Healthy behavior among Brazilian young adults

OBJECTIVE: To estimate the prevalence and factors associated to healthy behavior among young adults. METHODS: A total of 14,193 respondents aged 18–29 years who participated in the system Vigilância de Fatores de Risco e Proteção para Doenças Crônicas por Inquérito Telefônico (VIGITEL – Telephone-Based Surveillance of Risk and Protective Factors for Chronic Diseases) carried out in 27 Brazilian capitals in 2006 were studied. Healthy behavior was defi ned as non-smoking, reported regular physical activity and intake of fruits and vegetables fi ve days or more a week. Data analysis was based on prevalence ratios estimated using Poisson regression. RESULTS: The prevalence of healthy young adults was 8.0%; 39.6% reported two healthy behaviors, 45.3% one; and 7.0% none. In the multivariate analysis, healthy behavior was more commonly seen among those aged 25–29 years with 9 or more years of schooling and who reported engaging in physical activities near home. Inverse associations were found with non-white skin color, consumption of whole milk and fatty meat or poultry, being on a diet, and poor self-perception of health status. CONCLUSIONS: Young adults who show fewer healthy behaviors perceive their health as poor, which suggests that these behaviors negatively affect their own health perception. Positive associations with higher schooling, white skin color, and living near physical activity facilities indicate social inequalities in access to healthy behaviors. DESCRIPTORS: Health Knowledge, Attitudes, Practice. Gender and Health. Socioeconomic Factors. Risk Factors. Chronic Disease, prevention & control. Health Surveys. Brazil. Telephone interview. en_RSP43_S2.indb 1 13/11/2009 15:49:26 2 Healthy behavior among young adults Barreto SM et al Exposure to risk behaviors such as smoking, unhealthy diet, and physical inactivity usually starts in early adolescence and it consistently persists through adult life.6 These risk factors are associated to increased risk of chronic non-communicable diseases (CNCD) including cardiovascular diseases, diabetes, and cancer, which are leading causes of morbidity and mortality during adult life. Lifetime experiences and exposures, mostly from childhood to young adulthood, have long-term health impact and can contribute to health inequalities in adult and elder life.9,12,16 Epidemiological studies have evidenced that cardiovascular diseases would be a rare cause of death in the absence of main risk factors. The Interheart study was conducted in 52 countries and showed that factors such as dyslipidemia, smoking, hypertension, diabetes, abdominal obesity, psychosocial factors, low fruit and vegetable intake, excess alcohol intake, and physical inactivity account for 90% of population attributable risk in men and 94% in women, affecting all age groups and countries.25 Despite widespread knowledge, people fi nd it hard to have a healthier life. On one hand, several risk factors are associated to pleasure, promoted in the media and reinforced by modern consumerism. On the other hand, the health-related impact of behavioral risk factors is felt at more advanced ages as they are associated to conditions of slow and silent development that predominantly affect people after the age of 40. Public health interventions should promote healthy behavior among young people, preferably since childhood. The World Health Organization (WHO) launched in 2003 the Global Strategy on Diet, Physical Activity and Health that together with the Framework Convention on Tobacco Control constitute global recognition of the impact of these risk factors on people’s health and quality of life. These global agreements represent a commitment for political, economic, and social changes at national and international levels to support and promote healthier lifestyles. There are currently about 20 million adolescents aged 15–19 and 33 million young adults aged 20– 29 in Brazil. Young productive people are a major target of marketing strategies of consumption and leisure industry. The knowledge on the association between demographic, socioeconomic, and cultural factors and health behaviors among young adults can INTRODUCTION a Brazilian Ministry of Health. Informações de Saúde. Demográfi cas e Socioeconômicas. Brasília; 2006 [cited 2008 Jan 06]. Available from: http://tabnet.datasus.gov.br/cgi/deftohtm.exe?ibge/cnv/popuf.def b Brazilian Ministry of Health. VIGITEL Brazil 2006. Vigilância de fatores de risco e proteção para doenças crônicas por inquérito telefônico: estimativas sobre freqüência e distribuição sócio-demográfi ca de fatores de risco e proteção para doenças crônicas nas capitais dos 26 estados brasileiros e no Distrito Federal em 2006. Brasília; 2007. c The questionnaire is available from: http://hygeia.fsp.usp.br/nupens/questionario2006.pdf. provide input to support the implementation of public policies for health promotion and to meet specifi c needs of this population. The objective of the present study was to estimate the prevalence of and factors associated with healthy behaviors among young adults.

Exposure to risk behaviors such as smoking, unhealthy diet, and physical inactivity usually starts in early adolescence and it consistently persists through adult life. 6These risk factors are associated to increased risk of chronic non-communicable diseases (CNCD) including cardiovascular diseases, diabetes, and cancer, which are leading causes of morbidity and mortality during adult life.Lifetime experiences and exposures, mostly from childhood to young adulthood, have long-term health impact and can contribute to health inequalities in adult and elder life. 9,12,16idemiological studies have evidenced that cardiovascular diseases would be a rare cause of death in the absence of main risk factors.The Interheart study was conducted in 52 countries and showed that factors such as dyslipidemia, smoking, hypertension, diabetes, abdominal obesity, psychosocial factors, low fruit and vegetable intake, excess alcohol intake, and physical inactivity account for 90% of population attributable risk in men and 94% in women, affecting all age groups and countries. 25spite widespread knowledge, people fi nd it hard to have a healthier life.On one hand, several risk factors are associated to pleasure, promoted in the media and reinforced by modern consumerism.On the other hand, the health-related impact of behavioral risk factors is felt at more advanced ages as they are associated to conditions of slow and silent development that predominantly affect people after the age of 40.
Public health interventions should promote healthy behavior among young people, preferably since childhood.The World Health Organization (WHO) launched in 2003 the Global Strategy on Diet, Physical Activity and Health that together with the Framework Convention on Tobacco Control constitute global recognition of the impact of these risk factors on people's health and quality of life.These global agreements represent a commitment for political, economic, and social changes at national and international levels to support and promote healthier lifestyles.provide input to support the implementation of public policies for health promotion and to meet specifi c needs of this population.
The objective of the present study was to estimate the prevalence of and factors associated with healthy behaviors among young adults.VIGITEL has set a minimum sample of 2,000 individuals aged ≥18 years per city to estimate the prevalence of risk factors in adults with 95% confi dence and maximum error of two percent points.Considering 50% prevalence, the error would be 2%.b The sampling process consisted of systematic drawing of 5,000 telephone lines per city followed by redrawing of telephone lines in each city and grouping into 25 replicates of 200 lines.All (active) eligible home telephone lines were identifi ed.All individuals aged ≥18 years living in the household were listed and one of them was drawn to be interviewed.A total of 76,330 telephone calls were made, and 54,369 interviews were complete (71.1%) with response rates ranging between 64.4% (Porto Alegre, RS) and 81.0%(Palmas, TO).Unanswered calls were due to permanently busy lines, lines connected to fax or answering machines or impossibility to locate the adult selected from a household after 10 attempts.The refusal rate was 9.1%, ranging from 5.4% (Palmas) to 15.0% (Porto Alegre).Of 54,369 respondents VIGITEL in 2006, 21,294 were males and 33,075 were females.

Cross
Telephone interviews followed a computer-assisted structured questionnaire completed by trained interviewers who received constant supervision.c The survey method was described by Moura et al  (2008) and published elsewhere. 19om all interviews (54,369) made through VIGITEL, a sample comprising 14,193 respondents aged between 18 and 29, excluding pregnant women (n=280) and respondents of Asian (n=67) and indigenous (n= 9) ethnicity, was studied.
The dependent variable "healthy behavior" was defi ned as: mild to moderate leisure-time physical activity for at least 30 minutes fi ve or more days a week or intense activity for at least 20 minutes a day for three or more days a week; fruit and vegetable intake fi ve or more days a week; and non-smoking.Respondents who did not meet one or both criteria were considered unhealthy.

Independent
variables were grouped into sociodemographic and behavior characteristics and health indicators.Sociodemographic characteristics were: gender, age (18-24 and 25-29 years old), selfreferred skin color (white, black, and mixed), marital status (single, married, separated/divorced/widowed), schooling (0-8, 9-11, ≥12 years), employment status/ employed (yes, no), and physical activity facility in the neighborhood (yes, no).Behavior variables included: regular consumption of fatty meat and/or poultry (yes, no), consumption of whole milk (yes, no), and excess alcohol intake (>5 doses among men and >4 doses among women in the 30 days prior to the interview).Health indicators studied were: body mass index estimated based on self-reported weight and height (BMI <25, 25-29 and >30 kg/m 2 ); current dieting (yes, no) and dieting in the last 12 months (yes, no); medical diagnosis of hypertension (yes, no) and diabetes (yes, no); and self-perceived health status (excellent, good, fair, poor).
The association between healthy behaviors and independent variables was analyzed using Pearson's chi-square test at a 5% level of signifi cance.Univariate analysis was performed based on prevalence ratios obtained from Poisson's regression with robust variance.The multivariate analysis was fi rst carried out including independent variables associated to healthy behavior with p<0.20 in the univariate analysis.The fi nal model included all variables that remained statistically associated (p<0.05) in intermediate analyses.
To correct for different probability of selection due to number of telephone line in the household and number of people living in the same household, weighting factors were assigned to each respondent.Given that fi xed telephone coverage is not universal, a weighting factor was used to approach the sample's sociodemographic composition to that of people aged 18 years and more in each city.Additionally, a second weighting factor was applied to take into account differences between populations of each city and number of individuals selected per city. 19The analysis was conducted in Stata (version 9.2) using "svy" procedure (including weighting factors) for the analyses of data obtained in complex sampling.
As it was a telephone survey, the informed consent form was replaced by a verbal consent from all respondents at the time of phone contact.VIGITEL was approved by the Comitê Nacional de Ética em Pesquisa para Seres Humanos (Human Research Ethics Committee) of the Brazilian Ministry of Health. 19
Non-smoking was the most common healthy behavior (85.5%) reported followed by regular physical activity (46.4%).Adequate fruit and vegetable intake was the least frequently reported healthy behavior (16.6%).
In the univariate analysis, healthy behavior among young adults was positively associated to age 25-29 years, higher schooling and physical activity facility in the area.Having mixed or black skin color and being divorced/separated/widowed had a negative association with healthy behavior.Gender was not associated to healthy behavior (Table 1).
Young adults with healthy behaviors less frequently reported excess alcohol intake and consumption of fatty meat and/or poultry and whole milk.They also reported less current dieting and dieting in the 12 months prior to the interview (Table 2).
Healthy behavior was negatively associated with self-perceived health status, and a gradient was seen in this association.There was no association between healthy behavior and obesity, hypertension, and diabetes (Table 3).
In the multivariate analysis, healthy behavior was more common among adults aged 25-29 years with 9-11 and ≥12 years of schooling and who reported physical activity facility in their area.An inverse association was found with mixed or black skin color, consumption of whole milk and fatty meat and current dieting.A negative association remained between healthy behavior and self-perceived health status (Table 4).

DISCUSSION
The study results showed low prevalence of healthy behaviors among young adults studied.Healthy behavior was infl uenced by sociodemographic characteristics (gender, age, schooling, and marital status), availability of a physical activity facility in the neighborhood and self-perceived health status.In addition, it was negatively associated to other health behaviors such as risk alcohol consumption.
The prevalence of adequate physical activity was close to that found in other studies with young adults but the rate of non-smokers was higher.Fruit and vegetable intake could not be compared to other studies including young adults as they assessed daily intake.In a national survey investigating risk behaviors in the United States, 78% of respondents aged 18-24 consumed less than fi ve servings of fruits and vegetables a day, 43% reported inadequate physical activity and 29% were smokers. 24Among German college students, it was found that more than 95% consumed less than fi ve servings of fruits and vegetables a day, 60% had inadequate physical activity and 31% were smokers. 13oking prevalence in the sample studied was also lower than that found among young adults in Chile 5 and other Latin American countries. 23This lower rate refl ects a trend of reduced smoking prevalence in Brazil resulting from successful anti-smoking public policies that have been implemented for the last 20 years.Current ban of smoking in bars and restaurants can further reduce smoking prevalence among young adults.Studies have demonstrated that young adults are more responsive to anti-smoking actions than older adults, especially when they have to face increasing barriers that prevent them from smoking. 7A systematic review on the effectiveness of anti-smoking interventions among young adults showed higher success rates of combined individual and institutional/ community-based initiatives despite the small number of controlled studies and different approaches. 7ghting smoking is key due to its harmful health effect and association with other risk factors.A prospective study investigating women aged 20-29 showed increased risk of excess alcohol consumption among smokers, and this risk was even higher among those with sex initiation at the age of 15. 18 The association found in the present study between healthy behavior and lower alcohol consumption supports this fi nding.
Healthy eating and adequate physical activity are behaviors learned during childhood but reinforced during adolescence. 3In the present study, the fi nding of a large proportion of young adults with inadequate eating habits points toward widespread fast food consumption, great availability of industrialized foods, and high cost and diffi cult preservation of fresh foods.Moreover, healthy behavior was found to be inversely associated to the consumption of fatty meat and whole milk.Despite the existing initiatives to promote healthy eating in Brazil, problems such as income, the increasing practice of eating out and the low adherence to healthy food in workplace and school cafeterias are all barriers to changes of eating habits.
In Brazil, food insecurity is paradoxically associated with higher rates of obesity, especially among women, and changes are only likely to occur with policies for preventing poverty and reducing inequalities. 8,14e positive impact of nutritional education among low-income families was demonstrated in a controlled community study, which showed that an intervention including information and cooking classes increased fruit and vegetable intake. 10ysical activity can reduce lifetime morbidity.Despite being aware of that, many people claim they lack time, feel embarrassed to engage in collective physical activities or inapt to engage in exercises, and fi nd it not pleasurable. 1,2Physical activity is also infl uenced by opportunities and resources available.In the present study, the availability of adequate physical activity facility in the neighborhood had a positive effect on healthy behavior.A study with health service users found that living near areas suitable for walking and biking increases the chances of engaging in physical activity. 22And people engaging in physical activity in a community encourage other people to do that. 20r results indicate that healthy behaviors among young adults were associated to their level of education and skin color.A study with young Finnish adults reported that schooling was the most strongly associated factor to smoking. 15In the United Kingdom, healthier behaviors among children and young adults was found to be positively associated with more favorable socioeconomic conditions. 4In contrast with other studies, 11,21 we did not fi nd an association between employment status and healthy behavior.
In Finland, an increased prevalence of metabolic syndrome was seen among young adults aged 24-39, especially among males. 15VIGITEL and other population-based surveys carried out in Brazil will allow monitoring trends of risk behavior prevalence in the long run and targeting specifi c measures to this group.
Telephone-based surveys have the advantage of lower cost and greater access to remote places.However, there is a disadvantage regarding representativeness of respondents because access to fi xed telephone lines is not universal in Brazil and poor population can be underrepresented.Fixed telephone coverage rates in Brazil ranges between 31.4% and 89.9% in the lowest and highest quintiles, respectively, of per capita family income distribution of households in the cities participating in VIGITEL. 20In this sense, it was identifi ed in the United States lower participation of young adults in telephone-based surveys possibly due to increasing use of cellular phones among them. 16In Brazil, 4 the effect of telephone coverage on prevalence estimates in population-based surveys was minimal in metropolitan areas with coverage equal to or greater than 70% in the southeastern, southern, and central-western regions.But low telephone coverage in households in the northern and northeastern regions can result in biased estimates that do not correspond to the realities studied.Hence, although a weighting factor was applied to approach the sample's sociodemographic composition to that of the population in the analysis, it was decided not to stratify data per region in the present study.
Despite the relevance of different participation due to local characteristics, high response rate seen in VIGITEL contributed to good quality of data.Bearing in mind that healthy behaviors are even less prevalent among poor people who usually do not have fi xed telephone lines at home, the prevalence of healthy behaviors found in this study may be overestimated, which makes it even more concerning.
Due to its cross-sectional design, the study does not allow any inferences on causality.But it is crucial to know the characteristics associated to healthy behaviors for identifying groups with specifi c needs and guiding public policies for adequate monitoring of these behaviors.
Declining morbidity and mortality rates due to cardiovascular diseases seen worldwide is a result of prevention actions and better health care.
However, many CNCD determinants cannot be tackled with standard medical care.Only 4% of premature deaths occur due to inadequate care while 40% of them are attributed to individual behaviors such as dietary patterns, physical activity, smoking, consumption of alcohol and recreational drugs, and sexual behavior. 17e fi nding of a small proportion of young adults engaging in healthy behaviors points to the need to promote positive lifestyle changes among young adults.These actions can only be effective if access to healthy behaviors is also increased and facilitated.
Many prevention initiatives with impact on health rely on policies that are implemented outside the health sector such as taxes on tobacco products, regulation of food products, public safety policies, and investments to increase access to leisure and physical activities. 17As young adults' personal preferences are also strongly infl uenced by culture of immediate, easy, and purchasable pleasure, 17 understanding the social, cultural, and anthropological aspects related to healthy behaviors is also critical for successful prevention strategies.
-sectional study conducted based on data from the Telephone-Based Surveillance of Risk and Protective Factors for Chronic Diseases (VIGITEL) b including a probabilistic sample of adults living in households with fi xed telephone line in 26 Brazilian capitals and Federal District in 2006.

Table 1 .
Healthy behavior according to sociodemographic characteristics and availability of physical activity facility in the neighborhood among young adults aged18-29.Brazil, 2006.(N=14,193) a Information available for 13,181 young adults

Table 2 .
Healthy behavior according to lifestyle factors in young adults aged18-29.Brazil, 2006.(N=14,193) a Information available for 12,487 young adults

Table 3 .
Healthy behavior according to health indicators in young adults aged18-29.Brazil, 2006.(N=14,193) a Information available for 14,165 young adults b Information available for 13,329 young adults BMI: Body mars index