Mortalidade por doenças cardiorrespiratórias em idosos no estado de Mato Grosso , 1986 a 2006 Mortality due to cardiorespiratory diseases in elderly people in Mato Grosso state , 1986 to 2006

MÉTODOS: Estudo epidemiológico descritivo com delineamento ecológico de séries temporais realizado no estado do Mato Grosso, de 1986 a 2006. Foram utilizados dados sobre doenças dos aparelhos respiratório e circulatório obtidos do Sistema de Informação sobre Mortalidade do Ministério da Saúde. Modelos de regressão linear simples foram ajustados para avaliar a tendência das taxas específi cas de mortalidade por grupos específi cos de idade (60 a 69, 70 a 79 e 80 ou mais anos) e sexo.


INTRODUCTION
Like in many developing countries, Brazilian people is aging at an accelerated pace, especially the population group comprising elderly people aged 60 or older.The growth of this population group has been much faster than in European countries. 15Increased life expectancy has been progressively contributing to increase elderly population. 15The elderly accounted for 6% of total population in 1980 and this proportion increased to 8.5% in 2000, corresponding to 14 million elderly.
At the same time with population aging there has been a change in the profi le of deaths among the elderly, both in terms of proportional mortality by different causes and their related rates.According to data from the Brazilian Ministry of Health Mortality Database (SIM), in 1980, 38% of all deaths were among the elderly.In 1991 and 2000, they accounted for 50% and 55% of total deaths, respectively.a Cardiovascular and respiratory diseases stand out as major causes of mortality among the elderly.Cardiovascular diseases were the leading cause of death in Brazilian elderly (38%, n = 231,540) in 2006, followed by cancer (16%, n = 100,743) and respiratory diseases (13%, n = 79,410).a Mortality trends are influenced by several factors including improved data quality, access to health services, quality of health services, and individual variations in risk factors for cardiovascular and respiratory diseases. 2,6Trend analyses may show the effect of these factors on the mortality profi le of the elderly.
The elderly is a population group with specifi c characteristics due to lifetime accumulated disease burden, particularly cardiopulmonary diseases.Meeting the needs of this population group is a challenge to the Brazilian Health System.The objective of the present study was to analyze time trends in mortality rates due to cardiopulmonary diseases in elderly.

METHODS
Epidemiological study with an ecological time series approach of mortality data for the period 1986 to 2006 in people aged 60 or older in the state of Mato Grosso, Central-Western Brazil.Mato Grosso was the state of choice due to major social and environmental transformations that have occurred since early 1980s in the wake of the occupation of the Amazon region.Mato Grosso is a corridor for migration to the Northern states of Rondônia, Acre and Pará and 86 new towns were created between 1980 and 2006. 1,4rtality data were obtained from SIM database.Data from the elderly population living in the state of Mato Grosso by gender and age were obtained from the Brazilian Institute of Geography and Statistics (IBGE).b The elderly were divided into three age groups: 60-69; 70-79; and 80 and more.
Although mortality rates from ill-defi ned causes in Mato Grosso are decreasing, c it was decided to correct reported cardiopulmonary deaths for ill-defi ned deaths in each disease group by gender and age.These corrections were made proportional to all deaths, excluding ill-defi ned causes.It was assumed that the distribution of causes of death among deaths from ill-defi ned causes is similar to that of deaths from defi ned causes.This procedure was applied for all years and can be summarized by the equation Xc = X + M * X/(T -M), where X is the number of deaths due to a specifi c cause (pulmonary or cardiovascular), M is the number of deaths from ill-defi ned causes, T is the number of deaths from all causes, and Xc is the corrected number of deaths by a specifi c cause.After these corrections, adjustments were made for standardization resulting in mortality rates corrected for ill-defi ned causes and adjusted by gender and age.
It was then calculated the annual proportional mortality from pulmonary and cardiovascular diseases, by gender and age groups, to assess the weight of these diseases to total deaths 11  Mortality rates by gender and age groups were then estimated and standardized using a direct method, considering as standard population the harmonic mean of the population in each age group during the period studied.This standardization was necessary as age groups are decennial and there was an increase in population survival in the last 20 years. 8trend analysis of mortality rates was performed using simple linear regression models.The construction of scatter plots of deaths and years of study showed that a linear evolution could be assumed in all cases, which supported the use of linear models. 3e statistical model considered each mortality rate as a dependent variable (Y) and the years studied as an independent variable (X).The linear regression models were adjusted for each time series.It was decided to make the independent variable central by subtracting the midpoint value of each series (X -1996) to avoid autocorrelation between the terms of the regression equation. 10Thus, the estimated model can be written as: Y = β 0 + β 1 (X -1996) where Y is the mortality rate, β 0 is the mean rate in the period studied, β 1 is the average annual increment and X corresponds to each year studied in the series.
The residual analysis showed that the use of simple linear regression models was appropriate due to normal distribution, homoscedasticity and absence of outliers.The trend was signifi cant when the adjusted model obtained p<0.05.The coeffi cient of determination (r 2 ) was used as a measure of accuracy of the models.

RESULTS
The proportion of deaths from pulmonary diseases increased among the elderly compared to all deaths, with higher rates seen among the oldest group.There was a decrease of percentage increment in almost all age groups and by gender for cardiovascular disease (Table 1).
The percent change in proportional mortality varied by causes of death, age groups and gender.There was around 100% growth in deaths from pulmonary diseases in all age groups and both genders.For deaths related to cardiovascular diseases, almost all age groups showed a decrease of approximately 10% in both males and females.
The annual mortality rates due to pulmonary and cardiovascular diseases showed an increasing trend in both genders (Figure 1).However, death rates from cardiovascular disease were four times higher than those of pulmonary diseases.When comparing rates between genders, cardiovascular rates were 15% lower in women and pulmonary rates were similar in both males and females.An increase in mortality rates was also seen by age group (Figure 2).Those aged 60-69 showed the lowest rate increment for both causes.The highest mortality rates were seen among the oldest males.age groups of both genders for both pulmonary and cardiovascular deaths.
For both causes of deaths, the mean rate for the time period ( ) was higher in those aged 80 or more in both genders and mean rates were higher in males than females.The highest average annual increases ( ) seen in both male and female elderly aged 80 or more were: +1.99 and +1.76 for pulmonary and +3.43 and + 3.08 for cardiovascular deaths, respectively.The average annual increments for those males and females aged 60-69 years were 0.15 and 0.11 for pulmonary and 0.42 and 0.30 for cardiovascular rates, respectively.

DISCUSSION
The present study is one of the fi rst trend analyses of mortality from pulmonary and cardiovascular diseases in elderly people outside large Brazilian urban centers such as the city of São Paulo, Southeastern Brazil.
Campos & Rodrigues 2 studied mortality trends in elderly people in Southeastern Brazilian states during 1980 to 2000 and found a reduction in mortality rates from all causes in both men and women.Similar results to ours were reported in studies of mortality trends from pulmonary diseases among the elderly in the state of São Paulo between 1980 and 1998: increasing rate trend in both genders, especially in those aged 80 or more. 5,8owever, the average increments in the mortality of elderly in the state of Mato Grosso are higher than those found in São Paulo. 5,8nce death rates from cardiovascular disease increased in all age groups, this fi nding is likely to be affected by diagnosis quality and access to complementary exams.Lolio 13 claimed that the risk of cardiovascular death has been signifi cantly reduced due to improved diagnosis quality and emergency care.
A comparison of raw mortality data due to pulmonary or cardiovascular disease among elderly people available from SIM and the Inter-Agency Health Information Network (RIPSA) showed differences across the entire time series, from more than 20% to 6.5% in the last year.This difference arises from the correction of the number : estimate of the average annual increment.
of deaths from ill-defined cause made by RIPSA, especially in the elderly group.We estimated corrected mortality rates, adjusted by age and gender, before examining our data to reduce the effect of mortality from ill-defi ned in the state of Mato Grosso.
Duarte et al 6 stressed the weakness of mortality data in Northern and Central-Western Brazilian states, especially for small cities and remote years.For this reason we chose to analyze aggregated data for the state as a whole and without specifying diagnostic groups of pulmonary and cardiovascular diseases in an attempt to reduce differential systematic errors between years and municipalities.
The signifi cant increase in pulmonary and cardiovascular deaths found in this study corroborates other similar studies, even when controlling for the effect of changes in population age distribution during the study period.Although an increased mortality was seen in all age groups and both genders, this increase was more remarkable in cardiovascular deaths, especially in elderly men and in those aged 80 and older.
For pulmonary deaths, the average annual rate in males aged 80 years and more was 10 times higher than that seen in those aged 60-69 years.A similar trend was seen among females.The same was found for cardiovascular deaths: the average annual rate in males and females aged 80 and more was eight and nine times greater than that seen in those aged 60-69 years, respectively.
Besides population aging and consequent increase in the proportion of people aged 60 years or more, it is believed that this population cohort has experienced different prior occupational and environmental exposures that might also affect mortality rates.Mato Grosso is a state located in the Amazônia Legal region and has distinctive characteristics compared to the rest of Brazil such as land use practices with environmental and social impact and health risks such as cardiopulmonary disease. 7,9e trend of mortality rates for cardiopulmonary diseases in the elderly in Mato Grosso can be partially explained by higher susceptibility of the elderly, especially to pulmonary and cardiovascular infections and their associated complications. 8,12There is a progressive decrease of pulmonary and cardiovascular functions with reduced lung elasticity, vital capacity and forced expiratory volume, as well as reduced ciliary function and cough refl ex.Frail elderly, when exposed to some factors such as air pollution, may develop pulmonary and cardiovascular conditions that require constant care. 14cause the elderly is a population segment more vulnerable to cardiopulmonary conditions, the analysis of mortality rates in the elderly could be used to identify variations in the rates of these conditions.A comparison with similar analyses conducted in other Brazilian regions would allow to formulating hypotheses about the effect of external factors that can contribute to increased mortality rates over time and provide input for a more detailed assessment of the evolution of mortality rates.Further studies are needed to better understand the determinants of these trends.
In conclusion, Mato Grosso was found to have high mortality rates for pulmonary and cardiovascular diseases in the elderly with signifi cant increasing trends among the oldest-old groups.
and by specifi c mortality rates according to the International Classifi cation of Diseases (ICD): Diseases of the Respiratory System (Chapters VIII of ICD-9 and Chapter X of ICD-10) and Diseases of the Circulatory System (Chapters VII of ICD-9 and Chapter IX of ICD-10).
Source: Brazilian National Health System Mortality Database (SIM/SUS)

Table 2
Figure 2. Specifi c mortality rates for respiratory and cardiovascular diseases in elderly population, by age.State of Mato Grosso, Central-Western Brazil, 1986-2006.

Tabela 2 .
Estimates from the linear regression equation, by gender and age groups.State of Mato Grosso, Central-Western Brazil, 1986-2006.