Toward valid network meta-analysis: Challenges in clustering conservative interventions for musculoskeletal conditions

A network meta-analysis (NMA) is a statistical technique that combines and compares multiple interventions from direct and indirect evidence in one analysis, ranks them according to their effectiveness, and assists clinicians in making better-informed decisions within an EBP framework. Conducting an NMA is challenging and even harder for musculoskeletal conditions because of the complex pathophysiology and the abundance of complex conservative interventions comprising multiple components, each affecting the effectiveness of the intervention. Evidence for how nodes (clustering of interventions in a homogenous group) are constructed in an NMA remains largely unexplored and poorly reported. The clustering of interventions should be transparent, and the underlying assumptions of transitivity, including homogeneity and consistency, must be assessed accurately. Treatment effect modifiers should be balanced across the included study populations for a valid NMA. This thesis aims to address the issue of developing transparent and clinically relevant conservative intervention nodes for musculoskeletal conditions when designing an NMA. Neck pain is used as an example to identify methods and procedures to address the challenges of designing and conducting NMAs of conservative interventions for musculoskeletal conditions. The first study is a scoping review that showed wide variation in neck pain classification and conservative interventions and the challenges in grouping interventions in homogenous groups or nodes. A systematic review was conducted to explore these challenges further, evaluating how conservative interventions are grouped into nodes and how the assumption of transitivity has been reported in published NMAs for musculoskeletal conditions. The results indicate that node definitions and node-making methods for conservative interventions are not well outlined in NMAs for musculoskeletal conditions. To resolve the issue of how to combine conservative interventions into clinically relevant nodes, the node definitions developed in the scoping review were shared with neck pain experts to obtain external validation, feedback and finally establish consensus in a Delphi study. To further examine the reporting of interventions that possibly influence selecting effect modifiers in studies for patients with neck pain, a secondary analysis was conducted of massage studies from the scoping review, with additional studies from a search in PubMed. The findings of chapters two to five collectively provide a potential framework for developing clinically relevant nodes to conduct an NMA. The final study of this thesis is a protocol that describes the methods for an NMA for neck pain that builds on the previous work and concludes with a discussion and suggestions for future research.

Author dc:creator

• Ishaq, Iqra

Statement dc:rights

• info:eu-repo/semantics/openAccess
• The author owns the copyright in this thesis including all reproduction and reuse rights for the work. The work may not be altered without the permission of the copyright owner. Attribution is essential when quoting or paraphrasing from this thesis.
• © 2023 Iqra Ishaq
• au.edu.uts.lib/cph

Language dc:language.iso

en_US

Handle dc:identifier.uri

http://hdl.handle.net/10453/177602

OAI identifier oai:identifier

oai:opus.lib.uts.edu.au:10453/177602