ABSTRACT
There is consequently a need for methods which can be used to compare the results of a number of studies, so-called meta-analysis. A major problem with meta-analysis is that the studies which are to be compared with each other vary in design and execution. A good way of obtaining a basis for comparing the results of various studies is to calculate the confidence intervals for the individual studies and to display them in a figure. There are also methods for carrying out formal comparisons of individual study results, such as by calculating the mean value. It is actually this formal type of comparison which is called meta-analysis. The precondition upon which the calculation of the summary value described, is that every study provides an estimate of one and the same underlying relative risk. Thus, it is not necessary for all the studies to be of the same design.
