Can meta-analyses be trusted?

@article{Thompson1991CanMB,
  title={Can meta-analyses be trusted?},
  author={Simon G Thompson and Stuart J. Pocock},
  journal={The Lancet},
  year={1991},
  volume={338},
  pages={1127-1130}
}

A practical guide to meta-analysis.

The logic and role of meta-analysis in clinical research

The term meta-analysis refers to the quantitative combination of data from independent trials. Where the result of such combination is a descriptive summary of the weight of the available evidence,

Systematic Review: Why sources of heterogeneity in meta-analysis should be investigated

This paper distinguishes between the concepts of clinical and statistical heterogeneity and exemplifies the importance of investigating heterogeneity by using published meta-analyses of epidemiological studies of serum cholesterol concentration and clinical trials of its reduction.

Meta-Analysis: A Statistical Method to Integrate Information Provided by Different Studies

Meta-analyses are a supplement to RCTs, which remain the gold standard for evaluating the efficacy of therapeutic interventions, but need to be supplemented by meta-analyses in order to broaden the

Meta-Analysis: Methodology, Utility, and Limitations

This article will focus on the components of a properly performed meta-analysis and the many sources of bias inherent in a literature review of this magnitude.

Meta-Analysis

Meta-analysis has been employed in many clinical settings to evaluate efficacy and safety of a variety of therapeutic interventions and it is likely that it will continue to have a role in extrapolating data from clinical trials for use in the clinic.

A guide to understanding meta-analysis.

The purpose of this commentary is to assist clinicians in understanding meta-analysis as a statistical tool using both published articles and explanations of components of the technique.

The use and misuse of meta‐analysis in clinical medicine

In this chapter, it is demonstrated, using data from the literature, how such a study can be carried out and analysed, and what data are needed and some of the statistical methods that can be used are described.

The uses and abuses of meta-analysis.

Meta-analysis is a quantitative process of summary and interpretation which involves pooling information from independent studies concerning a single theme in order to draw conclusions and can only be directly applied to a target population when the ' meta-protocol' and 'meta-population' match the target situation in all relevant particulars.

Systematic Reviews and Meta-Analyses in Surgery

This chapter provides a structural framework to perform a meta-analysis that guides the clinician on a journey from the identification of the correct clinical question to data analysis and through to producing a structured report and highlights the limitations and pitfalls associated with the meta-analytical technique.
...

References

SHOWING 1-10 OF 18 REFERENCES

Meta-analysis in clinical trials.

A general parametric approach to the meta-analysis of randomized clinical trials.

A general parametric approach is presented, which utilizes efficient score statistics and Fisher's information, and relates this to different methods suggested by previous authors.

A cohort study of summary reports of controlled trials.

The hypotheses that clinical trials would be more likely to be followed by full reports if, on the basis of the information provided in the summary report, the trial was judged to be methodologically sound, the results favored the test treatment, and the sample size was relatively large are tested.

The potential and limitations of meta-analysis.

The differing applications and limitations of meta analysis are discussed, with a review of the analytic methods used and the problems and biases encountered.

Publication bias and clinical trials.

Statistical aspects of the analysis of data from retrospective studies of disease.

The role and limitations of retrospective investigations of factors possibly associated with the occurrence of a disease are discussed and their relationship to forward-type studies emphasized.

Confidence intervals rather than P values: estimation rather than hypothesis testing.

Some methods of calculating confidence intervals for means and differences between means are given, with similar information for proportions, and the paper also gives suggestions for graphical display.