The starting point of all meta-analyses of studies of effectiveness involves the identification of the data type for the outcome measurements. Throughout this chapter we consider outcome data to be of five different types:
dichotomous (or binary) data, where each individualâ€™s outcome is one of only two possible categorical responses;
continuous data, where each individualâ€™s outcome is a measurement of a numerical quantity;
ordinal data (including measurement scales), where the outcome is one of several ordered categories, or generated by scoring and summing categorical responses;
counts and rates calculated from counting the number of events that each individual experiences; and
time-to-event (typically survival) data that analyse the time until an event occurs, but where not all individuals in the study experience the event (censored data).
The ways in which the effect of an intervention can be measured depend on the nature of the data being collected. In this section we briefly examine the types of outcome data that might be encountered in systematic reviews of clinical trials, and review definitions, properties and interpretation of standard measures of intervention effect. In Sections 9.4.4.4 and 9.4.5.1 we discuss issues in the selection of one of these measures for a particular meta-analysis.