Box 2Cross-cutting methodological issues relevant to meta-analytic practice

  • Comparative effectiveness reviews of medical tests48
    -

    Often many alternative tests are applicable to a given testing scenario; systematic reviews may want to consider all relevant index tests that are applicable to the population and disease of interest.

    -

    In cases where multiple index tests are applicable, reviews that directly compare test performance may have the greatest impact on clinical practice.

  • Defining the setting and role of test use
    -

    The findings of systematic reviews can be meaningfully applied to clinical practice only if the role (add-on, triage, replacement) and setting (screening, diagnosis, prognosis/prediction, treatment selection) of test use is explicitly considered. These aspects of test use have implications for the study designs to be considered, the information to be extracted from each eligible study, the interpretation of individual study results, and the synthesis of findings across studies.49

  • Methods and reporting of systematic reviews of test performance50-53
    -

    A priori defined protocols, clearly delineating the scope of the review and outlining the proposed methods is in accordance with commonly held standards of research conduct, and probably applies to systematic reviews as much as other research enterprises.

    -

    Explicitly reporting the methodological approach followed by systematic reviews (including any deviations from the review protocol) promotes clarity. It is probably good practice for reviewers to consult existing (and continuously evolving) guidance on the optimal methods for searching the literature, identifying and selecting relevant studies, extracting data, assessing the validity of included studies, qualitatively and quantitatively synthesizing study results.

    -

    The assessment of study “quality” or “risk of bias” is an important component of systematic reviews of medical tests – yet exactly how these assessments (should) affect the conclusions of the systematic review is still a matter of research.

  • Examining the applicability of review findings and transferability of estimates
    -

    Based on our empirical assessment, many reviews did not adequately describe whether the applicability of research findings from individual studies was assessed, and if yes, how. Information for the assessment of applicability includes, but is not limited to, details about the index and reference standard tests used, and the demographics and disease-related characteristics of the population enrolled in each study. Assessment of applicability is critical in contextualizing the conclusions of a systematic review.

  • Interpreting and contextualizing review results
    -

    Reviews of test accuracy address an intermediate component of the effect of tests on clinical outcomes.54 Although studies assessing the overarching question of test effectiveness on clinical outcomes are rare, reviews of test performance often can only provide part of the information needed to fully assess the impact of tests.55 Other intermediate outcomes that could be considered include the impact of test results on physicians' diagnostic thinking, and on therapeutic decisionmaking.56

    -

    In the absence of studies assessing the direct effect of tests on clinical outcomes, formal modeling or simple (“back-of-the-envelope”) projections of the potential impact of tests may be informative.57

From: Discussion

Cover of Comprehensive Overview of Methods and Reporting of Meta-Analyses of Test Accuracy
Comprehensive Overview of Methods and Reporting of Meta-Analyses of Test Accuracy [Internet].
Dahabreh IJ, Chung M, Kitsios GD, et al.

NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.