Network meta-analysis of diagnostic test accuracy studies via factor copula mixed models (NIKOLOULOPOULA_U23CMP)

The identification of the most accurate diagnostic test for a particular disease contributes to the prevention of unnecessary risks to patients and healthcare costs. Clinical and policy decisions are usually made on the basis of the results from many diagnostic test accuracy studies on the same research question. The considerably large number of diagnostic test accuracy studies has led to the use of meta-analysis. The purpose of a meta-analysis of diagnostic test accuracy studies is to combine information over different studies and provide an integrated analysis that will have more statistical power to detect an accurate diagnostic test than an analysis based on a single study. As the accuracy of a diagnostic test is commonly measured by a pair of indices such as sensitivity and specificity, synthesis of diagnostic test accuracy studies is the most common medical application of multivariate meta-analysis. Most of the existing meta-analysis models and methods have mainly focused on a single test. As the meta-analysis of more than one diagnostic test can impact clinical decision-making and patient health, there is an increasing body of research in models and methods for meta-analysis of studies comparing multiple diagnostic tests. The application of the existing models to compare the accuracy of three or more tests suffers from the curse of multi-dimensionality. To overcome these issues in network meta-analysis of studies comparing multiple diagnostic tests, we will study parsimonious copula mixed models for comparing multiple diagnostic tests that can incorporate studies with different designs and studies with our without a gold standard. For the between-studies model, we will exploit the use of factor copula distributions. Factor copulas can provide a wide range of dependence and allow for different types of tail behaviour, different from assuming simple linear correlation structures, normality and tail independence.

Entry requirements & funding notes

This PhD project is in a competition for studentships allocated to the School of Computing Sciences as a direct result is increased PGT student fee income for the MSc Courses in Cyber Security, Data Science and Computing Sciences.  All successful candidates will be expected to support PGT Lab sessions from October 2023 and related activities as allocated in support of these programmes within the working hours permitted for full-time Postgraduate Researchers.

Funding comprises ‘home’ tuition fees and an annual stipend (2022/23 rate is £17,668, 2023/24 tbc) for a maximum of 3 years.

Applications are welcomed, and funding is available, to UK applicants only who have the right to work in the UK. 

REFERENCES

Nikoloulopoulos, A. K. (2015). A mixed effect model for bivariate meta-analysis of diagnostic test accuracy studies using a copula representation of the random effects distribution. Statistics in Medicine, 34:3842– 3865.

Nikoloulopoulos, A. K. (2019). A D-vine copula mixed model for joint meta-analysis and comparison of diagnostic tests. Statistical Methods in Medical Research, 28(10-11):3286–3300.

Nikoloulopoulos, A. K. (2020a). A multinomial quadrivariate D-vine copula mixed model for meta-analysis 

of diagnostic studies in the presence of non-evaluable subjects. Statistical Methods in Medical Research, 29(10):2988–3005.

Nikoloulopoulos, A. K. (2020b). A multinomial 1-truncated D-vine copula mixed model for meta-analysis and comparison of multiple diagnostic tests. ArXiv e-prints. arXiv:2010.08152.

Nikoloulopoulos, A. K. (2022). An one-factor copula mixed model for joint meta-analysis of multiple diagnostic tests.  Journal of the Royal Statistical Society: Series A (Statistics in Society), 185: 1398-1423.