Table 4.4.

Common methods for quantitative syntheses

	Statistical method	Reported data	Presentation
Meta-analysis
Aggregate data^a Individual participant data^c	Weighted average of effect estimates	Pairwise comparisons of effect estimates, CI Overall effect estimate, CI, P value Evaluation of heterogeneity	Forest plot^b with summary statistic for average effect estimate
Network^a	Variable^d	The interventions, which are compared directly versus indirectly	Network diagram or graph, tabular presentations
		Comparisons of relative effects between any pair of interventions	Effect estimates for intervention pairings
		Summary relative effects for pair-wise comparisons with evaluations of inconsistency and heterogeneity	Forest plot, other methods
		Treatment rankings (ie, probability that an intervention is among the best options)	Rankogram plot
Synthesis without meta-analysis ^e	Summarizing effect estimates from separate studies (without combination that would provide an average effect estimate)	Range and distribution of observed effects such as median, interquartile range, range	Box-and-whisker plot, bubble plot Forest plot (without summary effect estimate)
	Combining P values	Combined P value, number of studies	Albatross plot (study sample size against P values per outcome)
	Vote counting by direction of effect (eg, favors intervention over the comparator)	Proportion of studies with an effect in the direction of interest, CI, P value	Harvest plot, effect direction plot

CI, confidence interval (or credible interval, if analysis is done in Bayesian framework).

^{^a}

See text for descriptions of the types of data combined in each of these approaches.

^{^b}

See Supplementary File 4 (http://links.lww.com/JBJSREV/A980) for guidance on the structure and presentation of forest plots.

^{^c}

General approach is similar to aggregate data meta-analysis but there are substantial differences relating to data collection and checking and analysis¹⁶².

This approach to syntheses is applicable to intervention, diagnostic, and prognostic systematic reviews¹⁶³.

^{^d}

Examples include meta-regression, hierarchical and multivariate approaches¹⁶⁴.

^{^e}

In-depth guidance and illustrations of these methods are provided in Chapter 12 of the Cochrane Handbook¹⁶⁰.