Por meio de um conjunto de indicadores de saúde, pretendeu-se hierarquizar as delegacias regionais de saúde do Estado do Rio Grande do Sul (Brasil), no período de 1979 a 1982. Selecionaram-se cinco indicadores clássicos: mortalidade proporcional em menores de cinco anos, coeficiente de Swaroop-Uemura, coeficiente de mortalidade infantil, mortalidade proporcional por doenças infecciosas e percentual de óbitos mal definidos que apresentaram maior peso relativo usando metodologia de análise fatorial. Estes indicadores foram padronizados, utilizando a variável Z, ponderados, utilizando coeficiente de escore fatorial, e as regiões administrativas foram classificadas em sete grupos. Todos os grupos apresentaram melhora no nível de saúde, no período observado. Através do conceito de risco foram sugeridos alguns indicadores para integrarem a vigilância epidemiológica de saúde que permitiram hierarquizar as delegacias regionais de saúde, com vistas a estabelecer prioridades em termos de ações de saúde e aplicação de recursos.
By the use of a group of health statistics, this study classifies the health status of the administrative regions in Rio Grande do Sul (Brazil) for the period form 1979 to 1982. Five classic health parameters were chosen according to the following criteria: accessibility to a hisitorical series of three years; necessarily part of an information system currently in use and having a high rate of factorial score. Each parameter was weighed (CEF) for "Health Level", "Immunization Level" and "Medical Care Level". They were: proportional mortality under five years, Swaroop-Uemura, infant mortality rate, proportional mortality for infectious diseases and percentage of undefined deaths. The parameters were standardized by a Z-variable and weighed by factorial score rate. In this way the final score was obtained for each administrative region. Such a score is a fusion of five parameters in one, making the data comparison easier. The administrative regions were classified in seven groups. It was possible to identify regions where investment is needed, i.e., where health levels are low. Apart from this study's principal purpose, it was found that the regions with the lowest health levels are characteristically agricultural, with large rural properties and marked distinctions in social class. It is suggested that the parameters should be used as instruments in a model for epidemiological surveillance which would show the present situation of the health/disease ratio in a population.