prev next front |1 |2 |3 |4 |5 |6 |7 |8 |9 |10 |11 |12 |13 |14 |15 |16 |17 |18 |19 |20 |21 |review
Meta-analysis is essentially synthesis of available literature about a topic. Ideally, synthesis of randomized trials to arrive at a single summary estimate is used. Other types of studies are used for meta-analysis as well.

Begin with a systematic process of identifying your studies. Search your personal files and computerized databases, for example, Medline. Write down the search strategy.

After identifying all studies, define the ones you want to keep for the meta-analysis. This will help another researcher faced with the same body of literature applying the same criteria to find and work with the same studies.

Now create structured formats to key in information taken from the selected studies.

Finally, combine the data to arrive at a summary estimate of the effect, it’s 95% confidence interval, and a test of homogeneity of the studies.