TY - JOUR
AU - de Assis Guedes de Vasconcelos, Francisco
AB - Abstract Among the causes of obesity, environmental factors have also been studied, in addition to genetic, social, psychological, and hormonal factors. The distribution of food outlets, facilitating or hindering food acquisition, can promote body weight control by encouraging healthier food habits. The objective of this study was to investigate associations between environmental availability and utilization of food outlets and overweight/obesity in 7 to 14-year-old schoolchildren in Florianópolis, in the South of Brazil. A logistic regression analysis identified a positive association between overweight/obesity in 2195 schoolchildren and the presence of restaurants in the vicinity of their homes (buffer = 400 meters). Being a member of a family that utilizes public markets/greengrocers was also positively associated with overweight/obesity in the sample investigated. Identifying the distribution of these establishments in the vicinity of the homes of schoolchildren in middle-income countries is an important element in understanding the role played by the food environment in weight gain in a variety of different settings. buildings, constructions and points, epidemiology, young people Introduction Environmental characteristics such as the availability of food outlets in the area surrounding people's homes can have a positive or negative influence on eating habits and, as a result, on rates of overweight and obesity.1–3 People living near vegetable markets, greengrocers and other establishments that sell healthy foods (fruit and vegetables) or put these items in prominent positions can experience a positive influence on their eating habits.1,4,5 While supermarkets offer foods considered unhealthy, and do so at lower prices, which can involve a risk of body weight gain,6 access to large supermarkets has a preventative effect on obesity, because they also sell fruit and vegetables.5,6 Considering establishments that sell food for immediate consumption, the presence and high concentrations of fast food restaurants and convenience stores in the neighborhoods where children and adolescents live can have a negative impact on the quality of their diets5–8 and on body weight gain,3,6,7 because these establishments tend to sell less healthy food.9,10 In spite of this, a large proportion of these studies only investigate the availability of places that sell food and do not take into consideration whether the populations being investigated actually use them.11–15 Furthermore, the majority of studies of food environments and obesity have been conducted in North America, which means that findings cannot be extrapolated to middle income countries16,17 such as Brazil. Research shows that the amount of money that Brazilian families spend on eating away from home increased from 2002–03 to 2008–09.18,19 The more recent of these surveys found that 47.9% of adolescents habitually eat away from their homes and that this was responsible for 14.1% of their total daily energy intake. Moreover, more than one-third of the family food budget is spent on food consumed outside of the home.19 The objective of this paper is to investigate associations between environmental availability and utilization of places that sell food and overweight/obesity among 7–14-year-old schoolchildren in a city in the South of Brazil. Method Florianopolis is a coastal city situated in the South region of Brazil. Almost all of its citizens live in the urban area (96.2%).20 In 2010, Florianopolis’ Index of Human Development (IHD) was classified as very high (0.847), surpassing the national index for the same year (0.727), which was classified as high.21 Despite its high IHD, the city has evident social inequalities. In 2010, Florianopolis’ Gini Index was 0.5474 (the closer to 1.000, the greater the social inequalities between residents).22 In 2016, Florianopolis had a population of 477 798. This cross-sectional study is nested within a larger project that is based on a probabilistic sample of 7–14-year-old schoolchildren from public and private schools in Florianopolis. The methodological procedures have been described in detail elsewhere.23,24 Outcome prevalence data from previous studies conducted in Florianopolis in 200225 and 200726 were used to predict prevalence rates for the most recent cross-sectional survey, giving a final sample size of 2880 children. Additional simulations showed that even assuming a 30% prevalence of excess weight and a 5% prevalence of exposure, this sample size would still be large enough to detect odds ratios (ORs) ≥1.5 with 80% power and an alpha error of 5%. Data on individuals’ variables were collected from September 2012 to June 2013. The guardians of all schoolchildren selected for the study were sent free and informed consent forms. The criterion used to select anthropometrists was quality of anthropometric measurements, determined by technical error of measurement.27 Body mass and height were measured in order to calculate Body Mass Index (BMI), following procedures recommended in the literature.28 The cut-off for classification of overweight/obesity was BMI > z score +1 (equivalent to a BMI ≥25 kg/m2 at 19 years of age).29 Schoolchildren and their families self-reported socio-economic and demographic data and details about their lifestyle habits, which were categorized post hoc as follows: sex (male/female); age (7–10 or 11–14 years); type of school (public/private); whether (yes/no) schoolchildren and/or their families utilized the following food outlets: restaurants, snack bars/fast food outlets, street vendors, supermarkets, mini markets, butchers, bakeries and greengrocers/public markets; educational level of mother (≤completed primary education and ≥started secondary education); families’ income (terciles); means of transport used by the schoolchildren to get to school (active [walking or by bicycle] or passive [motorcycle, car or bus]). Environmental variables in the vicinity of the schoolchildren's homes were surveyed by secondary data collection. A list of all retailers within the municipal district was requested from the Florianopolis City Council Sanitary and Environmental Surveillance Department and a variety of strategies were used to supplement and confirm this information: consulting telephone directories, both printed and online, the websites of fast food chains, supermarkets, industry associations for tourism and restaurants, and commercial buildings and shopping centers to identify food outlets they hosted. Delivery-only food outlets and establishments that serve-specific populations were excluded (such as those located inside schools, companies, universities, hotels, bed and breakfasts, gyms, sports clubs and beauty salons). The locations of restaurants, snack bars/fast food outlets, street vendors, supermarkets, mini markets, butchers, bakeries and greengrocers/public markets in Florianopolis were recorded. Data for Florianopolis from the 2010 demographic census on monthly nominal income per household were assessed according to weighting areas, which are groups of census tracts in which results exhibit statistical significance and can be analyzed.30 Spatial analysis was conducted by viewing and exploring the data.31 Each establishment was manually spatialized by inputting its address on Google Earth™. The Street View™ tool was also used to identify the establishments’ locations with greater precision. As part of this manual spatialization process, when there was any doubt with relation to address, name, type of service or date of establishment, the firm's tax registration status was consulted. Additional Internet queries (on official websites and customer relationship websites) were conducted to determine whether establishments were still doing business and the types of services they provided, after which any corrections needed were made to the classification of retailer types. An automatic process was also used to conduct spatialization of the schoolchildren's homes using the open-access software package Quantum GIS (QGIS 2.0). Within this program, 400 m buffer zones were defined in the areas surrounding the schoolchildren's homes and the numbers of each different type of establishment within them were calculated. Other studies have also chosen similar buffers.32–35 The numbers of food outlets were used to calculate proportions, means, standard deviations (SD) and minimum and maximum values for each type of establishment within the buffers. The sample was described by constructing a Table of frequency distributions for the variables studied, by the schoolchildren's BMI. ORs and their respective 95% confidence intervals (95% CI), estimated by crude logistic regression analysis, were used to analyze factors associated with the outcome overweight/obesity, without adjustment for the other study variables. Estimates of association were then adjusted using a logistic regression model. All independent variables (individual—sociodemographic and individuals’ characteristics and environmental—tertiles of income for the weighting area in which the child's home was located and the presence/absence of establishments that sell food in the vicinity [within 400 m] of the schoolchildren's homes) were included in the adjusted model, independent of their P-values in the crude analysis. The two environmental variables studied (tertiles of income for the weighting area and the presence/absence of establishments that sell food in the vicinity of the schoolchildren's homes) were at different scales. The income variable is calculated per weighting area, which are larger-scale areas with sizes defined by the government for the purposes of surveys, and each of the 30 weighting areas studied has a different size, calculated in square kilometers.20 The measure (buffer) used for the second variable is a circle around the children's homes with a radius of 400 m, and hence has a smaller area than the first measure. Tests were conducted to identify possible interactions between variables on the individual level and between variables on the environmental level and no significant interactions were detected. Data were analyzed using Stata, version 13.0, considering complex sampling and sample weighting with the software's SVY command. The research protocol was approved by the Human Research Ethics Committee at Universidade Federal de Santa Catarina, hearing number 120341/2012. Results A total of 2506 7–14-year-old schoolchildren returned the free and informed consent forms correctly signed by their parents and took part in the research project. However, for this nested study, 4.1% of the total sample of schoolchildren for whom data were collected were not residents of Florianopolis, 4.2% did not provide their addresses, and a further 4.2% of the schoolchildren's homes could not be mapped because their addresses were incomplete or could not be found on the mapping. The total sample analyzed for this article was therefore 2195 schoolchildren (87.6% of those who returned the free and informed consent forms). As can be seen in Table 1, a majority of the study population were male, aged from 7 to 10 years, enrolled at public schools and reported using passive transport (car, motorcycle or bus) to commute from their homes to their schools and back. A majority of the schoolchildren's mothers had remained in education up to or beyond the start of secondary education. The prevalence of overweight/obesity observed was 29.0% for girls and 37.6% for boys (data not shown). Table 1 Distribution of the prevalence of overweight/obesity and crude logistic regression analysis results, by individual and environmental variables for the schoolchildren investigated, Florianopolis, SC, Brazil, 2012/13 Characteristic  Schoolchildren n (%)  Overweight/obese % (95% CI)  OR (95% CI)  P-value  Sex   Female  1046 (47.67)  29.01 (22.75–36.18)  1.00  <0.01*   Male  1149 (52.33)  37.61 (28.91–47.19)  1.46 (1.22–1.75)    Age group   7–10 years  1307 (59.55)  34.81 (27.68–42.70)  1.00  <0.01*   11–14 years  888 (40.45)  30.58 (20.72–42.61)  0.77 (0.64–0.93)    Type of school   Public  1346 (61.31)  33.94 (25.90–43.03)  1.00  0.86   Private  849 (38.69)  31.78 (0.94–95.83)  1.01 (0.83–1.23)    Transport to school   Active  746 (34.25)  31.96 (26.47–37.99)  1.00  0.33   Passive  1431 (65.75)  33.79 (24.91–43.98)  1.03 (0.85–1.24)    Mother's educational level   ≤Graduated primary education  540 (25.27)  31.71 (30.33–33.12)  1.00  0.34   ≥Started secondary education  1595 (74.73)  34.19 (24.92–44.85)  0.94 (0.84–1.06)    Monthly family income (tertile)   Top tertile  629 (32.20)  31.62 (18.38–48.71)  1.00  0.74   Middle tertile  606 (31.03)  37.16 (31.11–43.64)  1.16 (0.92–1.46)     Bottom tertile  718 (36.77)  32.48 (24.88–41.12)  1.04 (0.82–1.32)    Income for weighting area   High income  715 (32.56)  31.44 (21.08–44.04)  1.00  0.40   Middle income  811 (36.95)  33.36 (21.76–47.39)  1.07 (0.85–1.34)     Low income  669 (30.49)  34.40 (32.77–36.06)  1.10 (0.88–1.37)    Establishments present in the vicinity of home—400 m   Restaurant  1619 (73.75)  35.78 (27.82–44.61)  1.51 (1.22–1.86)  <0.01*   Greengrocers/public markets  650 (29.62)  34.45 (23.39–47.49)  1.11 (0.90–1.36)  0.29   Snack bar/fast food outlet  1503 (68.47)  34.81 (28.73–41.42)  1.21 (1.01–1.46)  0.04*   Street vendors  103 (4.67)  37.60 (32.15–43.38)  0.93 (0.63–1.40)  0.76   Supermarket  445 (20.27)  38.98 (32.07–46.35)  1.23 (0.98–1.54)  0.07   Mini market  1596 (72.72)  34.85 (28.99–41.22)  1.21 (0.99–1.47)  0.06   Bakery  1019 (46.42)  35.83 (28.06–44.43)  1.19 (0.99–1.43)  0.05   Butcher  169 (7.70)  34.71 (20.65–52.05)  1.06 (0.75–1.49)  0.71  Establishments utilized by schoolchildren and/or family   Restaurant    33.19 (24.77–42.83)  1.01 (0.59–1.74)  0.95   Greengrocers/public markets    33.85 (27.10–41.33)  1.32 (0.74–2.34)  0.22   Snack bar/fast food outlet    34.39 (27.90–41.53)  1.39 (0.78–2.45)  0.17   Street vendors    35.02 (28.87–41.72)  1.25 (1.08–1.46)  0.02*   Supermarket    32.82 (26.39–39.96)  0.86 (0.34–2.13)  0.62   Mini market    31.75 (24.22–40.36)  0.87 (0.62–1.10)  0.13   Bakery    33.26 (26.06–41.33)  1.16 (0.69–1.96)  0.42   Butcher    31.85 (26.30–37.98)  0.92 (0.74–1.14)  0.29  Characteristic  Schoolchildren n (%)  Overweight/obese % (95% CI)  OR (95% CI)  P-value  Sex   Female  1046 (47.67)  29.01 (22.75–36.18)  1.00  <0.01*   Male  1149 (52.33)  37.61 (28.91–47.19)  1.46 (1.22–1.75)    Age group   7–10 years  1307 (59.55)  34.81 (27.68–42.70)  1.00  <0.01*   11–14 years  888 (40.45)  30.58 (20.72–42.61)  0.77 (0.64–0.93)    Type of school   Public  1346 (61.31)  33.94 (25.90–43.03)  1.00  0.86   Private  849 (38.69)  31.78 (0.94–95.83)  1.01 (0.83–1.23)    Transport to school   Active  746 (34.25)  31.96 (26.47–37.99)  1.00  0.33   Passive  1431 (65.75)  33.79 (24.91–43.98)  1.03 (0.85–1.24)    Mother's educational level   ≤Graduated primary education  540 (25.27)  31.71 (30.33–33.12)  1.00  0.34   ≥Started secondary education  1595 (74.73)  34.19 (24.92–44.85)  0.94 (0.84–1.06)    Monthly family income (tertile)   Top tertile  629 (32.20)  31.62 (18.38–48.71)  1.00  0.74   Middle tertile  606 (31.03)  37.16 (31.11–43.64)  1.16 (0.92–1.46)     Bottom tertile  718 (36.77)  32.48 (24.88–41.12)  1.04 (0.82–1.32)    Income for weighting area   High income  715 (32.56)  31.44 (21.08–44.04)  1.00  0.40   Middle income  811 (36.95)  33.36 (21.76–47.39)  1.07 (0.85–1.34)     Low income  669 (30.49)  34.40 (32.77–36.06)  1.10 (0.88–1.37)    Establishments present in the vicinity of home—400 m   Restaurant  1619 (73.75)  35.78 (27.82–44.61)  1.51 (1.22–1.86)  <0.01*   Greengrocers/public markets  650 (29.62)  34.45 (23.39–47.49)  1.11 (0.90–1.36)  0.29   Snack bar/fast food outlet  1503 (68.47)  34.81 (28.73–41.42)  1.21 (1.01–1.46)  0.04*   Street vendors  103 (4.67)  37.60 (32.15–43.38)  0.93 (0.63–1.40)  0.76   Supermarket  445 (20.27)  38.98 (32.07–46.35)  1.23 (0.98–1.54)  0.07   Mini market  1596 (72.72)  34.85 (28.99–41.22)  1.21 (0.99–1.47)  0.06   Bakery  1019 (46.42)  35.83 (28.06–44.43)  1.19 (0.99–1.43)  0.05   Butcher  169 (7.70)  34.71 (20.65–52.05)  1.06 (0.75–1.49)  0.71  Establishments utilized by schoolchildren and/or family   Restaurant    33.19 (24.77–42.83)  1.01 (0.59–1.74)  0.95   Greengrocers/public markets    33.85 (27.10–41.33)  1.32 (0.74–2.34)  0.22   Snack bar/fast food outlet    34.39 (27.90–41.53)  1.39 (0.78–2.45)  0.17   Street vendors    35.02 (28.87–41.72)  1.25 (1.08–1.46)  0.02*   Supermarket    32.82 (26.39–39.96)  0.86 (0.34–2.13)  0.62   Mini market    31.75 (24.22–40.36)  0.87 (0.62–1.10)  0.13   Bakery    33.26 (26.06–41.33)  1.16 (0.69–1.96)  0.42   Butcher    31.85 (26.30–37.98)  0.92 (0.74–1.14)  0.29  *P < 0.05. Restaurants were the most common type of food outlet in the vicinity of the children's homes (buffer = 400 m), followed by mini markets and then snack bars/fast food outlets. The crude logistic regression indicated associations between overweight/obesity and being male and between overweight/obesity and the 7–10 years age group. There were also associations between overweight/obesity and the presence of restaurants and snack bars/fast food outlets close to home and with utilization of street vendors by schoolchildren (Table 1). A total of 2286 food outlets in Florianopolis were mapped. The number of food outlets in the vicinity of the home of each schoolchild (buffer = 400 m) exhibited considerable variability, especially with relation to places to eat when not at home, such as restaurants, snack bars/fast food outlets and street vendors, as shown in Table 2. Table 2 Numbers of different types of food outlets in the vicinity of the homes of the 7–14-year-old schoolchildren investigated, considering a 400 meter buffer, Florianopolis, SC, Brazil, 2012/13 Type of establishment  n (%)  Mean (SD)  Range  Restaurants  1009 (44.1)  4.8 (±9.7)  0–87  Greengrocers  78 (3.4)  0.4 (±0.6)  0–4  Butchers  21 (0.9)  0.1 (±0.3)  0–3  Snack bars/fast food outlets  682 (29.8)  3.7 (±8.8)  0–105  Street vendors  57 (2.5)  0.3 (±2.3)  0–45  Supermarkets  33 (1.4)  0.2 (±0.5)  0–3  Mini markets  260 (11.8)  1.3 (±1.3)  0–7  Bakeries  146 (6.4)  0.9 (±1.6)  0–12  TOTAL  2286 (100%)      Type of establishment  n (%)  Mean (SD)  Range  Restaurants  1009 (44.1)  4.8 (±9.7)  0–87  Greengrocers  78 (3.4)  0.4 (±0.6)  0–4  Butchers  21 (0.9)  0.1 (±0.3)  0–3  Snack bars/fast food outlets  682 (29.8)  3.7 (±8.8)  0–105  Street vendors  57 (2.5)  0.3 (±2.3)  0–45  Supermarkets  33 (1.4)  0.2 (±0.5)  0–3  Mini markets  260 (11.8)  1.3 (±1.3)  0–7  Bakeries  146 (6.4)  0.9 (±1.6)  0–12  TOTAL  2286 (100%)      We have estimated a multivariate logistic regression model (Table 3). Male schoolchildren were more likely to be overweight/obese than the females. Schoolchildren from families who utilized public markets/greengrocers also exhibited a greater likelihood of being overweight/obese compared with schoolchildren whose families did not utilize these retailers. With regard to the environmental characteristics, the presence of restaurants in the vicinity of a schoolchild's home was associated with overweight/obesity, but their utilization by the schoolchildren investigated was not associated with this condition. Table 3 Adjusted logistic regression models for identifying factors associated with overweight/obesity among the 7–14-year-old schoolchildren investigated, Florianopolis, SC, Brazil, 2012/13.   Adjusted analysisa  OR (95% CI)  P-value  Individual characteristics   Sex (male)  1.42 (1.13–1.77)  <0.01*   Age group (11–14 years)  0.90 (0.71–1.14)  0.38   Type of school (private)  1.10 (0.80–1.52)  0.56   Mother's educational level (≥Started secondary education)  0.95 (0.79–1.15)  0.61   Monthly family income        Middle tertile  1.34 (0.96–1.84)  0.42    Bottom tertile  1.20 (0.83–1.74)     Food outlets utilized        Restaurant  0.86 (0.60–1.25)  0.43    Snack bar/fast food outlet  1.13 (0.79–1.61)  0.48    Street vendor  1.16 (0.88–1.53)  0.28    Supermarket  0.71 (0.42–1.26)  0.26    Mini market  0.81 (0.64–1.04)  0.10    Bakery  1.11 (0.78–1.58)  0.58    Greengrocer/public market  1.54 (1.06–2.24)  0.02*    Butcher  0.91 (0.71–1.16)  0.43  Environmental characteristics   Income of weighting area        Middle tertile  1.09 (0.81–1.45)  0.58    Bottom tertile  1.07 (0.79–1.45)     Presence of establishments        Restaurant  1.52 (1.13–2.06)  <0.01*    Snack bar/fast food outlet  1.00 (0.75–1.34)  0.99    Street vendor  0.80 (0.46–1.39)  0.44    Supermarket  1.13 (0.83–1.54)  0.44   Mini market  1.22 (0.93–1.60)  0.14    Bakery  1.07 (0.84–1.37)  0.57    Greengrocer/public market  0.92 (0.71–1.19)  0.52    Butcher  0.92 (0.60–1.42)  0.72    Adjusted analysisa  OR (95% CI)  P-value  Individual characteristics   Sex (male)  1.42 (1.13–1.77)  <0.01*   Age group (11–14 years)  0.90 (0.71–1.14)  0.38   Type of school (private)  1.10 (0.80–1.52)  0.56   Mother's educational level (≥Started secondary education)  0.95 (0.79–1.15)  0.61   Monthly family income        Middle tertile  1.34 (0.96–1.84)  0.42    Bottom tertile  1.20 (0.83–1.74)     Food outlets utilized        Restaurant  0.86 (0.60–1.25)  0.43    Snack bar/fast food outlet  1.13 (0.79–1.61)  0.48    Street vendor  1.16 (0.88–1.53)  0.28    Supermarket  0.71 (0.42–1.26)  0.26    Mini market  0.81 (0.64–1.04)  0.10    Bakery  1.11 (0.78–1.58)  0.58    Greengrocer/public market  1.54 (1.06–2.24)  0.02*    Butcher  0.91 (0.71–1.16)  0.43  Environmental characteristics   Income of weighting area        Middle tertile  1.09 (0.81–1.45)  0.58    Bottom tertile  1.07 (0.79–1.45)     Presence of establishments        Restaurant  1.52 (1.13–2.06)  <0.01*    Snack bar/fast food outlet  1.00 (0.75–1.34)  0.99    Street vendor  0.80 (0.46–1.39)  0.44    Supermarket  1.13 (0.83–1.54)  0.44   Mini market  1.22 (0.93–1.60)  0.14    Bakery  1.07 (0.84–1.37)  0.57    Greengrocer/public market  0.92 (0.71–1.19)  0.52    Butcher  0.92 (0.60–1.42)  0.72  *P < 0.05. aAnalysis adjusted for all independent variables. Discussion Main finding of this study The presence of restaurants in the vicinity of the schoolchildren's homes and a child's family utilizing public markets/greengrocers were both positively and significantly associated with overweight/obesity. What is already known on this topic An earlier study conducted in 2002 in Florianopolis with 7–10-year-old children found that 27.6% of the girls and 32.9% of the boys were overweight and/or obese.25 In 2007, the prevalence rates observed were 32.5% for girls and 36.2% for boys, among 7–14-year-old schoolchildren in the same city.26 Our current data apparently reflect a trend among boys in Florianopolis towards increasing prevalence of overweight/obesity over the last decade, and oscillation in the prevalence rates among girls. The findings reported in the literature-related to overweight/obesity in schoolchildren and the distances from their homes to restaurants are still contradictory. A study conducted in Virginia (USA) did not detect any relationship between the proximity of restaurants and the BMI of 5–19-year-old schoolchildren.34 In contrast, a study conducted in Minnesota (United States) found that proximity of restaurants to home was associated with a higher BMI among female adolescents.36 Along the same lines, lower BMI values were observed among pediatric patients (4–18 years old) who lived farther from restaurants in Massachusetts (United States).14 Among adults, a study conducted in Canada identified that the density of restaurants had a negative influence on the BMI.12 In contrast, a study conducted in Brazil with adults did not identify any associations between the prevalence of excess body weight and the presence of supermarkets or fruit and vegetable stores near to people's homes.11 The differences between these results could be due to different analytical methods or the characteristics of products sold by food retailers in different cities. In relation to farmer's market availability, a study conducted in North Carolina (USA) found that the habit of utilizing this type of retailer where healthy foods were available was associated with lower BMI.37 Easy access to and availability of a wide range of fresh foods at a relatively low price and of good quality can encourage adoption of healthier dietary habits and, as a consequence, facilitate maintenance of body weight.38 Those results contradict our findings. What this study adds We have noticed that there are not many studies in the literature investigating the association between environmental availability of restaurants and overweight/obesity in children and adolescents. This may be because investigations of fast food restaurants and convenience stores have been given priority, since these establishments are very common in the USA, which is where a large proportion of studies of the food environment have been conducted. In Brazil, there are full service restaurants and buffet restaurants that sell food by weight in all areas, serving highly varied clienteles. These types of restaurants offer a very diverse range of foods and preparation styles,39 which can include both healthy and unhealthy foods and beverages. Studies investigating the foods bought in these establishments near students’ homes could help to elucidate these results. There is a possibility that the association between overweight/obesity and farmer's markets reflects reverse causality. Utilization of public markets/greengrocers by the families of schoolchildren who are overweight/obese is a strategy for purchasing healthier foods for consumption at home, since these public markets sell fruit and vegetables. On the other hand, farmer's markets in Florianopolis also sell biscuits, cheeses and cold cuts, that the schoolchildren's families may be buying. However, in this study we are unable to state which foods the study population purchases at these establishments. Future investigations could elucidate this better. The presence of other types of food outlets in the vicinity of the children's homes was not associated with overweight/obesity, in contrast with what has been observed in other studies, in which supermarkets6,13,40 and snack bars/fast food outlets have been identified.13,37,41 Notwithstanding, in one review study investigating the relationship between the built environment and nutrition, it was observed that, in general, studies that report a positive relationship between these variables may be more likely to be published than those with findings indicating null associations, which could explain the low number of articles identified in the literature with findings reporting a non-relationship between these retailers and overweight/obesity.42 Another reason could be that students are choosing different stores and places for their meals that were not examined in the study (schools, shopping malls and friends’ homes, for example). However, it is expected that the school meals provided at public schools and school canteens found at private schools do not promote overweight/obesity in schoolchildren because there are strong regulations against sweetened and high-caloric foods.43,44 Limitations of this study This study did not consider establishments that only sell food via delivery and as electronic shopping increases the implications of this type of commerce for locations, sizes and types of food outlets merits consideration.45 Also, we did not assess businesses that were not primarily dedicated to selling food, such as rentals, utility stores, pharmacies or news stands. However, commercial establishments that are not traditionally associated with selling food and drink may provide opportunities to buy unhealthy items.46 Although the information used to construct the database of food outlets in Florianopolis was triangulated, the sources of information used to supplement the database are primarily maintained by the retailers themselves, which could lead to lower rates of coverage of establishments that are not fully legal and of businesses in lower income neighborhoods.47 In counterpoint, not employing self-report anthropometric measurements, training the data collection team and conducting quality control of the information collected were essential to guarantee the quality of the data. Another strong point of this study is the methodology employed to survey the municipal district for food outlets. The methods employed to list these places using secondary data involved using several different sources of data, improving the validity of the information used.48 Conclusion We did not observe associations between overweight/obesity among the schoolchildren investigated and the presence of snack bars/fast food outlets, supermarkets, mini markets or bakeries in the vicinity of their homes. The association with the presence of restaurants but not with their utilization must be considered within a given context, specific to the scenario in Florianopolis, which is a city that attracts large numbers of tourists and in which there are very many establishments of this type throughout the municipal district. Future studies could investigate the food groups sold at each type of establishment, especially in restaurants and greengrocers, in order to more objectively specify which establishments can be considered healthy, unhealthy or mixed. Authors’ Contributions Designed the study: ENC and FAGV; Collection of data: ENC; Performed the analyses: DASS and ENC; Wrote the article: ENC and CER; Reviewed the manuscript: FAGV, DASS and JN. All authors contributed to writing the manuscript and approved the manuscript. Conflict of interest The authors declare no conflict of interest. Funding The authors are grateful to Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq - National Council for Scientific and Technological Development), a grant agency affiliated to the Ministery of Science, Technology and Inovation of Brazil (MCTI/CNPq n. 014/2011 – CNPq n° 483955/2011-6) for its financial support. ENC received a research Grant from the Fundo de Apoio à Manutenção e ao Desenvolvimento da Educação Superior , a grant from the Education Secretary of Santa Catarina State (FUMDES/UNIEDU) and from the Programa Ciência Sem Fronteiras . References 1 Brazil, Guia Alimentar para a população Brasileira . Ministério da Saúde, Brasília, Brazil, 2014. Available online: http://portalsaude.saude.gov.br/images/pdf/2014/novembro/05/Guia-Alimentar-para-a-pop-brasiliera-Miolo-PDF-Internet.pdf (25 November 2016, date last accessed). 2 Lydecker JA, Cotter E, Gow RW et al.  . Preventing childhood obesity. In: Choate LH (ed). Eating Disorders and Obesity—A Couselor´s Guide to Prevention and Treatment . Alexandria, USA: American Couseling Association, 2013, 169– 200. 3 Carroll-Scott A, Gilstad-Haydena K, Rosenthala L et al.  . Disentangling neighborhood contextual associations with child body mass index, diet, and physical activity: the role of built, socioeconomic, and social environments. Soc Sci Med  2013; 95: 106– 14. Google Scholar CrossRef Search ADS PubMed  4 Black C, Ntani G, Cooper C et al.  . Measuring the healthfulness of food retail stores: variations by store type and neighbourhood deprivation. Intern J Behav Nutr and Phys Activ  2014; 11: 1. Google Scholar CrossRef Search ADS   5 Burton LM, Kemp SP, Leung M et al.  . Communities, Neighborhoods, and Health: Expanding the Boundaries of Place . New York, USA: Springer, 2011. Google Scholar CrossRef Search ADS   6 Shier V, An R, Sturm R. Is there a robust relationship between neighbourhood food environment and childhood obesity in the USA? Pub Health  2012; 126: 723– 30. Google Scholar CrossRef Search ADS   7 Pabayo R, Spence JC, Cutumisu N et al.  . Sociodemographic, behavioural and environmental correlates of sweetened beverage consumption among pre-school children. Pub Health Nutr  2012; 15: 1338– 46. Google Scholar CrossRef Search ADS   8 He M, Tucker P, Gilliland J et al.  . The influence of local food environments on adolescents’ food purchasing behaviors. Intern J Environ Res Pub Health  2012; 9: 1458– 71. Google Scholar CrossRef Search ADS   9 Centers for Disease Control and Prevention. Childhood Obesity: Causes and Consequences, in: Division of Nutrition, Physical Activity, and Obesity. Available online: http://www.cdc.gov/obesity/childhood/causes.html (1 march 2016, date last accessed). 10 Corrêa EN, Schmitz BAS, Vasconcelos FAG. Aspects of the built environment associated with obesity in children and adolescents: a narrative review. Rev Nutr  2015; 28: 327– 40. Google Scholar CrossRef Search ADS   11 Velásquez-Meléndez G, Mendes LL, Padez CMP. Built environment and social environment: associations with overweight and obesity in a sample of Brazilian adults. Cad Saúde Pública  2013; 29: 1988– 96. Google Scholar CrossRef Search ADS PubMed  12 Hollands S, Campbell MK, Gilliland J et al.  . A spatial analysis of the association between restaurant density and body mass index in Canadian adults. Prev Med  2013; 57: 258– 64. Google Scholar CrossRef Search ADS PubMed  13 Cetateanu A, Jones A. Understanding the relationship between food environments, deprivation and childhood overweight and obesity: evidence from a cross sectional England-wide study. Health Place  2014; 27: 68– 76. Google Scholar CrossRef Search ADS PubMed  14 Fiechtner L, Sharifi M, Sequist T et al.  . Food environments and childhood weight status: effects of neighborhood median income. Child Obes  2015; 11: 260– 8. Google Scholar CrossRef Search ADS PubMed  15 Chen H-J, Wang Y. Changes in the neighborhood food store environment and children's body mass index at Peripuberty in the United States. J Adolesc Health  2016; 58: 111– 9. Google Scholar CrossRef Search ADS PubMed  16 Johnson F, Michelle P, Jane W. Socio-economic status and obesity in childhood. In: Moreno LA, Pigeot I, Wolfgang A (eds). Epidemiology of Obesity in Children and Adolescents - Prevalence and Etiology . New York, USA: Springer, 2011: 377– 90. Google Scholar CrossRef Search ADS   17 Mayne SL, Auchincloss AH, Michael YL. Impact of policy and built environment changes on obesity-related outcomes: a systematic review of naturally occurring experiments. Obes Rev  2015; 16: 362– 75. Google Scholar CrossRef Search ADS PubMed  18 Brazil, Instituto Brasileiro de Geografia e Estatística. Pesquisa de orçamentos familiares 2002–03: análise da disponibilidade domiciliar de alimentos e do estado nutricional no Brasil / IBGE, Coordenação de Trabalho e Rendimento. Rio de Janeiro: IBGE, 2004.74 p. Available online: http://downloads.ibge.gov.br/downloads_estatisticas.htm (25 November 2016, date last accessed). 19 Brazil, Instituto Brasileiro de Geografia e Estatística. Pesquisa de orçamentos familiares 2008–09: análise do consumo alimentar pessoal no Brasil / IBGE, Coordenação de Trabalho e Rendimento. Rio de Janeiro: IBGE, 2011.150 p. Available online: http://downloads.ibge.gov.br/downloads_estatisticas.htm (25 November 2016, date last accessed). 20 Brazil, Instituto Brasileiro de Geografia e Estatística. Censo Demográfico 2010: Resultados Gerais da Amostra por áreas de ponderação - municípios com áreas redefinidas. Available online: http://www.ibge.gov.br/home/estatistica/populacao/censo2010/resultados_gerais_amostra_areas_ponderacao/default_redefinidos.shtm (25 November 2016, date last accessed). 21 Programa das Nações Unidas para o Desenvolvimento (PNUD)/ Instituto de Pesquisa Econômica Aplicada (IPEA) / Fundação João Pinheiro. Índice de Desenvolvimento Humano Municipal Brasileiro – Brasília: PNUD, Ipea, FJP, 2013. 96 p. Available online: http://www.ipea.gov.br/portal/index.php?option=com_content&id=24037 (25 November 2016, date last accessed). 22 Brazil, Ministério da Saúde. Índice de Gini da renda domiciliar per capita. Available online: http://tabnet.datasus.gov.br/cgi/idb2011/b09capc.htm (25 November 2016, date last accessed). 23 D'Avila GL, Muller RL, Gonsalez PS et al.  . Associação entre estado nutricional da mãe e a frequência, local e companhia durante as refeições com o sobrepeso/obesidade de adolescebtes da cidade de Florianópolis, Brasil. Braz J Mother and Child Health  2015; 15: 289– 99. 24 Motter AF, Vasconcelos FAG, Corrêa EN et al.  . Pontos de venda de alimentos e associação com sobrepeso/obesidade em escolares de Florianópolis, Santa Catarina, Brasil. Cad Saúde Pública  2015; 31: 620– 32. Google Scholar CrossRef Search ADS PubMed  25 Assis MA, Rolland-Cachera MF, Grosseman S et al.  . Obesity, overweight and thinness in schoolchildren of the city of Florianopolis, Southern Brazil. Eur J Clin Nutr  2005; 59: 1015– 21. Google Scholar CrossRef Search ADS PubMed  26 Bernardo CO, Fernandes PS, Campos RMMB et al.  . Associação entre o índice de massa corporal de pais e de escolares de 7 a 14 anos de Florianópolis, SC, Brasil. Braz J Mother and Child Health  2010; 10: 183– 90. 27 Habicht JP. Estandarizacion de metodos epodemiologicos cuantitativos sobre el terreno. Boletin de la Oficina Sanitaria Panamericana  1974; 76: 375– 84. Google Scholar PubMed  28 Lohman TG, Roche AFE, Martoreli R Anthropometric Standardization Reference Manual . Illinois, USA: Human Kintics Books, 1991. 29 De Onis M, Onyango AW, Borghi E et al.  . Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Org  2007; 85: 649– 732. Google Scholar CrossRef Search ADS   30 Brazil, IBGE. Censo Demográfico 2010: características da população e dos domicílios. Instituto Brasileiro de Geografia e estatística, Rio de Janeiro. Brazil, 2011. Available online: http://downloads.ibge.gov.br/downloads_estatisticas.htm (25 November 2016, date last accessed). 31 Brazil, Introdução à Estatística Espacial para a saúde Pública . Secretaria de Vigilância em Saúde. Ministério da Saúde, Brasília, Brazil, 2007. Available online: http://www.escoladesaude.pr.gov.br/arquivos/File/TEXTOS_CURSO_VIGILANCIA/capacitacao_e_atualizacao_em_geoprocessamento_em_saude_3.pdf (25 November 2016, date last accessed). 32 Kwate NO, Loh JM. Separate and unequal: the influence of neighborhood and school characteristics on spatial proximity between fast food and schools. Prev Med  2010; 51: 153– 6. Google Scholar CrossRef Search ADS PubMed  33 Leung CW, Gregorich SE, Laraia BA et al.  . Measuring the neighborhood environment: associations with young girls’ energy intake and expenditure in a cross-sectional study. Int J Behav Phys Act  2010; 7: 1. Google Scholar CrossRef Search ADS   34 Ellaway A, Macdonald L, Lamb K et al.  . Do obesity-promoting food environments cluster around socially disadvantaged schools in Glasgow, Scotland? Health Place  2012; 18: 1335– 40. Google Scholar CrossRef Search ADS PubMed  35 Wall MM, Larson NI, Forsyth A et al.  . Patterns of obesogenic neighborhood features and adolescent weight: a comparison of statistical approaches. Am J Prev Med  2012; 42: e65– 75. Google Scholar CrossRef Search ADS PubMed  36 Mellor JM, Dolan CB, Rapoport RB. Child body mass index, obesity, and proximity to fast food restaurants. Int J Pediatr Obes  2011; 6: 60– 8. Google Scholar CrossRef Search ADS PubMed  37 Pitts SBJ, Acheson MLM, Ward RK et al.  . Disparities in healthy food zoning, farmers’ market availability, and fruit and vegetable consumption among North Carolina residents. Arch Pub Health  2015; 73: 1. Google Scholar CrossRef Search ADS   38 Cerin E, Frank LD, Sallis JF et al.  . From neighborhood design and food options to residents’ weight status. Appetite  2011; 56: 693– 703. Google Scholar CrossRef Search ADS PubMed  39 Santos MV, Proença RP, Fiates GMR et al.  . Os restaurantes por peso no contexto de alimentação saudável fora de casa. Rev Nutr  2011; 24: 641– 50. Google Scholar CrossRef Search ADS   40 Harrison F, Jones AP, Van Sluijs EM et al.  . Environmental correlates of adiposity in 9–10 year old children: considering home and school neighbourhoods and routes to school. Soc Sci Med  2011; 72: 1411– 9. Google Scholar CrossRef Search ADS PubMed  41 Fraser LK, Edwards KL. The association between the geography of fast food outlets and childhood obesity rates in Leeds, UK. Health Place  2010; 16: 1124– 8. Google Scholar CrossRef Search ADS PubMed  42 Caspi CE, Sorensen G, Subramanian SV et al.  . The local food environment and diet: a systematic review. Health Place  2012; 18: 1172– 87. Google Scholar CrossRef Search ADS PubMed  43 Brazil, Lei n. 12061, de 18 de dezembro de 2001. Dispõe sobre critérios de concessão de serviços de lanches e bebidas nas unidades educacionais, localizadas no estado de Santa Catarina. 2001. http://www.leisestaduais.com.br/sc/lei-ordinaria-n-12061-2001-santa-catarina-dispoe-sobre-criterios-de-concessao-de-servicos-de-lanches-e-bebidas-nas-unidades-educacionais-localizadas-no-estado-de-santa-catarina/ (14 November 2016, date last accessed). 44 Brazil, Lei n. 11947 de 16 de junho de 2009. PNAE – Programa Nacional de Alimentação Escolar. http://www.planalto.gov.br/ccivil_03/_ato2007-2010/2009/lei/l11947.htm/ (14 November 2016, date last accessed). 45 Ratclife J. Built environment futures research: the need for foresight and scenario learning. In: Knight A, Ruddock L (eds). Advanced Research Methods in the Built Environment . USA: Wiley-Blackwell, 2008: 13. 46 Farley TA, Rice J, Bodor N et al.  . Measuring the food environment: shelf space of fruits, vegetables and snack food in stores. J Urban Health  2010; 86: 672– 82. Google Scholar CrossRef Search ADS   47 Zick CD, Smith KR, Fan JX et al.  . Running to the store? The relationship between neighborhood environments and the risk of obesity. Soc Sci Med  2009; 69: 1493– 1500. Google Scholar CrossRef Search ADS PubMed  48 Wang MC, Kim S, Gonzalez AA et al.  . Socioeconomic and food-related physical characteristics of the neighbourhood environment are associated with body mass index. J Epid Comm Health  2007; 61: 491– 8. Google Scholar CrossRef Search ADS   © The Author 2017. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)
TI - Utilization and environmental availability of food outlets and overweight/obesity among schoolchildren in a city in the south of Brazil
JO - Journal of Public Health
DO - 10.1093/pubmed/fdx017
DA - 2018-03-01
UR - https://www.deepdyve.com/lp/oxford-university-press/utilization-and-environmental-availability-of-food-outlets-and-tvKBlZOYkU
SP - 106
EP - 113
VL - 40
IS - 1
DP - DeepDyve
ER -