Calculation of prefectural disability-adjusted life expectancy (DALE) using long-term care prevalence and its socioeconomic correlates in Japan

Health Policy. 2006 May;76(3):346-58. doi: 10.1016/j.healthpol.2005.06.017. Epub 2005 Aug 2.


Objectives: The objectives of this study were: (1) to determine the disability weight, "utility", for calculation of disability-adjusted life expectancy (DALE) using the prevalence of long-term care; (2) to calculate prefectural DALE; and (3) to clarify the relation between DALE and area socioeconomic conditions in Japan.

Methods: Disability utility by care level (support and levels I-V) of long-term care insurance was determined by a survey of 236 professionals with four standard utility measures: EuroQol-5D, time trade-off, standard gamble, and visual analogue scale. DALE at age 65 (DALE65) and age-adjusted weighted disability prevalence (WDP) of 47 prefectures were calculated using the determined utilities, prevalence of long-term care, and life tables. The relationships of DALE and WDP to mortality from major causes and socioeconomic indicators were examined by correlation analysis.

Results: The determined utilities were: support, 0.78; level I, 0.68; level II, 0.64; level III, 0.44; level IV, 0.34; and level V, 0.21. The prefectural DALE65 ranged from 17.11 to 15.29 years for men and from 20.21 to 18.42 years for women. Strong correlations were found between DALE65 and mortality for both sexes. Male DALE65 was correlated with no socioeconomic indicators, while female DALE65 was correlated with some indicators. WDP was positively associated with indicators representing socioeconomic disadvantage, such as unemployment rate and percentage of elderly single households.

Conclusions: The socioeconomic correlates of DALE and WDP suggested that favorable socioeconomic policies, in addition to a decrease in mortality from major causes, will contribute to significant extension of the independence period in the elderly. The method proposed here encourages the practical use of health expectancy in health policy, especially at local and regional levels.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Aged, 80 and over
  • Data Collection
  • Disabled Persons*
  • Female
  • Humans
  • Japan
  • Long-Term Care / statistics & numerical data
  • Male
  • Quality-Adjusted Life Years*
  • Social Class*