Authors

  1. Blegen, Mary A.

Article Content

Building a body of useful knowledge takes many individual studies followed by an integration of those studies to identify the pieces of knowledge that have strong support and those with only weak support. That second step is as important as the first. Systematic integrative reviews then provide the basis for practice guidelines and strategic planning for future research. Systematic reviews include attempts to summarize all empirical evidence that addresses a specific research question. Meta-analyses use statistical techniques to summarize the results of the studies in the systematic review.

 

Learning how to integrate research results with systematic reviews and meta-analyses has moved from an optional, advanced course in research training to a prominent place in basic research courses. A new set of guidelines is now available to guide instruction in and the production and publication of systematic reviews and meta-analyses (Liberati et al., 2009). These guidelines were produced by a multidisciplinary, multinational group seeking to "ensure the transparent and complete reporting of systematic review and meta-analyses" (p. W74). They have been endorsed by Nursing Research and a long list of organizations and journals in healthcare.

 

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines contain a 27-item checklist-addressing seven elements including title, abstract, introduction, methods, results, discussion, and funding-and a four-step flow diagram (http://prisma-statement.org/statement.htm). These directly address writing reports of these reviews and inform the conduct of studies beginning with the search for the literature, continuing through assessment of bias in the studies and analytic approach, and ending with the interpretation of results. The PRISMA guidelines supersede the Quality of Reporting of Meta-analyses (QUOROM) statement (Moher et al., 1999) that spoke only to meta-analyses.

 

The PRISMA group emphasized the assessment of bias in the systematic review itself as well as in each individual study included in the review. Risk of bias originates in the clinical and methodological aspects of each study and should be rigorously assessed with a standardized assessment tool. In addition, the potential risk of bias of the systematic review itself should be assessed and reported. These risks for bias focus on the potential for non publication of original studies or incomplete reporting of results in the published works. Systematic reviews and meta-analyses have become vital contributions to scientific knowledge and the new PRISMA guidelines will improve the quality of this work.

 

Mary A. Blegen, RN, PhD, FAAN

 

Associate Editor

 

[email protected]

 

References

 

Liberati, A., Altman, D. G., Tetzlaff, J., Mulrow, C., Gotzsche, P. C., Ioannidis, J. P., et al. (2009). The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate health care interventions: Explanation and elaboration. Annals of Internal Medicine, 151(4), W65-W94. [Context Link]

 

Moher, D., Cook D. J., Eastwood, S., Olkin, I., Rennie, D., et al, for the QUOROM group. (1999). Improving the quality of reporting of meta-analysis of randomized controlled trials: The QUOROM statement. Lancet, 354, 1896-1900. [Context Link]