Revista Brasileira de Epidemiologia, Volume: 4, Issue: 3, Published: 2001
• ## Editorial

Carvalheiro, José da Rocha
• ## Time series analysis in epidemiology: an introduction to methodological aspects

Latorre, Maria do Rosário Dias de Oliveira; Cardoso, Maria Regina Alves

Abstract in Portuguese:

Abstract in English:

A time series, also denominated historical series, is a sequence of data obtained in regular intervals of time during a specific period. In the analysis of a time series, one first wants to model the study phenomenon and, from this, to describe the behaviour of the series, to make estimates, and, in the end, to evaluate the factors that may have influenced the behaviour of the series, with the objective of defining cause-effect relationships between two or more series. For this, there is a set of available statistical techniques which depend upon the defined model (or that estimated for the series), the type of the study series, and of the objective of the work. To analyse trends, it is possible to adjust polynomial regression models based on the whole series or on the neighbourhood of a specific point. This can also be done with mathematical functions. A seasonal phenomenon is defined as the one that occurs regularly in fixed periods of time and, if there is seasonality considered as deterministic in the series, one can use regression models which include functions like seno or cosseno to the variable time. In the analysis of the behaviour of a time series without trend and seasonality, the auto-regressive models (AR) or models which incorporate moving averages (ARMA) can be used. When trend is present, one can use auto-regressive models integrated with moving averages (ARIMA) and to incorporate the seasonality component the SARIMA models are used. The generalized linear models constitute another class of models. In this group of statistical models, the response variable is a counting process and the independent variables are those which are candidates to explain the behaviour of the series throughout the time. This class of models is indicated when the study variables do not follow the Normal distribution, mainly because they are counting processes. These models represent a group of probability distributions known as exponential family of distributions that incorporates many additive functions like the linear regression, Poisson, logistic, log-linear, etc. The generalized additive models are an extension of this class of models, in which each independent variable analysed does not enter in the model with its own value, but adopting a non parametric function in a non specific manner, which is estimated from smoothing curves.
• ## Trends of infant mortality in the city of Guarulhos: 1971 to 1998

Tomé, Ermelinda Antonia; Latorre, Maria do Rosário Dias de Oliveira

Abstract in Portuguese:

Abstract in English:

The infant mortality rates have presented a decreasing trend in the last decades especially due to the post-neonatal component. The objective of this paper was to analyze the trend of infant mortality rate and its components in the city of Guarulhos from 1971 to 1998. Data on infant mortality rates, early neonatal, late neonatal and post-neonatal mortality rates are from Fundação SEADE and trends were analyzed in three different periods: from 1971 to 1980, from 1981 to 1990 and from 1991 to 1998. Polynomial regression models were used in order to analyze trends. All infant mortality rates had statistically significant decreasing trends only in the first period (1971-80). In the second period (1981 to 1990) the infant mortality rate (p= 0.0058), the late neonatal mortality (p= 0.0105) and post-neonatal mortality (p= 0.0045) presented statistically significant decreasing trends. In the period from 1991 to 1998, the infant mortality rate (p< 0.0001), neonatal mortality (p= 0.0173) and post-neonatal mortality rate (p= 0.0044) had decreasing trends. We concluded that it is necessary to improve mother-child care especially in the neonatal period in order to reduce infant mortality rates in Guarulhos.
• ## Time trends in cancer mortality in Rio Grande do Sul, Brazil, 1979-1995

Hallal, Ana Luiza Curi; Gotlieb, Sabina Léa Davidson; Latorre, Maria do Rosário Dias de Oliveira

Abstract in Portuguese:

Abstract in English:

The aim of this study was to analyze cancer mortality in Rio Grande do Sul (RS), Brazil, during the period from1979 to 1995. Study variables were sex, age, year and underlying cause of death. The simple linear regression technique was used to evaluate the trend of standardized death rates (direct method, using the population of RS in 1996 as the standard). The most frequent sites of tumors in males were lung, esophagus, prostate, stomach and colon/rectum; in females they were breast, cervix of the uterus, lung, colon/rectum and stomach. Standardized death rates presented a stable trend for all malignant neoplasms in both sexes, as did cancer of cervix of the uterus/ non-specified uterus and colon/rectum tumors in females. A significant rising trend was observed in mortality rates due to lung cancer in both sexes, breast cancer in females, prostate and colon/rectum cancer in males. The rates of stomach cancer presented a significant decreasing trend in both sexes as did esophagus cancer rates in men.
• ## Trend of the AIDS epidemic in intravenous drug users in the Municipality of São Paulo from 1985 to 1997

Bastos, Maria do Socorro Castelo Branco de Oliveira; Latorre, Maria do Rosário Dias de Oliveira; Waldman, Eliseu Alves

Abstract in Portuguese:

OBJETIVO: Analisar a tendência da epidemia de AIDS em adultos, no período de 1985 a 1997, no Município de São Paulo, focalizando principalmente os usuários de drogas injetáveis (UDI), considerando o grande impacto da epidemia nesse grupo populacional. METODOLOGIA: A fonte dos dados foi a vigilância de AIDS do Estado de São Paulo. O método estatístico utilizado para a análise de tendência temporal dos 8.558 casos de AIDS entre UDI, e dos 16.756 entre não UDI, foi a regressão polinomial. RESULTADOS: No período de 1985 a 1992 a tendência foi de ascensão para os casos de AIDS entre UDI e não UDI a partir desse ponto e, observou-se um declínio para UDI e manutenção em platô elevado para os não UDI, em conseqüência do crescimento constante do número de casos entre as mulheres e homens heterossexuais. Os coeficientes de regressão ajustados (R²a) de todos os modelos ficaram acima de 70%, e os modelos que melhor explicaram as tendências foram os de segunda ordem (parábola), exceto para heterossexuais não UDI cuja tendência foi de aumento linear. CONCLUSÃO: No final do período considerado, os dados sugerem um declínio da epidemia em UDI, a estabilização em níveis elevados para o total de não UDI, e a manutenção da tendência de crescimento da epidemia para pessoas heterossexuais não UDI.

Abstract in English:

OBJECTIVE: To analyze trends of AIDS epidemic among adults, from 1985 to 1997, in the city of São Paulo, focussing primarily on intravenous drug users (IDU), owing to the major impact of the epidemic on this population group. METHODOLOGY: Data from the AIDS surveillance of São Paulo State were analyzed. The statistical method applied to analyze the time series of the 8,558 cases of AIDS among IDU and of the 16,756 cases among non-IDU was polynomial regression. RESULTS: From 1985 to 1992 there was a increase in AIDS cases both in IDU and non-IDU. Thereafter there was a decrease in cases among IDU, while cases among non-IDU leveled off owing to the constant growth trend among heterosexual men and women. Adjusted regression coefficients (R²a) for all models remained higher than 70%, and the second-order (parabolic) models explained better the trends, except in the case of heterosexual non-IDU, in which the model was linear. CONCLUSION: At the end of the period under study the data suggest a decrease in the epidemic among IDU, while the trend among non-IDU keep flat at high level, and showed a constant increase among heterosexual.
• ## Deaths due to malnutrition in the elderly, São Paulo e Rio de Janeiro: time series analyse. 1980-1996

Otero, Ubirani Barros; Rozenfeld, Suely; Gadelha, Angela Jourdan

Abstract in Portuguese:

Abstract in English:

OBJECTIVES: To describe mortality evolution of malnutrition in elderly in Rio de Janeiro (RJMA) and São Paulo Metropolitan Areas (SPMA); to verify it's tendecies, between 1980-1996; and to propose a model wich permit us to foresee the cases occurances. METHODOLOGY: It's a time series analysis, using ecological data. People from RJMA and SPMA around 60 year-old and older ones, of both sexes were included in this research. The source of those data were the Mortality Information System (1980-1998). The series of monthly cases were analyzed, in a period from January of 1980 to December of 1996. To model the series and to predict the malnutrition cases in elderly. The method Box and Jenkins was used, SARIMA (Seasonal Autoregressive Integrated Moving Average). RESULTS: The results show the existence of seasonable pattern with the largest number of deaths during the months of June and July in SPMA probably because of the cold weather in São Paulo in the winter and during the month of january, in Rio de Janeiro, probably because of the high temperature in this city in the summer. These results can be explained by the difficult of elderly to mainten their bodies in a good temperature during those months what represents an additional stress to their organism poor in nutritional reservations. CONCLUSIONS: The series presented a rising tendency of death and showed seasonable pattern in winter in SPRM and in the summer in RJRM. To estimate the SARIMA models to the Regions for the year of 1996, the results reforced the hypothesis that this method can be efficient to make future forecasts. It's probably that a research wich considers so many death causes will permit to verify the real contribution of the malnutrition as the basic cause or at least a contributive cause of those deaths.
• ## GLM and GAM model for analyzing the association between atmospheric pollution and morbidity-mortality markers: an introduction based on data from the city of São Paulo

Conceição, Gleice Margarete de Souza; Saldiva, Paulo Hilário Nascimento; Singer, Julio da Motta

Abstract in Portuguese:

Abstract in English:

In this study, we have described and compared two classes of models - the Generalized Linear Models (GLM) and the Generalized Additive Models (GAM) _ which can be used to evaluate the association between atmospheric pollution and morbidity and mortality counts. We consider GAM as an alternative for modeling unspecific nonlinear relationships, and show that this class of models constitute a good option to represent both seasonality and the relation between the outcome and the meteorological variables. As an illustration, we evaluate the association between mortality in elderly people and air pollution in São Paulo, Brazil for the period ranging from 1994 to 1997. The two modeling classes have produced coherent results, but the most sophisticated models were more powerful to detect significant effects. Associations between mortality and the levels of CO, SO2 and, in to a lesser extent, PM10 were observed.
• ## Air pollution and emergency room visits for upper airway respiratory infection disease in São Paulo city: evaluation of vehicle restriction

Martins, Lourdes Conceição; Latorre, Maria do Rosário Dias de Oliveira; Saldiva, Paulo Paulo Hilário Nascimento; Braga, Alfésio Luís Ferreira

Abstract in Portuguese: