Uma proposta metodológica para estimar o padrão etário das transições de incapacidade e tendências na expectativa de vida ativa dos idosos: um estudo para o Brasil entre 1998 e 2008

• Autor(a) principal: Marcos Roberto Gonzaga
• Data de Publicação: 2012
• Tipo de documento: Tese
• Idioma: por
• Título da fonte: Repositório Institucional da UFMG
• Texto Completo: http://hdl.handle.net/1843/AMSA-94NFL9
• topic: Tábua de vida multiestado; Expectativa de vida ativa; Compressão da morbidade; Idosos Brasil; Expectativa de vida Brasil; Idosos Saude e higiene

Healthy life expectancy (HALE) is an empirical indicator commonly used in studies about the compression of morbidity and health status. Good estimates of HALE usually requires multi-state life tables and information about age-specific transitions in health status and mortality during the period of study. However, in Brazil, the inexistence of these tables limits the studies about trends in HALE since the country lacks longitudinal data about the healthy conditions of the population. In this context, this dissertation aims to present a new methodology to estimate health life expectancy in the absence of longitudinal data. The proposed method estimates age-specific transition rates of cross-sectional data according to well-documented longitudinal age-specific disability transition rates of other populations, and the age-specific proportion of active and functional disability individuals, reported in cross-sectional Brazilian datasets. In order to estimate disability-free life expectancy, this research makes use of three cross-sectional Brazilian household surveys from 1998, 2003 and 2008. The results show that the estimated disability transition rates are consistent with the literature. Moreover, the estimated parameters for the basic model produces very reliable results and good estimates of transition rates for Brazil. In 1998, 2003 and 2008 the estimated life expectancy with and without any disability do not show significant statistical differences from other estimates produced by other authors. The second exercise, which includes other covariates to the baseline model, presents less satisfactory estimates due to the lack of robustness in parameters estimation. However, the results of this exercise are promising and are indicative of other more complex solutions.