A statistical analysis conducted according to Bayesian principles. It involves incorporation of existing information regarding the likelihood of an event (i.e., “priors”) to estimate the likelihood based on additional information (i.e., “posteriors”).

A type of network in which all elements are connected to one another.

The interval in which a population parameter lies, based on a random sample of the population. The most commonly reported confidence interval is the 95% confidence interval.

In Bayesian statistics, an interval in which the actual value of a parameter of interest lies with a defined probability.

A measure of model comparison and accuracy. Smaller DIC values indicate a better-fitting model, with a difference greater than two indicating a much better-fitting model.

The extent to which an intervention, procedure, regimen, or service produces the intended outcomes when deployed under routine (“real world”) circumstances.

Methods of fixed effects meta-analysis are based on the mathematical assumption that a single common (or “fixed”) effect underlies every study in the meta-analysis. In other words, if we were doing a meta-analysis of odds ratios, we would assume that every study is estimating the same odds ratios. Under this assumption, if every study were infinitely large, every study would yield an identical result. This is the same as assuming there is no (statistical) heterogeneity among the studies.

Variation in treatment effects between RCTs within a pairwise contrast. Heterogeneity is likely to occur if trials have been undertaken on different patient groups, and/or different settings and/or methodological differences in the design and conduct of the trials.

Statistical synthesis of the results of individual studies that examine the same question to produce a single estimate of effect.

A Bayesian approach that combines direct and indirect evidence in a single analysis, thus enabling simultaneous comparison of multiple treatment interventions.

A distribution that will not influence the posterior distribution.

A distribution that embodies both the prior distribution and the observed data information.

A distribution that expresses information available to the researcher before any “data” are involved in the statistical analysis.

A random effects analysis makes the assumption that individual studies are estimating different treatment effects. In order to make some sense of the different effects they assume, they have a distribution with some central value and some degree of variability.

A prospective experimental study designed to test the efficacy of an intervention in which patients are randomly allocated to either a treatment group or the control group.

A measure of the variability or spread of the data.

A summary of the medical literature that uses explicit methods to identify, select, appraise, and analyze studies relevant to a particular clinical question.

Episodes of stroke symptoms that last only briefly.