionAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Ethnicity distribution Recommended	No value held	Not published	Searched — not published UNSOURCEABLE in this pass: ethnicity JSON not transcribed (Nature HTML abstract mentions ethnicity but not structured JSON).	Response for Ethnicity distributionAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Smoking status distribution Recommended	No value held	No source yet	No citation linked to this field yet.	Response for Smoking status distributionAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Median age Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no median age field extracted for the platform row.	Response for Median ageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Percent female Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no percent female extracted for the platform row.	Response for Percent femaleAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Percent non-white Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: no percent non-white extracted for the platform row.	Response for Percent non-whiteAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Geographic coverage Recommended	national	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: application/ld+json description “covers 40% of all patients in England” Page: https://doi.org/10.1038/s41586-020-2521-4 geographic_coverage=national because platform scope described as England-wide patient coverage.	Response for Geographic coverageAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Follow-up duration Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: platform has no single follow-up window.	Response for Follow-up durationAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Last data collection date Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: rolling EHR feeds do not present one last_data_collection_date for the platform entity.	Response for Last data collection dateAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Recruitment start date Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: not a recruited cohort.	Response for Recruitment start dateAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Recruitment end date Recommended	No value held	Not published	Searched — not published UNSOURCEABLE: not a recruited cohort.	Response for Recruitment end dateAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Diagnosis areas

Mental health diagnosis areas represented in this cohort.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Diagnosis area: Schizophrenia Recommended Confirm this diagnosis area applies and estimated N if shown.	Schizophrenia · Psychotic Disorders	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: dc.description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 Schizophrenia and other psychotic disorders are routinely coded in English primary care; retained diagnosis_area_id=1 as representative psychotic-disorder EHR phenotyping.	Response for Diagnosis area: SchizophreniaAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Diagnosis area: Major depressive disorder Recommended Confirm this diagnosis area applies and estimated N if shown.	Major depressive disorder · Mood Disorders	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: dc.description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 Depressive disorders are core high-prevalence MH phenotypes in GP-record research; retained diagnosis_area_id=5.	Response for Diagnosis area: Major depressive disorderAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Diagnosis area: Generalised anxiety disorder Recommended Confirm this diagnosis area applies and estimated N if shown.	Generalised anxiety disorder · Anxiety Disorders	Field-cited	Williamson et al. — Factors associated with COVID-19-related death using OpenSAFELY (Nature, 2020) https://doi.org/10.1038/s41586-020-2521-4 Section: dc.description “primary care electronic health records” Page: https://doi.org/10.1038/s41586-020-2521-4 Anxiety disorders are core high-prevalence MH phenotypes in GP-record research; retained diagnosis_area_id=8.	Response for Diagnosis area: Generalised anxiety disorderAccurateNeeds correctionNot sureDoes not apply No response selected	No notes
Diagnosis area: Long COVID neuropsychiatric Recommended Confirm this diagnosis area applies and estimated N if shown.	Long COVID neuropsychiatric · Other	Field-cited	ClinicalTrials.gov https://clinicaltrials.gov Section: conditions “Long COVID” Page: https://clinicaltrials.gov/study/NCT05552612 NEW row: OpenPROMPT (OpenSAFELY-linked) lists Long COVID as condition -> diagnosis_area_id=33.	Response for Diagnosis area: Long COVID neuropsychiatricAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Linked sites

Sites contributing participants or data to this cohort.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Linked sites Optional	No value held	No source yet	No citation linked to this field yet.	Response for Linked sitesAccurateNeeds correctionNot sureDoes not apply No response selected	No notes

Verification metadata

Internal catalogue status — for reference only.

Field	Current value	Evidence	Source evidence	Response	Notes / correction
Verification status Recommended	unverified	Internal	Internal catalogue field	—	—
Last verified Recommended	16 Apr 2026	Internal	Internal catalogue field	—	—
Data source Recommended	Verified Apr 2026 from OpenSAFELY in brief, OpenSAFELY in detail, OpenSAFELY documentation (data sources overview), Williamson et al. Nature 2020 (DOI 10.1038/s41586-020-2521-4), ClinicalTrials.gov API (OpenSAFELY query), ISRCTN API (OpenSAFELY query), Herrett et al. BMJ Open 2023 protocol (OpenPROMPT; 10.1136/bmjopen-2022-071261), DIGIT data-dictionary.md and data-population-plan.md. HDR UK Gateway API returned HTTP 500; DPUK cohort directory URL fetch returned 404 — both logged in entity_sources.	Internal	Internal catalogue field	—	—