On Analysis of the Difference of Two Exposure-Adjusted Poisson Rates

• Activity Number: 501
• Type: Topic Contributed
• Date/Time: Wednesday, August 3, 2011 : 10:30 AM to 12:20 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Ivan S. F. Chan*+ and William W. B. Wang
• Companies: Merck Research Laboratories and Merck Research Laboratories
• Keywords: Approximate exact method ; Exact test ; Poisson rate ; Stratified analysis ; Risk difference

Abstract

Exposure adjusted incidence rates are often used to account for differential follow-up in evaluating treatment effects from long-term studies. For rare diseases, the Poisson distribution is a reasonable approximation for the disease occurrence. The difference of rates is sometimes used to quantify the treatment's public health impact because the reciprocal of this difference can be interpreted as "the number needed to treat (or number needed to vaccinate) in order to cure (or prevent) 1 case of disease". In this paper we focus on the analysis of two exposure-adjusted rates for both superiority and noninferiority hypothesis testing. We will present several asymptotic and exact methods for analyzing the difference of two Poisson rates with stratification. Performance of these methods will be compared in terms of type I error control and power. An example is used to illustrate these methods. Finally, we will make some general practical recommendations.