Table 3. Current and planned future data available in the UK Biobank resource.
| Data type | Number of participants | Details | Date of data acquisition | Date first available for research‡ |
|---|---|---|---|---|
| Baseline assessment | Whole cohort | Questionnaire, physical measures, samples (see Table 2); haematological assays done on fresh blood samples | 2006–2010 | Q2 2012 |
| Repeat of baseline assessment | 20,000–25,000 | As above every few years, to allow correction for regression dilution due to measurement error and within person fluctuations in exposure levels [12]. | 2013– | Q3 2013 |
| Biochemical assays (of baseline samples) | Whole cohort | Biomarkers with known disease associations (e.g., lipids for vascular disease), diagnostic value (e.g., HbA1c for diabetes), or ability to characterize phenotypes not otherwise well assessed (e.g., renal and liver function tests). | 2014–2015 | 2015 |
| Genotyping (of baseline samples) | Whole cohort | Dense genotyping chip with >800,000 markers including: approximately 250,000 SNPs in a whole-genome array; approximately 200,000 markers covering CNV, loss of function, insertions, deletions, and previously identified risk factor or disease associations; approximately 150,000 exome markers covering a high proportion of non-synonymous coding variants with allele frequency >0.02%. | 2013–2015 | 2015 |
| Dietary Web questionnaire | 210,000 | Automatically coded dietary recall questionnaire, providing estimates of nutrient intake. 80,000 respondents completed it ≥ three times. | 2011–2012 | Q2 2013 |
| Other Web questionnaires | 350,000 to be approached | Participants invited by email to provide additional information via Web questionnaires about exposures (e.g., occupation) and health outcomes (cognitive function, depression) that are not readily identified from health record linkages. | 2014– | 2015 |
| Accelerometry | 100,000 | Wrist-worn tri-axial accelerometers record information on type, intensity, and duration of physical activity. | 2013–2015 | 2015 |
| Multimodal imaging | 100,000 | MRI brain, heart, and abdomen (for lipid distribution); ultrasound of carotid arteries; whole body DXA scan of bones and joints | Pilot phase: 2014–2015 Main phase: 2016–2019 | 2015 |
| Health record linkage | Whole cohort | |||
| Death registrations | ICD-coded cause specific mortality | 2006– | Q2 2013 | |
| Cancer registrations | ICD-coded cancer diagnoses | 1971–* | Q2 2013 | |
| Hospital inpatient episodes | ICD-coded diagnoses, OPCS-coded procedures | 1997–* | Q4 2013 | |
| Hospital outpatient episodes | Limited ICD and OPCS coding | 2003–* | 2015 | |
| Primary care | Read-coded information including diagnoses, measurements, referrals, prescriptions | Variable | 2015 | |
| Other | UK Biobank will obtain data from national mental health care, residential history, laboratory and disease audit datasets and is considering the value of further linkages (e.g., imaging, cancer screening, dental). | Variable | Not yet determined | |
| Adjudicated health outcomes | Whole cohort | Expert-led confirmation and subclassification of outcomes in a range of disease areas, including cancer, diabetes, heart disease, stroke, mental health, musculoskeletal, respiratory, neurodegenerative, and ocular disorders. | 2015 | |
Hb: haemoglobin; SNPs: single nucleotide polymorphisms; CNV: copy number variations; MRI: magnetic resonance imaging; DXA: dual-energy X-ray absorptiometry; ICD: International Classification of Diseases; OPCS: Office of Population Censuses and Surveys Classification of Interventions and Procedures
‡ Future dates are estimated. Data available may be all or part of the relevant dataset.
* available from an earlier date from health record systems in Scotland