Key points are not available for this paper at this time.
There is evidence that 'health life expectancy' (expected number of years to be lived in health) differs by socioeconomic status. Time spent in health or disability plays a critical role in the use of health care services. The objective of this study was to estimate 'disability life expectancy' by age, gender and education attainment for the elderly of the city of São Paulo, Brazil, in the year 2000. Data came from the SABE database, population censuses and mortality statistics (SEADE Foundation). Life expectancy with disability was calculated using Sullivan's method on the basis of the current probability of death and prevalence of disability by educational level. The prevalence of disability increased with age, for both sexes and both levels of educational attainment studied. Men showed a lower prevalence of disability, in general, and persons with lower educational attainment showed a higher prevalence of disability. Regarding life expectancy, women could expect to live longer than men, with and without disability. For both sexes, the percentage of life expectancy lived with disability decreased with increasing educational attainment. With increasing educational attainment, the sex differences in the percentage of remaining years to be lived with disability increased for most ages. Finally, the percentage of remaining years to be lived with disability increased with age for males and females, except for males with high educational attainment between the ages 70-75 and 75-80. The results may serve as a guide for public policies in the country, since health problems faced by older persons, such as disability, are the result of a number of past experiences during their life-times, such as health care, housing conditions, hygiene practices and education. Education influences health behaviours and is related, to some extent, to all these factors. Therefore, improvements in education for the disadvantaged may improve health.
Camargos et al. (2006) studied this question.