David L. Streiner

Understanding meta-analysis

ing the studies For the meta-analyst, this is where the work begins. Each article must be read and the key elements abstracted into a database. At a minimum, these elements consist of the magnitude of the difference between the groups (called an “effect size”) and the sample size of the study. We may also want to look at any other factor that may affect the magnitude of the findings, such as whether the study was done in a tertiary care setting or a community hospital; the mean age of the women; whether they were a mix of people who had had previous cancers or were all first-timers; the dosage and duration of therapy; and so on. We’ll see later how these factors are used.