This is an archived version. For the current version, please go to training.cochrane.org/handbook/current.

Dichotomous data are described in Chapter 9, Section 9.2.2, and their meta-analysis is described in Chapter 9, Section 9.4.4. The only data required for a dichotomous outcome are the numbers in each of the two outcome categories in each of the intervention groups (the numbers needed to fill in the four boxes S_{E}, F_{E}, S_{C}, F_{C} in Chapter 9, Box 9.2.a). These are entered into RevMan as the numbers with the outcomes and the total sample sizes for the two groups. It is most reliable to collect dichotomous outcome data as the numbers who specifically did, and specifically did not, experience the outcome in each group. Although in theory this is equivalent to collecting the total numbers and the numbers experiencing the outcome, it is not always clear whether the reported total numbers are those on whom the outcome was measured. Occasionally the numbers incurring the event need to be derived from percentages (although it is not always clear which denominator to use, and rounded percentages may be compatible with more than one numerator).

Sometimes the numbers of participants and numbers of events are not available, but an effect estimate such as an odds ratio or risk ratio may be reported, for example in a conference abstract. Such data may be included in meta-analyses using the generic inverse variance method only if they are accompanied by measures of uncertainty such as a standard error, 95% confidence interval or an exact P value: see Section 7.7.7.