A SAS program for calculating cumulative incidence of events (with confidence limits) and number at risk at specified time intervals with partially censored data

Correct analysis and interpretation of longitudinal (cohort) studies with partially censored time-to-event data requires that the cumulative count of events and censored observations as well as the number at risk be calculated at appropriate time points (for example, every year), by baseline group or stratum. We present here a simple SAS program, for use in situations in which competing risks do not need to be accounted for, that calculates, by baseline group or stratum, the cumulative event count, cumulative event probability (with upper and lower 95% confidence limits), and number at risk at selected time points that can be chosen by the user. We demonstrate the use of the program in the analysis of longitudinal time-to-event data from a prospective study, the Atherosclerosis Risk In Communities (ARIC) Study, for four groups and a 10-year follow-up. The SAS code presented here is easy to follow and modify and can be incorporated quickly by the user for immediate use. It provides an especially valuable tool for less experienced SAS users.