J. Norberto W. Dachs

Determinantes das desigualdades na auto-avaliação do estado de saúde no Brasil: análise dos dados da PNAD/1998

O trabalho apresenta um panorama geral das desigualdades nos determinantes considerados na explicacao das desigualdades na auto-avaliacao do estado de saude a partir dos dados da PNAD/1998. Mostra-se como existem gradientes na classificacao do estado de saude autopercebido de acordo com os niveis de educacao, de renda per capita, de acordo com a raca ou cor de pele das pessoas, por grandes regioes do pais e especialmente com o aumento da idade. Usando modelos de regressao logistica, tenta-se explicar quais as determinacoes importantes dessa autoclassificacao. Os resultados mais importantes indicam que educacao e rendimento tem efeitos que se somam e que ha diferencas entre homens e mulheres e de acordo com populacoes urbanas e rurais. As desigualdades na classificacao do estado de saude de acordo com a raca ou cor de pele das pessoas deixam de ser estatisticamente significativas depois de se controlar por nivel de educacao e de renda. Discute-se a utilidade desse tipo de informacao sobre classificacao autopercebida de saude e a importância de melhorar, em futuros inqueritos, a qualidade dos dados por meio de sugestoes sobre alteracoes nos procedimentos de entrevista.

Strengthening health information systems to address health equity challenges.

Special studies and isolated initiatives over the past several decades in low-, middle- and high-income countries have consistently shown inequalities in health among socioeconomic groups and by gender, race or ethnicity, geographical area and other measures associated with social advantage. Significant health inequalities linked to social (dis)advantage rather than to inherent biological differences are generally considered unfair or inequitable. Such health inequities are the main object of health development efforts, including global targets such as the Millennium Development Goals, which require monitoring to evaluate progress. However, most national health information systems (HIS) lack key information needed to assess and address health inequities, namely, reliable, longitudinal and representative data linking measures of health with measures of social status or advantage at the individual or small-area level. Without empirical documentation and monitoring of such inequities, as well as country-level capacity to use this information for effective planning and monitoring of progress in response to interventions, movement towards equity is unlikely to occur. This paper reviews core information requirements and potential databases and proposes short-term and longer term strategies for strengthening the capabilities of HIS for the analysis of health equity and discusses HIS-related entry points for supporting a culture of equity-oriented decision-making and policy development.

Acesso à educação por faixas etárias segundo renda e raça/cor

Sao analisadas as contribuicoes da situacao economica e da raca/cor da pele no acesso a escola, ate o ensino superior, considerando as principais transicoes escolares e grupos etarios, usando dados da Pesquisa Nacional por Amostra de Domicilios - PNAD 2003. As variaveis renda e raca/cor afetam com intensidades diferentes as diversas faixas etarias e transicoes escolares. As restricoes que levam ao reduzido percentual de jovens com acesso ao ensino superior dependem do numero de vagas neste nivel de ensino, mas parecem ser mais determinadas pelo pequeno contingente de jovens brasileiros que consegue completar o ensino medio, atingindo a qualificacao formal necessaria para o acesso ao ensino superior. Somente 40% dos jovens de 18 a 24 anos possuem o ensino medio completo, sendo que 13% tiveram acesso ao ensino superior. A situacao economica e um determinante mais importante do que a variavel raca/cor, embora essa ultima variavel apresente influencia em todas as faixas de renda. Finalmente, estuda-se a distribuicao dos alunos de ensino superior nesta faixa etaria nos estabelecimentos publicos e privados.

Menstrual pattern and ovarian function in women with hyperprolactinemia

Prolactine (PRL) and progesterone were systematically measured in all women presenting with amenorrhea, oligomenorrhea or galactorrhea at the Infertility Clinic of the Department of Obstetrics and Gynecology, State University of Campinas, Brazil, during a period of 34 months. The same hormonal assays were done to all infertile patients presenting for the first time during the last 6 months of the same period, for a total of 190 subjects. Fifty‐five patients with amenorrhea, 38 with oligomenorrhea and 97 with normal cycles, 20 of whom had galactorrhea, were included in the study. Fifty‐five percent of amenorrheic patients, 37% of oligomenorrheic and 9% of those with normal menses had elevated PRL. The mean PRL was higher the greater the menstrual disturbance but was not influenced by presence or absence of galactorrhea. Short luteal phase was the ovarian function condition most frequently associated with high PRL among women with normal menses.

Comparison of tests for the two-way layout

In this paper three methods of testing for treatment effects, in two-factor experiments, are compared. Both asymptotic results and tables and graphs for a finite situation, under several different distributions, are presented for the power of the tests. The three procedures compared are: F-test, Friedman (1937) test and Mehra-Sarangi (1967) test. This latter one is an extension of a proposal by Hodges and Lehmann (1962) of ranking after alignment to the case of k treatments.