A primer and comparative review of major US mortality databases

• Published in: Annals of epidemiology Vol. 12; no. 7; pp. 462 - 468
• Main Authors: Cowper, Diane C, Kubal, Joseph D, Maynard, Charles, Hynes, Denise M
• Format: Journal Article
• Language: English
• Published: United States 01.10.2002
• Subjects: Cause of Death; Databases as Topic - statistics & numerical data; Epidemiologic Methods; Humans; Mortality; National Center for Health Statistics (U.S.); United States; United States Department of Veterans Affairs; United States Social Security Administration; Vital Statistics; United States; USA
• ISSN: 1047-2797; 1873-2585
• DOI: 10.1016/S1047-2797(01)00285-X

Mortality data are important tools for research requiring vital status information. We reviewed the major mortality databases and mortality ascertainment services available in the United States, including the National Death Index (NDI), the Social Security Administration (SSA) files, and the Department of Veterans Affairs databases. The content, reliability, and accuracy of mortality sources are described and compared. We also describe how investigators can gain access to these resources and provide further contact information. We reviewed the accuracy of major mortality sources. The sensitivity (i.e., the proportion of the true number of deaths) of the NDI ranged from 87.0% to 97.9%, whereas the sensitivity for the VA Beneficiary Identification and Records Locator System (BIRLS) ranged between 80.0% and 94.5%. The sensitivity of SSA files ranged between 83.0% and 83.6%. Sensitivity for the VA Patient Treatment File (PTF) was 33%. While several national mortality ascertainment services are available for vital status (i.e., death) analyses, the NDI information demonstrated the highest sensitivity and, currently, it is the only source at the national level with a cause of death field useful for research purposes. Researchers must consider methods used to ascertain vital status as well as the quality of the information in mortality databases.