Introduction

Overweight is considered as a relevant Public Health issue. The fundamental cause of obesity and overweight is an energy imbalance between calories consumed and calories expended,^1,2 and it presents a favorable condition to the emergence of chronic non-communicable diseases (NCDs) - for example, heart diseases, cancers, and diabetes mellitus, among others.³

The global prevalence of obese individuals has increased considerably when compared to obesity prevalence in 1975 and 2016, which nearly tripled. In 2016, the number of individuals who were overweight was greater than those who were underweight, a result observed in almost all continents except in parts of Sub-Saharan Africa and Asia. In that same year, nearly 2 billion adults aged 18 years and over were overweight, of these over 650 million were obese.¹ In Brazil, in 2019, 55% of adults were overweight and 20% were obese.⁴

Given the high prevalence of overweight and obesity, in 2011, the Ministry of Health implemented a Strategic Action Plan for Tackling Non-communicable Diseases (NCDs), aimed to face these diseases and associated factors, such as sociodemographic (sex, age and schooling) and behavioral (poor diet, physical inactivity, smoking and alcohol consumption) characteristics.⁵

Assessing nutritional status and its associated factors is of fundamental importance for the analysis of the population health conditions.⁶ This study provides updates on the health status of Brazilian adults regarding nutritional status, and may serve as a guide for the planning of strategic actions in order to enable changes in the country’s morbidity and mortality scenario.

The aim of this study was to analyze the prevalence and factors associated with overweight in the Brazilian state capitals and Federal District in 2019.

Methods

This was a cross-sectional study, based on data from the Surveillance System for Risk and Protective Factors for Chronic Diseases by Telephone Survey (Vigitel), collected by telephone interviews between January and December 2019.⁴

Vigitel is a telephone survey coordinated by the Ministry of Health. In the sampling process it used the ZIP codes (CEPs) - of 5 million telephone numbers per Brazilian state capital and the Federal District in order to randomly select household samples. Adult participants aged 18 years or older, living in the Brazilian state capitals and Federal District and who had a landline telephone at home were eligible for the 2019 Vigitel.⁴ In each state capital and the Federal District, the eligible telephone numbers were randomly selected again, and fractionated into replicas of 200 lines.

Among the eligible households there was another random selection, the last one, for the selection of one of the adult residents (≥18 years old). After the random selection stage, and before starting the interview, the resident was asked about his interest in participating in the study. The objective of the interview was to identify the frequency of factors associated with the development of NCD, considering a 95% confidence interval (95%CI) and 2% maximum error.⁴

The study outcome was overweight: yes; no.

The following independent variables were searched:

All variables were obtained from self-reported information collected by a data collection company, hired to conduct telephone interviews.⁴

Overweight was checked by calculating body mass index - BMI≥25.0kg/m² -,⁴ obtained by dividing self-reported weight (in kilograms: kg) by self-reported height (in meters squared: m).

The independent variables of food consumption were estimated based on the questions answered by the participants:

Participants who gave a positive answer to the question Are you currently smoking?, without taking into account the frequency, quantity and duration of smoking, were classified as smokers.

For the construction of the indicator ‘physical inactivity’, questions were asked about physical activity during his/her free time, occupational activity, locomotion, and hygiene in his/her residence.

The indicator ‘television viewing habit’ (less than 3 hours/day; greater than or equal to 3 hours/day) was constructed considering the answer to the following question: How many hours a day do you usually watch television?

Regarding the variables on health status, the indicator of morbidity reported for diabetes mellitus and hypertension was assessed by the following questions: Has a doctor ever told you that you have high blood pressure? and Has a doctor ever told you that you have diabetes?, respectively. For self-rated health, the following question was asked: How would you classify your health status: very good, good, regular, poor or very poor?

The calculation of the research sample was based on the minimum sample size (about 2,000 individuals per city), 95%CI, 2% maximum error and the periodicity of each risk factor in the adult individuals who were interviewed. A number of 197,600 landline telephones were randomly selected, aimed at achieving at least 2,000 interviews per state capital. Only 75,789 eligible telephones were identified, and the final sample included n=52,443 individuals interviewed and an average of 1,943 interviews conducted by state capital.⁴

Descriptive statistics were used to calculate the relative frequency of the variables, and the prevalence of overweight-obesity was estimated with the respective 95%CI. Poisson regression with robust variance was used in order to calculate the association between the variables, stratified according to gender, through the crude and adjusted prevalence ratio (PR) and its 95%CI, with a 5% significance level for the tests (p≤0.05). In the analyses of the adjusted model, only the independent variables that, in the crude model, presented p<0.20 were considered. The data analyses were conducted with Stata software, version 14.0, using the complex sample module (survey), taking as a reference the raking method.⁴

Vigitel project was evaluated and approved by the National Research Ethics Committee (CONEP), linked to the National Health Council (CNS): Opinion No. 4324071, published on June 6, 2017; Certificate of Submission for Ethical Appraisal (CAAE) No. 65610017.1.0000.0008. The individual’s consent was obtained orally, at the time of the telephone interview.

Results

A total of 197,600 landline telephones were randomly selected, but only 75,789 telephones were eligible. 52,443 individuals were interviewed by the end of the survey. The losses occurred, given the impossibility of the first approach to the selected household because either the telephone lines remained busy and nobody answered the phone calls or the call went straight to voicemail; furthermore, it was not possible to contact the adult resident selected for the individual interview, even after a great number of calls on different days, at different times.⁴

Among the 52,443 adults interviewed, the characteristics that presented the highest percentages were: female gender (54.0% - 95%CI 53.1;55.0), white race/skin color (43.8% - 95%CI 42.9;44.8), aged 25 to 34 years (25.0% - CI95% 24.1;26.1), 9-11 years of schooling (38.4% - 95%CI 37.5;39.3) and single marital status (42.8% - CI95% 41.9;43.8) (Table 1). In 2019, 55.4% (95%CI - 54.4;56.3) of adults were overweight or obese, with the highest percentage observed in males: 57.1% - 95%CI 55.6;58.7.

Brown men (58.5% - 95%CI 56.1;60.9) in the age group 35-44 (67.7% - 95%CI 64.3;70.8) had a higher percentage of overweight when compared to brown women (55.0 - 95%CI 53.2;57.0) in the age group 35-44 (55.6 - 95%CI 53.0;58.1). The highest percentage of overweight was observed in women who had 0 to 8 years of schooling (62.0% - 95%CI 60.0;64.1). The other results on percentages of overweight can be observed in Table 2.

A statistically significant association with overweight was observed in men aged 35 to 44 (PR=1.87 - 95%CI 1.65;2.12), 12 years of schooling or more (PR=1.15 - CI95% 1.07;1.24), married (PR=1.23 - 95%CI 1.15;1.32), who consumed alcoholic beverages (PR=1.09 - CI95% 1.03;1.15), with arterial hypertension (PR=1.24 - 95%CI 1.17;1.31) and diabetes mellitus (PR=1.09 - CI95% 1.01;1.19) (Table 3). Men who reported consuming fruit regularly (PR=0.92 - 95%CI 0.87;0.97) and smokers (PR=0.86 - CI95% 0.78;0.96) had a lower prevalence of the outcome.

Overweight was more frequent in self-reported black women (PR=1.25 - 95%CI 1.04;1.51), aged 45 to 54 (PR=2.03 - 95%CI 1.77;2,32), married (PR=1.15 - 95%CI 1.08;1.22), who reported consuming alcoholic beverages (PR=1.05 - 95%CI 1.01;1.11), and having television viewing habit, 3 hours/day or over (PR=1.09 - 95%CI 1.04;1.15), with arterial hypertension (PR=1.21 - CI95% 1.15;1.27) and diabetes mellitus (PR=1.09 - CI95% 1.03;1.15), and self-rated health as poor/very poor (PR=1.31 - 95%CI 1.20;1.43). The lowest prevalence of the outcome was observed in women with 12 years of schooling or more (PR=0.87 - 95%CI 0.81;0.93) and smokers (PR=0.84 - CI95% 0.76;0.94) (Table 4).

Discussion

Based on data from the 2019 Vigitel survey, it could be seen that overweight was found in most of the interviewees. The highest percentage was observed in men. In both sexes, the highest prevalence of the outcome was associated with excessive alcohol consumption, high blood pressure and glycemic index. Being married, with television viewing habit for more than 3 hours a day and self-rated health as negative were associated with high prevalence of overweight and obesity in females.

The main limitation of the study was due to the fact that it was based on self-reported data. Self-reported data may not represent reality, due to the possibility of the respondents’ perception bias, such as overestimation of their height and underestimation of their weight, resulting in an underestimation of BMI and, consequently, underestimation of overweight prevalence.⁷ Therefore, we highlight the importance of using correction methods for self-reported data. Another limitation of the study was the sample representativeness, restricted to the inhabitants of the state capitals and Federal District, where few individuals have a landline telephone, although it was minimized with the use of data weighting methods adopted by Vigitel, through post-stratification, aimed at estimating the prevalence, taking into consideration the different demographic characteristics of the participants, compared to those of the other Brazilians.⁸

The high prevalence of the outcome in this study is similar to the results presented in the National Health Survey (PNS),⁹ which observed the highest representation of overweight among its participants. The 2019 Vigitel⁴ record reported that between 2006 and 2019, in all Brazilian regions, there was a significant temporal variation of increase in the prevalence of overweight and obesity, possibly explained by the worsening of this health problem. It is worth emphasizing the relevance of knowledge of the associated factors and the quest for strategies for their control and prevention.

The highest prevalence of the outcome was found in men aged 35 to 44 years, and in women aged 45 to 54 years. These findings corroborate the results of a systematic review and data analysis from 183 countries, in the period between 1980 and 2013, which observed that people of all ages gained weight; and those aged 20 and 40 gained weight faster.¹⁰

However, aging is a factor that influences weight gain. During this phase of life, the body undergoes several physiological changes, such as the menopause process in women, the decline of physical activities, muscle mass and metabolic rate.¹¹ A cross-sectional study collected data from 981 Brazilian adult women (20 to 60 years old) and found the highest prevalence of obesity in those who were older (50 to 60 years old), pointing to a possible association between high prevalence of obesity and advanced stages of life.¹²

Married men and women had the highest prevalence of overweight and obesity. Marriage may lead to excessive weight gain, due to the fact that married people have less involvement in practices that could contribute to weight control, such as physical activity and a healthy diet.¹³ Prioritizing people and their partners regarding overweight prevention initiatives is important for the improvement of the well-been of couples and the community.

Women who reported having a higher level of education showed low prevalence of overweight-obesity; regarding men, the opposite was observed. Data from Vigitel (2006 and 2016), also presented a lower prevalence of overweight and obesity in women who had higher educational level, while in men, the factor ‘education’ did not present the same result.¹⁴

In a study that used information from Vigitel (2006 and 2012), when analyzing the variation in body weight and the change in nutritional status of Brazilians older than 20 years of age, it could be seen a greater weight gain in women interviewed and who had a higher education level. This finding can be attributed to social factors, such as changes in work relations between genders and the incorporation of women into higher education and better-paying occupation.¹⁵ These factors influence weight gain, which are directly related to the increase in the acquisition and intake of industrialized foods and fast food.¹⁶

A lower prevalence of overweight and obesity was observed in men who reported consuming fruits regularly. Regular consumption of healthy foods, such as fruits, prevents NCDs, and these foods are sources of nutrients, fiber and water. A diet based on the consumption of healthy foods contributes to the stabilization of ideal body weight, in addition to contributing to the improvement of well-being.¹⁷

The high prevalence of individuals who are overweight was also associated with alcohol consumption. Similar results were found in the National Food Survey, conducted between 2008 and 2009, when the habit of drinking alcoholic beverages was positively associated with body fat accumulation. There is also the possibility that alcohol consumption is directly related to the intake of foods with high caloric density.¹⁸

The lower prevalence of the outcome was associated with smokers who were interviewed. This finding confirms data from the 2015 Vigitel,¹⁹ which observed a lower prevalence of overweight in smokers when compared to nonsmokers. This inverse relationship is attributed to the effect of nicotine, a substance present in cigarettes: it causes decreased appetite and increased satiety, leading to reduced body weight.²⁰

Obesity is a multifactorial disease. In addition, it is also associated with a reduction of physical activity level, resulting from changes in the forms of leisure, the increase in the use of computers, smartphones and tablets, and television viewing.²¹ The results of this study are in line with this behavioral change: it could be seen a direct relationship between overweight and obesity and the time women spend watching TV.

The study showed a significant association between overweight and reported high blood pressure for both sexes. A positive relationship between overweight and obesity and hypertension was also found in a study on prevalence and factors associated with hypertension among Brazilian adults.²² Obese individuals are up to three times more likely to develop hypertension.²³

The results of this study showed a significant association between overweight and self-reported diabetes mellitus, regardless of gender. The 2013 PNS, in turn, showed a strong association between overweight and obesity and diabetes mellitus, although with a higher prevalence of diabetes in women.⁹ A review study,²⁴ which sought evidence on the relationship between overweight and diabetes mellitus, found that overweight and obesity are directly related to onset of type 2 diabetes, and also to the implications of metabolic imbalance, blood glucose levels and the development of complications.

This study showed that poor or very poor self-rated health is an indication of a higher percentage of overweight among women, a result similar to that found in a survey conducted with adults in a city in the state of Santa Catarina.²⁵ In a population-based household survey in Belo Horizonte, state capital of Minas Gerais, there was a significant association between poor self-rated health, low intake of healthy foods and physical activity level lower than recommended, showing that people with negative self-rated health may present low adherence to a healthy life style.²⁶

Taking these results, it can be concluded that the study allowed us to understand the prevalence of overweight and its main associated factors. The results showed an association between overweight and obesity and interviewees’ sociodemographic, behavioral factors and health status. The knowledge of these factors is relevant for the development of intersectoral strategies, encouraging a healthy lifestyle, prioritizing health interventions for modifiable risk factors.