Written and maintained by CASRAI Editorial Board
Last updated
UK Biobank is a large-scale prospective cohort study — roughly 500,000 UK volunteers aged 40–69 at recruitment between 2006 and 2010, followed for at least 30 years — that has become one of the most heavily used population-health data resources in biomedical research. For a research administrator or PI evaluating it as a secondary-data source, the practical question is rarely “what is UK Biobank” so much as “what does it actually take to get access, and what does approval commit our institution to.” This guide covers that mechanics: the tiered access model, the Access Management System, cost, timeline, and what an application actually requires.
What data is available
UK Biobank’s holdings go well beyond a single data type, which is part of why it supports such a broad range of research questions from one application:
- Genomic data — whole-genome sequencing across the full cohort, plus genome-wide SNP genotyping (over 800,000 markers) and exome sequencing.
- Imaging — brain, heart, and abdominal MRI; DXA bone density scans; carotid ultrasound; ECG.
- Biological samples — more than 15 million stored blood, urine, and saliva aliquots available for further assay or biomarker measurement.
- Linked health records — NHS hospital inpatient records, cancer registry data, and mortality records, updated on an ongoing basis as participants age.
- Phenotype and lifestyle data — more than 10,000 variables drawn from baseline and follow-up questionnaires covering diet, occupation, mental health, and physical activity.
Who can apply, and on what basis
Access is not limited to UK academic institutions. Researchers from academia, industry, charities, and government bodies worldwide may apply, subject to two conditions that UK Biobank actually checks: the proposed research must be health-related, and it must be judged to be in the public interest — UK Biobank’s access decisions are not a rubber stamp, and applications that fail to articulate a clear health research rationale, or that propose uses inconsistent with participants’ original consent, can be declined.
The Access Management System (AMS) and the application process
Applications are submitted and managed through UK Biobank’s online Access Management System. A complete application requires a defined research protocol (not an open-ended data-exploration request), identification of the specific data fields or categories needed, and named collaborators. Because UK Biobank reviews the scientific and ethical basis of each application rather than granting blanket dataset access, the specificity of the protocol materially affects both approval speed and what gets released — a vague request for “genomic and phenotype data” is a slower, harder approval than a request scoped to the specific fields a pre-registered analysis plan actually needs.
Cost
Unlike a fully open repository such as GenBank, UK Biobank access carries a fee that reflects the cost of curating, hosting, and releasing linked data at this scale; the exact amount depends on the data categories and volume requested, and researchers should budget for it explicitly in a grant application rather than assume the data is free simply because the underlying cohort was publicly funded. Because the fee structure is reviewed periodically, confirm the current schedule directly against UK Biobank’s own access-cost page at proposal-writing time rather than relying on a remembered figure from a prior application cycle.
Where analysis happens: the Research Analysis Platform
Since 2021, UK Biobank has also operated a cloud-based Research Analysis Platform, hosted on Amazon Web Services, that lets approved researchers analyze sensitive data (particularly genomic data) within a secure, UK Biobank-controlled compute environment rather than downloading raw files to institutional infrastructure. This mirrors the same near-the-data analysis logic behind data-commons infrastructure more broadly: for the largest and most sensitive data categories, UK Biobank increasingly expects analysis to happen inside its own environment, not on a downloaded copy.
What approval commits the institution to
An approved application is bound by UK Biobank’s Material Transfer and access agreement terms, which typically require published outputs to be openly accessible, restrict onward redistribution of the raw data, and obligate the research team to destroy or return data at the end of the approved project unless a renewal is granted. A research administrator reviewing an incoming UK Biobank-based proposal should treat these terms as a real compliance obligation on the award, not a formality — the same discipline applied to a dbGaP Data Use Certification or a GISAID Database Access Agreement.
Frequently asked questions
How long does UK Biobank access approval take?
Timelines vary with application complexity and completeness, and UK Biobank publishes current review-cycle guidance directly; a well-scoped application with a clear protocol and named collaborators is consistently faster than an open-ended request, since less clarification back-and-forth is needed during review.
Is UK Biobank data fully anonymized?
Data released to researchers is de-identified, but because it combines genomic, imaging, and linked health-record data at individual level, UK Biobank treats it as sensitive data requiring a signed access agreement and (for the most sensitive categories) analysis within its own Research Analysis Platform, rather than as anonymous data safe for unrestricted redistribution.
Can industry researchers apply, not just academics?
Yes — UK Biobank explicitly accepts applications from industry, charities, and government bodies as well as academia, provided the proposed research is health-related and judged to be in the public interest.
Related resources
UK Biobank’s access model sits in the same broader category as PhysioNet’s credentialed-access system and dbGaP’s Data Access Request process — all three gate sensitive individual-level data behind an application and a binding use agreement rather than either fully open release or a one-off data-sharing negotiation. See also biobank for the general concept UK Biobank is a specific, large-scale instance of.
