An inferential procedure for the Poisson intervention parameter.
To describe a health chance mechanism whose incidence rate is altered in the middle of a data collection period due to preventive treatments taken by health service agencies, a model called Intervened Poisson Distribution (cf. Shanmugam, 1985, Biometrics 41, 1025-1029) was introduced. In this article, we derive a statistic to test the hypothesis concerning the effectiveness of such preventive treatments, and compare its power with another suggested test (cf. Streit, 1987, Biometrics 43, 999-1000).