Network Meta-Analysis for Complex Social Interventions: Problems and Potential

Abstract

When multiple interventions exist for the same problem, evidence-based decision making requires clear and systematic syntheses of available evidence to determine which interventions are most effective. Network meta-analysis extends standard pairwise meta-analytic models to allow direct and indirect evidence from primary studies comparing alternative interventions to be statistically combined in a single analysis. The use of indirect evidence can increase the precision of estimates and allow the relative effectiveness of interventions to be estimated even if particular interventions have never been directly compared in primary studies. The validity of network meta-analyses requires assumptions that may be difficult to verify in reviews of complex social interventions, and the method has rarely been applied outside of medicine. This paper provides a nontechnical introduction to network meta-analysis and to some of the advantages and challenges involved in its application to complex social interventions. We conclude that network meta-analysis is an important and needed tool for social researchers. However, we also find that applications to complex social interventions require more extensive and theory-driven assessments of validity than those needed for either pairwise meta-analyses or for applications to biomedical treatments. Existing guidance on reporting of network meta-analyses could be usefully expanded to require more detailed reporting of these additional measures in the case of complex interventions.

Dressings and topical agents for treating pressure ulcers

“…'Transitivity' refers to the situation in which an intervention effect measured using an indirect contrast is valid and equivalent to the intervention effect measured using a direct contrast. Where there are differences in (known or unknown) effect modifiers across contrasts, the transitivity assumption may not be met which may generate statistical inconsistency in the network (Grant 2013;Jansen 2013). We did not identify any potential effect modifiers from the literature, so there was no evidence that the transitivity assumption was not met.…”

“…the combination of trials comparing A with C and trials comparing B with C), and a synthesis of both when available. When pooling relative effect estimates, NMAs preserve within-trial randomisation (Grant 2013;Thorlund 2012;Tu 2012).…”

Network meta-analysis in health psychology and behavioural medicine: a primer

“…However, it is clearly not feasible to conduct such a study, as the complexity of the study design and the resources required would be too great (Catalá-López, Tobías, Cameron, Moher, & Hutton, 2014). For example, whereas several types of behaviour change interventions are known to be effective in reducing blood pressure, including increased physical activity, smoking cessation and dietary modifications (Mancia et al, 2013), it would be impractical to attempt implementing even one multi-arm RCT that compared the effects of changes to one of these behaviours on blood pressure, let alone an RCT that compared the different techniques used to change each of these behaviours (Grant & Calderbank-Batista, 2013). Furthermore, even if such complex studies could be conducted, the pairwise evidence synthesis methods normally employed in health psychology could not coherently synthesise their results.…”

Recent Research Trends in Meta-analysis

“…Network MA requires special assumption and analysis methods, such as heterogeneity, transitivity, and consistencies. Furthermore, it is also applicable to social science research fields [8]. Additionally, network MA can be used to provide valuable information to patients, practitioners, and decision makers.…”