English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

BACKGROUNDSocioeconomic mortality differentials are known to exist almost universally. Many studies show a trend towards convergence with increasing age. Information about the highest ages is very rare, though.OBJECTIVEWe want to find out whether socioeconomic factors determine the chance of death in the United States among the oldest people.METHODSBased on official death count records, we employ the extinct cohort method to estimate the age-specific probability of dying by level of education.RESULTSWe present evidence that socioeconomic differentials in mortality exist even at the highest ages (95+), although the gap is small.1. IntroductionFew relationships in demography are as well established as the one between socioeco- nomic status (SES) and mortality: people with lower SES suffer from higher mortal- ity (and vice versa). This finding is rather universal, irrespective how SES is measured (e.g., income, education, occupation) or for which countries the study is conducted (e.g., Kitagawa and Hauser 1973; Rogers, Hummer, and Nam 1995; Kunst 1997; Hum- mer, Rogers, and Eberstein 1998; Pappas et al. 1993; Mackenbach et al. 1999; Goldman 2001).4 Many studies show that excess mortality of disadvantaged groups decreases with increasing age (e.g., Doblhammer, Rau, and Kytir 2005). This tendency towards conver- gence is at least partly caused by compositional changes as a result of mortality selection. Other explanations typically refer to policies, such as Medicare, that reduce inequalities or to the larger influence of biological factors at older ages (see, for instance, Hoffmann (2008) for an overview of causal explanations).In this paper we analyze whether socioeconomic mortality differentials still exist at ages 95 and higher in the United States. Our goal is not to specify whether selection effects are stronger than policy or biological effects, or are potentially counteracted by an accumulation of detrimental effects acquired over the life-course (Ross and Wu 1996).Despite the fact that the number of people at the highest ages is increasing rapidly (e.g. Rau et al., 2008), little is known about whether socioeconomic mortality differences still persist at those ages. Using data from the "National Long Term Care Survey" (NLTCS) of the United States, Manton, Stallard, and Corder (1997) found differences in remaining life expectancy at age 95 of less than one year between people with high and low education. More recently Zhu and Xie (2007) found that education - in conjunction with urbanity - plays a major role in Chinese mortality at ages 90+ and 100+ based on data from the Chinese Longitudinal Healthy Longevity Survey.2. Data and methodFor this study, we used the "Multiple Cause of Death Data" from the National Vital Statis- tics System of the National Center for Health Statistics of the United States. These data can be downloaded from the website of the National Bureau of Economic Research (Na- tional Center for Health Statistics 2013). The data are currently available for the years 1959-2010 and list individual deaths by a multitude of covariates such as sex, age at death, cause of death, year of death, or state of residence. Since 1989, these data also contain information on education, which we used as our indicator for socioeconomic sta- tus. Besides this practical reason of data availability, education might be the best marker for SES as it can be expected to remain relatively stable throughout the adult life course, especially among the oldest-old. Additionally, from a causal perspective, it can be ex- pected to predate other measures of SES (such as occupation or income), which might be difficult to measure at ages 90+ at all.To estimate mortality from such right-truncated data, we employed the so-called Extinct-Cohort method pioneered by Vincent (1951) and Depoid (1973): Starting at the highest age the last surviving member of a birth cohort has reached, this method estimates the age-specific probability of dying, q (x), backwards using the number of people who died at a given age as the numerator and the cumulative number of people who died at that age or older as the denominator. …

Minor gradient in mortality by education at the highest ages