China Kadoorie Biobank (CKB) prospective study of 0.5 million adults

Lead Research Organisation: University of Oxford

Abstract

Chronic diseases, such as stroke, heart disease, and cancer, are the leading causes of disability and death worldwide. Understanding what causes these diseases in different populations can lead to improvements in disease prevention, risk prediction and development of new treatments. One way of identifying causes of disease is through blood-based “prospective cohort” studies, in which large numbers of apparently healthy individuals from the general population are interviewed and measured, and have blood collected and tested for genetic and non-genetic biomarkers. The health status of the study participants is then monitored “prospectively” to see who develops what disease. When sufficiently large numbers of people have developed a particular disease (“cases”), their blood and other characteristics are compared with those from “controls” who have not developed the disease. We have established one of the world’s largest studies of this kind, involving 512,000 adults in China. Ongoing follow-up work ensures that disease outcomes are correctly identified, checked and carefully classified before the data is analysed by researchers. We are also expanding the use of this valuable resource by the wider research community, so that many important new findings will emerge which will enable better disease prevention and treatment, benefiting populations worldwide.

Technical Summary

China Kadoorie Biobank is a blood-based prospective study of 512,000 adults, recruited during 2004-8 from 10 diverse regions of China, with extensive data collected at baseline and subsequent resurveys using questionnaires, physical measurements, and stored biological samples.
Chronic diseases are major causes of death and disability in China and worldwide. Large-scale prospective studies of multiple risk factors, with blood storage for testing novel hypotheses and detailed follow-up of health outcomes, will help assess the causes of chronic diseases. Studies in China, given the special patterns of exposures and diseases, will yield novel findings, complementing similar studies elsewhere (e.g. UK Biobank). The CKB will improve understanding of disease aetiology, risk prediction and the development of new therapies, benefitting populations worldwide.
By Q4 2016, follow-up through linkage to death and disease registries and to nationwide health insurance databases of all hospitalised events has accumulated 0.5 million hospitalised events and 35,500 deaths among CKB participants, including incident cases of stroke (42,000), IHD (35,000), cancer (23,000), diabetes (13,500) and COPD (9,500). Only ~4,000 (<1%) participants were lost to follow-up. Genome-wide data are being generated (first phase: ~100,000 participants by Q3/2016), along with blood biochemistry and multi-omics data for nested case-control studies of specific diseases.
CKB will prioritise (i) continued follow-up of cause-specific morbidity and mortality and hospital records through electronic linkage to health insurance systems; (ii) validation, clinical adjudication and detailed sub-phenotyping for selected diseases (e.g. stroke, IHD, cancer); (iii) maintenance and management of extensive and uniquely large and complex datasets; (iv) enhancement of collaboration and data sharing with the wider scientific community.
Future research based on data and samples collected to date is likely to lead to the identification of novel genetic and blood-based biomarkers involved in various chronic diseases. These can help develop improved algorithms for disease risk prediction. Biomarkers associated with early onset and/or progression of disease may also be used for early diagnosis and prediction of prognosis. Improved understanding of disease pathways and mechanisms may lead to the identification of novel therapeutic and pharmacogenetic targets that may be patentable or commercially exploitable.
Health insurance data already collected (2006-2016), includes 600,000 coded hospitalised disease events of many different types and 60 million in-hospital treatment, test and procedures-related data. Analysis of this will help evaluate the delivery of effective healthcare by quantifying over-hospitalization and inappropriate use of investigative procedures and treatments for common causes of hospitalizations.

Publications


10 25 50