The development of test research methodology

Medical tests are an important part of the armoury of clinicians when diagnosing, monitoring and assessing the risk of patients in health care. Many of these tests have existed for decades and advances in technology have meant a steady flow of more sophisticated tests becoming available. They are an integral part of healthcare.

With the advances made in the development of new tests, the methods used to evaluate, and combine test research have also advanced. In particular, the statistical methodology used to combine test research from multiple primary studies has moved from simple linear models to bivariate random effects models and Bayesian models which are able to accommodate the heterogeneity or variation between different study estimates.

A problem that remains outstanding for clinicians and methodologists is when to translate the results of test research into practice. This requires robust methods many of which are still in their infancy.

The objective of the doctorate would be to investigate existing methods and provide the basis for new methodologies. Other proposals related to this area of research would also be considered.

Applicants should hold or realistically expect to obtain at least an Upper Second Class Honours Degree in a quantitative subject such as mathematics/physics/computer science or statistics and have skills in data analysis.