In orthodontics and dentistry in general, we are frequently working on studies where groups of patients are followed over time with the aim to record the occurrence of an event. Subjects may enter a study at various times and be followed until (1) the event of interest occurs, (2) the study is completed (end of follow-up time), or (3) they are lost to follow-up. The analysis of this kind of study is based on incidence rates, or more simply, rates (λ). Rate is the total number of events per person per unit of time (ie, person-time) and reflects the probability that a subject will experience the event of interest at a prespecified time.
Rate λ = total number of ( new ) events total person – time at risk
The unit of rate is the number of events per person-time; depending on what is more sensible for interpretation purposes, time can be measured in months, years, and so on, and persons can be counted per 100 or 1000. For example, if the number of implant failures is on average 20 in a population of 100 persons over a 2-year period, then
Rate λ = 20 100 * 2 = 0.1 events per person – years = 10 events per 100 person – years