Sampling Strategies Based on Existing Information in Nested Case Control Study

• Activity Number: 288
• Type: Contributed
• Date/Time: Tuesday, August 5, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Yi Luo*+
• Companies: University of Southern California
• Keywords: Nested case control study ; sampling method ; counter matching ; outside caliper matching

Abstract

This paper presents a new sampling strategy called Outside Caliper Matching (OCM), and the comparison of OCM to two established sampling strategies, Simple Random Sampling (SRS) and Counter Matching (CM). OCM is similar to CM but is appropriate for sampling using continuous exposure data. Like CM, OCM increases the diversity of exposure values and retains more information from the cohort. To achieve this goal, OCM puts restrictions on the caliper (minimum difference in exposure rank orders between the case and the control) instead of using exposure levels as CM. This avoids categorizing the continuous exposure to get exposure levels needed to implement CM. Simulation results indicate that using a wider caliper yields better relative efficiency. And comparison between OCM, CM and SRS shows that OCM has the highest relative efficiency when the rate ratio is within a practical range of 0.05 to 20. Therefore, we conclude that OCM with a caliper of half the maximum rank order would be the best of the three sampling strategies when exposure variable is continuous.