Crimmins, Zhang, and Saito 2016 — Trends Over 4 Decades in Disability-Free Life Expectancy in the United States

Summary

Crimmins, Zhang, and Saito compute disability-free life expectancy (DFLE) at birth and at age 65 for U.S. men and women using 40+ years of National Health Interview Survey (NHIS) data (1970–2011) merged with period life tables via the Sullivan method. At birth, DFLE grew but no faster than total life expectancy — no compression of morbidity at the population level. At age 65, DFLE grew faster than life expectancy (LE) for women and about equally for men, suggesting modest compression at older ages. Working-age men consistently had higher DFLE than women despite lower total LE, reflecting the gender disability gap.

Key Claims

• Between 1970 and 2011, life expectancy at birth increased $\approx 6$–$7$ years for both sexes, but DFLE at birth increased by a nearly identical amount — no compression of morbidity at birth
• At age 65, DFLE increased more than total LE for women (compression), but roughly equally for men (dynamic equilibrium)
• Men have higher DFLE at birth than women throughout the period — their shorter lives contain proportionally more disability-free years
• DFLE at birth was flat or declining during the 1970s, then rose after 1980, mirroring changes in NHIS disability measurement and underlying health trends
• Results are sensitive to NHIS disability definition used (activities of daily living (ADL) vs. activity limitation vs. work limitation); the paper uses activity limitation as the primary indicator
• Policy implication: raising the retirement age is poorly supported by evidence — DFLE gains are concentrated at age 65+, not in the working-age years where older workers would be affected

Concepts Introduced or Extended

• Compression of Morbidity
• Disability-Free Life Expectancy

Entities Mentioned

• Eileen M. Crimmins
• Yuan Zhang
• Yasuhiko Saito

Quotes

"Increases in life expectancy at birth have not been accompanied by similar increases in years free of disability... This indicates a lack of compression of morbidity at birth."

"The lack of evidence that the working-age population is gaining more years free of disability makes us cautious about raising the retirement age."

My Take

A landmark descriptive paper providing the longest continuous DFLE time series for the U.S. Its main contribution is methodological discipline — consistent Sullivan method application across four decades — rather than causal identification. The sensitivity to disability definition is an important caveat: different NHIS measures yield different trend conclusions, a persistent challenge in the compression-of-morbidity literature. The gender paradox (men have shorter but proportionally more disability-free lives) is a robust finding that complicates simple narratives about aging and disability. The retirement-age policy implication is stated cautiously but is the most practically significant takeaway.