Funnel plots, and tests for funnel plot asymmetry, have been widely used to examine bias in the results of meta-analyses. Funnel plot asymmetry should not be equated with publication bias, because it has a number of other possible causes. This article describes how to interpret funnel plot asymmetry, recommends appropriate tests, and explains the implications for choice of meta-analysis model

Objective Meta-analysis is of fundamental importance to obtain an unbiased assessment of the available evidence. In general, the use of meta-analysis has been increasing over the last three decades with mental health as a major research topic. It is then essential to well understand its methodology and interpret its results. In this publication, we describe how to perform a meta-analysis with the freely available statistical software environment R, using a working example taken from the field of mental health. Methods R package meta is used to conduct standard meta-analysis. Sensitivity analyses for missing binary outcome data and potential selection bias are conducted with R package metasens. All essential R commands are provided and clearly described to conduct and report analyses. Results The working example considers a binary outcome: we show how to conduct a fixed effect and random effects meta-analysis and subgroup analysis, produce a forest and funnel plot and to test and adjust for funnel plot asymmetry. All these steps work similar for other outcome types. Conclusions R represents a powerful and flexible tool to conduct meta-analyses. This publication gives a brief glimpse into the topic and provides directions to more advanced meta-analysis methods available in R.

Network meta-analysis is used to compare three or more treatments for the same condition. Within a Bayesian framework, for each treatment the probability of being best, or, more general, the probability that it has a certain rank can be derived from the posterior distributions of all treatments. The treatments can then be ranked by the surface under the cumulative ranking curve (SUCRA). For comparing treatments in a network meta-analysis, we propose a frequentist analogue to SUCRA which we call P-score that works without resampling. P-scores are based solely on the point estimates and standard errors of the frequentist network meta-analysis estimates under normality assumption and can easily be calculated as means of one-sided p-values. They measure the mean extent of certainty that a treatment is better than the competing treatments. Using case studies of network meta-analysis in diabetes and depression, we demonstrate that the numerical values of SUCRA and P-Score are nearly identical. Ranking treatments in frequentist network meta-analysis works without resampling. Like the SUCRA values, P-scores induce a ranking of all treatments that mostly follows that of the point estimates, but takes precision into account. However, neither SUCRA nor P-score offer a major advantage compared to looking at credible or confidence intervals.

The heterogeneity statistic I2, interpreted as the percentage of variability due to heterogeneity between studies rather than sampling error, depends on precision, that is, the size of the studies included.

Standard generic inverse variance methods for the combination of single proportions are based on transformed proportions using the logit, arcsine, and Freeman‐Tukey double arcsine transformations. Generalized linear mixed models are another more elaborate approach. Irrespective of the approach, meta‐analysis results are typically back‐transformed to the original scale in order to ease interpretation. Whereas the back‐transformation of meta‐analysis results is straightforward for most transformations, this is not the case for the Freeman‐Tukey double arcsine transformation, albeit possible.In this case study with five studies, we demonstrate how seriously misleading the back‐transformation of the Freeman‐Tukey double arcsine transformation can be. We conclude that this transformation should only be used with special caution for the meta‐analysis of single proportions due to potential problems with the back‐transformation. Generalized linear mixed models seem to be a promising alternative.

This is an updated systematic review of 57 trials and 9353 cancer patients from articles, abstracts, and reports published between January 1, 1985, and April 30, 2005, on the effects of epoetin alfa and beta (i.e., epoetin) and darbepoetin alfa (i.e., darbepoetin). We included randomized controlled trials comparing epoetin or darbepoetin plus red blood cell transfusion with red blood cell transfusion alone for prophylaxis or treatment of anemia in cancer patients with or without concurrent antineoplastic therapy. The Cochrane Library, MEDLINE, EMBASE, and conference proceedings were searched. Effect estimates and 95% confidence intervals (CIs) were calculated with fixed-effects models. Treatment with epoetin or darbepoetin statistically significantly reduced the risk for red blood cell transfusions (relative risk [RR] = 0.64, 95% CI = 0.60 to 0.68; 42 trials and 6510 patients) and improved hematologic response (RR = 3.43, 95% CI = 3.07 to 3.84; 22 trials and 4307 patients). Treatment with epoetin or darbepoetin increased the risk of thrombo-embolic events (RR = 1.67, 95% CI = 1.35 to 2.06; 35 trials and 6769 patients). Uncertainties remain as to whether and how epoetin or darbepoetin affects overall survival (hazard ratio = 1.08, 95% CI = 0.99 to 1.18; 42 trials and 8167 patients). Caution is advised when using epoetin or darbepoetin in combination with thrombogenic chemotherapeutic agents or for cancer patients who are at high risk for thrombo-embolic events.