Desemprego, trabalho sem proteção social e saúde: uma análise do indivíduo e do contexto

Abstract

The social and institutional changes occurring in the last decades due to globalization, employment crisis and erosions in social welfare conquests, are printing new faces to social inequalities, with important repercussions in population health. This thesis investigated healthinequalities related to individuals´ labor market position. The labor market status was defined by the following categories: employment with social protection, employment without social protection and unemployment, both short and long terms unemployment. At first, it wasinvestigated the association between employment with no social protection, short and long terms unemployment and health status and healthcare use indicators, after considering the effects of age and education. This work has also investigated the household and neighborhood effects on heath according to individuals´ labor market status. Two main hypothesis were tested: 1) whether sharing the same household with an unemployed or/and an informal worker is associated with a worse self rated health, regardless of individuals own characteristics and household material conditions and; 2) if the association between worse self rated health and unemployment is modified by socio-economic characteristics of the neighborhood. The twocontextual indicators used to identify neighborhood were living in areas classified as slam, and living in areas with high proportion of poor household heads. This study is based on two population based surveys: the National Health Survey (PNAD) carried out in 1998 and in 2003 by the Brazilian Institute for Geography and Statistics (IBGE) and the Household Survey on Risk Behaviors and Reported Morbidity on Non-Communicable Diseases carried out in 2002/2003 by Brazilian Ministry of Health. It alsoused data from the Brazilian 2000 Census to identify neighborhood characteristics. The analysis employed multiple logistic regression and Generalized Estimation Equations (GEE), which considers the correlation among individuals within the household and models the average population response as a function of covariables. Results show that Brazilian males employed with no social protection, in short and longterms unemployment have worse health indicators after considering the effects of age, schooling and race. Moreover, in spite of their worse health indicators, men in informal work and unemployed, both in short and long terms, also reported lower health care use, especially medical visits. When we studied the household effect, it was observed that living in ahousehold which had someone unemployed or who had an unprotected job was associated with residents' poor self rated health, regardless of individual's own labor market position and of other individual and household factors. The study of the effects of neighborhood characteristics on self rated health showed that individuals who lived in slums or in areas with greater proportion of lower income household heads reported worse self rated health more often than those living in better areas. It also confirmed that unemployed individuals reportworse self rated health than employed ones. The magnitude of this association was attenuated by sociodemographic characteristics, behavioral risk factors and heath status conditions of individuals. However, there was no evidence that the association between unemployment and self rated health is modified by the neighborhood characteristics investigated in the present work. Our study showed that employment with no social protection, unemployment and length ofunemployment characterize heterogeneous situations in relation to sociodemographic characteristics as well as to health conditions and healthcare use. It also highlights that self rated health of individuals sharing the same household was negatively affected by the presence of unemployed and/or unprotected worker in the household. But, although the proportions of individuals who rated their health as poor were greater among unemployed living in socially disadvantaged neighborhoods, the association between self rated health and unemployment was not different according to neighborhood context. Adverse work situations are increasing, especially unprotected work and unemployment. They are highly important questions for public health research agendas aiming at health to improve populations´ health and reduce current health inequalities.