Abstract
Background: UK Biobank is a large prospective study that recruited 500,000 participants aged 40 to 69 years, between 2006-2010.The study has collected (and continues to collect) extensive phenotypic and genomic data about its participants. In order to enhance further the value of the UK Biobank resource, a wide range of biochemistry markers were measured in all participants with an available biological sample. Here, we describe the approaches UK Biobank has taken to minimise error related to sample collection, processing, retrieval and assay measurement. Methods: During routine quality control checks, the laboratory team observed that some assay results were lower than expected for samples acquired during certain time periods. Analyses were undertaken to identify and correct for the unexpected dilution identified during sample processing, and for expected error caused by laboratory drift of assay results. Results: The vast majority (92%) of biochemistry serum assay results were assessed to be not materially affected by dilution, with an estimated difference in concentration of less than 1% (i.e. either lower or higher) than that expected if the sample were unaffected; 8.3% were estimated to be diluted by up to 10%; very few samples appeared to be diluted more than this. Biomarkers measured in urine (creatinine, microalbumin, sodium, potassium) and red blood cells (HbA1c) were not affected. In order to correct for laboratory variation over the assay period, all assay results were adjusted for date of assay, with the exception of those that had a high biological coefficient of variation or evident seasonal variability: vitamin D, lipoprotein (a), gamma glutamyltransferase, C-reactive protein and rheumatoid factor. Conclusions: Rigorous approaches related to sample collection, processing, retrieval, assay measurement and data analysis have been taken to mitigate the impact of both systematic and random variation in epidemiological analyses that use the biochemistry assay data in UK Biobank.
Highlights
UK Biobank is a population-based prospective study designed to allow the reliable assessment of a wide range of different types of exposure to multiple diseases, including those that cause much morbidity and disability but have not previously been extensively investigated[1]
The availability of data on a wide range of key biochemistry markers for all 500,000 participants, measured in a highly standardised and systematic manner are a valuable enhancement to UK Biobank, as they enable comparisons of biomarker results across the entire cohort
Epidemiological considerations for the analysis of the biochemistry data In order to help minimise the impact of measurement error in assay data for a very large number of samples, it is important to apply robust quality control (QC) procedures and to identify and mitigate variations that arise
Summary
UK Biobank is a population-based prospective study designed to allow the reliable assessment of a wide range of different types of exposure (including lifestyle, environment and genes) to multiple diseases, including those that cause much morbidity and disability but have not previously been extensively investigated[1]. The assessment visit included the collection of a vast amount of self-reported data via a touchscreen questionnaire and nurse interview, a wide range of physical measures (e.g. blood pressure, anthropometry, spirometry) and biological samples (blood, urine and saliva). Such data depth and breadth was made possible by implementing purposefully-designed, high-throughput processes. This included carefully piloted sample collection and processing protocols designed to collect and store participant samples for maximum scientific return over the long-term, and automation of sample aliquoting and storage to provide a consistent and fully auditable process.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
