Examining the Effects of Differential Coverage of Three Mortality Record Sources

• Activity Number: 282
• Type: Topic Contributed
• Date/Time: Tuesday, July 31, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Social Statistics Section
• Author(s): Sandy Eyster*+ and Celeste Stone and Susan J. Lapham
• Companies: American Institutes for Research and American Institutes for Research and American Institutes for Research
• Keywords: administrative records ; data linkages ; mortality research ; survival analysis ; bias

Abstract

Key to research on mortality is determining mortality status via linking to administrative records. Several methods for ascertaining mortality status exist, including the Social Security Administration's Death Master File (DMF), the National Center for Health Statistics' National Death Index (NDI), and consumer and commercial credit bureau databases. This study examines the effect of differential coverage of three mortality record sources for a random subsample of 4,159 Project Talent participants. Project Talent is a longitudinal study that began in 1960 and measured the cognitive abilities, interests, personality, and demographics of approximately 440,000 9th-12th grade students. In this study we summarize findings on coverage rates, mortality rates, and potential bias related to the different mortality record sources. Furthermore, we investigate whether and to what extent relying on different mortality record sources affects the results of survival analyses. In addition to key demographics, we examine the relationship among early life personality and cognitive ability factors and mortality, which are rarely available for a large, nationally-representative sample.