National Longitudinal Mortality Study (NLMS)

Overview

To estimate probabilities of death by cause for a representative United States population, according to age, sex, race, occupation, industry, income, education, residence, country of birth, calendar year, and a number of other demographic and epidemiologic factors. The representative population is derived from selected Bureau of the Census files and is matched to the National Death Index maintained by the National Center for Health Statistics. The Bureau of the Census provides the data processing, coding and management necessary to complete this objective.

BACKGROUND: The current National Longitudinal Mortality Study began in 1983 with agreements with the Census Bureau which prepared baseline files from surveys representative of the United States and with the National Center for Health Statistics which prepared the National Death Index identifying deaths in the baseline populations. The baseline population of over one million persons has been matched to the NDI on several occasions, resulting in approximately 100,000 deaths occurring in 1979 to 1989. Using available data, an analysis was conducted relating socio-economic characteristics to subsequent mortality. DESIGN NARRATIVE: The NLMS is a national study of mortality over time among selected Census Bureau population samples numbering about 2.4 million. The census samples are matched to the National Death Index (NDI) maintained by the National Center for Health Statistics. The NDI is a file of all U.S. deaths since 1979 and is used to determine which individuals in the Current Population Surveys (CPS) have died. The samples are matched every other year to obtain deaths among these cohorts. Death certificates are then purchased from the states and coded for causes of death and other data. Mortality rates by age, sex, race, national origin, occupation, industry, income, education, state of residence and other factors are then obtained. The follow-up period begins with 1979, the first year covered by the NDI and ends with 1998. The total number of deaths for these cohorts is estimated to be about 250,000. Census samples in the NLMS are being matched to the Centers for Medicare \& Medicaid Services (CMS) medicare database. Hospitalization, physician, outpatient, and other CMS data will be incorporated into the NLMS database for all participants meeting the Medicare age eligibility criteria. The study continues through December, 2009.

Conditions

Eligibility