There's a very simple test to compare two incidence density ratios that is available in such software as OpenEpi (in www.openepi.com, in the menu on the left, in the "Person Time" area, "Compare 2 Rates") or the DOS True EpiStat or PEPI programs. It uses a 2x2 table, with counts of exposed and unexposed cases in the top row and counts of exposed and unexposed person-time in the bottom row, and generates exact and asymptotic tests for the rate ratio based on a null hypothesis that the rates are equal and that hence each observed case is an independent Bernoulli trial with probability of being in the exposed (resp. unexposed) category proportional to the total fraction of person-time that is exposed (resp. unexposed). (See, e.g., D. G. Kleinbaum, L. L. Kupper, & H. Morgenstern, Epidemiologic Research (1982), pp. 285-288.)
I would like to do this sort of analysis in SAS so I don't have to retype data in OpenEpi or one of the DOS programs. Any ideas on how to do this?