Understanding and modelling the geographical variation in relative risks for smoking and other major risk factors for burden of disease analyses
Smoking is a leading global risk factor for premature death and disability. Estimates of the burden of disease attributable to major risk factors such as smoking are critical for policymaking, resource allocation, intervention development, and health system planning, but current approaches to calculating such estimates are still far from precise, particularly for low- and middle-income countries. For example, analyses for the Global Burden of Disease (GBD) study continue to rely predominantly on estimates of the effects of exposures from studies conducted in high-income countries, despite several studies demonstrating significant variation in relative risks by geography. This variation may be the result of differences in exposure patterns, effect modification by other common risk factors, and genetics, but this requires investigation by standardised analyses across geographically diverse cohort studies.
The aims of this project are:
- to understand the relevance of differences in smoking exposure patterns to the observed geographic variation in relative risks;
- assess whether effect modification by other major risk factors may account for some of the variation; and,
- given these findings, refine the methodology currently used to estimate the burden of disease from smoking worldwide.
This project focuses on smoking but may be expanded to include other major risk factors (such as adiposity and alcohol). It will inform an analytic framework and statistical software that can be applied to other risk factors in burden of disease analyses, ultimately better informing local clinical practice, public health interventions, and policymaking.
RESEARCH EXPERIENCE, RESEARCH METHODS AND TRAINING
This project will involve detailed analysis and interpretation of existing data from large-scale prospective studies. The student will work within a multi-disciplinary team and will gain research experience in literature review, epidemiological and statistical methodology, programming and data analysis. Regular research meetings and workshops will be held which the candidate will be expected to attend and to present their research findings.
FIELD WORK, SECONDMENTS, INDUSTRY PLACEMENTS AND TRAINING
The project will provide a range of training opportunities in statistical analysis, data visualisation and interpretation, and in statistical programming. By the end of the DPhil, it is expected that the candidate will be competent to plan, undertake and interpret statistical analysis of large-scale epidemiological data, and to report their findings in a clear and concise manner. The project will be based in CTSU, Nuffield Department of Population Health, which has excellent facilities and a world-class community of statistical, epidemiological and clinical scientists.
Candidates should have a strong background in a mathematical or biomedical discipline and postgraduate training in epidemiology, statistics or public health (or be willing to do the MSc in Global Health Science and Epidemiology at Oxford in preparation for the project). The project will involve large-scale data and statistical analyses. Candidates should therefore have an interest and aptitude in extending these skills as well as a strong interest in non-communicable disease epidemiology.