Bibliographic record
Abstract
To the Editor: In 2015, for the first time in nearly 25 years, life expectancy decreased in the United States.1 The decrease was small—from 78.2 to 78.1 years—but is nevertheless a cause for concern given recent studies2,3 showing adverse trends in mortality (although these studies were limited to whites). Understanding the components of changes in life expectancy, and how they differ across demographic groups, is an important first step toward identifying root causes and potential ameliorative interventions. We abstracted data on deaths and population from the US National Vital Statistics System by age, cause-of-death, and race ethnicity for 2014 and 2015.4 We limited our analysis to non-Hispanic blacks and non-Hispanic whites because of longstanding concerns for black-white differences in life expectancy. We created abridged life tables and used Arriaga’s5 method for decomposing changes in life expectancy by age and cause-of-death. We selected International Classification of Disease, 10th edition codes (Figure) to capture leading causes of death among gender and race groups. We used Stata software (version 14) to analyze the data and do not present measures of precision because the mortality data are available for the entire population.FIGURE: Contribution of cause-of-death groups (International Classification of Diseases, Tenth Revision [ICD-10] categories taken from National Center for Health Statistics list of 113 selected causes of death: Cardiovascular diseases [I00–I78]; Cancers [C00–C97]; Diabetes [E10–E14]; Alzheimer’s disease (G30); Influenza and pneumonia [J09–J18]; Human immunodeficiency virus [B20–B24]; Chronic lower respiratory disease [J40–J47]; Liver disease [K70, K73–K74]; Kidney disease [N00–N07, N17–N19, N25–N27]; Motor vehicle crashes [V02–V04, V09.0, V09.2, V12–V14, V19.0–V19.2, V19.4–V19.6, V20–V79, V80.3–V80.5, V81.0–V81.1, V82.0–V82.1, V83–V86, V87.0–V87.8, V88.0–V88.8, V89.0, V89.2]; Unintentional poisoning [X40–X49]; Suicide [*U03, X60–X84, Y87.0]; Homicide [*U01–*U02, X85–Y09, Y87.1]; All other causes (all other codes). Available from: http://www.cdc.gov/nchs/data/dvs/Part9InstructionManual2011.pdf) to the change in life expectancy between 2014 and 2015, by gender and race ethnicity (Race and Hispanic origin were classified by the funeral director for death certificates and self-reported for population estimates, and were reported separately on the death certificate in accordance with standards set forth by the US Office of Management and Budget).Among non-Hispanic men, life expectancy at birth decreased from 76.6 to 76.5 years for whites and from 72.7 to 72.4 for blacks. For non-Hispanic women, life expectancy decreased from 81.3 to 81.1 years for whites and remained essentially constant (78.5 years) for blacks. The largest absolute increases in age-adjusted death rates between 2014 and 2015 were for Alzheimer’s and cardiovascular disease among women, unintentional poisoning for men, and homicide for black men (eTable 1; https://links.lww.com/EDE/B202). Among white women, increases in cardiovascular and Alzheimer’s disease accounted for 71% of the decrease in life expectancy (Figure). Alzheimer’s disease also made a notable (20%) contribution among white men, but the majority (50%) of the decline was due to unintentional poisoning, in addition to suicide (12%) and motor vehicle crashes (11%). For black men, however, increases in homicide accounted for nearly 60% of the life expectancy decrease, alongside contributions from unintentional poisoning (23%) and motor vehicle crashes (16%). Improvements in cancer survival kept life expectancy from decreasing further in all groups. Analysis by age group showed that the increase in mortality among the oldest group (85 and over) accounted for 49% of the decrease in life expectancy for white women, whereas mortality increases among those 15–44 accounted for 80% and 65% of the decrease for black and white men, respectively (eFigure 1 and eTable2; https://links.lww.com/EDE/B202). The decline in US life expectancy resulted from a heterogeneous group of causes of death, and did not affect all demographic groups equally. Black men lost nearly twice as many years of life expectancy than did white men, and black women showed virtually no change in life expectancy. The increase in cardiovascular disease is worrisome and consistent with other reports of stalling progress in mortality declines, but our estimates show that this primarily affected life expectancy among white women. Considerable attention has also focused on middle-aged whites who have experienced sustained increases in mortality, largely due to the rise in opioid overdose deaths.2,3 However, the rise in homicide, which disproportionately affects young black men, has received little attention. Although the homicide rate has shown impressive declines since peaking in the 1970s, the 2015 increase requires additional investigation.6 If this trend were sustained it could erode the substantial progress made in reducing the black-white life expectancy gap.7 On a more positive note, improvements in cancer survival kept life expectancy from decreasing by more than it otherwise would have. Our analysis used the underlying cause-of-death and may underestimate the contribution of factors involved in multiple causes. Our findings demonstrate that paths to decreased life expectancy differ substantially by gender and race. Sam Harper Jay S. Kaufman Department of Epidemiology Biostatistics & Occupational Health McGill University Montreal, QC, Canada [email protected] Richard S. Cooper Loyola University Chicago Chicago, IL
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.012 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.002 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 0.005 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".