University of Ioannina
Statistical models for quantitative evidence synthesis with an emphasis on network meta-analysis and on meta-analysis of diagnostic tests
Abstract
dc:descriptionAs the body of health-related research literature is rapidly growing, systematic reviews and meta-analyses of intervention and diagnostic test accuracy studies provide a transparent overview and summary of the available evidence in each case respectively, improving healthcare decision-making. An important first step is diagnosis: A good diagnosis can guide effective treatment choices. The aim of this thesis was to discuss and address issues related to summarizing evidence from intervention or diagnostic test accuracy studies, with more emphasis on the latter, employing meta-analysis and network meta-analysis. Diagnostic accuracy measures refer to the ability of a test to discriminate between the presence and the absence of a target condition. A variety of diagnostic accuracy measures can be used to quantify this discriminatory ability. In the first part of thesis an introduction to the systematic reviews and diagnostic test accuracy studies is presented; afterwards, the models for conducting a meta-analysis of interventions and diagnostic tests, as well as their key assumptions are discussed. The second part refers to network meta-analysis (NMA) of interventions and the completeness of reporting of published NMAs in the literature, based on the extension of PRISMA statement for NMA. The results of the empirical assessment of changes in reporting before and after the publication of PRISMA extension statement for network meta-analysis are described. Finally, the last part contains the results of a scoping review conducted to identify methodological and application papers reporting the use of a model to conduct diagnostic test network meta-analysis of at least three index tests: the characteristics of the identified methods, a descriptive analysis of characteristics of the application articles, and an application of methods to the available data are presented. Additionally, the P-scores for providing a treatment hierarchy in a network meta-analysis of interventions for a single or more outcomes are described and more particularly, their extension to diagnostic tests is presented.
Degree
thesis:*- Grantor dc:publisher
- University of Ioannina
- Year dc:date
- 2022
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsokani, Sofia
- Τσοκάνη, Σοφία
Subjects
dc:subject × 20- Βιοστατιστική
- Διαγνωστικά τεστ
- Μετα-ανάλυση
- Μετα-ανάλυση δικτύου
- Biostatistics
- Diagnostic tests
- Meta-analysis
- Network meta-analysis
- Μαθηματικά
- Φυσικές Επιστήμες
- Στατιστική και Πιθανότητες
- Επιστήμες Υγείας
- Ιατρική και Επιστήμες Υγείας
- Επιδημιολογία
- Mathematics
- Natural Sciences
- Statistics and Probability
- Health Sciences
- Medical and Health Sciences
- Epidemiology
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/53157