This is an archived version of the Handbook. For the current version, please go to training.cochrane.org/handbook/current or search for this chapter here.

9.1.5  What does a meta-analysis entail?

While the use of statistical methods in reviews can be extremely helpful, the most essential element of an analysis is a thoughtful approach, to both its narrative and quantitative elements. This entails consideration of the following questions:

  1. Which comparisons should be made?

  2. Which study results should be used in each comparison?

  3. What is the best summary of effect for each comparison?

  4. Are the results of studies similar within each comparison?

  5. How reliable are those summaries?

The first step in addressing these questions is to decide which comparisons to make (see Section 9.1.6) and what sorts of data are appropriate for the outcomes of interest (see Section 9.2). The next step is to prepare tabular summaries of the characteristics and results of the studies that are included in each comparison (extraction of data and conversion to the desired format is discussed in Chapter 7, Section 7.7). It is then possible to derive estimates of effect across studies in a systematic way (Section 9.4), to measure and investigate differences among studies (Sections 9.5 and 9.6) and to interpret the findings and conclude how much confidence should be placed in them (see Chapter 12).