The Role of Meta Analyses and Literature Reviews

For many clinical questions, the practitioner will find that much of the work of reviewing and evaluating research has already been done, or at least addressed, and that summaries of literature exist. The end of this chapter offers a number of readily available sources of comprehensive literature reviews, along with other convenient sources of evidence-based medical knowledge. By way of definition, an overview or review of literature may be any type of summary, whereas the term meta-analysis describes a quantitative summary of results from different studies. Because a meta-analysis attempts to combine results from a number of studies into a summary statistic, it is important to be sure that the studies have a common outcome measure. Also, the underlying assumption is made that the studies are drawn from the same pool or universe of studies and can therefore be legitimately combined. Statistical tests for homogeneity are therefore often used to test this assumption. Also, the validity of a meta-analysis depends upon the assumption that there has not been substantial selective reporting of studies depending on their result (e.g., greater likelihood to report small positive studies than small negative studies). Funnel plots are used to assess this possibility graphically. In a funnel plot, the outcome variable for each study is graphed versus its variance.

Literature reviews may be systematic or rely on the recall of the author. In a systematic review, a complete computer search is made of the relevant literature using specific search terms and prospective rules for inclusion or exclusion of studies found in the systematic search.

To begin evaluating a review, the practitioner should ask two questions: does it ask a carefully focused clinical question, and is the method for including studies reasonable and appropriate? The latter question can be expanded as follows: are methodological standards articulated (for example, those laid out in this chapter), and do the studies chosen address the research question articulated by the reviewer? Stating inclusion criteria up front helps avoid any biases toward preconceived conclusions a reviewer might hold. Aspects of oncology studies to keep in mind when reviewing a review are outcomes (as described previously) and latency periods. Review writers should demonstrate that they exhaustively searched appropriate bibliographic databases (such as Medline), but also that they contacted experts in the area who might be aware not only of published studies not yet appearing in Medline, but of unpublished studies; this is important because studies with negative findings are less likely to reach publication.

Practitioners reading reviews should consider whether results of reasonably comparable studies are similar. Meta-analyses study formally and quantitatively whether results of similar studies differ more than would be expected by chance alone. If so, it is likely that study designs differ enough to account for observed differences in results.

A review should not simply compare the number of positive and negative studies of a given question to obtain an answer, as this fails to give different weight to large and small studies, large studies being more likely to show a positive result because of increased statistical power or efficiency. In addition, such a comparison ignores effects other than the primary outcome of interest, and says nothing about the magnitude of an effect, its clinical importance, or the relatively quality of individual studies.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook


Post a comment