- Email: [email protected]
Temporal changes in the prevalence of disordered eating behaviors among adolescents living in the metropolitan area of Rio de Janeiro, Brazil
Psychiatry Research 253 (2017) 64–70
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Temporal changes in the prevalence of disordered eating behaviors among adolescents living in the metropolitan area of Rio de Janeiro, Brazil
MARK
⁎
Danilo Dias Santana , Erica Guimarães Barros, Rosana Salles da Costa, Gloria Valeria da Veiga Department of Nutrition Josué de Castro, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
A R T I C L E I N F O
A BS T RAC T
Keywords: Binge eating Strict dieting Fasting Compensatory behaviors Low socioeconomic status
To investigate temporal changes in the prevalence of disordered eating behaviors among adolescents, and their association with socio demographic factors and overweight. Using probability sampling, two population-based cross-sectional surveys were conducted: one in 2005 (n=511) and the other in 2010 (n=314). The frequency of disordered eating behaviors (binge eating, strict dieting or fasting and compensatory behaviors) was investigated using a self-administered questionnaire. The presence of binge eating increased by 18.4% in the 5 years between the two surveys. In 2005, girls were 1.95 times more likely to engage in strict dieting or fasting than boys, and this difference increased to 7.02 times in 2010. Overweight adolescents were 2.29 times more likely to undertake strict dieting than non-overweight adolescents in 2005 and 3.65 times more likely to do so in 2010. No significant associations were found for compensatory behaviors. A pronounced increase in the prevalence of binge eating was observed, and girls and overweight adolescents were more likely to engage in strict dieting or fasting.
1. Introduction Eating disorders are psychiatric conditions that may be associated with morbidity and mortality and result in both socioemotional impairment as well as damage to the metabolic and endocrine systems (AAP, 2003). Eating disorders have a multifactorial etiology characterized by disturbed eating behaviors, and excessive preoccupation with weight and body shape (APA, 2006). According to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (APA, 2013), the types of eating disorders are anorexia nervosa, bulimia nervosa, binge eating, and other specified feeding or eating disorders (OSFED). The disordered eating behaviors, namely, strict dieting or fasting, binge eating, and compensatory behaviors (laxative and diuretic misuse and self-induced vomiting), are more frequently observed than full syndrome of eating disorders (Neumark-Sztainer et al., 2011; Leal et al., 2013). Although they can be present at all ages (Hay et al., 2008; Mitchison et al., 2014), they are more frequent during adolescence (Smink et al., 2012; Ferreira et al., 2013). Adolescents are more vulnerable to specific problems such as dissatisfaction with body image, especially body weight, and may develop inadequate eating behaviors with the aim of achieving an idealized body image, which increases the risk of developing eating disorders (Galindo and Carvalho, 2007). This dissatisfaction is more
pronounced in the modern society, where in on one hand, the prevalence of overweight and obesity is increasing and on the other hand, there is social pressure for individuals to remain thin as a form of achieving success and personal and professional accomplishment (Oliveira et al., 2003). The increase in the prevalence of obesity in young people has been observed worldwide (WHO, 2013). In Brazil, the number of overweight adolescents increased by approximately 15% points over the last three decades, with the highest increase occurring between 2003 and 2008 (IBGE, 2010). Therefore, this increase could be associated with an increase in disordered eating behaviors in a short period of time. Other countries have also reported an increase in the prevalence of disordered eating behaviors (Hay et al., 2008; Neumark-Sztainer et al., 2011; Mitchison et al., 2014; Nakai et al., 2014), however in Brazil no information regarding changes in behaviors associated with eating disorders is available. Studies in Brazil have focused exclusively on increasing excess weight, without information about changes in disordered eating behaviors. In 2005, a cross-sectional survey performed in the municipality of Rio de Janeiro, where the population predominantly belongs to the lower social classes, showed that 16.8% of adolescents were overweight and 7.2% were obese. Furthermore, 20% of adolescents engaged in binge eating, 3.3% in compensatory behaviors engaged in to compen-
⁎ Correspondence to: Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro, Avenida Carlos Chagas Filho, n° 373, Centro de Ciências da Saúde, Bloco J, 2° andar, Ilha do Fundão, Rio de Janeiro CEP 21941-590, Brazil. E-mail address: [email protected] (D.D. Santana).
http://dx.doi.org/10.1016/j.psychres.2017.03.042 Received 13 July 2016; Received in revised form 26 January 2017; Accepted 21 March 2017 Available online 21 March 2017 0165-1781/ © 2017 Elsevier B.V. All rights reserved.
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
participated. The study was approved by the Research Ethical Council of the Institute of Social Medicine of the State University of Rio de Janeiro. The questions used to identify disordered eating behaviors were adapted from the interview script by Hay (1998), which was developed for analyzing the prevalence of these behaviors in an Australian community; this interview script had previously shown good reproducibility in a study including students from public schools from another municipality of Rio de Janeiro (Ferreira and Veiga, 2008b). The questions were aimed at identifying the occurrence of the following behaviors over the previous 6 months: binge eating, compensatory behaviors (laxative and diuretic misuse, and self-induced vomiting), and strict dieting or fasting to control weight. Binge eating was assessed based on the definition proposed in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (APA, 1994) using the following: “Have you ever eaten an amount of food greater than that most people would eat in a period of two hours or less? If so, did you feel unable to stop eating or to control how much you were eating?” In order to determine whether respondents engaged in strict dieting or fasting and purging behaviors at least once per week, respondents were asked the following question: “Over the last six months, did you regularly use any methods to control your weight such as laxatives, diuretics, self-induced vomiting, or very strict diets, or fasting?” In the 2005 study, the response options were yes and no; in the 2010 study for the same question, the response options were as follows: never, less than once a week, once a week and two or more times a week. For the purpose of comparison of the two studies, the option of “never” in the 2010 study was considered as the option of “no” in the 2005 study; the other options in the 2010 study were considered as the option of “yes” in the 2005 study. Weight and height were measured in both surveys, and measurement were taken with the participants barefoot and dressed in light clothing. Weight was measured using an electronic scale (Kratos PPS®; São Paulo, Brazil), with a 150 kg capacity and 50 g readability. Height was measured twice to the nearest 0.1 cm using a portable stadiometer (Leicester®; United Kingdom), with a maximum variation of 0.5 cm between twomeasurements, and the average was calculated. Body mass index (BMI) was calculated using the following equation: BMI = weight (Kg)/height (m2). Adequacy of weight was classified according to the sex- and age- specific BMI cut-off points proposed by the World Health Organization (low weight: Z-score < −2; adequate weight: Z-score ≥−2 and ≤1; overweight: Z-score > 1 and ≤2; and obesity: Z-score > 2) (de Onis et al., 2007). The sociodemographic factors evaluated were gender, age, and skin color (black/brown and white). The monthly family income per capita (total household income divided by the number of household members) was used as a socioeconomic indicator and was expressed as multiples of the minimum wage: R$300.00 in 2005 (US $111.4 in January 2005) and R$510.00 in 2010 (US $286.8 in January 2010).
sate for potential weight gain resulting from binge eating (e.g., selfinduced vomiting and use of laxatives or diuretics), and 18.9% in strict dieting. Binge eating and strict dieting were more prevalent in overweight adolescents (Ferreira et al., 2013). In 2010, another study similar to the 2005 study was held in the same city, which evaluated the temporal changes in these behaviors. The results of the study may contribute to future interventions for obesity prevention, as well as prevention of disordered eating behaviors, which can pose a serious health risk to adolescents. The present study aimed at investigating whether temporal changes occurred in the prevalence of disordered eating behaviors in adolescents over a period of 5 years, and determining the association of these behaviors with sociodemographic variables and overweight. 2. Methods 2.1. Population and sampling The present study used data from two population-based crosssectional surveys, one conducted in 2005 and the other, in 2010 in the second district of the municipality of Duque de Caxias, state of Rio de Janeiro. The chosen variables were investigated by household interviews. The surveys were performed using probability samples of 1125 households, that were selected in 3 stages (census sector, household, and individual). In both 2005 and 2010, sample size was determined on the basis of an estimated prevalence of extreme poverty of 14.5%, with a relative maximum error of 5%. More details about the sampling criteria were described in a previous study (Salles-Costa et al., 2008). The second survey was performed in 2010, using inverse cluster sampling (census sector, households, and individuals) (Haldane, 1945). During the first selection stage, the same 75 census sectors were sampled using a new screening to update the age groups. Households with children were then randomly selected (maximum 8 households per sector), followed by random selection of households with adults and/or adolescents, until a total of 15 households per census sector (second selection stage) were selected. During the third selection stage, one individual from each group (child, adolescent, or adult) was selected from each households. For the present study, data from adolescents aged 12–18.9 years were considered. For the two surveys, eligibility criteria included absence of physical deficiencies that would prevent the performance of anthropometric measurements and application of the questionnaires, and absence of pregnancy. In 2005, 573 adolescents residing in households randomly selected were invited to participate in the study, however 561 eligible adolescents were initially interviewed (response rate, 97.9%). Fifty adolescents were excluded due to inconsistencies in answering the questionnaires; the data from 511 adolescents were analyzed (91.0% of questionnaire respondents). In 2010, 347 adolescents were invited to participate in the study, however 314 eligible adolescents were interviewed (response rate, 90.5%); their data were then analyzed. Considering an estimated prevalence of approximately 20% for disordered eating behaviors (Dunker and Philippi, 2003; Sampei et al., 2009; Ferreira et al., 2013), 95% confidence interval, with 5% absolute precision for 2005 and 4% absolute precision for 2010, and the design effect for cluster sampling, the used sample sizes of 511 and 314 allowed estimation of the prevalence of disordered eating behaviors for the adolescent population evaluated.
2.3. Data analysis Data from the two surveys were entered into CSPro 2.5 version (Washington DC, USA) (2005survey) and CSPro 3.3 (Washington DC, USA) version (2010 survey), with restriction mechanism of improbable data entry, and were analyzed using the SPSS 19.0 version software (Chicago, IL, USA). The sampling weight of each individual (calculated by the inverse of the probability of selection, i.e., sample weight =1 probability of selection of each individual in the sample) was used to expand the sample, and the design effect for cluster sampling was taken into account for all analyses (Beckett et al., 1992; Sousa and Silva, 2003), using the Complex Sample procedure from SPSS. A descriptive analysis was performed considering the frequencies and 95% confidence intervals (95% CI) for the variables investigated for the two surveys (2005 and 2010): presence or absence of binge eating, compensatory behaviors (self-induced vomiting, use of laxatives
2.2. Data collection In both the 2005 and 2010 surveys, data collection was performed by a team of trained interviewers. Household interviews and measurements were performed following the signing of a free and informed consent form by the household head, and only willing adolescents 65
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
3.65 times higher in 2010, 95% CI: 1.45–9.22) and for black/brown adolescents than for white adolescents in 2005 (ORadj: 1.87, 95% CI: 1.02–3.43). No significant association between the prevalence of strict dieting or fasting and skin color was observed in 2010 (Table 3). The prevalence of compensatory behaviors was significantly higher for overweight adolescents than for non-overweight ones in 2005 (5.9% vs 2.5%, p=0.029) and was not significantly associated with any of the remaining variables, in both 2005 and 2010 (Table 2).
or diuretics), and strict dieting orfasting over the previous 6 months; gender; age (12–14.9 years and 15–18.9 years); skin color (black/ brown and white). For the per capita income classification, 3 categories were considered for the descriptive analysis (up to half minimum wage; half to one minimum wage; and > 1 minimum wage), and 2 categories were used for the association analysis (up to half minimum wage, which approaches the definition of the poverty line in Brazil (Brasil, 2014), and higher than half minimum wage). For the BMI classification, 4 categories were considered for the descriptive analysis (low weight, adequate weight, overweight, and obesity), and 2 for the association analysis: overweight (overweight and obesity) and nonoverweight (low weight and adequate weight). Frequencies were compared between the two surveys using the chisquare test. For variables with more than 2 categories, a partitioning chi-square test was performed using the WinPepi software (Abramson, 2004). Associations between disordered eating behaviors and the other variables were analyzed for each survey using the chi-square test. The changes in prevalence of disordered eating behaviors in each sub-group of investigated variables were evaluated by logistic regression. To statistically compare the 2005–2010 odds ratios of disordered eating behavior prevalence between sub-groups (e.g. boys vs. girls), the method described by Altman and Bland (2003) to compare parameter estimates from separate analyses was used. The strength of associations were analyzed by estimating the odds ratio (OR) and confidence interval 95% (CI 95%) using logistic regression for the variables found to be significantly associated with the disordered eating behaviors (p < 0.05) in each survey. These variables were considered in a multivariate analysis to investigate their associations with disordered eating behaviors after adjusting for the remaining variables (OR adjusted: ORadj). Variables with p values < 0.05 were retained in the final model.
4. Discussion The most important finding of the present study was an 18.4% increase in the prevalence of binge eating over a period of 5 years; however, the prevalence of compensatory behaviors and strict dieting varied only slightly between the two surveys in this period. For the same period, the proportion of adolescents with low and adequate weight decreased, and the proportion of overweight and obese adolescents increased. These changes are probably not related to changes in the socioeconomic and demographic profile of the studied population, since no significant differences were observed for these variables between the years of research. This finding could indicate that the increase in inadequate eating habits, especially in binge eating, may be related to the increase in the prevalence of overweight subjects in this population. The present study is the first in Brazil to investigate temporal changes in the prevalence of disordered eating behaviors in adolescents, which makes its comparison with other Brazilian studies difficult. However, international studies have investigated these questions. Hay et al. (2008) conducted a population-based study in Australia, with individuals older than 15 years of age and observed a 7.4% increase in the prevalence of binge eating for those aged 15–24 years over a period of 10 years (1995–2005), while the prevalence of strict dieting and compensatory behaviors decreased. Also in Australia, Mitchison et al. (2014) found that between 1998 and 2008, binge eating increased significantly (by 4.9% points) in adolescents and young adults aged 15–24 years. Restrictive diet and compensatory behaviors did not show significant changes between the years of research, as in the present study. Neumark-Sztainer et al. (2011), in a longitudinal study with American individuals with a mean age of 12.8 ± 0.7 (early adolescence) and 15.9 ± 0.8 years (middle adolescence), and after a 10-year followup with a mean age of 23.2 ± 1.0 (early young adulthood) and 26.2 ± 0.9 years (middle young adulthood), found that binge eating significantly increased from adolescence to adulthood in the older follow-up cohort. However, Nakai et al. (2014), investigating temporal changes in the disordered eating behaviors of female Japanese students between the ages of 16 and 23 years, found a significant increase in binge eating, as in the others disordered behaviors, in 20 years 1982–2002). In the above-mentioned studies, binge eating was the disordered behavior that increased more remarkably, as observed in the present study. The increase in the prevalence of overweight observed in the present study (7.5%) is in accordance with nationwide findings (approximately 5%) (IBGE, 2010), which revealed that the prevalence of overweight has been continuously increasing among Brazilian adolescents over the last three decades. The increase in the prevalence of disordered eating behaviors and overweight observed for the population of adolescents living in Duque de Caxias therefore reinforces our hypothesis that the two factors - eating disorders and overweight - may be interrelated. In this study, the prevalence of binge eating was not significantly associated with gender in either survey, but increased significantly in both boys and girls. In contrast, Ferreira and Veiga (2008a) observed that binge eating was more frequent in girls, although its prevalence was not low in boys. According to Mitchison and Mond (2015), in a narrative review, binge eating appears to be the most common eating disorder behavior in men, with a prevalence almost equal to that of women.
3. Results No significant differences in the sociodemographic factors were observed between the participants of the two surveys. Fifty percent of the evaluated adolescents were male, approximately 80% were of black/brown color, and approximately 50% had a per capita income of up to 1 minimum wage. A significant difference in BMI classification was observed (p=0.033), with a decreased prevalence of low weight (by 3.6%) and an increased prevalence of overweight (by 6.5%). Prevalence of binge eating increased by 18.4% in 2010 compared to the prevalence observed in 2005 (p=0.001). Significant differences were neither observed for the remaining disordered eating behaviors evaluated nor for the presence of 2 or more behaviors (Table 1). The prevalence of binge eating increased significantly for both genders and age groups, for both skin colors (black/brown and white), and all levels of income per capita. However, a significant change was observed between 2005 and 2010, for non-overweight adolescents, but not for overweight ones. Using the Altman and Bland method, the odds ratios calculated across the two surveys for binge eating did not differ significantly between all variables. No significant changes were found for the other behaviors (Table 2). In 2005, the prevalence of binge eating was associated with skin color and BMI classification (Table 2), so that, the above mentioned prevalence was higher for black/brown participants than for white participants (ORadj: 1.91; 95% CI: 1.04–3.52) and for overweight adolescents than for non-overweight adolescents (ORadj: 2.02; 95% CI: 1.13–3.61). In 2010, none of the variables were significantly associated with binge eating (Table 3). In both 2005 and 2010, prevalence of strict dieting or fasting was associated with gender and BMI classification (Table 2). This behavior was more prevalent in girls than in boys in 2005 (ORadj: 1.95; 95%: 1.07–3.53), and this difference increased to 7.02 times in 2010 (95% CI: 2.81–17.54). Moreover, in both surveys, the prevalence of strict dieting or fasting was higher for overweight adolescents than for nonoverweight adolescents (2.29 times higher in 2005, 95% CI: 1.14–4.64; 66
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 1 Sociodemographic factors, body mass index classification (BMI), and disordered eating behaviors among adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. 2005 na Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb Up to half minimum wage Half to one minimum wage > 1 minimum wage BMI classification Low weight Adequate weight Overweight Obesity DEBc Binge eating Strict dieting or fasting Compensatory behaviors 2 or more DEB
2010 % (95% CI)
511
na
511
0.668 51.0 (44.3–57.7) 49.0 (42.3–55.7)
+ 1.7 − 1.7
50.3 (42.7–57.9) 49.7 (42.1–57.3)
+ 5.4 − 5.4
78.9 (70.9–85.1) 21.1 (14.9–29.1)
− 2.3 + 2.3
50.5 (40.7–60.2) 36.5 (28.5–45.3) 13.1 (7.4–22.1)
+ 2.8 − 5.8 + 3.0
0.9 (0.3–2.8) 68.2 (60.5–74.9) 22.8# (17.3–29.5) 8.1 (5.1–12.8)
− − + +
3.6 4.0 6.5 1.0
38.4 (30.3–47.1) 17.4 (12.3–23.9) 4.9 (2.6–9.1) 13.2 (8.8–19.5)
+ − + +
18.4 1.5 1.6 4.0
314 44.9 (39.3–50.6) 55.1 (49.4–60.7)
510
0.178
308 81.2 (76.7–85.0) 18.8 (15.0–23.3)
503
0.572
304 47.7 (41.2–54.3) 42.3 (36.8–47.9) 10.1 (7.1–14.1)
511
0.470
301 4.5# (2.7–7.2) 72.2 (66.9–76.9) 16.3 (12.7–20.6) 7.1 (4.8–10.5) 20.0 (15.0–26.1) 18.9 (14.2–24.6) 3.3 (1.8–6.0) 9.2 (6.1–13.5)
312 312 311 312
p valued
% (95% CI)
314 49.3 (44.5–54.1) 50.7 (45.9–55.5)
506 504 506 511
Variation
0.033
< 0.001 0.701 0.380 0.181
a
Different totals due to missing values in each variable. Minimum wage: 2005= R$ 300.00; 2010= R$ 510.00. c DEB: disordered eating behaviors. d Chi-square test. # Partitioning chi-square test p < 0.05. b
2010 may be explained by the significant and more pronounced increase in the prevalence of binge eating among non-overweight adolescents (2-fold) than among overweight adolescents (1.5-fold), which reduced the difference between the two groups in 2010. The prevalence of strict dieting was relatively constant, with a tendency to decrease over the period between the two surveys, similar to that observed by Hay et al. (2008) in Australia. An exception was the girls, for whom an increase of 4.9% in the prevalence of strict dieting was observed, which was higher than the value noted for boys. Other studies performed in Brazil (Vilela et al., 2004; Ferreira and Veiga, 2008a) and other countries (Ata et al., 2007; Keel et al., 2007; NuñoGutiérrez et al., 2009; Bilali et al., 2010) also reported a higher prevalence of strict dieting in girls than in boys. This may be explained by the fact that girls are more concerned with their body image (Vilela et al., 2004; Goldenberg, 2011) and adopt inadequate eating habits to attain the desired bodyshape. In addition, there was also a significant association (p=0.039) for have 2 or more disordered eating behaviors among girls in 2010, but not in 2005. This suggests a worsening of this problem in females, since the prevalence increased from 11.1% to 19.2%, whereas that in boys remained constant (7.1% to 7.4%) (data not shown). The prevalence of strict dieting was also associated with overweight, similar to the findings of a previous study including adolescents who attended public schools in Niterói, RJ (Ferreira and Veiga, 2008a). In addition, compensatory behaviors were associated with overweight in 2005, when the prevalence of these behaviors were two times higher among overweight adolescents than among non-overweight adolescents. Overweight individuals are undervalued, undergo psychological suffering, and may therefore opt for extreme alternatives for weight control, such as strict dieting, fasting, or use of compensatory behaviors (Bernardi et al., 2005). This may explain the association between overweight and these behaviors. The hypothesis of the present study was that increased prevalence of disordered eating behaviors was concomitant with an increase in the prevalence of overweight in adolescents. The prevalence of overweight
None of the analyzed disordered eating behaviors was significantly associated with the per capita income, indicating that they are homogenously distributed throughout economic levels. Similarly, another study performed in Brazil did not find any association between disordered eating behaviors and the socioeconomic status of adolescents (Dunker et al., 2009), similar to the findings in other countries (Bauer and Kirchengast, 2006; Bergström and Elfhag, 2007; Preti et al., 2007; Soh et al., 2007). The fact that binge eating was significantly associated with the black/brown skin color in 2005 but not in 2010 can be explained by the more pronounced increase in its prevalence amongst white adolescents (approximately 3 times, from 11.6% in 2005 to 34.7% in 2010) as compared to that amongst black/brown adolescents (approximately 2 times, from 21.8% in 2005 to 40.2% in 2010), thereby reducing the difference between the two groups in 2010. The divergent results found regarding the association of binge eating with income and skin color may be intriguing considering that in Brazil, skin color is used as a socioeconomic level marker (Ferreira et al., 2015; Moreira et al., 2015). The comparison of this result with other studies is difficult, considering that previous investigations regarding associations of disordered eating behaviors with skin color have focused on the issue of ethnicity (Javier et al., 2016), which differs from the context addressed in this study and that of other studies conducted in Brazil. Therefore, this issue needs to be better investigated in other studies. The prevalence of binge eating was associated with overweight in 2005, confirming the results of previous studies performed in Brazil and other countries (Stice et al., 2002; Pivetta and Silva, 2010; Prisco et al., 2013, Sonneville et al., 2013). This association may be explained by the fact that although binge eating and obesity present different symptoms, they have a few similarities; both are usually accompanied by compulsive eating, i.e., consumption of large quantities of food in short periods of time, leading to feelings of guilt and lack of control (Pinheiro et al., 2006). The fact that a significant association between the prevalence of binge eating and overweight was not observed in 67
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 2 Prevalence of binge eating, strict dieting or fasting and compensatory behaviors with odds ratio and 95% confidence intervals (95%CI), and its association with sociodemographic factors and body mass index (BMI) classification for adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. Binge eating
Var. (%)
2005 a
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
2010
n
%
249 258
21.8 18.1
228 279
21.0 19.1
414 91
21.8 11.6
240 258
19.6 20.7
388 119
17.2 28.8
p
Odds ratio (95%CI)
a
n
p
0.432
0.757 158 153
39.7 37.0
158 153
34.3 42.5
241 65
40.2 34.7
151 151
36.5 38.1
206 93
36.4 45.7
0.653
0.026
246 258
13.7 23.7
228 276
17.5 20.0
408 95
20.6 11.6
238 258
21.1 16.8
385 119
15.4 30.1
p 0.025
2010 na
%
+ 16.9# + 17.4#
2.35 (1.20 – 4.62) 2.36 (1.27 – 4.37)
+ 19.8# + 16.5
2.74 (1.57 – 4.80) 2.08 (0.89 – 4.83)
Odds ratio (CI95%)
158 153
6.5 28.6
158 153
16.7 18.1
241 65
19.7 9.2
151 151
18.2 15.6
206 93
13.4 28.6
p < 0.001 − 7.2 + 4.9
0.44 (0.16 – 1.16) 1.28 (0.67 – 2.45)
− 0.8 − 1.9
0.94 (0.41 – 2.17) 0.88 (0.47 – 1.66)
− 0.9 − 2.4
0.94 (0.52 – 1.71) 0.77 (0.26 – 2.22)
− 2.9 − 1.2
0.83 (0.40 – 1.70) 0.91 (0.47 – 1.77)
− 2.4 − 1.9
0.85 (0.44 – 1.63) 0.93 (0.38 – 2.25)
0.825
0.094
0.340
0.601
0.014
0.027
Compensatory behaviors
Var. (%)
2005
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
2.41 (1.56 – 3.71) 4.07 (1.53 – 10.82)
Var. (%)
0.031
n
+ 18.4# + 23.1#
0.298
0.522
a
1.96 (1.07 – 3.59) 3.14 (1.62 – 6.07)
0.842
Strict dieting or fasting
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
+ 13.3# + 23.4# 0.564
0.805
%
2.35 (1.22 – 4.56) 2.65 (1.39 – 5.05)
0.280
0.020
2005 na
+ 17.9# + 18.9#
2010 %
p
a
Odds ratio (95%CI)
n
%
158 153
3.8 6.0
157 154
3.2 6.6
240 65
3.4 10.8
150 151
5.2 4.9
206 92
5.8 3.5
0.101 246 259
2.0 4.6
228 277
2.9 3.6
410 95
3.6 2.1
239 259
3.9 2.9
386 120
2.5 5.9
p 0.484
0.679
+ 1.8 + 1.4
1.99 (0.44 – 8.90) 1.33 (0.46 – 3.85)
+ 0.3 + 3.0
1.09 (0.29 – 4.19) 1.89 (0.59 – 6.01)
− 0.2 + 8.7
0.95 (0.31 – 2.86) 5.63 (0.84 – 37.52)
+ 1.3 + 2.0
1.36 (0.41 – 4.49) 1.75 (0.46 – 6.63)
+ 4.0 − 2.5
2.42 (0.77 – 7.62) 0.58 (0.14 – 4.48)
0.222
0.433
0.080
0.569
0.941
0.029
0.445
a
Different totals due to missing values in each variable. Minimum wage: 2005= R$ 300.00; 2010= R$ 510.00. # Logistic regression p < 0.05. b
2010 most variables. These results indicate that these extreme behaviors, which could lead to in important nutritional deficiencies, may not be as widespread amongst adolescents of lower social classes such as those evaluated in the present study. This study did not aim at diagnosing eating disorders, but at screening changes in disordered eating behaviors over a period of 5
increased by 7.5% and that of binge eating increased by 18.4% between the two surveys. Binge eating increased remarkably, even in groups that did not present significant increase in prevalence of overweight (data not shown), which may indicate an even higher increase in the risk for overweight. However, the prevalence of strict dieting decreased, and the use of compensatory behaviors varied little between 2005 and 68
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 3. Gross and adjusted Odds Ratio and 95% confidence intervals (95%CI) for association between disordered eating behaviors with sociodemographic factors and body mass index (BMI) classification for adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. Variables
2005
Variables
OR (95%CI) Gross Strict dieting or fasting Gender Male Female Skin color Black/Brown White
OR (95%CI) Adjusted
1.00 1.95 (1.08 – 3.53)
1.00 1.95 (1.07 – 3.53)
1.00 1.97 (1.06 – 3.69)
1.00 1.87 (1.02 – 3.43)
1.00 2.37 (1.19 – 4.74)
1.00 2.29 (1.14 – 4.64)
Skin color Black/Brown White
1.00 2.14 (1.12 – 4.08)
1.00 1.91 (1.04 – 3.52)
BMI classification Non-overweight Overweight
1.00 1.94 (1.08 – 3.49)
1.00 2.02 (1.13 – 3.61)
BMI classification Non-overweight Overweight
2010
Strict dieting or fasting Gender Male Female BMI classification Non-overweight Overweight
Gross
Adjusted
1.00 5.75 (2.17 – 15.20)
1.00 7.02 (2.81 – 17.54)
1.00 2.60 (1.09 – 6.17)
1.00 3.65 (1.45 – 9.22)
Binge eating
Acknowledgments
years in adolescents predominantly of lower social classes. Because it used two cross-sectional surveys, we could not determine the causality of the associations between these behaviors and the studied variables, which is a limitation of the study. Future longitudinal studies may help clarify these associations. Another limitation of the present study was related to the questions used to investigate the disordered eating behaviors. Since the questions were simplified and did not include all eating disorder symptoms, they may have failed to reliably detect these behaviors. However, when compared to the wider and more frequently used questionnaires, the questions used addressed important issues about the main behaviors related to eating disorders. In addition, the reproducibility of these questions was considered moderate to excellent in a previous study with adolescents (Ferreira and Veiga, 2008b). This simplified questionnaire can therefore be considered useful for screening such behaviors in population studies, especially in populations with low socioeconomic status, which presents greater difficulties in the interpretation of more complex questions such as the one investigated in the present study. Furthermore, the difference in response options to questions that investigated disordered eating behaviors in the 2005 and 2010 studies is other limitation of the study as it is conceivable that some participants who used eating behaviors less than once a week, but more than never (as answered in 2010), could have answered "no" to the question in 2005. In conclusion, there was a pronounced increase in the prevalence of binge eating over the 5-year period between the two surveys in the studied population, and girls and overweight adolescents were more vulnerable to strict dieting. The present study shows that health professionals should pay special attention to adolescents with these behaviors, which may persist into adulthood or eventually evolve to a subthreshold, full eating disorder, or a weight disorder as overweight and obesity, which could result in serious health risks.
Grant support: The authors thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq (National Council for Scientific and Technological Development) grant numbers 503139/ 2003-3 and 476344/2008-5 and Instituto Nacional de Câncer (National Institute of Cancer) which funded the project. References Abramson, J.H., 2004. WINPEPI (PEPI-for-Windows): computer programs for epidemiologists. Epidemiol. Perspect. Innov. 1, 6. Altman, D.G., Bland, J.M., 2003. Interaction revisited: the difference between two estimates. BMJ (Clinical research ed.) 326, 219. American Academy of Pediatrics (AAP) - (US), 2003. Children and adolescents with eating disorders: the state of the art. Pediatrics 111, 1–13. American Psychiatric Association (APA), 2006. Treatment of patients with eating disorders. Am. J. Psychiatry 163, 4–54, (Third edition). American Psychiatric Association (APA), 2013. Diagnostic and Statistical Manual of Mental Disorders: DSM-5 5th edition. American Psychiatric Publishing Inc, Arlington. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders DSM-IV. Washington, DC. Ata, R.N., Ludden, A.B., Lally, M.M., 2007. The effects of gender and family, friend, and media influences on eating behaviors and body image during adolescence. J. Youth Adolesc. 36, 1024–1037. Bauer, M., Kirchengast, S., 2006. Body composition, weight status, body image and weight control practices among female adolescents from eastern Austria. Anthropol. Anz. 64, 321–331. Beckett, L.A., Scherr, P.A., Evans, D.A., 1992. Population prevalence estimates from complex samples. J. Clin. Epidemiol. 45, 393–402. Bergström, H., Elfhag, K., 2007. Socioeconomic differences in obese patients with and without eating disorders. Eat. Weight Disord. 12, 98–103. Bernardi, F., Cichelero, C., Vitolo, M.R., 2005. Comportamento de restrição alimentar e obesidade. (restrained eating behavior and obesity). Rev. De. Nutr. 18, 85–93. Bilali, A., Galanis, P., Velonakis, E., Katostaras, T., 2010. Factors associated with abnormal eating attitudes among greek adolescents. J. Nutr. Educ. Behav. 42, 292–298, (Journal of Racial Ethnic Health Disparities). Brasil, 2014. Ministério do Desenvolvimento Social. Decreto n° 8.232, de 30 de abril de 2014. Altera o Decreto n° 5.209, de 17 de setembro de 2004, que regulamenta o Programa Bolsa Família, e o Decreto n° 7.492, de 2 de junho de 2011, que institui o Plano Brasil Sem Miséria. (It amends Decree No. 5,209, dated September 17, 2004, which regulates the Bolsa Família Program, and Decree No. 7,492, of June 2, 2011, which establishes the Brazil Without Misery Plan). Brasília; Presidência da República. Dunker, K.L., Philippi, S.T., 2003. Hábitos e comportamentos alimentares de adolescentes com sintomas de anorexia nervosa. (food habits and feeding behavior in
Conflicts of interest The authors declare no conflict of interest. 69
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
students in 1982, 1992 and 2002. Psychiatry Res. 219, 151–156. Neumark-Sztainer, D., Wall, M., Larson, N.I., Eisenberg, M.E., Loth, K., 2011. Dieting and disordered eating behaviors from adolescence to young adulthood: findings from a 10-year longitudinal study. J. Am. Diet. Assoc. 111, 1004–1011. Nuño-Gutiérrez, B.L., Rosa, Celis-De. La, Unikel-Santoncini, C, A., 2009. Prevalence and associated factors related to disordered eating in student adolescents of Guadalajara across sex. Rev. Investig. Clínica 61, 286–293. Oliveira, F.P., Bosi, M.L.M., Vigario, O.S., Vieira, R.S., 2003. Eating behavior and body image in athletes. Revista Brasileira de Medicina do Esporte 9, 348–356. de Onis, M., Onyango, A.W., Borghi, E., Siyam, A., Nishida, C., Siekmann, J., 2007. Development of a WHO growth reference for school-aged children and adolescents. Bull. World Health Organ. 85, 660–667. Pinheiro, A.P., Sullivan, P.F., Bacaltchuck, J., Prado-Lima, P.A.S., Bulik, C.M., 2006. Genética em transtornos alimentares: Ampliando os horizontes de pesquisa. (Genetics in eating disorders: extending the boundaries of research). Revista Brasileira de Psiquiatria 28, 218–225. Pivetta, L.A., Gonçalves-Silva, R.M.V., 2010. Compulsão alimentar e fatores associados em adolescentes de Cuiabá, Mato Grosso, Brasil. (binge eating and associated factors among teenagers in Cuiabá, Mato Grosso state, Brazil). Cad. De. Saúde Pública 26, 337–346. Preti, A., Pinna, C., Nocco, S., Pilia, S., Mulliri, E., Micheli, V., Casta, M.C., Petretto, D.R., Masala, C., 2007. Rural/urban differences in the distribution of eating disorder symptoms among adolescents from community samples. Aust. N.Z. J. Psychiatry 41, 525–535. Prisco, A.P.K., Araujo, T.M., Almeida, M.M.G., Santos, K.O.B., 2013. Prevalência de transtornos alimentares em trabalhadores urbanos de município do Nordeste do Brasil. (Prevalence of eating disorders in urban workers in a city of the northeast of Brazil). Ciência e Saúde Coletiva 18, 1109–1118. Salles-Costa, R., Pereira, R.A., Vasconcellos, M.T.L., Veiga, G.V., Marins, V.M.R., Jardim, B.C., Gomes, F.S., Sichieri, R., 2008. Associação entre fatores socioeconômicos e insegurança alimentar: estudo de base populacional na Região Metropolitana do Rio de Janeiro, Brasil. (Association between socioeconomic factors and food insecurity: a population-based study in the Rio de Janeiro metropolitan area, Brazil). Rev. Nutr. 21 (Supl), 99s–109s. Sampei, M.A., Sigulem, D.M., Novo, N.F., Juliano, Y., Colugnati, F.A.B., 2009. Eating attitudes and body image in ethnic Japanese and Caucasian adolescent girls in the city of São Paulo, Brazil. J. Pediatr. 85, 122–128. Smink, F.R., van Hoeken, D., Hoek, H.W., 2012. Epidemiology of eating disorders: incidence, prevalence and mortality rates. Curr. Psychiatry Rep. 14, 406–414. Soh, N.L., Touyz, S., Dobbins, T.A., Surgenor, L.J., Clarke, S., Kohn, M.R., Lee, E.L., Leow, V., Rieger, E., Ung, K.E., Walter, G., 2007. Restraint and eating concern in North European and East Asian women with and without eating disorders in Australia and Singapore. Aust. N.Z. J. Psychiatry 41, 536–545. Sonneville, K.R., Horton, N.J., Micali, N., Crosby, R.D., Swanson, S.A., Solmi, F., Field, A.E., 2013. Longitudinal associations between binge eating and overeating and adverse outcomes among adolescents and young adults: Does loss of control matter? JAMA Pediatr. 167, 149–155. Sousa, M.H., Silva, N.N., 2003. Estimativas obtidas de um levantamento complexo (estimates from a complex survey). Rev. De. Saúde Pública 37, 662–670. Stice, E., Presnell, K., Spangler, D., 2002. Risk factors for binge eating onset in adolescent girls: a 2-year prospective investigation. Health Psychol. 21, 131–138. Vilela, J.E.M., Lamounier, J.A., Dellaretti, M.A., Neto, B.J.R., Horta, G.M., 2004. Eating disorders in school children. J. De. Pediatr. 80, 49–54. World Health Organization, 2013. Global Database on Body Mass Index. 〈http://apps. who.int/bmi/index.jsp/〉 (Accessed 18 February 2016).
adolescents with symptoms of anorexia nervosa). Rev. Nutr. 16, 51–60. Dunker, K.L.L., Fernandes, C.P.B., Carreira Filho, D., 2009. Influência do nível socioeconômico sobre comportamentos de risco para transtornos alimentares em adolescentes. (Socioeconomic influence on eating disorders risk behaviors in adolescentes). J. Bras. Psiquiatr. 58, 156–161. Ferreira, J.E.S., Veiga, G.V., 2008a. Eating disorder risk behavior in Brazilian adolescents from low socioeconomic level. Appetite 51, 249–255. Ferreira, J.E.S., Veiga, G.V., 2008b. Confiabilidade (teste-reteste) de um questionário simplificado para avaliar comportamentos de risco para transtornos alimentares em adolescentes. (test-retest reliability of a simplified questionnaire for screening adolescents with risk behaviors for eating disorders in epidemiologic studies). Rev. Bras. Epidemiol. 11, 393–401. Ferreira, J.E.S., Souza, P.R.B., Costa, R.S., Sichieri, R., Veiga, G.V., 2013. Disordered eating behaviors in adolescents and adults living in the same household in metropolitan area of Rio de Janeiro, Brazil. Psychiatry Res. 210, 612–617. Ferreira, M.F., Moraes, C.L., Reichenheim, M.E., Verly, E., Marques, E.S., Salles-Costa, R., 2015. Effect of physical intimate partner violence on body mass index in lowincome adult women. Cadernos de Saúde Pública 31, 161–172. Galindo, E.M.C., Carvalho, A.M.P., 2007. Tradução, adaptação e avaliação da consistência interna do eating behaviours and body image test para uso com crianças do sexo feminino. (translation, adaptation and internal consistency evaluation of the eating behaviors and body image test for female children). Rev. Nutr. 20, 20–21. Goldenberg, M., 2011. Gênero, “o Corpo” e “Imitação Prestigiosa” na Cultura brasileira. (Gender, "the body" and "Prestigious imitation" in the Brazilian culture). Saúde e Sociedade 20, 543–553. Haldane, J.B.S., 1945. On a method of estimating frequencies. Biometrika 33, 222–225. Hay, P., 1998. The Epidemiology of eating disorder behaviors: an Australian community – based survey. J. Eat. Disord. 23, 371–382. Hay, P.J., Mond, J.M., Buttner, P., Darby, A., 2008. Eating disorder behaviors are increasing: findings from two sequential community surveys in South Australia. PLoS ONE 3, 1–5. Instituto Brasileiro de Geografia e Estatística (IBGE), 2010. . Pesquisa de Orçamentos Familiares 2008–2009. Antropometria e estado nutricional de crianças, adolescentes e adultos no Brasil (Household Budget Survey 2008–2009). (Anthropometry and nutritional status of children, adolescents and adults in Brazil). Instituto Brasileiro de Geografia e Estatística, Rio de Janeiro〈http://www.ibge.gov.br/〉, (Accessed 13 February 2016). Javier, S.J., Moore, M.P., Belgrave, F.Z., 2016. Racial comparisons in perceptions of maternal and peer attitudes, body dissatisfaction, and eating disorders among African American and White women. Women Health 56, 615–633. Keel, P.K., Baxter, M.G., Heatherton, T.F., Joiner, T.E.J.R., 2007. A 20-year longitudinal study of body weight, dieting, and eating disorder symptoms. J. Abnorm. Psychol. 116, 422–432. Leal, G.V.S., Philippi, S.T., Polacow, V.O., Cordás, T.A., Alvarenga, M.S., 2013. O que é comportamento de risco para transtornos alimentares em adolescentes? (What is adolescents' eating disorder risk behavior?). J. Bras. Psiquiatr. 62, 62–75. Mitchison, D., Mond, J., 2015. Epidemiology of eating disorders, eating disordered behaviour, and body image disturbance in males: a narrative review. J. Eat. Disord. 23, 20. Mitchison, D., Hay, P., Slewa-Younan, S., Mond, J., 2014. The changing demographic profile of eating disorder behaviors in the community. BMC Public Health 14, 943. Moreira, N.F., Sichieri, R., Reichenheim, M.E., Oliveira, A.S., Veiga, G.V., 2015. The associations of BMI trajectory and excessive weight gain with demographic and socio-economic factors: the Adolescent Nutritional assessment longitudinal study cohort. Br. J. Nutr. 114, 2032–2038. Nakai, Y., Nin, K., Noma, S., 2014. Eating disorder symptoms among Japanese female
70
Contents lists available at ScienceDirect
Psychiatry Research journal homepage: www.elsevier.com/locate/psychres
Temporal changes in the prevalence of disordered eating behaviors among adolescents living in the metropolitan area of Rio de Janeiro, Brazil
MARK
⁎
Danilo Dias Santana , Erica Guimarães Barros, Rosana Salles da Costa, Gloria Valeria da Veiga Department of Nutrition Josué de Castro, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil
A R T I C L E I N F O
A BS T RAC T
Keywords: Binge eating Strict dieting Fasting Compensatory behaviors Low socioeconomic status
To investigate temporal changes in the prevalence of disordered eating behaviors among adolescents, and their association with socio demographic factors and overweight. Using probability sampling, two population-based cross-sectional surveys were conducted: one in 2005 (n=511) and the other in 2010 (n=314). The frequency of disordered eating behaviors (binge eating, strict dieting or fasting and compensatory behaviors) was investigated using a self-administered questionnaire. The presence of binge eating increased by 18.4% in the 5 years between the two surveys. In 2005, girls were 1.95 times more likely to engage in strict dieting or fasting than boys, and this difference increased to 7.02 times in 2010. Overweight adolescents were 2.29 times more likely to undertake strict dieting than non-overweight adolescents in 2005 and 3.65 times more likely to do so in 2010. No significant associations were found for compensatory behaviors. A pronounced increase in the prevalence of binge eating was observed, and girls and overweight adolescents were more likely to engage in strict dieting or fasting.
1. Introduction Eating disorders are psychiatric conditions that may be associated with morbidity and mortality and result in both socioemotional impairment as well as damage to the metabolic and endocrine systems (AAP, 2003). Eating disorders have a multifactorial etiology characterized by disturbed eating behaviors, and excessive preoccupation with weight and body shape (APA, 2006). According to the fifth edition of the Diagnostic and Statistical Manual of Mental Disorders (APA, 2013), the types of eating disorders are anorexia nervosa, bulimia nervosa, binge eating, and other specified feeding or eating disorders (OSFED). The disordered eating behaviors, namely, strict dieting or fasting, binge eating, and compensatory behaviors (laxative and diuretic misuse and self-induced vomiting), are more frequently observed than full syndrome of eating disorders (Neumark-Sztainer et al., 2011; Leal et al., 2013). Although they can be present at all ages (Hay et al., 2008; Mitchison et al., 2014), they are more frequent during adolescence (Smink et al., 2012; Ferreira et al., 2013). Adolescents are more vulnerable to specific problems such as dissatisfaction with body image, especially body weight, and may develop inadequate eating behaviors with the aim of achieving an idealized body image, which increases the risk of developing eating disorders (Galindo and Carvalho, 2007). This dissatisfaction is more
pronounced in the modern society, where in on one hand, the prevalence of overweight and obesity is increasing and on the other hand, there is social pressure for individuals to remain thin as a form of achieving success and personal and professional accomplishment (Oliveira et al., 2003). The increase in the prevalence of obesity in young people has been observed worldwide (WHO, 2013). In Brazil, the number of overweight adolescents increased by approximately 15% points over the last three decades, with the highest increase occurring between 2003 and 2008 (IBGE, 2010). Therefore, this increase could be associated with an increase in disordered eating behaviors in a short period of time. Other countries have also reported an increase in the prevalence of disordered eating behaviors (Hay et al., 2008; Neumark-Sztainer et al., 2011; Mitchison et al., 2014; Nakai et al., 2014), however in Brazil no information regarding changes in behaviors associated with eating disorders is available. Studies in Brazil have focused exclusively on increasing excess weight, without information about changes in disordered eating behaviors. In 2005, a cross-sectional survey performed in the municipality of Rio de Janeiro, where the population predominantly belongs to the lower social classes, showed that 16.8% of adolescents were overweight and 7.2% were obese. Furthermore, 20% of adolescents engaged in binge eating, 3.3% in compensatory behaviors engaged in to compen-
⁎ Correspondence to: Instituto de Nutrição Josué de Castro, Universidade Federal do Rio de Janeiro, Avenida Carlos Chagas Filho, n° 373, Centro de Ciências da Saúde, Bloco J, 2° andar, Ilha do Fundão, Rio de Janeiro CEP 21941-590, Brazil. E-mail address: [email protected] (D.D. Santana).
http://dx.doi.org/10.1016/j.psychres.2017.03.042 Received 13 July 2016; Received in revised form 26 January 2017; Accepted 21 March 2017 Available online 21 March 2017 0165-1781/ © 2017 Elsevier B.V. All rights reserved.
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
participated. The study was approved by the Research Ethical Council of the Institute of Social Medicine of the State University of Rio de Janeiro. The questions used to identify disordered eating behaviors were adapted from the interview script by Hay (1998), which was developed for analyzing the prevalence of these behaviors in an Australian community; this interview script had previously shown good reproducibility in a study including students from public schools from another municipality of Rio de Janeiro (Ferreira and Veiga, 2008b). The questions were aimed at identifying the occurrence of the following behaviors over the previous 6 months: binge eating, compensatory behaviors (laxative and diuretic misuse, and self-induced vomiting), and strict dieting or fasting to control weight. Binge eating was assessed based on the definition proposed in the Diagnostic and Statistical Manual of Mental Disorders, 4th edition (APA, 1994) using the following: “Have you ever eaten an amount of food greater than that most people would eat in a period of two hours or less? If so, did you feel unable to stop eating or to control how much you were eating?” In order to determine whether respondents engaged in strict dieting or fasting and purging behaviors at least once per week, respondents were asked the following question: “Over the last six months, did you regularly use any methods to control your weight such as laxatives, diuretics, self-induced vomiting, or very strict diets, or fasting?” In the 2005 study, the response options were yes and no; in the 2010 study for the same question, the response options were as follows: never, less than once a week, once a week and two or more times a week. For the purpose of comparison of the two studies, the option of “never” in the 2010 study was considered as the option of “no” in the 2005 study; the other options in the 2010 study were considered as the option of “yes” in the 2005 study. Weight and height were measured in both surveys, and measurement were taken with the participants barefoot and dressed in light clothing. Weight was measured using an electronic scale (Kratos PPS®; São Paulo, Brazil), with a 150 kg capacity and 50 g readability. Height was measured twice to the nearest 0.1 cm using a portable stadiometer (Leicester®; United Kingdom), with a maximum variation of 0.5 cm between twomeasurements, and the average was calculated. Body mass index (BMI) was calculated using the following equation: BMI = weight (Kg)/height (m2). Adequacy of weight was classified according to the sex- and age- specific BMI cut-off points proposed by the World Health Organization (low weight: Z-score < −2; adequate weight: Z-score ≥−2 and ≤1; overweight: Z-score > 1 and ≤2; and obesity: Z-score > 2) (de Onis et al., 2007). The sociodemographic factors evaluated were gender, age, and skin color (black/brown and white). The monthly family income per capita (total household income divided by the number of household members) was used as a socioeconomic indicator and was expressed as multiples of the minimum wage: R$300.00 in 2005 (US $111.4 in January 2005) and R$510.00 in 2010 (US $286.8 in January 2010).
sate for potential weight gain resulting from binge eating (e.g., selfinduced vomiting and use of laxatives or diuretics), and 18.9% in strict dieting. Binge eating and strict dieting were more prevalent in overweight adolescents (Ferreira et al., 2013). In 2010, another study similar to the 2005 study was held in the same city, which evaluated the temporal changes in these behaviors. The results of the study may contribute to future interventions for obesity prevention, as well as prevention of disordered eating behaviors, which can pose a serious health risk to adolescents. The present study aimed at investigating whether temporal changes occurred in the prevalence of disordered eating behaviors in adolescents over a period of 5 years, and determining the association of these behaviors with sociodemographic variables and overweight. 2. Methods 2.1. Population and sampling The present study used data from two population-based crosssectional surveys, one conducted in 2005 and the other, in 2010 in the second district of the municipality of Duque de Caxias, state of Rio de Janeiro. The chosen variables were investigated by household interviews. The surveys were performed using probability samples of 1125 households, that were selected in 3 stages (census sector, household, and individual). In both 2005 and 2010, sample size was determined on the basis of an estimated prevalence of extreme poverty of 14.5%, with a relative maximum error of 5%. More details about the sampling criteria were described in a previous study (Salles-Costa et al., 2008). The second survey was performed in 2010, using inverse cluster sampling (census sector, households, and individuals) (Haldane, 1945). During the first selection stage, the same 75 census sectors were sampled using a new screening to update the age groups. Households with children were then randomly selected (maximum 8 households per sector), followed by random selection of households with adults and/or adolescents, until a total of 15 households per census sector (second selection stage) were selected. During the third selection stage, one individual from each group (child, adolescent, or adult) was selected from each households. For the present study, data from adolescents aged 12–18.9 years were considered. For the two surveys, eligibility criteria included absence of physical deficiencies that would prevent the performance of anthropometric measurements and application of the questionnaires, and absence of pregnancy. In 2005, 573 adolescents residing in households randomly selected were invited to participate in the study, however 561 eligible adolescents were initially interviewed (response rate, 97.9%). Fifty adolescents were excluded due to inconsistencies in answering the questionnaires; the data from 511 adolescents were analyzed (91.0% of questionnaire respondents). In 2010, 347 adolescents were invited to participate in the study, however 314 eligible adolescents were interviewed (response rate, 90.5%); their data were then analyzed. Considering an estimated prevalence of approximately 20% for disordered eating behaviors (Dunker and Philippi, 2003; Sampei et al., 2009; Ferreira et al., 2013), 95% confidence interval, with 5% absolute precision for 2005 and 4% absolute precision for 2010, and the design effect for cluster sampling, the used sample sizes of 511 and 314 allowed estimation of the prevalence of disordered eating behaviors for the adolescent population evaluated.
2.3. Data analysis Data from the two surveys were entered into CSPro 2.5 version (Washington DC, USA) (2005survey) and CSPro 3.3 (Washington DC, USA) version (2010 survey), with restriction mechanism of improbable data entry, and were analyzed using the SPSS 19.0 version software (Chicago, IL, USA). The sampling weight of each individual (calculated by the inverse of the probability of selection, i.e., sample weight =1 probability of selection of each individual in the sample) was used to expand the sample, and the design effect for cluster sampling was taken into account for all analyses (Beckett et al., 1992; Sousa and Silva, 2003), using the Complex Sample procedure from SPSS. A descriptive analysis was performed considering the frequencies and 95% confidence intervals (95% CI) for the variables investigated for the two surveys (2005 and 2010): presence or absence of binge eating, compensatory behaviors (self-induced vomiting, use of laxatives
2.2. Data collection In both the 2005 and 2010 surveys, data collection was performed by a team of trained interviewers. Household interviews and measurements were performed following the signing of a free and informed consent form by the household head, and only willing adolescents 65
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
3.65 times higher in 2010, 95% CI: 1.45–9.22) and for black/brown adolescents than for white adolescents in 2005 (ORadj: 1.87, 95% CI: 1.02–3.43). No significant association between the prevalence of strict dieting or fasting and skin color was observed in 2010 (Table 3). The prevalence of compensatory behaviors was significantly higher for overweight adolescents than for non-overweight ones in 2005 (5.9% vs 2.5%, p=0.029) and was not significantly associated with any of the remaining variables, in both 2005 and 2010 (Table 2).
or diuretics), and strict dieting orfasting over the previous 6 months; gender; age (12–14.9 years and 15–18.9 years); skin color (black/ brown and white). For the per capita income classification, 3 categories were considered for the descriptive analysis (up to half minimum wage; half to one minimum wage; and > 1 minimum wage), and 2 categories were used for the association analysis (up to half minimum wage, which approaches the definition of the poverty line in Brazil (Brasil, 2014), and higher than half minimum wage). For the BMI classification, 4 categories were considered for the descriptive analysis (low weight, adequate weight, overweight, and obesity), and 2 for the association analysis: overweight (overweight and obesity) and nonoverweight (low weight and adequate weight). Frequencies were compared between the two surveys using the chisquare test. For variables with more than 2 categories, a partitioning chi-square test was performed using the WinPepi software (Abramson, 2004). Associations between disordered eating behaviors and the other variables were analyzed for each survey using the chi-square test. The changes in prevalence of disordered eating behaviors in each sub-group of investigated variables were evaluated by logistic regression. To statistically compare the 2005–2010 odds ratios of disordered eating behavior prevalence between sub-groups (e.g. boys vs. girls), the method described by Altman and Bland (2003) to compare parameter estimates from separate analyses was used. The strength of associations were analyzed by estimating the odds ratio (OR) and confidence interval 95% (CI 95%) using logistic regression for the variables found to be significantly associated with the disordered eating behaviors (p < 0.05) in each survey. These variables were considered in a multivariate analysis to investigate their associations with disordered eating behaviors after adjusting for the remaining variables (OR adjusted: ORadj). Variables with p values < 0.05 were retained in the final model.
4. Discussion The most important finding of the present study was an 18.4% increase in the prevalence of binge eating over a period of 5 years; however, the prevalence of compensatory behaviors and strict dieting varied only slightly between the two surveys in this period. For the same period, the proportion of adolescents with low and adequate weight decreased, and the proportion of overweight and obese adolescents increased. These changes are probably not related to changes in the socioeconomic and demographic profile of the studied population, since no significant differences were observed for these variables between the years of research. This finding could indicate that the increase in inadequate eating habits, especially in binge eating, may be related to the increase in the prevalence of overweight subjects in this population. The present study is the first in Brazil to investigate temporal changes in the prevalence of disordered eating behaviors in adolescents, which makes its comparison with other Brazilian studies difficult. However, international studies have investigated these questions. Hay et al. (2008) conducted a population-based study in Australia, with individuals older than 15 years of age and observed a 7.4% increase in the prevalence of binge eating for those aged 15–24 years over a period of 10 years (1995–2005), while the prevalence of strict dieting and compensatory behaviors decreased. Also in Australia, Mitchison et al. (2014) found that between 1998 and 2008, binge eating increased significantly (by 4.9% points) in adolescents and young adults aged 15–24 years. Restrictive diet and compensatory behaviors did not show significant changes between the years of research, as in the present study. Neumark-Sztainer et al. (2011), in a longitudinal study with American individuals with a mean age of 12.8 ± 0.7 (early adolescence) and 15.9 ± 0.8 years (middle adolescence), and after a 10-year followup with a mean age of 23.2 ± 1.0 (early young adulthood) and 26.2 ± 0.9 years (middle young adulthood), found that binge eating significantly increased from adolescence to adulthood in the older follow-up cohort. However, Nakai et al. (2014), investigating temporal changes in the disordered eating behaviors of female Japanese students between the ages of 16 and 23 years, found a significant increase in binge eating, as in the others disordered behaviors, in 20 years 1982–2002). In the above-mentioned studies, binge eating was the disordered behavior that increased more remarkably, as observed in the present study. The increase in the prevalence of overweight observed in the present study (7.5%) is in accordance with nationwide findings (approximately 5%) (IBGE, 2010), which revealed that the prevalence of overweight has been continuously increasing among Brazilian adolescents over the last three decades. The increase in the prevalence of disordered eating behaviors and overweight observed for the population of adolescents living in Duque de Caxias therefore reinforces our hypothesis that the two factors - eating disorders and overweight - may be interrelated. In this study, the prevalence of binge eating was not significantly associated with gender in either survey, but increased significantly in both boys and girls. In contrast, Ferreira and Veiga (2008a) observed that binge eating was more frequent in girls, although its prevalence was not low in boys. According to Mitchison and Mond (2015), in a narrative review, binge eating appears to be the most common eating disorder behavior in men, with a prevalence almost equal to that of women.
3. Results No significant differences in the sociodemographic factors were observed between the participants of the two surveys. Fifty percent of the evaluated adolescents were male, approximately 80% were of black/brown color, and approximately 50% had a per capita income of up to 1 minimum wage. A significant difference in BMI classification was observed (p=0.033), with a decreased prevalence of low weight (by 3.6%) and an increased prevalence of overweight (by 6.5%). Prevalence of binge eating increased by 18.4% in 2010 compared to the prevalence observed in 2005 (p=0.001). Significant differences were neither observed for the remaining disordered eating behaviors evaluated nor for the presence of 2 or more behaviors (Table 1). The prevalence of binge eating increased significantly for both genders and age groups, for both skin colors (black/brown and white), and all levels of income per capita. However, a significant change was observed between 2005 and 2010, for non-overweight adolescents, but not for overweight ones. Using the Altman and Bland method, the odds ratios calculated across the two surveys for binge eating did not differ significantly between all variables. No significant changes were found for the other behaviors (Table 2). In 2005, the prevalence of binge eating was associated with skin color and BMI classification (Table 2), so that, the above mentioned prevalence was higher for black/brown participants than for white participants (ORadj: 1.91; 95% CI: 1.04–3.52) and for overweight adolescents than for non-overweight adolescents (ORadj: 2.02; 95% CI: 1.13–3.61). In 2010, none of the variables were significantly associated with binge eating (Table 3). In both 2005 and 2010, prevalence of strict dieting or fasting was associated with gender and BMI classification (Table 2). This behavior was more prevalent in girls than in boys in 2005 (ORadj: 1.95; 95%: 1.07–3.53), and this difference increased to 7.02 times in 2010 (95% CI: 2.81–17.54). Moreover, in both surveys, the prevalence of strict dieting or fasting was higher for overweight adolescents than for nonoverweight adolescents (2.29 times higher in 2005, 95% CI: 1.14–4.64; 66
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 1 Sociodemographic factors, body mass index classification (BMI), and disordered eating behaviors among adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. 2005 na Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb Up to half minimum wage Half to one minimum wage > 1 minimum wage BMI classification Low weight Adequate weight Overweight Obesity DEBc Binge eating Strict dieting or fasting Compensatory behaviors 2 or more DEB
2010 % (95% CI)
511
na
511
0.668 51.0 (44.3–57.7) 49.0 (42.3–55.7)
+ 1.7 − 1.7
50.3 (42.7–57.9) 49.7 (42.1–57.3)
+ 5.4 − 5.4
78.9 (70.9–85.1) 21.1 (14.9–29.1)
− 2.3 + 2.3
50.5 (40.7–60.2) 36.5 (28.5–45.3) 13.1 (7.4–22.1)
+ 2.8 − 5.8 + 3.0
0.9 (0.3–2.8) 68.2 (60.5–74.9) 22.8# (17.3–29.5) 8.1 (5.1–12.8)
− − + +
3.6 4.0 6.5 1.0
38.4 (30.3–47.1) 17.4 (12.3–23.9) 4.9 (2.6–9.1) 13.2 (8.8–19.5)
+ − + +
18.4 1.5 1.6 4.0
314 44.9 (39.3–50.6) 55.1 (49.4–60.7)
510
0.178
308 81.2 (76.7–85.0) 18.8 (15.0–23.3)
503
0.572
304 47.7 (41.2–54.3) 42.3 (36.8–47.9) 10.1 (7.1–14.1)
511
0.470
301 4.5# (2.7–7.2) 72.2 (66.9–76.9) 16.3 (12.7–20.6) 7.1 (4.8–10.5) 20.0 (15.0–26.1) 18.9 (14.2–24.6) 3.3 (1.8–6.0) 9.2 (6.1–13.5)
312 312 311 312
p valued
% (95% CI)
314 49.3 (44.5–54.1) 50.7 (45.9–55.5)
506 504 506 511
Variation
0.033
< 0.001 0.701 0.380 0.181
a
Different totals due to missing values in each variable. Minimum wage: 2005= R$ 300.00; 2010= R$ 510.00. c DEB: disordered eating behaviors. d Chi-square test. # Partitioning chi-square test p < 0.05. b
2010 may be explained by the significant and more pronounced increase in the prevalence of binge eating among non-overweight adolescents (2-fold) than among overweight adolescents (1.5-fold), which reduced the difference between the two groups in 2010. The prevalence of strict dieting was relatively constant, with a tendency to decrease over the period between the two surveys, similar to that observed by Hay et al. (2008) in Australia. An exception was the girls, for whom an increase of 4.9% in the prevalence of strict dieting was observed, which was higher than the value noted for boys. Other studies performed in Brazil (Vilela et al., 2004; Ferreira and Veiga, 2008a) and other countries (Ata et al., 2007; Keel et al., 2007; NuñoGutiérrez et al., 2009; Bilali et al., 2010) also reported a higher prevalence of strict dieting in girls than in boys. This may be explained by the fact that girls are more concerned with their body image (Vilela et al., 2004; Goldenberg, 2011) and adopt inadequate eating habits to attain the desired bodyshape. In addition, there was also a significant association (p=0.039) for have 2 or more disordered eating behaviors among girls in 2010, but not in 2005. This suggests a worsening of this problem in females, since the prevalence increased from 11.1% to 19.2%, whereas that in boys remained constant (7.1% to 7.4%) (data not shown). The prevalence of strict dieting was also associated with overweight, similar to the findings of a previous study including adolescents who attended public schools in Niterói, RJ (Ferreira and Veiga, 2008a). In addition, compensatory behaviors were associated with overweight in 2005, when the prevalence of these behaviors were two times higher among overweight adolescents than among non-overweight adolescents. Overweight individuals are undervalued, undergo psychological suffering, and may therefore opt for extreme alternatives for weight control, such as strict dieting, fasting, or use of compensatory behaviors (Bernardi et al., 2005). This may explain the association between overweight and these behaviors. The hypothesis of the present study was that increased prevalence of disordered eating behaviors was concomitant with an increase in the prevalence of overweight in adolescents. The prevalence of overweight
None of the analyzed disordered eating behaviors was significantly associated with the per capita income, indicating that they are homogenously distributed throughout economic levels. Similarly, another study performed in Brazil did not find any association between disordered eating behaviors and the socioeconomic status of adolescents (Dunker et al., 2009), similar to the findings in other countries (Bauer and Kirchengast, 2006; Bergström and Elfhag, 2007; Preti et al., 2007; Soh et al., 2007). The fact that binge eating was significantly associated with the black/brown skin color in 2005 but not in 2010 can be explained by the more pronounced increase in its prevalence amongst white adolescents (approximately 3 times, from 11.6% in 2005 to 34.7% in 2010) as compared to that amongst black/brown adolescents (approximately 2 times, from 21.8% in 2005 to 40.2% in 2010), thereby reducing the difference between the two groups in 2010. The divergent results found regarding the association of binge eating with income and skin color may be intriguing considering that in Brazil, skin color is used as a socioeconomic level marker (Ferreira et al., 2015; Moreira et al., 2015). The comparison of this result with other studies is difficult, considering that previous investigations regarding associations of disordered eating behaviors with skin color have focused on the issue of ethnicity (Javier et al., 2016), which differs from the context addressed in this study and that of other studies conducted in Brazil. Therefore, this issue needs to be better investigated in other studies. The prevalence of binge eating was associated with overweight in 2005, confirming the results of previous studies performed in Brazil and other countries (Stice et al., 2002; Pivetta and Silva, 2010; Prisco et al., 2013, Sonneville et al., 2013). This association may be explained by the fact that although binge eating and obesity present different symptoms, they have a few similarities; both are usually accompanied by compulsive eating, i.e., consumption of large quantities of food in short periods of time, leading to feelings of guilt and lack of control (Pinheiro et al., 2006). The fact that a significant association between the prevalence of binge eating and overweight was not observed in 67
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 2 Prevalence of binge eating, strict dieting or fasting and compensatory behaviors with odds ratio and 95% confidence intervals (95%CI), and its association with sociodemographic factors and body mass index (BMI) classification for adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. Binge eating
Var. (%)
2005 a
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
2010
n
%
249 258
21.8 18.1
228 279
21.0 19.1
414 91
21.8 11.6
240 258
19.6 20.7
388 119
17.2 28.8
p
Odds ratio (95%CI)
a
n
p
0.432
0.757 158 153
39.7 37.0
158 153
34.3 42.5
241 65
40.2 34.7
151 151
36.5 38.1
206 93
36.4 45.7
0.653
0.026
246 258
13.7 23.7
228 276
17.5 20.0
408 95
20.6 11.6
238 258
21.1 16.8
385 119
15.4 30.1
p 0.025
2010 na
%
+ 16.9# + 17.4#
2.35 (1.20 – 4.62) 2.36 (1.27 – 4.37)
+ 19.8# + 16.5
2.74 (1.57 – 4.80) 2.08 (0.89 – 4.83)
Odds ratio (CI95%)
158 153
6.5 28.6
158 153
16.7 18.1
241 65
19.7 9.2
151 151
18.2 15.6
206 93
13.4 28.6
p < 0.001 − 7.2 + 4.9
0.44 (0.16 – 1.16) 1.28 (0.67 – 2.45)
− 0.8 − 1.9
0.94 (0.41 – 2.17) 0.88 (0.47 – 1.66)
− 0.9 − 2.4
0.94 (0.52 – 1.71) 0.77 (0.26 – 2.22)
− 2.9 − 1.2
0.83 (0.40 – 1.70) 0.91 (0.47 – 1.77)
− 2.4 − 1.9
0.85 (0.44 – 1.63) 0.93 (0.38 – 2.25)
0.825
0.094
0.340
0.601
0.014
0.027
Compensatory behaviors
Var. (%)
2005
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
2.41 (1.56 – 3.71) 4.07 (1.53 – 10.82)
Var. (%)
0.031
n
+ 18.4# + 23.1#
0.298
0.522
a
1.96 (1.07 – 3.59) 3.14 (1.62 – 6.07)
0.842
Strict dieting or fasting
Gender Male Female Age (years) 12–14.9 15–18.9 Skin color Black/Brown White Per capita incomeb ≤Half minimum wage > Half minimum wage BMI classification Non-overweight Overweight
+ 13.3# + 23.4# 0.564
0.805
%
2.35 (1.22 – 4.56) 2.65 (1.39 – 5.05)
0.280
0.020
2005 na
+ 17.9# + 18.9#
2010 %
p
a
Odds ratio (95%CI)
n
%
158 153
3.8 6.0
157 154
3.2 6.6
240 65
3.4 10.8
150 151
5.2 4.9
206 92
5.8 3.5
0.101 246 259
2.0 4.6
228 277
2.9 3.6
410 95
3.6 2.1
239 259
3.9 2.9
386 120
2.5 5.9
p 0.484
0.679
+ 1.8 + 1.4
1.99 (0.44 – 8.90) 1.33 (0.46 – 3.85)
+ 0.3 + 3.0
1.09 (0.29 – 4.19) 1.89 (0.59 – 6.01)
− 0.2 + 8.7
0.95 (0.31 – 2.86) 5.63 (0.84 – 37.52)
+ 1.3 + 2.0
1.36 (0.41 – 4.49) 1.75 (0.46 – 6.63)
+ 4.0 − 2.5
2.42 (0.77 – 7.62) 0.58 (0.14 – 4.48)
0.222
0.433
0.080
0.569
0.941
0.029
0.445
a
Different totals due to missing values in each variable. Minimum wage: 2005= R$ 300.00; 2010= R$ 510.00. # Logistic regression p < 0.05. b
2010 most variables. These results indicate that these extreme behaviors, which could lead to in important nutritional deficiencies, may not be as widespread amongst adolescents of lower social classes such as those evaluated in the present study. This study did not aim at diagnosing eating disorders, but at screening changes in disordered eating behaviors over a period of 5
increased by 7.5% and that of binge eating increased by 18.4% between the two surveys. Binge eating increased remarkably, even in groups that did not present significant increase in prevalence of overweight (data not shown), which may indicate an even higher increase in the risk for overweight. However, the prevalence of strict dieting decreased, and the use of compensatory behaviors varied little between 2005 and 68
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
Table 3. Gross and adjusted Odds Ratio and 95% confidence intervals (95%CI) for association between disordered eating behaviors with sociodemographic factors and body mass index (BMI) classification for adolescents living in Rio de Janeiro, Brazil, in 2005 and 2010. Variables
2005
Variables
OR (95%CI) Gross Strict dieting or fasting Gender Male Female Skin color Black/Brown White
OR (95%CI) Adjusted
1.00 1.95 (1.08 – 3.53)
1.00 1.95 (1.07 – 3.53)
1.00 1.97 (1.06 – 3.69)
1.00 1.87 (1.02 – 3.43)
1.00 2.37 (1.19 – 4.74)
1.00 2.29 (1.14 – 4.64)
Skin color Black/Brown White
1.00 2.14 (1.12 – 4.08)
1.00 1.91 (1.04 – 3.52)
BMI classification Non-overweight Overweight
1.00 1.94 (1.08 – 3.49)
1.00 2.02 (1.13 – 3.61)
BMI classification Non-overweight Overweight
2010
Strict dieting or fasting Gender Male Female BMI classification Non-overweight Overweight
Gross
Adjusted
1.00 5.75 (2.17 – 15.20)
1.00 7.02 (2.81 – 17.54)
1.00 2.60 (1.09 – 6.17)
1.00 3.65 (1.45 – 9.22)
Binge eating
Acknowledgments
years in adolescents predominantly of lower social classes. Because it used two cross-sectional surveys, we could not determine the causality of the associations between these behaviors and the studied variables, which is a limitation of the study. Future longitudinal studies may help clarify these associations. Another limitation of the present study was related to the questions used to investigate the disordered eating behaviors. Since the questions were simplified and did not include all eating disorder symptoms, they may have failed to reliably detect these behaviors. However, when compared to the wider and more frequently used questionnaires, the questions used addressed important issues about the main behaviors related to eating disorders. In addition, the reproducibility of these questions was considered moderate to excellent in a previous study with adolescents (Ferreira and Veiga, 2008b). This simplified questionnaire can therefore be considered useful for screening such behaviors in population studies, especially in populations with low socioeconomic status, which presents greater difficulties in the interpretation of more complex questions such as the one investigated in the present study. Furthermore, the difference in response options to questions that investigated disordered eating behaviors in the 2005 and 2010 studies is other limitation of the study as it is conceivable that some participants who used eating behaviors less than once a week, but more than never (as answered in 2010), could have answered "no" to the question in 2005. In conclusion, there was a pronounced increase in the prevalence of binge eating over the 5-year period between the two surveys in the studied population, and girls and overweight adolescents were more vulnerable to strict dieting. The present study shows that health professionals should pay special attention to adolescents with these behaviors, which may persist into adulthood or eventually evolve to a subthreshold, full eating disorder, or a weight disorder as overweight and obesity, which could result in serious health risks.
Grant support: The authors thank the Conselho Nacional de Desenvolvimento Científico e Tecnológico – CNPq (National Council for Scientific and Technological Development) grant numbers 503139/ 2003-3 and 476344/2008-5 and Instituto Nacional de Câncer (National Institute of Cancer) which funded the project. References Abramson, J.H., 2004. WINPEPI (PEPI-for-Windows): computer programs for epidemiologists. Epidemiol. Perspect. Innov. 1, 6. Altman, D.G., Bland, J.M., 2003. Interaction revisited: the difference between two estimates. BMJ (Clinical research ed.) 326, 219. American Academy of Pediatrics (AAP) - (US), 2003. Children and adolescents with eating disorders: the state of the art. Pediatrics 111, 1–13. American Psychiatric Association (APA), 2006. Treatment of patients with eating disorders. Am. J. Psychiatry 163, 4–54, (Third edition). American Psychiatric Association (APA), 2013. Diagnostic and Statistical Manual of Mental Disorders: DSM-5 5th edition. American Psychiatric Publishing Inc, Arlington. American Psychiatric Association, 1994. Diagnostic and Statistical Manual of Mental Disorders DSM-IV. Washington, DC. Ata, R.N., Ludden, A.B., Lally, M.M., 2007. The effects of gender and family, friend, and media influences on eating behaviors and body image during adolescence. J. Youth Adolesc. 36, 1024–1037. Bauer, M., Kirchengast, S., 2006. Body composition, weight status, body image and weight control practices among female adolescents from eastern Austria. Anthropol. Anz. 64, 321–331. Beckett, L.A., Scherr, P.A., Evans, D.A., 1992. Population prevalence estimates from complex samples. J. Clin. Epidemiol. 45, 393–402. Bergström, H., Elfhag, K., 2007. Socioeconomic differences in obese patients with and without eating disorders. Eat. Weight Disord. 12, 98–103. Bernardi, F., Cichelero, C., Vitolo, M.R., 2005. Comportamento de restrição alimentar e obesidade. (restrained eating behavior and obesity). Rev. De. Nutr. 18, 85–93. Bilali, A., Galanis, P., Velonakis, E., Katostaras, T., 2010. Factors associated with abnormal eating attitudes among greek adolescents. J. Nutr. Educ. Behav. 42, 292–298, (Journal of Racial Ethnic Health Disparities). Brasil, 2014. Ministério do Desenvolvimento Social. Decreto n° 8.232, de 30 de abril de 2014. Altera o Decreto n° 5.209, de 17 de setembro de 2004, que regulamenta o Programa Bolsa Família, e o Decreto n° 7.492, de 2 de junho de 2011, que institui o Plano Brasil Sem Miséria. (It amends Decree No. 5,209, dated September 17, 2004, which regulates the Bolsa Família Program, and Decree No. 7,492, of June 2, 2011, which establishes the Brazil Without Misery Plan). Brasília; Presidência da República. Dunker, K.L., Philippi, S.T., 2003. Hábitos e comportamentos alimentares de adolescentes com sintomas de anorexia nervosa. (food habits and feeding behavior in
Conflicts of interest The authors declare no conflict of interest. 69
Psychiatry Research 253 (2017) 64–70
D.D. Santana et al.
students in 1982, 1992 and 2002. Psychiatry Res. 219, 151–156. Neumark-Sztainer, D., Wall, M., Larson, N.I., Eisenberg, M.E., Loth, K., 2011. Dieting and disordered eating behaviors from adolescence to young adulthood: findings from a 10-year longitudinal study. J. Am. Diet. Assoc. 111, 1004–1011. Nuño-Gutiérrez, B.L., Rosa, Celis-De. La, Unikel-Santoncini, C, A., 2009. Prevalence and associated factors related to disordered eating in student adolescents of Guadalajara across sex. Rev. Investig. Clínica 61, 286–293. Oliveira, F.P., Bosi, M.L.M., Vigario, O.S., Vieira, R.S., 2003. Eating behavior and body image in athletes. Revista Brasileira de Medicina do Esporte 9, 348–356. de Onis, M., Onyango, A.W., Borghi, E., Siyam, A., Nishida, C., Siekmann, J., 2007. Development of a WHO growth reference for school-aged children and adolescents. Bull. World Health Organ. 85, 660–667. Pinheiro, A.P., Sullivan, P.F., Bacaltchuck, J., Prado-Lima, P.A.S., Bulik, C.M., 2006. Genética em transtornos alimentares: Ampliando os horizontes de pesquisa. (Genetics in eating disorders: extending the boundaries of research). Revista Brasileira de Psiquiatria 28, 218–225. Pivetta, L.A., Gonçalves-Silva, R.M.V., 2010. Compulsão alimentar e fatores associados em adolescentes de Cuiabá, Mato Grosso, Brasil. (binge eating and associated factors among teenagers in Cuiabá, Mato Grosso state, Brazil). Cad. De. Saúde Pública 26, 337–346. Preti, A., Pinna, C., Nocco, S., Pilia, S., Mulliri, E., Micheli, V., Casta, M.C., Petretto, D.R., Masala, C., 2007. Rural/urban differences in the distribution of eating disorder symptoms among adolescents from community samples. Aust. N.Z. J. Psychiatry 41, 525–535. Prisco, A.P.K., Araujo, T.M., Almeida, M.M.G., Santos, K.O.B., 2013. Prevalência de transtornos alimentares em trabalhadores urbanos de município do Nordeste do Brasil. (Prevalence of eating disorders in urban workers in a city of the northeast of Brazil). Ciência e Saúde Coletiva 18, 1109–1118. Salles-Costa, R., Pereira, R.A., Vasconcellos, M.T.L., Veiga, G.V., Marins, V.M.R., Jardim, B.C., Gomes, F.S., Sichieri, R., 2008. Associação entre fatores socioeconômicos e insegurança alimentar: estudo de base populacional na Região Metropolitana do Rio de Janeiro, Brasil. (Association between socioeconomic factors and food insecurity: a population-based study in the Rio de Janeiro metropolitan area, Brazil). Rev. Nutr. 21 (Supl), 99s–109s. Sampei, M.A., Sigulem, D.M., Novo, N.F., Juliano, Y., Colugnati, F.A.B., 2009. Eating attitudes and body image in ethnic Japanese and Caucasian adolescent girls in the city of São Paulo, Brazil. J. Pediatr. 85, 122–128. Smink, F.R., van Hoeken, D., Hoek, H.W., 2012. Epidemiology of eating disorders: incidence, prevalence and mortality rates. Curr. Psychiatry Rep. 14, 406–414. Soh, N.L., Touyz, S., Dobbins, T.A., Surgenor, L.J., Clarke, S., Kohn, M.R., Lee, E.L., Leow, V., Rieger, E., Ung, K.E., Walter, G., 2007. Restraint and eating concern in North European and East Asian women with and without eating disorders in Australia and Singapore. Aust. N.Z. J. Psychiatry 41, 536–545. Sonneville, K.R., Horton, N.J., Micali, N., Crosby, R.D., Swanson, S.A., Solmi, F., Field, A.E., 2013. Longitudinal associations between binge eating and overeating and adverse outcomes among adolescents and young adults: Does loss of control matter? JAMA Pediatr. 167, 149–155. Sousa, M.H., Silva, N.N., 2003. Estimativas obtidas de um levantamento complexo (estimates from a complex survey). Rev. De. Saúde Pública 37, 662–670. Stice, E., Presnell, K., Spangler, D., 2002. Risk factors for binge eating onset in adolescent girls: a 2-year prospective investigation. Health Psychol. 21, 131–138. Vilela, J.E.M., Lamounier, J.A., Dellaretti, M.A., Neto, B.J.R., Horta, G.M., 2004. Eating disorders in school children. J. De. Pediatr. 80, 49–54. World Health Organization, 2013. Global Database on Body Mass Index. 〈http://apps. who.int/bmi/index.jsp/〉 (Accessed 18 February 2016).
adolescents with symptoms of anorexia nervosa). Rev. Nutr. 16, 51–60. Dunker, K.L.L., Fernandes, C.P.B., Carreira Filho, D., 2009. Influência do nível socioeconômico sobre comportamentos de risco para transtornos alimentares em adolescentes. (Socioeconomic influence on eating disorders risk behaviors in adolescentes). J. Bras. Psiquiatr. 58, 156–161. Ferreira, J.E.S., Veiga, G.V., 2008a. Eating disorder risk behavior in Brazilian adolescents from low socioeconomic level. Appetite 51, 249–255. Ferreira, J.E.S., Veiga, G.V., 2008b. Confiabilidade (teste-reteste) de um questionário simplificado para avaliar comportamentos de risco para transtornos alimentares em adolescentes. (test-retest reliability of a simplified questionnaire for screening adolescents with risk behaviors for eating disorders in epidemiologic studies). Rev. Bras. Epidemiol. 11, 393–401. Ferreira, J.E.S., Souza, P.R.B., Costa, R.S., Sichieri, R., Veiga, G.V., 2013. Disordered eating behaviors in adolescents and adults living in the same household in metropolitan area of Rio de Janeiro, Brazil. Psychiatry Res. 210, 612–617. Ferreira, M.F., Moraes, C.L., Reichenheim, M.E., Verly, E., Marques, E.S., Salles-Costa, R., 2015. Effect of physical intimate partner violence on body mass index in lowincome adult women. Cadernos de Saúde Pública 31, 161–172. Galindo, E.M.C., Carvalho, A.M.P., 2007. Tradução, adaptação e avaliação da consistência interna do eating behaviours and body image test para uso com crianças do sexo feminino. (translation, adaptation and internal consistency evaluation of the eating behaviors and body image test for female children). Rev. Nutr. 20, 20–21. Goldenberg, M., 2011. Gênero, “o Corpo” e “Imitação Prestigiosa” na Cultura brasileira. (Gender, "the body" and "Prestigious imitation" in the Brazilian culture). Saúde e Sociedade 20, 543–553. Haldane, J.B.S., 1945. On a method of estimating frequencies. Biometrika 33, 222–225. Hay, P., 1998. The Epidemiology of eating disorder behaviors: an Australian community – based survey. J. Eat. Disord. 23, 371–382. Hay, P.J., Mond, J.M., Buttner, P., Darby, A., 2008. Eating disorder behaviors are increasing: findings from two sequential community surveys in South Australia. PLoS ONE 3, 1–5. Instituto Brasileiro de Geografia e Estatística (IBGE), 2010. . Pesquisa de Orçamentos Familiares 2008–2009. Antropometria e estado nutricional de crianças, adolescentes e adultos no Brasil (Household Budget Survey 2008–2009). (Anthropometry and nutritional status of children, adolescents and adults in Brazil). Instituto Brasileiro de Geografia e Estatística, Rio de Janeiro〈http://www.ibge.gov.br/〉, (Accessed 13 February 2016). Javier, S.J., Moore, M.P., Belgrave, F.Z., 2016. Racial comparisons in perceptions of maternal and peer attitudes, body dissatisfaction, and eating disorders among African American and White women. Women Health 56, 615–633. Keel, P.K., Baxter, M.G., Heatherton, T.F., Joiner, T.E.J.R., 2007. A 20-year longitudinal study of body weight, dieting, and eating disorder symptoms. J. Abnorm. Psychol. 116, 422–432. Leal, G.V.S., Philippi, S.T., Polacow, V.O., Cordás, T.A., Alvarenga, M.S., 2013. O que é comportamento de risco para transtornos alimentares em adolescentes? (What is adolescents' eating disorder risk behavior?). J. Bras. Psiquiatr. 62, 62–75. Mitchison, D., Mond, J., 2015. Epidemiology of eating disorders, eating disordered behaviour, and body image disturbance in males: a narrative review. J. Eat. Disord. 23, 20. Mitchison, D., Hay, P., Slewa-Younan, S., Mond, J., 2014. The changing demographic profile of eating disorder behaviors in the community. BMC Public Health 14, 943. Moreira, N.F., Sichieri, R., Reichenheim, M.E., Oliveira, A.S., Veiga, G.V., 2015. The associations of BMI trajectory and excessive weight gain with demographic and socio-economic factors: the Adolescent Nutritional assessment longitudinal study cohort. Br. J. Nutr. 114, 2032–2038. Nakai, Y., Nin, K., Noma, S., 2014. Eating disorder symptoms among Japanese female
70