Cookies on this website

We use cookies to ensure that we give you the best experience on our website. If you click 'Accept all cookies' we'll assume that you are happy to receive all cookies and you won't see this message again. If you click 'Reject all non-essential cookies' only necessary cookies providing core functionality such as security, network management, and accessibility will be enabled. Click 'Find out more' for information on how to change your cookie settings.


Frailty is associated with higher risks of disability and death. Identification of frailty is recommended for routine clinical care of older people. Frailty is strongly correlated with age (10% vs 40% at age 60 vs 80 yr), and the prevalence of frailty in people with a prior history of cardiovascular disease is 4-fold higher among those with versus without prior cardiovascular disease (CVD). Among those without prior CVD, the prevalence of frailty is linearly and positively related to the number of established CVD risk factors. However, there is substantial uncertainty about the natural history of individuals identified with frailty and the role of reversible modifiable risk factors for frailty on subsequent risk of hospitalisation or death.

This DPhil proposal seeks to analyse the determinants and consequences of frailty in a 10-year follow-up of 500,000 adults in the China Kadoorie Biobank (CKB). 

A frailty scale will be developed using electronic records of all hospitalisations in the population studies.  All incident diseases requiring hospitalisation will be identified using linkage to health insurance, disease and mortality registers. The CKB study has almost 300,000 incident events (heart disease: 48,000; stroke: 54,000; Diabetes: 31,000; cancer 27,000; COPD: 19,000, respectively) in the first 10 years of follow-up. Novel software has been developed by CKB to classify individual diseases using disease standardisation programme to assess clustering of diseases to define individuals with frailty. All participants have detailed data on past medical history, lifestyle, socioeconomic factors, physical activity and standard CVD risk factors. Statistical modelling will be used to assess clustering of ICD-10 coded diagnoses using modern statistical approaches including electronic tree structure analysis.


This thesis will (i) examine the associations of cardiovascular risk factors with frailty; (ii) validate diagnosis of frailty using hand-grip strength in random subsets; (iii) assess comorbidity associated with frailty using modern multivariate statistical methods; and (iv) estimate the relative and absolute risks for recurrent hospital admission and death among people with frailty.

The candidate will be expected to have an MSc in Global Health and Epidemiology or other degree in Statistics or Mathematics. The candidate will be supervised by an Epidemiologist and a Statistician and be part of the CKB study group within Nuffield Department of Population Health and Big Data Institute.

This project will involve working within a multi-disciplinary team and the candidate will gain research experience in systematic reviews, study design and planning, epidemiological and modern statistical methods. 


By the end of their DPhil, it is expected that the candidate will be able to plan, undertake, interpret, and report their findings in a clear and concise manner. The candidate will have acquired transferable skills such as the writing of project proposals and presenting the research findings at local, national, and international meetings. The candidate will be expected to publish several peer-reviewed papers as the lead author by the end of their DPhil.


The candidate should have a 2.1 or higher degree in a quantitative subject and an MSc in Global Health Science, Epidemiology, Medical Statistics or Mathematics. The candidate should also have a strong interest in cardiovascular disease epidemiology.