A Comparison of Group Sequential Approach and MaxSPRT Approach in a Rapid Cycle Analysis of 13-Valent Pneumococcal Conjugate Vaccine Safety

• Activity Number: 35
• Type: Contributed
• Date/Time: Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Lei Qian*+ and Hung Fu Tseng and Steve J. Jacobsen and Lina Somsouk Sy and Eric Weintraub and Jennifer Nelson
• Companies: Kaiser Permanente Southern Californiaz and Kaiser Permanente Southern California and Southern California Permanente Medical Group and Kaiser Permanente Southern California and Centers for Disease Control and Prevention and Group Health Cooperative
• Keywords: Vaccine safety ; Group sequential test ; MaxSPRT ; PCV13 ; Post-marketing safety surveillance ; RCA

Abstract

Rapid cycle analysis (RCA) is an active surveillance system lunched by the Vaccine Safety Datalink team to conduct population-based research on rare and serious adverse events following immunization in a timely manner. Maximized sequential probability ratio test (MaxSPRT), which is designed for data continuously collected and tested, has been widely used in RCA studies. A group sequential likelihood ratio test was recently adopted due to its flexibility on boundary shape and test frequency. PCV13 vaccine safety is the first RCA study conducted both group sequential test and MaxSPRT analysis. We compared the surveillance length, statistical power and time to signal between two approaches in the study. The experience in implementation the group sequential test and MaxSPRT analysis will also be shared.