Hostname: page-component-586b7cd67f-2brh9 Total loading time: 0 Render date: 2024-11-26T05:17:29.239Z Has data issue: false hasContentIssue false

Prevalence and risk factors for obesity in Balearic Islands adolescents

Published online by Cambridge University Press:  12 August 2009

Maria del Mar Bibiloni
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
Elisa Martinez
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
Rosa Llull
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
Maria Daniela Juarez
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
Antoni Pons
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
Josep A. Tur*
Affiliation:
Research Group on Community Nutrition and Oxidative Stress, University of Balearic Islands, Guillem Colom Building, Campus, E-07122Palma de Mallorca, Spain
*
*Corresponding author: Dr Josep A. Tur, fax +34 971 173184, email [email protected]
Rights & Permissions [Opens in a new window]

Abstract

The aim of this work was to assess the prevalence and risk factors of obesity in the Balearic Islands' adolescents. A cross-sectional nutritional survey was carried out in the Balearic Islands (2007–2008). A random sample (n 1231) of the adolescent population (12–17 year old) was interviewed. Anthropometric measurements, two non-consecutive 24 h recalls and a general questionnaire incorporating questions related to sociodemographic and lifestyle variables including the physical activity questionnaire were used. The prevalence of overweight was 19·9 % (boys) and 15·5 % (girls), and obesity 12·7 % (boys) and 8·5 % (girls). The main risk factors associated with a higher prevalence of obesity were low parental education level (boys OR: 3·47; 95 % CI: 1·58, 7·62; girls OR: 3·29; 95 % CI: 1·38, 7·89), to skip meals (boys OR: 4·99; 95 % CI: 2·1, 11·54; girls OR: 2·20; 95 % CI: 0·99, 4·89), age (12–13-year-old boys; OR: 2·75; 95 % CI: 1·14, 6·64), attention to mass media (television (TV)+radio; boys OR: 1·50; 95 % CI: 0·81, 2·84; girls OR: 2·06; 95 % CI: 0·91, 4·68), short sleep (boys OR: 3·42; 95 % CI: 0·88, 13·26), low parental socioeconomic status (girls OR: 3·24, 95% CI: 1·04, 10·05 ) and smoking (girls OR: 2·51; 95 % CI: 0·88, 7·13). A programme of action including school healthy education and promotion programmes targeted at parents and adolescents are needed. These programmes may be mainly focused to increase educational level, to make the adolescents to be aware of to skip meals and to smoke are not appropriate methods to reduce the risk of obesity, but the usefulness is to do not eat while watching TV, to sleep 8–10 h/d and to be physically active.

Type
Full Papers
Copyright
Copyright © The Authors 2009

The prevalence of overweight and obesity among children and adolescents has risen greatly worldwide, making them one of the most serious public health problems in this age group and in adulthood(Reference Ebbeling, Pawlak and Ludwig1). Childhood and adolescent excessive weight (overweight and obesity) are associated with significant immediate and long-term health risks(Reference Ebbeling, Pawlak and Ludwig1, Reference Dehghan, Akhtar-Danesh and Merchant2). About 70 % of obese adolescents grow up to become obese adults(Reference Dehghan, Akhtar-Danesh and Merchant2). Childhood and adolescent obesity predict obesity in adulthood and increase the risk of adult morbidity and mortality(Reference Ebbeling, Pawlak and Ludwig1, Reference Bibbins-Domingo, Coxson and Pletcher3).

Although there are both genetic and environmental causes of obesity, the increase in prevalence of obesity worldwide is likely to be more closely associated with changes in environmental factors(Reference Dehghan, Akhtar-Danesh and Merchant2). Sociocultural factors as parental occupational status, maternal level of education, cultural and/or religious habits and the role of family and patterns of beauty are factors that have a strong influence on eating habits in adolescents(Reference Neumark-Sztainer, Hannan and Story4). Changes on physical activity patterns in youth as a result of adopting a major inactive lifestyle, increasing time spent watching television (TV), playing video games and Internet are also associated with obesity increases(Reference Ebbeling, Pawlak and Ludwig1, Reference Dehghan, Akhtar-Danesh and Merchant2).

The aim of present study was to assess the prevalence and risk factors of obesity in adolescents of the Balearic Islands, a Mediterranean region.

Methods

Study design

The study is a population based cross-sectional nutritional survey carried out in the Balearic Islands between 2007 and 2008.

Selection of participants, recruitment and approval

The target population consisted of all inhabitants living in the Balearic Islands aged 12–17 years. The sample population was derived from residents aged 12–17 years registered in the scholar census of the Balearic Islands. The theoretical sample size was set at 1500 individuals in order to provide a specific relative precision of 5 % (type I error = 0·05; type II error = 0·10), taking into account an anticipated 70 % participation rate. The sampling technique included stratification according to municipality size, age and sex of inhabitants, and random selection within subgroups, with Balearic Islands' municipalities being the primary sampling units, and individuals within these municipalities comprising the final sample units. The interviews were performed at the schools. The final sample size was 1231 individuals (82 % participation). The reasons to not participate were (a) the subject declined to be interviewed; (b) the parents did not authorise the interview.

Anthropometric measurements

Height was determined using a mobile anthropometer (Kawe 44 444, Asperg, Germany) to the nearest millimetre, with the subject's head in the Frankfurt plane. Body weight was determined to the nearest 100 g using a digital scale (Tefal, sc9210, Rumilly, France). The subjects were weighed in bare feet and light underwear, which was accounted for by subtracting 300 g from the measured weight. Triceps and subscapular skinfold thickness were measured using a Holtain skinfold calliper (Tanner/Whitehouse, Crosswell, Crymych, UK), and the mean of three measurements (right arm) was used, and used to calculate body fat as described previously(Reference Tur, Serra-Majem and Romaguera5). Anthropometric measurements were performed by well-trained observers in order to avoid the inter-observer coefficients of variation. According to the WHO growth standards for children and adolescents, the prevalence of overweight (BMI ≥ P85 < P97) and obesity (BMI ≥ P97) was age- and sex-specific calculated(Reference De Onis, Onyango and Borghi6). Waist circumference and waist:hip ratio were calculated according to cut-off limits described elsewhere(Reference Cook, Weitzman and Auinger7).

Energy and nutrient intakes

For energy intake (EI), two non-consecutive 24 h diet recalls were obtained from the participants. Well-trained dietitians administered the recalls and verified and quantified the food records. To estimate volumes and portion sizes, the household measures found in the subjects' own homes were used. Conversion of food into nutrients was made using a self-made computerised program based on Spanish(Reference Moreiras, Carvajal and Cabrera8, Reference Mataix, Mañas and Llopis9) and European(Reference Feinberg, Favier and Ireland-Ripert10) food composition tables and complemented with food composition data available for Majorcan food items(Reference Ripoll11). Misreporters were determined using the Black's modification(Reference Black12) of the Goldberg et al. cut-off(Reference Livingstone and Black13) for EI. To define underreporting, the lower cut-off value for the ratio between EI and BMR (EI/BMR) was calculated for boys and girls separately as 3 sd below the mean of the ratio between total energy expenditure(14) and BMR (total energy expenditure/BMR) for the 99 % lower confidence limit. With this calculation, the variation of EI, BMR and energy requirements are taken into account. In large studies (n>500), the number of days of dietary assessment has a negligible effect on the cut-off values, and for measured BMR and 2 d of assessment, the coefficient variation used was 28·74 %(Reference Black12). The cut-off values were >1·56 EI/BMR for boys and >1·41 for girls. An EI/BMR ≥ 2·4 was considered to represent overreporting(Reference Johansson, Solvoll and Bjørneboe15, Reference Mendez, Wynter and Wilks16). Underreporters (20 %) and overreporters (2 %) were excluded from the analysis of dietary patterns.

General questionnaire

The study population was divided into two categories according to their BMI: obese (BMI ≥ P97) and non-obese (BMI < P97). A questionnaire incorporating the following questions was used: age group; region of origin (defined as being born in the Balearic Islands, East of Spain as a representative of the Spanish Mediterranean coast, other parts of Spain and other countries); parental educational level (grouped according to years and type of education: low, < 6 years at school; medium, 6–12 years of education; high, >12 years of education); parental socioeconomic level (based on the occupation of parents and classified as low, medium and high, according to the methodology described by the Spanish Society of Epidemiology(Reference Alvarez, Alonso and Domingo17)).

Information about smoking habits, alcohol intake, sweets and salty snacks (potato chips, maize chips, ready-to-eat pop maize, pork rinds, potato sticks, extruded snacks and cheese puffs) consumption and breakfast habits were collected as described: smoking habit (no; yes; occasionally; < 1 cigarette/d); alcohol consumption (no; frequently; occasionally; < 1 drink/week); sweets and salty snacks consumption (yes; occasionally; no); breakfast daily (yes; occasionally; no).

The number of daily meals was calculated using the total of eating occasions that subjects declared to made among the following: breakfast; mid-morning snack; lunch; mid-afternoon snack; dinner; before going to sleep; others. Three groups of eating frequency were considered: ≤ 3; 4; ≥ 5 times/d.

Distraction during mealtime was studied using different possibilities: (1) watching the TV; (2) listen to the radio; (3) get up from the table; (4) conversation during mealtime. These variables were joined and expressed as (1) attention to mass media (TV+radio); (2) others.

Physical activity was evaluated according to the guidelines for data processing and analysis of the international physical activity questionnaire(18) in the short form and its specific modification for adolescents (IPAQ-A)(Reference Hagströmer, Bergman and De Bourdeaudhuij19). The specific types of activity assessed were walking, moderate-intensity activities (i.e. physical activity at school) and vigorous-intensity activities (i.e. sport practice) and an additional question about sitting time was using as an indicator variable of time spent in sedentary activity. The questionnaire also included information on hours per day of TV viewing, computer use, video games, another leisure time physical activity practice and typical sleep duration to the nearest 10 min. The physical activity assessed by the international physical activity questionnaire was correlated with physical activity level (PAL) according to the estimation of the individualised activity coefficient. Each subject was classified taking into account its PAL value(20) as: sedentary (PAL ≥ 1·0 < 1·4); low active (PAL ≥ 1·4 < 1·6); active (PAL ≥ 1·6 < 1·9); very active (PAL ≥ 1·9). Each subject was then classified into no active (sedentary or low active) and active (active or very active).

Differences in EI between BMI < P97 and BMI ≥ P97 subjects were also studied. Total EI (MJ), energy (kJ) per kg of weight, energy (kJ) per kg of lean body mass and percentage of energy from macronutrients were taken into consideration for this analysis.

Statistics

Analyses were performed with SPSS version 16.0 (SPSS, Inc., Chicago, IL, USA). All tests were stratified by sex. Significant differences in prevalence were calculated by means of χ2. Difference between groups' means was tested using ANOVA. Logistic regression models with the calculations of corresponding adjusted OR and 95 % CI were used to examine possible differences between those with BMI below or above P97. Univariate analysis was first carried out for all the sociodemographic and lifestyle variables that could be associated with the frequency of obesity. Any factor that was significantly associated was considered as a candidate for the multivariate model. Multiple logistic regression analyses were used to simultaneously examine the effect of different sociodemographic and lifestyle variables on the prevalence of obesity. The study of possible differences in food patterns between those obese and non-obese was adjusted by confounding factors. All correlations were investigated with Spearman's rank correlation coefficient (r). Level of significance for acceptance was P < 0·05.

Ethics

The present study was conducted according to the guidelines laid down in the Declaration of Helsinki, and all procedures involving human subjects/patients were approved by the Balearic Islands Ethics Committee. Written informed consent was obtained from all subjects and their parents or legal tutors.

Results

The overall prevalence of overweight and obesity estimated for the adolescent population of the Balearic Islands was 17·5 and 10·4 %, respectively (Table 1). Boys had a higher prevalence of overweight and obesity than girls. Excessive weight considered overall (BMI ≥ P85) was 32·6 % in boys and 24·0 % in girls. Prevalence of obesity (BMI ≥ P97) was higher in boys (12·7 %) than girls (8·5 %), and this difference was evident at 12–13 years old. Boys also showed higher waist circumference values and waist:hip ratio than girls, and girls showed higher hip circumference than boys after they were 14 years old. More girls than boys were above the cut-off values established for the waist circumference indicators of cardiovascular risk. Girls showed higher body fat mass than boys.

Table 1 Anthropometric characteristics and prevalence of obesity and overweight of study participants

(Mean values and standard deviations or percentages)

WC, waist circumference; WHR, waist:hip ratio.

* Significant differences between men and women by ANOVA.

Significant differences between men and women by χ2.

The univariate analysis with sociodemographic and lifestyle variables (Table 2) showed that the risk of obesity was associated with parental education level and number of daily meals in both sexes, age in boys, and parental socioeconomic status and current smoking habit in girls. The multivariate analysis showed that the risk of obesity increased in younger boys and in girls who were not active, in boys and girls who usually distracted during mealtime by means of mass media (TV+radio) and in boys who sleep < 7 h daily, despite the lack of statistical significance; an inverse association between parental education level and number of daily meals, and BMI in boys and girls was also found. None of the other lifestyle variables considered in the present study (place of birth, breakfast habit, sweets or salty snacks consumption and alcohol consumption) was significantly associated with obesity in any sex.

Table 2 Sociodemographic and lifestyle characteristics of non-obese (BMI<P97) and obese (BMI≥P97) adolescents

(OR and 95 % CI values)

Ref., reference.

Univariate analysis (logistic regression analysis considering the effect of one explanatory variable): significant differences (* P < 0·05, ** P < 0·01, *** P < 0·001).

Multivariate analysis (multiple logistic regression analysis considering the simultaneous effect of all explanatory variables): significant differences (* P < 0·05, ** P < 0·01, *** P < 0·001).

A positive correlation between sugar-sweetened beverages consumption and watching TV was found (r 0·123, P < 0·001). However, no significant association between sweetened-beverages consumption and body weight status was found.

Although no significant difference in sedentary behaviour was found between obese and non-obese adolescents, high body fat ( ≥ P85) was associated with a less time spent in vigorous-intensity physical activities in both boys (r − 0·174, P < 0·001) and girls (r − 0·102, P < 0·05).

Table 3 shows dietary characteristics of obese and non-obese boys and girls excluding misreporters. No statistical significant difference in prevalence of misreporters was observed between obese and non-obese subjects (data not shown). Obese boys and girls reported a significant lower intake of energy than non-obese adolescents. There were no differences in the dietary origin of the energy between obese and non-obese adolescents.

Table 3 Energy and nutrient intakes and selected food group consumption in non-obese (BMI<P97) and obese (BMI≥P97) adolescents

(Mean values and standard deviations)

EI, energy intake; CHO, carbohydrates.

Crude analysis: significant differences (* P < 0·01, ** P < 0·001) between group means by ANOVA.

Adjusted analysis: significant differences (** P < 0·001) after controlling for age by multiple logistic regression.

§ Boys: n 306 and 36 for BMI < P97 and BMI ≥ P97, respectively.

Girls: n 351 and 34 for BMI < P97 and BMI ≥ P97, respectively.

Discussion

The present results reveal the magnitude of overweight and obesity among 12–17-year-old adolescents in Balearic Islands. The observed percentages of prevalence of overweight (boys 19·9 %; girls 15·5 %) and obesity (boys 12·7 %; girls 8·5 %) among adolescents in Balearic Islands are higher in boys than girls as in most European countries(Reference Carroquino21). In comparison with other data, despite differences in ages studied and the percentile considered as indicator of obesity in adolescents (we considered the 97th percentile), prevalence of overweight and obesity is higher than in Portugal (overweight 16·6 % boys and 12·9 % girls; obese 3·6 % boys and 2·7 % girls), Belgium (overweight 8·6 % boys and 7·4 % girls; obese 1·8 % boys and 1·5 % girls) and Ireland (overweight 17 % boys – exception girls, which has a high prevalence, 21 %; – obese 6 % boys and 7 % girls), whereas the prevalence of obesity is lower and the prevalence of overweight is higher than in the enKid Study of the Spanish population (obese 15·8 % boys and 9·1 % girls; overweight 10·4 % boys and 8·0 % girls), Canary Islands (obese 12·0 % boys and 17·6 % girls; overweight 10·8 % boys and 11·9 % girls), Germany (obese 13·9 % boys and 11·7 % girls; overweight 12·5 % boys and 10·2 % girls) and the United States (obese 16·7 % boys and 15·4 % girls; overweight 14·5 % boys and 15·1 % girls)(Reference Elmadfa and Weichselbaum22Reference Henríquez, Doreste and Laínez26).

The association between sociodemographic and lifestyle factors and BMI in adolescents of the Balearic Islands showed that parental education level is one of the strength factors associated with prevalence of obesity in boys and girls, which agrees data reported by other authors(Reference Fernández27).

Multiple logistic regression analysis also revealed an inverse association between age and obesity in boys. Previous studies pointed out that the trend in the prevalence of overweight (including obesity) decreases with age in adolescents(Reference Lazzeri, Rossi and Pammolli28).

Previous works reported that girls paid more attention to foods than boys as a way to influence health and to meet nutritional recommendations and then to prevent or to reverse the obese state(Reference Sweeting29). We have observed that both obese boys and girls reduced the number of meals per day, skipped the breakfast and avoided sweets and salty snacks consumption, which is more evident in girls, as a method to counteract the obesity.

We observed an inverse association between obesity and the number of daily meals, which is observed in boys and girls. This inverse association between obesity and meal frequency is in agreement with previous studies developed in children(Reference Nicklas, Yang and Baranowski30, Reference Toschke, Küchenhoff and Koletzko31).

We detected a greater risk to be obese in adolescents that do not have breakfast, despite the lack of statistical significance. The omission or consumption of an inadequate breakfast is usual in adolescents(Reference De Rufino, Redondo and Amigo32), although its consumption has been identified as an important factor in the nutritional well being of children and it is rarely compensated for in other meals of the day(Reference Fernández27). Adolescents must be conscience that to skip meals is not an appropriate method to reduce the risk of obesity(Reference Toschke, Küchenhoff and Koletzko31).

We also found that adolescents showed low risk of obesity when they usually consume sweets and salty snacks, which may sound controversial. It may be possible that they underreported their sweets and salty snacks consumption. However, both obese boys and girls consume proportionally less sweets and salty snacks than their lean pairs; it may be a method to self-control their body weight.

A positive association between BMI and distraction during mealtime has been detected, and many adolescents are distracting especially watching the TV (68·4 % of boys; 72·1 % of girls) and listen to the radio (13·0 % of boys; 14·6 % of girls), compared with those that declared to get up from the table (14·0 % of boys; 18·2 % of girls) and to speak during mealtime (57·4 % of boys; 67·5 % of girls). Different studies have detected a positive association between BMI and body fatness and TV among children and adults(Reference Vioque, Torres and Quiles33, Reference Jackson, Djafarian and Stewart34). Despite TV viewing being linked to lower physical activity, Jackson et al. (Reference Jackson, Djafarian and Stewart34) suggested that the relation between TV viewing and fatness is more likely to be due to an effect on food intake. TV viewing during mealtime is inversely associated with consumption of products not advertised, such as fruits and vegetables(Reference Ebbeling, Pawlak and Ludwig1); leads to more eating and encourages the consumption of ‘socially prestigious’, ‘healthy’ and/or ‘tasty’ foods and drinks(Reference Fernández27); and a large proportion of televised food advertisements are for highly processed ‘junk foods’(Reference Fernández27). In agreement with a previous study(Reference Ochoa, Moreno-Aliaga and Martínez-González35), sugar-sweetened beverages' consumption was positively correlated to watching TV, but no significant association between sweetened-beverages' consumption and body weight status was found. Controversial results about sugar-sweetened beverage consumption and body weight can be found in the literature(Reference Ochoa, Moreno-Aliaga and Martínez-González35, Reference Libuda, Alexy and Sichert-Hellert36), which may be explained in part by differences in the design of the study, in definitions of sugar-sweetened beverages and methodological bias in self-reported data. Overall, adolescents must be conscience that watching the TV during mealtime is not a good practice to have a healthy diet, and health promotion plans must be devoted of it.

Short sleep ( ≤ 7 h/d) is associated with obesity compared with longer sleep in boys, whereas no association has been found in girls. These results agree previous studies that suggested a negative association between usual sleep duration and obesity or body weight in children and adults(Reference Marshall, Glozier and Grunstein37, Reference Capuccio, Taggart and Kandala38). Other studies suggest that boys are more susceptible to sleep less than girls(Reference Patel and Hu39). However, recent reviews of the epidemiological evidence have now pointed out that it is still precipitous to suggest that sleep is a cause or solution to the obesity epidemic and further studies are needed(Reference Marshall, Glozier and Grunstein37, Reference Capuccio, Taggart and Kandala38).

Adolescence is a time of rapid physiological, psychological and social development, and also a key period for the adoption of tobacco use. Is there any relationship between obesity and smoking habit in adolescents? Smoker girls showed higher risk of obesity than non-smokers, whereas usually smoking habit was associated with a low risk of obesity in boys. Previous studies showed similar BMI, energetic intake and fat intake among smokers and non-smokers, in spite of 40 % adolescents believed that smoking will decrease or control their weight(Reference Strauss and Mir40). This belief may be supported by the fact that smoke decreases the taste threshold and also girls exhibit greater taste sensitivity than boys, and these differences are of hormonal nature(Reference Gromysz-Kałkowska, Wójcik and Szubartowska41). Consequently, girls apply this method to ‘reduce’ their weight because girls are usually more worried about their image than boys; however, it cannot be discarded that to smoke may also be a facilitating factor of social relationships(Reference Paniagua, García and Castellano42). Adolescents need to be more aware about this unhealthy behaviour, so education programmes related to healthy behaviours are needed in adolescence to prevent future health problems.

We also observed that EI is lower in non-obese than in obese boys and girls, with no differences in the dietary origin of the energy. This paradoxical effect could be attributed to the validity of the instruments to measure food consumption in the population. It is also possible that obese adolescents overestimated the consumption of healthy food items and underestimated the consumption of unhealthy foods, because it has been well documented that people with greater relative weight usually underreported their food intake(Reference Tur, Serra-Majem and Romaguera5).

The 24 h recalls provide information on food intake, and because the data collection occurs after consumption, this method does not affect an individual's food choices on a given day(Reference Tucker43). At least, two non-consecutive administrations are necessary to assess usual intakes, to reduce dependency on intake from the previous day and by household food availability(Reference Tucker43). Accordingly, we applied two 24 h non-consecutive recalls in the present study.

Although 24 h recalls collect data soon after intake, recalls have also limitations related to memory and bias(Reference Tucker43). Difficulties assessing portion sizes could also contribute to underestimation of EI(Reference Waling and Larsson44). Many adolescents have found it difficult to estimate their consumption of some foods, and the consequence of difficulties in assessing high energy-dense foods may be that consumed amounts are underestimated, which, in turn, may have a great effect on the validity of the reported EI(Reference Waling and Larsson44). Underreporting and overreporting are significant contributors to the systematic bias of self-reported dietary assessments, increasing or decreasing estimates of the incidence of inadequate intake and distorting the relationships between nutrient intake and health. Therefore, there are real shortcomings in dietary studies.

To solve these shortcomings, it has been pointed out(Reference Black12) that dietary studies should include an internal validation procedure. Biomarkers of EI have been suggested to play a useful role in dietary assessment, especially for components of foods that are highly variable within different samples of the same food. However, biomarkers, even when available, also have many limitations(Reference Willett45). Advantages and limitations of each must be carefully considered in any specific application, but the mainstay of nutritional epidemiology will remain assessments of dietary intake.

Since 1995, the Goldberg et al. cut-off(Reference Livingstone and Black13) has been used to identify EI misreporters. However, the Goldberg cut-off has been usually misinterpreted. Black(Reference Black12) restates the principles underlying the Goldberg cut-off for identifying underreporters of EI. According to Black contributions(Reference Black12), the Goldberg cut-off can be used to determine the degree of overall bias to reported EI in a dietary assessment. Comparison should be made with a PAL value appropriate to the study population based on information about physical activity or lifestyle. A short questionnaire on home, leisure and occupational activity such that subjects may be assigned to low, medium or high PAL for calculating the Goldberg cut-off should be included as routine. One comparison of a simple questionnaire and a detailed activity diary has suggested that the former might be adequate for large-scale studies(Reference Almendingen, Trygg and Pedersen46). Therefore, we used the Black's modification of Goldberg cut-off to solve the EI misreporting, in spite of that we are conscious that this is just an approximation to the misreporting problem, and that the solution of EI misreporting still needs further research. A late validation study(Reference Waling and Larsson44) of dietary assessment methods using double-labelled water in children and adolescents founded, however, similar percentages of EI misreporters to ours.

Physical activity is a complex multidimensional exposure, which is difficult to measure by self-reported questionnaires in epidemiological studies(Reference Wareham and Rennie47). Self-report of physical activity can lead to overreport it due to a social desirability bias, and therefore the number of inactive individuals may be greater than that reported(Reference Tammelin, Näyhä and Laitinen48, Reference Motl, McAuley and DiStefano49), especially among children and adolescents, and also among obese(Reference Tammelin, Näyhä and Laitinen48). Questionnaires have inherent limitations, mainly because they are subjective in nature. Limitations in the validity of physical activity questionnaires are considered the main reasons for inaccuracies in epidemiological studies.

An extensive range of instruments for measuring physical activity has been reported in the literature, but critical elements in the utility of an instrument to measure physical activity are that it will be relatively inexpensive, cause minimal inconvenience to the participant and be able to be administered with relative ease(Reference Hills50).

Furthermore, Black's modification of Goldberg cut-off and many of the methodological challenges that face nutritional epidemiology are mirrored in the assessment of physical activity. It is well established that a sedentary lifestyle with a low level of energy expended on physical activities is a risk factor of weight gain(Reference Tur, Serra-Majem and Romaguera5). We found that the risk of obesity is higher in physically inactive adolescents than in active adolescents, despite the lack of statistical significance. Balearic Islands' obese and non-obese adolescents showed no different sedentary behaviour, but high body fat was associated with a less time devoted to vigorous-intensity physical activities in both boys and girls. One of the worldwide suggested factors for obesity prevention is to encourage the physical activity practice, which decreases annually about 2·7 % among boys and 7·4 % among girls between the ages of 10 and 17(Reference Sweeting29).

However, the capacity to convert these physiological data to estimates of energy expenditure at low and high intensities of effort requires clarification(Reference Byrne51). The intensity of the physical activity needs to be assessed relative to each individual's physiologic capacity, and referenced to resting metabolism or maximal aerobic capacity. There are still shortcomings in the employment of instruments to measure physical activity. So, questionnaires on detailed activity diary might be adequate for large-scale studies(Reference Almendingen, Trygg and Pedersen46).

Conclusions

To sum up, the prevalence of overweight and obesity in the Balearic Islands' adolescents should take into consideration. The main risk factors associated with prevalence of obesity in adolescents are the age, the parental education level, to skip meals, the distraction during mealtime, a short sleep and the current smoking habit. A programme of action including school healthy nutritional education and promotion programmes targeted at parents and adolescents is needed. These programmes may be mainly focused to increase educational level, to make the adolescents to be aware of to skip meals and to smoke are not useful methods to reduce the risk of obesity, but to do not eat while watching TV, to sleep 8–10 h/d and to be physically active.

Acknowledgements

Sources of funding. Spanish Ministry of Health and Consumption Affairs (Programme of Promotion of Biomedical Research and Health Sciences, Project 05/1276, and Red Predimed-RETIC RD06/0045/1004) and Spanish Ministry of Education and Science (FPU Programme, PhD fellowship to M. M. B.). Authors' contributions. The authors' contributions were as follows: A. P. and J. A. T. conceived, designed and devised the study; M. M. B., E. M., R. L., M. D. J. and J. A. T. collected and supervised the samples. M. M. B. and J. A. T. analysed the data and wrote the manuscript. A. P. and J. A. T. supervised the study. A. P. and J. A.T. obtained funding. Conflict of interests. The authors state that there are no conflicts of interest.

References

1 Ebbeling, CB, Pawlak, DB & Ludwig, DS (2002) Childhood obesity: public-health crisis, common sense cure. Lancet 360, 473482.CrossRefGoogle ScholarPubMed
2 Dehghan, M, Akhtar-Danesh, N & Merchant, AT (2005) Childhood obesity, prevalence and prevention. Nutr J 4, 24.CrossRefGoogle ScholarPubMed
3 Bibbins-Domingo, K, Coxson, P, Pletcher, MJ, et al. (2007) Adolescent overweight and future adult coronary heart disease. N Engl J Med 357, 23712379.CrossRefGoogle ScholarPubMed
4 Neumark-Sztainer, D, Hannan, PJ, Story, M, et al. (2003) Family meal patterns: associations with sociodemographic characteristics and improved dietary intake among adolescents. J Am Diet Assoc 103, 317322.CrossRefGoogle ScholarPubMed
5 Tur, JA, Serra-Majem, Ll, Romaguera, D, et al. (2005) Profile of overweight and obese people in a Mediterranean region. Obes Res 13, 527536.CrossRefGoogle Scholar
6 De Onis, M, Onyango, AW, Borghi, E, et al. (2007) Development of a WHO growth reference for school-aged children an adolescents. Bull WHO 85, 660667.Google ScholarPubMed
7 Cook, S, Weitzman, M, Auinger, P, et al. (2003) Prevalence of Metabolic Syndrome Phenotype in Adolescents Findings from the Third National Health and Nutrition Examination Survey, 1988–1994. Arch Pediatr Adolesc Med 157, 821827.CrossRefGoogle ScholarPubMed
8 Moreiras, O, Carvajal, A, Cabrera, L, et al. (2003) Tablas de composición de alimentos (Food Composition Tables), 7th ed. Madrid: Pirámide.Google Scholar
9 Mataix, J, Mañas, M, Llopis, J, et al. (2004) Tablas de composición de alimentos españoles (Spanish Food Composition Tables), 4th ed. Granada: INTA-Universidad de Granada.Google Scholar
10 Feinberg, M, Favier, JC & Ireland-Ripert, J (1995) Répertoire géneral des aliments (Food Composition Tables). Paris: Tec & Doc Lavoisier.Google Scholar
11 Ripoll, L (1992) Cocina de las Islas Baleares (The Balearic Islands Cookery), 5th ed. Palma de Mallorca: L. Ripoll.Google Scholar
12 Black, AE (2000) Critical evaluation of energy intake using the Goldberg cut-off for energy intake: basal metabolic rate. A practical guide to its calculation, use and limitations. Int J Obes 24, 11191130.CrossRefGoogle Scholar
13 Livingstone, MBE & Black, AE (2003) Biomarkers of nutritional exposure and nutritional status. J Nutr 133, 895S920S.CrossRefGoogle Scholar
14 FAO/WHO/UNU (1985) Energy and Protein Requirements. Report of a Joint. FAO/WHO/UNU Consultation. Technical Report Series 724. Geneva: World Health Organization.Google Scholar
15 Johansson, L, Solvoll, K, Bjørneboe, GA, et al. (1998) Under- and overreporting of energy intake related to weight status and lifestyle in a nationwide sample. Am J Clin Nutr 68, 266274.CrossRefGoogle ScholarPubMed
16 Mendez, MA, Wynter, S, Wilks, R, et al. (2003) Under- and overreporting of energy is related to obesity, lifestyle factors and food group intakes in Jamaican adults. Public Health Nutr 7, 919.CrossRefGoogle Scholar
17 Alvarez, C, Alonso, J, Domingo, A, et al. (1995) La medición de la clase social en ciencias de la salud (Social Class Measurements in Health Sciences). Barcelona: SG-Sociedad Española de Epidemiología.Google Scholar
18 IPAQ – International Physical Activity Questionnaire. Guidelines for Data (2008) Processing and Analysis of the International Physical Activity Questionnaire (IPAQ) available at http://www.ipaq.ki.es (accessed October 2008).Google Scholar
19 Hagströmer, M, Bergman, P, De Bourdeaudhuij, I, et al. (2008) Concurrent validity of a modified version of the International Physical Activity Questionnaire (IPAQ-A) in European adolescents: The HELENA Study. Int J Obes 32, 542548.CrossRefGoogle ScholarPubMed
20 Food and Nutrition Board, Institute of Medicine of the National Academies (2005) Dietary Reference Intakes. Chapter 12, Physical Activity, pp. 880935. Washington, DC: The National Academies Press.Google Scholar
21 Carroquino, MJ (2007) Prevalence of Excess Body Weight and Obesity in Children and Adolescents, Fact Sheet 2.3. Copenhagen: WHO Europe.Google Scholar
22 Elmadfa, I & Weichselbaum, E (2005) European nutrition and health report 2004. Forum Nutr 58, 4761.CrossRefGoogle Scholar
23 Whelton, H, Harrington, J, Crowley, E, et al. (2007) Prevalence of overweight and obesity on the island of Ireland: results from the north south survey of children's height, weight and body mass index, 2002. BMC Pub Health 7, 187195.CrossRefGoogle ScholarPubMed
24 Hedley, AA, Ogden, C, Johnson, CL, et al. (2004) Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA 291, 28472850.CrossRefGoogle ScholarPubMed
25 Serra-Majem, L, Ribas, L, Aranceta, J, et al. (2003) Childhood and adolescent obesity in Spain. Results of the enKid Study (1998–2000). Med Clin (Barc) 121, 725732.Google ScholarPubMed
26 Henríquez, P, Doreste, J, Laínez, P, et al. (2008) Prevalence of obesity and overweight in adolescents from Canary Islands, Spain. Relationship with breakfast and physical activity. Med Clin (Barc) 130, 606610.Google Scholar
27 Fernández, PM (2006) Dietary habits and nutritional status of school aged children in Spain. Nutr Hosp 21, 374378.Google Scholar
28 Lazzeri, G, Rossi, S, Pammolli, A, et al. (2008) Underweight and overweight among children and adolescents in Tuscany (Italy). Prevalence and short-term trends. J Prev Med Hyg 49, 1321.Google ScholarPubMed
29 Sweeting, HN (2008) Gendered dimensions of obesity in childhood and adolescence. Nutr J 7, 114.CrossRefGoogle ScholarPubMed
30 Nicklas, TA, Yang, SJ, Baranowski, T, et al. (2003) Eating patterns and obesity in children: The Bogalusa Heart Study. Am J Prev Med 25, 916.CrossRefGoogle ScholarPubMed
31 Toschke, AM, Küchenhoff, H, Koletzko, B, et al. (2005) Meal frequency and childhood obesity. Obes Res 13, 19321938.CrossRefGoogle ScholarPubMed
32 De Rufino, P, Redondo, C, Amigo, T, et al. (2005) Desayuno y almuerzo de los adolescentes escolarizados de Santander (Breakfast and snack of schooled adolescents in Santander). Nutr Hosp 20, 217222.Google Scholar
33 Vioque, J, Torres, A & Quiles, J (2000) Time spent watching television, sleep duration and obesity in adults living in Valencia, Spain. Int J Obes. Relat Metab Disord 24, 16831688.CrossRefGoogle ScholarPubMed
34 Jackson, DM, Djafarian, K, Stewart, J, et al. (2009) Increased television viewing is associated with elevated body fatness but not with lower total energy expenditure in children. Am J Clin Nutr 89, 10311036.CrossRefGoogle Scholar
35 Ochoa, MC, Moreno-Aliaga, MJ, Martínez-González, MA, et al. (2007) Predictors factors for childhood obesity in a Spanish case–control study. Nutrition 23, 379384.CrossRefGoogle Scholar
36 Libuda, L, Alexy, U, Sichert-Hellert, W, et al. (2008) Pattern of beverage consumption and long-term association with body-weight status in German adolescents – results from the DONALD Study. Br J Nutr 99, 13701379.CrossRefGoogle ScholarPubMed
37 Marshall, NS, Glozier, N & Grunstein, RR (2008) Is sleep duration related to obesity? A critical review of the epidemiological evidence. Sleep Med Rev 12, 289298.CrossRefGoogle ScholarPubMed
38 Capuccio, FP, Taggart, FM, Kandala, NB, et al. (2008) Meta-analysis of short sleep duration and obesity in children and adults. Sleep 31, 619626.CrossRefGoogle Scholar
39 Patel, SR & Hu, FB (2008) Short sleep duration and weight gain: a systematic review. Obesity 16, 643653.CrossRefGoogle ScholarPubMed
40 Strauss, RS & Mir, HM (2001) Smoking and weight loss attempts in overweight and normal-weight adolescents. Int J Obes 25, 13811385.CrossRefGoogle ScholarPubMed
41 Gromysz-Kałkowska, K, Wójcik, K, Szubartowska, E, et al. (2002) Taste perception of cigarette smokers. Ann Univ Mariae Curie-Sklodowska 57, 143154.Google ScholarPubMed
42 Paniagua, H, García, S, Castellano, G, et al. (2001) Tobacco, alcohol and illegal drug consumption among adolescents, relationship with lifestyle and environment. An Esp Pediatr 55, 121128.Google Scholar
43 Tucker, KL (2007) Assessment of usual dietary intake in population studies of gene–diet interaction. Nutr Met Card Dis 17, 7481.CrossRefGoogle ScholarPubMed
44 Waling, MU & Larsson, CL (2009) Energy intake of Swedish overweight and obese children is underestimated using a diet history interview. J Nutr 139, 522527.CrossRefGoogle ScholarPubMed
45 Willett, W (2006) Does the future of nutritional epidemiology depend on biomarkers? Sixth International Conference on Diet and Activity Methods Abstracts Book, PL03-02, Copenhagen.Google Scholar
46 Almendingen, K, Trygg, K & Pedersen, JI (1998) An assessment of the use of simple methods to predict individual energy intakes for intervention studies. Eur J Clin Nutr 52, 5459.CrossRefGoogle ScholarPubMed
47 Wareham, NJ & Rennie, KL (1998) The assessment of physical activity in individuals and populations: why try to be more precise about how physical activity is assessed? Int J Obes 22, S30S38.Google ScholarPubMed
48 Tammelin, T, Näyhä, S, Laitinen, J, et al. (2003) Physical activity and social status in adolescence as predictors of physical inactivity in adulthood. Prev Med 37, 375381.CrossRefGoogle ScholarPubMed
49 Motl, RW, McAuley, E & DiStefano, C (2005) Is social desirability associated with self-reported physical activity? Prev Med 40, 735739.CrossRefGoogle ScholarPubMed
50 Hills, AP (2006) Objective instruments for measuring physical activity – an overview. Sixth International Conference on Diet and Activity Methods Abstracts Book, SY05-01, Copenhagen.Google Scholar
51 Byrne, NM (2006) Objective instruments for measuring physical activity – one size does not fit all. Sixth International Conference on Diet and Activity Methods Abstracts Book, SY05-02, Copenhagen.Google Scholar
Figure 0

Table 1 Anthropometric characteristics and prevalence of obesity and overweight of study participants(Mean values and standard deviations or percentages)

Figure 1

Table 2 Sociodemographic and lifestyle characteristics of non-obese (BMI(OR and 95 % CI values)

Figure 2

Table 3 Energy and nutrient intakes and selected food group consumption in non-obese (BMI(Mean values and standard deviations)