Some types of event can happen to a person more than once, for example, a myocardial infarction, fracture, an adverse reaction or a hospitalization. It may be preferable, or necessary, to address the number of times these events occur rather than simply whether each person experienced any event (that is, rather than treating them as dichotomous data). We refer to this type of data as count data. For practical purposes, count data may be conveniently divided into counts of rare events and counts of common events.

Counts of rare events are often referred to as ‘Poisson data’ in statistics. Analyses of rare events often focus on rates. Rates relate the counts to the amount of time during which they could have happened. For example, the result of one arm of a clinical trial could be that 18 myocardial infarctions (MIs) were experienced, across all participants in that arm, during a period of 314 person-years of follow-up. The rate is 0.057 per person-year or 5.7 per 100 person-years. The summary statistic usually used in meta-analysis is the rate ratio (also abbreviated to RR), which compares the rate of events in the two groups by dividing one by the other. It is also possible to use a difference in rates as a summary statistic, although this is much less common.

Counts of more common events, such as counts of decayed, missing or filled teeth, may often be treated in the same way as continuous outcome data. The intervention effect used will be the mean difference which will compare the difference in the mean number of events (possibly standardized to a unit time period) experienced by participants in the intervention group compared with participants in the control group.