Trends in self-reported arterial hypertension in Brazilian adults : an analysis of data from the Brazilian National Household Sample Survey , 1998-2008

This study investigated the time trend in selfreported arterial hypertension from 1998 to 2008 in individuals older than 20 years in Brazil. Data were analyzed on prevalence of hypertension from the National Household Sample Survey (PNAD) for the years 1998, 2003, and 2008. The analysis considered the sampling complexity and included 203,419, 238,591, and 257,816 individuals in the years 1998, 2003, and 2008, respectively. The time trend in hypertension was analyzed by sex, family income, geographic regions of Brazil, and area of residence (urban versus rural), and the prevalence rates were adjusted for age. Prevalence of self-reported hypertension was 12.5% in 1998, 13% in 2003, and 13.9% in 2008, with an annual increase of 1.07%. The highest increases occurred in the Southeast Region (1.41%), in men (1.82%), and in urban areas (1.15%). The study showed a rise in the prevalence of self-reported hypertension in all income brackets and with increasing age. Knowledge of the prevalence of hypertension is essential for backing activities to prevent and control the problem, especially among the more vulnerable subgroups. Hypertension; Adult; Chronic Disease; Prevalence Introduction Systemic arterial hypertension (SAH) is a risk factor for cardiovascular, cerebrovascular, and renal diseases and accounts for some 40% of deaths due to stroke and 25% of deaths from coronary artery disease 1,2,3. SAH affects more than 30 million Brazilians, with an annual treatment cost of approximately US$ 398.9 million for the public Brazilian Unified National Health System (SUS) and US$ 272.7 million for the private healthcare system, representing 1.11% of overall healthcare expenditures 4. Negative effects have also been reported on quality of life for individuals with SAH, including physical, functional, emotional, and mental problems and various forms of pain 5. The detection, treatment, and control of systemic arterial hypertension are essential for the prevention and management of health outcomes related to the disease. Population-based surveys have thus been conducted in various countries to monitor trends in arterial hypertension (measured or self-reported), showing important increases in its prevalence. A study in Mexico based on three nationwide health surveys showed that prevalence of arterial hypertension increased from 23.8% in 1993 to 43.2% in 2006 6. In Brazil, the Telephone Survey for Surveillance of Risk and Protective Factors for Chronic Non-Communicable Diseases (VIGITEL) showed that self-reported arterial hypertension increased evenly and ARTIGO ARTICLE Muniz LC et al. 1600 Cad. Saúde Pública, Rio de Janeiro, 28(8):1599-1607, ago, 2012 significantly in the 27 State capitals included in the study, varying from 21.5% to 24.4% from 2006 to 2009 7. Developed countries like Switzerland, Canada, and the United States have also shown increases in the prevalence of self-reported arterial hypertension 8,9,10,11,12. The National Household Sample Survey (PNAD) provides representative data on self-reported arterial hypertension in the major geographic regions, States, and some metropolitan areas of Brazil. Although self-reported measures are subject to classification errors, Brazilian and international studies have shown that selfreported arterial hypertension appears to be a valid indicator for estimating prevalence of the disease 13,14,15,16. Monitoring the prevalence of systemic arterial hypertension, as assessed in the PNAD, helps identify the distribution of the disease and thus supports the implementation of appropriate health measures. In this context, the current study aimed to describe the time trend in the prevalence of self-reported arterial hypertension in Brazil, in adults twenty years and older, according to demographic and socioeconomic characteristics, using data from the PNAD for the years 1998, 2003, and 2008.


Introduction
Systemic arterial hypertension (SAH) is a risk factor for cardiovascular, cerebrovascular, and renal diseases and accounts for some 40% of deaths due to stroke and 25% of deaths from coronary artery disease 1,2,3 .SAH affects more than 30 million Brazilians, with an annual treatment cost of approximately US$ 398.9 million for the public Brazilian Unified National Health System (SUS) and US$ 272.7 million for the private healthcare system, representing 1.11% of overall healthcare expenditures 4 .Negative effects have also been reported on quality of life for individuals with SAH, including physical, functional, emotional, and mental problems and various forms of pain 5 .
The detection, treatment, and control of systemic arterial hypertension are essential for the prevention and management of health outcomes related to the disease.Population-based surveys have thus been conducted in various countries to monitor trends in arterial hypertension (measured or self-reported), showing important increases in its prevalence.A study in Mexico based on three nationwide health surveys showed that prevalence of arterial hypertension increased from 23.8% in 1993 to 43.2% in 2006 6 .In Brazil, the Telephone Survey for Surveillance of Risk and Protective Factors for Chronic Non-Communicable Diseases (VIGITEL) showed that self-reported arterial hypertension increased evenly and significantly in the 27 State capitals included in the study, varying from 21.5% to 24.4% from 2006 to 2009 7 .Developed countries like Switzerland, Canada, and the United States have also shown increases in the prevalence of self-reported arterial hypertension 8,9,10,11,12 .
The National Household Sample Survey (PNAD) provides representative data on self-reported arterial hypertension in the major geographic regions, States, and some metropolitan areas of Brazil.Although self-reported measures are subject to classification errors, Brazilian and international studies have shown that selfreported arterial hypertension appears to be a valid indicator for estimating prevalence of the disease 13,14,15,16 .Monitoring the prevalence of systemic arterial hypertension, as assessed in the PNAD, helps identify the distribution of the disease and thus supports the implementation of appropriate health measures.In this context, the current study aimed to describe the time trend in the prevalence of self-reported arterial hypertension in Brazil, in adults twenty years and older, according to demographic and socioeconomic characteristics, using data from the PNAD for the years 1998, 2003, and 2008.

Methods
The study was conducted with data from the health supplements included in the PNAD in  1998, 2003, and 2008.PNAD is a survey conducted by the Brazilian Institute of Geography and Statistics (Instituto Brasileiro de Geografia e Estatística -IBGE) and whose objective is to furnish basic information on the country's socioeconomic development.The survey has been conducted every five years since 1998 and has included a health supplement that provides a five-year historical series with national population-based information on health conditions and access, utilization, and financing of health services in Brazil 17 .
The sampling plan in the PNAD is conducted in up to three stages (municipalities or counties, census tracts, and households), thereby guaranteeing the sample's representativeness for the country, geographic regions, States, and some metropolitan areas 18 .The sample consisted of 344,975, 384,834, and 391,868 individuals in 1998,  2003, and 2008, respectively.Additional information on methodology and sampling can be found on the IBGE website 19,20 .
The current study population consisted of individuals 20 years of age or over.The question "Has a physician or other health professional ever said (...) that you have arterial hypertension (high blood pressure)?"was used to estimate the prevalence of self-reported arterial hypertension for each year.The variable was used dichotomously, as a "yes" or "no" answer.
Prevalence of self-reported arterial hypertension was analyzed according to sex, age, monthly per capita family income, major geographic regions, and area of residence (urban versus rural).Age was categorized in brackets: 20-29, 30-39, 40-49, 50-59, and 60 years or older.Monthly per capita family income was measured in reais and later categorized in income quintiles for Brazil.The area of residence variable was dichotomized as "urban" (city or town in an urbanized area, city or town in a non-urbanized area, and isolated urban area) and "rural" (rural cluster as an urban extension, isolated rural cluster or village, isolated rural cluster -nucleus, isolated rural cluster -other clusters and rural areas other than rural clusters).
The databases were obtained from the IBGE.Data were analyzed in the Stata software, version 11.0 (Stata Corp., College Station, USA), using the svyset command to define the sampling weights and clusters and the svy prefix for all the analyses, given the complexity of the sampling process.Prevalence rates for self-reported arterial hypertension were estimated for Brazil and its geographic regions, as well as annual percentage ranges.This information was also stratified by area of residence and sex.The prevalence rates for systemic arterial hypertension were adjusted for age.
The mean annual percentage variation was calculated using the prevalence values observed in 1998 and 2008.The total percentage variation was calculated, and the mean annual variation was estimated as its 10 th root: where p 1998 and p 2008 are the prevalence rates calculated for the years 1998 and 2008, respectively.The statistical test used to verify the time trend was the Wald test for linear trend.

Results
The current study only evaluated individuals twenty years or older, with information on selfreported arterial hypertension, totaling 203,419, 238,591, and 257,816 individuals in 1998, 2003, and 2008, respectively.
Table 1 shows the sample's characteristics for each year of the PNAD.In all three years there was a slight predominance of females and young adults (20-29 years).As for the geographic region and area of residence, the majority lived in the Southeast region and in urban areas in all three years, while there was an increase from 1998 to 2008 in the proportion of individuals living in the North of Brazil.The year 2008 showed a lower proportion of individuals from the lowest income quintile as compared to 1998.Table 2 describes the prevalence of selfreported arterial hypertension and the annual variation for Brazil and its geographic regions, according to sex and area of residence.At any rate, from 1998 to 2008 there was an increase in the prevalence of arterial hypertension in Brazil as a whole and in all the geographic regions, except for the North and Central-West.Except for the North (for males) and the North and Central-West (for females), all the other regions and the country as a whole showed an increase in the prevalence of SAH in both sexes from 1998 to 2008.In Brazil and in all the regions, the prevalence of self-reported arterial hypertension was always higher among women; however, the annual variation was always higher in men, reaching some 2% per year in the South and Southeast.For Brazil as a whole, the annual increase was lower in the rural as compared to urban areas (0.7% and 1.2%, respectively).In the rural areas in the South and Southeast, the annual increases in the prevalence of arterial hypertension were 1% and 1.6%, respectively.
Figure 1 shows the trends in the overall and sex-specific prevalence of self-reported systemic arterial hypertension according to age.From 20 to 49 years of age there was a trend towards stability in prevalence of arterial hypertension, with an upward trend in both sexes starting at 50 years.Prevalence of systemic arterial hypertension was higher in women than in men, especially from the 40-49-year bracket upward.
Figure 2 shows the trends in the overall and sex-specific prevalence of self-reported arterial hypertension according to per capita income Cad.Saúde Pública, Rio de Janeiro, 28(8):1599-1607, ago, 2012 quintiles.There is a clear upward trend in the prevalence rates for self-reported systemic arterial hypertension in all the income quintiles, except for women in the third quintile.In all the income groups, the prevalence was higher among women than men.

Discussion
Prevalence of self-reported arterial hypertension increased from 1998 to 2008 in Brazil and in its various geographic regions (1.07%per year).Although the estimated prevalence for all three study years was higher among women, the annual increase was higher in men.Self-reported arterial hypertension was proportional to age, and the highest increase was observed in age groups 50 years and older.Higher prevalence rates were also seen in residents of the South and in urban areas of the country.Nationwide surveys such as the PNAD are useful for understanding the panorama of public health problems such as arterial hypertension, by providing information capable of supporting public policies and strategies for the control of arterial hypertension, thus reinforcing the importance of using such nationwide data and studies 17 .The third health supplement of the PNAD allowed verifying a 10-year time trend in self-reported systemic arterial hypertension.
Despite the inherent limitations in obtaining self-reported morbidity data, there is a consensus in the literature that self-reported systemic arterial hypertension is a valid and appropriate indicator for estimating the frequency of the disease, with the advantage of rapid data collection and low cost 13,14,15,16 .Another fact to be considered in the utilization of the selfreported measure is the influence of access to health services, such that the prevalence rates may be overestimated.
The current study shows an increase in the prevalence of self-reported arterial hypertension in adults, over the course of 10 years (1998-2008).This finding corroborates other nationwide surveys, many of which in developed countries, using the same measure.However, the variation in definition of age brackets hinders any direct comparison between studies.In Switzerland, the prevalence of self-reported arterial hypertension in individuals 15 years or older increased from 22.1% to 24.2% in a 10-year period 12 .In the United States, self-reported arterial hypertension among individuals 18 years of age and older increased from 68.5% to 80.6% between 1988 and 2008 9,10,11 .Canada showed an increase from 8.2% in 1994 to 14.6% in 2005, among individuals 12 years or older 8 .In Mexico, among individuals 20 to 69 years of age, prevalence of self-reported SAH increased from 23.8% in 1993 to 43.2% in 2006 6 , and in the Czech Republic there was an increase from 41.4% in 1985 to 68.4% in 2008, among participants from 25 to 64 years of age 21 .
In the current study, the prevalence rates for self-reported arterial hypertension were higher among women in all three study years.Other studies on self-reported arterial hypertension have also reported higher prevalence among women 8,9,11,13,14,16,21,22 .Some studies that evaluated hypertension using more accurate methods such as repeat blood pressure measurements found a higher prevalence among men 23,24 .The current study's findings can be partially explained by women's greater use of health services 25 , which gives them more opportunity to receive a medical diagnosis of arterial hypertension.Another explanation is that obesity 26,27,28 and physical inactivity are more common among women 29 .However, a larger increase in prevalence, as observed in men, may mean that there is an increase in the risk of cardiovascular diseases in this group.This trend was observed in other studies 6,9,11,21 and may be related to the increase in obesity among men 30 .
Prevalence of self-reported systemic arterial hypertension increased with age, especially after 50 years, which is consistent with other studies showing more frequent reporting of arterial hypertension in older persons 8,9,24,31 .Older individuals tend to use health services more often due to the higher prevalence of disabilities and chronic illnesses in this age bracket.This increases the likelihood that these individuals will receive a medical diagnosis for some disease, like arterial hypertension, which could explain the findings.The current study also showed differences between the regions of Brazil and between areas of residence (urban versus rural).Urban residents presented higher prevalence rates.The relationship between arterial hypertension and urbanization may be associated with increased access to industrialized products by part of the population, lifestyle changes, daily stress 32 , and even greater access to medical services when compared to residents of rural areas of Brazil.
The current study showed a direct relationship between arterial hypertension and socioeconomic status.Travassos et al. 33 demonstrated that in Brazil, access to and utilization of the health system increased in proportion to income and schooling, reflecting the country's pattern of social inequalities, which could explain the finding.
Knowledge on the prevalence of arterial hypertension in Brazil is essential for supporting prevention and control measures that will contribute to reducing the impact of cardiovascular diseases at the population level, especially the subgroups most exposed to the disease.In conclusion, the study showed that overall prevalence of self-reported arterial hypertension increased during the 10-year period.This trend not only reflects increased access to health services, but also exposes the inadequacy of current strategies to fight arterial hypertension.The challenge thus remains to make health promotion and prevention measures more effective, through legislation and regulation, as well as educational activities, focusing especially on the reduction of sodium in foods, increased access to and consumption of fruits and vegetables, and treatment of arterial hypertension, among other effective measures.

Figure 1 TrendFigure 2
Figure 1Trend in prevalence of self-reported arterial hypertension according to age, for the total sample and stratifi ed by sex.National Household Sample Survey (PNAD),Brazil, 1998-2008.
Contributors L. C. Muniz was responsible for the literature review, data analysis, and writing of the article.A. M. Cascaes collaborated in the literature review and writing of the article.F. C. Wehrmeister collaborated in the literature review and data analysis.J. Martínez-Mesa collaborated in the literature review and data analysis.A. J. D. Barros oriented the study and revised the article.A. M. B. Menezes oriented the study and conducted the final revision of the article.
* Adjusted for cluster effect and sampling weight.

Table 2
Prevalence of self-reported arterial hypertension and annual variation for Brazil and regions according to sex and area of residence.National Household Sample Survey (PNAD),Brazil, 1998-2008.