Food and nutrient intake

The present dietary data on European children and adolescents are flawed with serious methodological problemsReference Lambert, Agostoni, Elmadfa, Hulshof, Krause and Livingstone¹³. For instance, various collection methods are used and the ages or age cut-off points of the children surveyed are inconsistent. Therefore, there is a need for harmonisation and standardisation of methods to be used in nutrition surveys in children and adolescents in EuropeReference Lambert, Agostoni, Elmadfa, Hulshof, Krause and Livingstone¹³.

In addition, for dietary evaluations each country uses a different set of food composition data which differ in definitions, analytical methods, units and coding. This makes comparisons between countries difficult and inaccurate.

It is well known that adolescents have particular food choices and meal habits compared with younger children and adultsReference Hoglund, Samuelson and Mark^14, Reference Alexy, Sichert-Hellert and Kersting¹⁵. Furthermore, a number of validation studies have shown that, in addition to the general problem of misreporting in dietary surveys in children and adolescents, underreporting of energy intake is found more often in adolescents, especially in girls, than in children.

A limited number of dietary assessment instruments that are specifically designed for adolescents have been found to be valid and reproducible. Thus there is a demand for a short, easily administered, inexpensive and accurate instrument that can be used in a broad range of adolescent subpopulationsReference Rockett, Berkey and Colditz^16, Reference Moreno, Kersting, de Henauw, González-Gross, Sichert-Hellert and Matthys¹⁷ and that should take advantage of modern computer-based dietary assessment toolsReference Bakker, Twisk, van Mechelen, Mensink and Kemper¹⁸. For nutritional epidemiological studies in Europe, the EFCOSUM (‘European Food Consumption Survey Method’) project has proposed to use repeated 24-hour recalls from the age of 10 years onwardsReference Biro, Hulshof, Ovesen and Amorin Cruz¹⁹.

In summary, there are insufficient data to draw any conclusions about the nutritional quality of the diets of European children and adolescents. The HELENA Study will provide harmonised and comparable data on food intake among male and female European adolescents, taking advantage of a computer-based and attractive dietary assessment tool for 24-hour recalls, with a central standardised protocol. For this purpose, an existing tool called YANA-C (Young Adolescents’ Nutrition Assessment on Computer)Reference Vereecken, Covents, Matthys and Maes²⁰ has been adapted to local conditions (food lists, portion sizes, etc.) for all participating countries and will be linked to local food composition databases through collaboration with another ongoing Sixth Framework Programme project called EUROFIR (‘European Food Information Resource Network’; www.eurofir.net), using a standardised protocol.

Cognitive determinants of eating habits

Despite the intuitive appeal of education as a means of improving diet, many studies in this area have failed to find significant associations between nutritional knowledge and dietary behaviourReference Parmenter and Wardle²¹. Since the way in which nutrition knowledge transforms into dietary behaviour and nutrient intake may vary among populations, it appears important to assess whether nutrition knowledge is associated with particular food choices and nutrient intakes before any nutrition intervention is initiated in a given population.

Studies in adults suggest that nutrition knowledge influences dietary behaviourReference Wardle, Parmenter and Waller^22, Reference Dallongeville, Marécaux, Cottel, Bingham and Amouyel²³. Studies in children and adolescents are scarce. Increasing age and type of school were found to correlate significantly with nutritional knowledge but not with degree of overweightReference Reinehr, Kersting, Chahda and Andler²⁴.

Previous studies revealed that large proportions of populations have misconceptions about personal dietary intake levelsReference Brug, Hospers and Kok^25, Reference Brug, Van Assema, Kok, Lenderink and Glanz²⁶ and may misunderstand general dietary information, e.g. to decrease fat intake. Therefore, personal feedback on dietary intake, such as delivered through computer-tailoring, together with tailored information about food choices, has considerable potential to communicate a personal need to changeReference De Bourdeaudhuij and Brug²⁷.

The HELENA Study will provide data about nutrition knowledge, attitudes towards nutrition and the main determinants of food choice and preference, among male and female European adolescents.

Physical activity, exercise and physical fitness

The detailed relationship between physical activity, fitness, fatness and health in adolescence remains to be clarified. Cross-sectional studies have documented the relationship between physical activity, physical fitness and health, and a number of cardiovascular risk factors already during childhood and adolescenceReference Ekelund, Poortvliet, Nilsson, Yngve, Holmberg and Sjostrom^28–Reference Wedderkopp, Froberg, Hansen, Riddoch and Andersen³⁰. Similarly, longitudinal studies have shown that the degree of physical fitness during childhood and adolescence may determine one’s physical fitness as an adult. In addition, poor physical fitness during these stages of life seems to be associated with later cardiovascular risk factors such as hyperlipidaemia, hypertension and obesityReference Boreham, Twisk, Murray, Savage, Strain and Cran^31–Reference Janz, Dawson and Mahoney³⁴. Recent observations suggest that preventive efforts focusing on maintaining physical fitness (especially muscular and cardiorespiratory fitness) and physical activity through puberty may have favourable health benefits in later yearsReference Janz, Dawson and Mahoney³⁵. The HELENA project has the ambition and potential to explore these relationships further.

Physical activity is a behaviour that is, because of its complex nature, difficult to assess under free-living conditions. No single method is available to quantify all dimensions of the activity (total amount, intensity, frequency, duration, etc.). Lack of comparable data, especially regarding young individuals, hampers the understanding of the complex relationship between these characteristics of physical activity, physical fitness and health outcomes.

Physical fitness encompasses all the physical qualities of a person, among young as well as adult individuals. The state of physical fitness can be considered an integrated part of all of the functions and structures involved in the performance of physical exertionReference Castillo, Ortega and Ruiz³⁶. Health-related physical fitness involves cardiorespiratory fitness, muscle strength, speed and agility, coordination, flexibility and body composition.

Physical activity can be measured using subjective methods, such as questionnaires, or objective methods, such as motion sensors or heart-rate monitorsReference Sjöström, Ekelund, Yngve, Gibney, Margetts, Kearney and Arab³⁷. Motion sensors can be either simple pedometers or more advanced accelerometers. Accelerometers measure the acceleration of the body movement in one or more directions and can quantify physical activity data in terms of time and intensityReference Esliger, Copeland, Barnes and Tremblay^38, Reference Trost, McIver and Pate³⁹. Accelerometers can store physical activity data for several weeks and have been found to be a valid and reliable measure of intensity, duration and frequency of the physical activity, as well as a measure of total physical activity, in adolescentsReference Esliger, Copeland, Barnes and Tremblay³⁸.

In the HELENA Study, levels and patterns of physical activity are measured using a uniaxial accelerometer. The accelerometer is worn for one week and summarises the physical activity data each time period of 15 s or more, for every waking hour except during water-based activities.

One advantage – especially in an international setting like that of the HELENA Study – is that an accelerometer does not know any geographical, linguistic or cultural boundaries. However, accelerometers are costly and advanced devices, and even if they provide data of good quality they are not a possible alternative for many research groups and organisations.

The International Physical Activity Questionnaire (IPAQ) was originally developed as an instrument for cross-national monitoring of physical activity and inactivity in adults. The IPAQ instrument has been tested against accelerometers and has been found to have acceptable measurement properties, at least as good as other established self-reportsReference Craig, Marshall, Sjöström, Bauman, Booth and Ainsworth⁴⁰. The HELENA Study will develop and test a questionnaire for use among adolescents, based on the long format of the IPAQ, which will provide internationally comparable data.

The health-related physical fitness components will be assessed by means of the following physical fitness tests (the rationale for the selection of all these tests has been previously publishedReference Ruiz, Ortega, Gutierrez, Meusel, Sjöström and Castillo⁴¹):

1. 1. Back-saver sit-and-reach test, to assess flexibility.

2. 2. Handgrip test, to assess maximum handgrip strength.

3. 3. Standing broad jump test, to assess lower-limb explosive strength.

4. 4. Jump tests according to the Bosco protocol (three different jumps) – the squat jump, to assess lower-limb explosive strength; the counter movement jump, to assess lower-limb explosive strength and elastic component assessment; and the Abalakov jump, to assess lower-limb explosive strength, elastic component and inter-muscular coordination capacity assessment.

5. 5. Bent arm hang test, to assess upper-limb endurance strength.

6. 6. Shuttle run test, 4 × 10 m, to assess speed of movement, agility and coordination.

7. 7. Shuttle run test, 20 m, to assess cardiorespiratory fitness.

The HELENA Study will provide harmonised and comparable data about physical activity level and patterns, physical fitness and health outcomes among male and female European adolescents, using an objective measure of physical activity and a standardised valid set of fitness measures. In this regard, validation and reliability studies for the physical activity and physical fitness tools used in the HELENA Study are being currently developed. In fact, mathematical equations have already been reported that may improve the reliability and accuracy of the results, and may guide clinicians and researchers in selecting the optimal grip span on the hand dynamometer when measuring handgrip strength in adolescentsReference Ruiz, Espana-Romero, Ortega, Sjostrom, Castillo and Gutierrez⁴².

Body composition, obesity and related risk factors in adolescents

Rates of cardiovascular diseases and diabetes have been found to increase in both men and women who were obese during adolescenceReference Dietz⁴³. Approximately 50% of obese adolescents with a body mass index at or above the 95th percentile become obese adultsReference Dietz⁴³. The prevalence of obesity in US children and adolescents has increased dramatically in the last decadesReference Ogden, Flegal, Carroll and Johnson⁴⁴. In Europe, there are few representative data about obesity prevalence in adolescents and the existing ones are not comparable, because different definitions for obesity were used. However, available results point out that there is also a dramatic increase in the prevalence of obesity in European adolescents, with differences according to gender and socio-economic statusReference Moreno, Sarría, Fleta, Rodríguez, Pérez-González and Bueno⁴⁵.

Recently, Lobstein and FrelutReference Lobstein and Frelut⁴⁶ reported estimates of the prevalence of overweight in children and adolescents in various European countries based on 20 surveys with the data recalculated where necessary to conform to international definitions. They detected two apparent trends. The first was the generally lower levels of overweight in the countries of Central and Eastern Europe whose economies suffered varying degrees of recession during the period of economic and political transition in the 1990s. The second trend was for the prevalence of overweight to be higher among the southern countries of Europe, especially those outside the former Eastern Bloc. These data were also included in a very recent report from the International Association of the Study of Obesity’s International Obesity Task ForceReference Lobstein and Baur⁴⁷.

In cross-sectional, nationally representative, school-based surveys in 1997–1998 that used identical data collection methods, Lissau et al.Reference Lissau, Overpeck, Ruan, Due and Holstein⁴⁸ assessed overweight prevalences in 13- and 15-year-old adolescents in 13 European countries, Israel and the USA. The highest prevalence was found in the USA and the lowest in Lithuania. The highest prevalences in Europe were found in Ireland, Greece and Portugal.

Very recently, it was observed in a representative sample of Spanish adolescents (AVENA Study) that the rate of change in overweight and obesity prevalences seems to have increased in recent years; from 0.88 (1985 to 1995) to 2.33% per year (1995 to 2000–2002) in males and from 0.5 (1985 to 1995) to 1.83% per year (1995 to 2000–2002) in femalesReference Moreno, Mesana, Fleta, Ruiz, González-Gross and Sarría⁴⁹. These findings confirm the idea of a real epidemic situation for paediatric obesity in EuropeReference Jackson-Leach and Lobstein⁵⁰.

The Working Group Report of the Second World Congress of Paediatric Gastroenterology, Hepatology, and Nutrition has recommended that large-scale prospective population studies should incorporate careful anthropometric measurements and regular monitoring of selected obesity-associated complications using some or all of the following tests: fasting blood tests (glucose, insulin, lipids, liver function tests), blood pressure measurement, glucose tolerance tests, liver ultrasound, measurements of psychosocial functioning, lung function tests, and biomechanical or podiatric assessmentReference Fisberg, Baur, Chen, Hoppin, Koletzko and Lau⁵¹.

Atherosclerosis starts in childhood and adolescence, but clinical manifestations can appear 30 to 50 years later. This is the reason why it is so important to identify risk factors as early as possible. Risk factors for several of the major chronic diseases, such as cardiovascular diseases, hypertension, diabetes, obesity and cancer, are often observed during childhood. Metabolic syndrome seems to be present already during adolescence, especially if there is a predisposing genetic background. Prevalence of multi-metabolic syndrome among obese children and adolescents in Hungary was 7.7% for boys and 9.1% for girls. Only 14.4% of the obese children and adolescents were free from any cardiovascular risk factorsReference Csábi, Török, Jeges and Molnár⁵².

Adolescents at risk for this condition could easily be identifiedReference Moreno, Pineda, Rodríguez, Fleta, Sarría and Bueno⁵³. Unfortunately no consensus concerning the definition of the metabolic syndrome in childhood has been establishedReference Molnár⁵⁴. Therefore the results of those few studies investigating the prevalence of the metabolic syndrome in childhood as a primary goal are not comparable.

A growing body of evidence implicates adipose tissue as a key regulator of inflammation, and chronic, subclinical inflammation implicated in the pathophysiological mechanisms of atherosclerotic disease has already been observed in adolescenceReference Wärnberg, Moreno and Mesana⁵⁵. The inflammatory biomarkers selected for this study are sensitive enough to investigate the health of European adolescents and to establish determinants for chronic inflammation caused by unhealthy body weight or nutritional and physical activity habits.

The HELENA Study will provide comparable data on body composition, blood lipids, fasting glucose, insulin, adiponectin, leptin and blood pressure, in order to establish the prevalence of obesity and analyse the clustering of cardiovascular risk factors (metabolic syndrome) among male and female European adolescents.

Micronutrient status in adolescents

In developed countries, nutritional studies dealing with vitamin and mineral status are facing new challenges. In spite of the huge availability and variety of foods and the possibility of good nutrition, malnutrition is affecting several population subgroups for selected nutrients, with adolescents being one of these groups at riskReference Casas, González-Gross, Marcos and Tojo⁵⁶. Similar to the situation mentioned in other parts of this review, representative data are lacking for most micronutrients in Europe. In addition, the lack of reference values for adolescents for most of the blood parameters makes the analysis of available data quite difficult, as adult reference values have to be used. Several review articles have tried to establish micronutrient status in European adolescentsReference Lambert, Agostoni, Elmadfa, Hulshof, Krause and Livingstone^13, Reference Al-Tahan, González-Gross and Pietrzik⁵⁷. With the necessary caution that arises when analysing data from different studies performed with different methodologies, there seems to be an agreement that folateReference Al-Tahan, González-Gross and Pietrzik^57–Reference McNulty, Eaton-Evans, Cran, Woulahan, Boreham and Savage⁶⁰, calciumReference Hoglund, Samuelson and Mark^14, Reference Alexy, Sichert-Hellert and Kersting^15, Reference Casas, González-Gross, Marcos and Tojo⁵⁶, vitamin DReference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala and Viikari^61, Reference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala, Leino and Viikari⁶² and ironReference Samuelson, Bratteby, Berggren, Elverby and Kempe⁶³ are nutrients at risk in the adolescent population in most of the countries.

Another aspect that needs more in-depth study is the possible differences in vitamin and mineral status by gender. As Bergstrom et al.Reference Bergstrom, Hernell, Lonnerdal and Persson⁶⁴ established in Swedish adolescents, the differences in iron status between boys and girls result primarily from biological differences other than menstrual bleeding or insufficient iron intake. In the UK, adolescent girls showed the highest prevalence of low iron intake and poor iron status, with the latter independently associated with non-Caucasian ethnicity and vegetarianismReference Thane, Bates and Prentice⁶⁵. Several studies show serum cobalamin values to be higher in girls than in boysReference Al-Tahan, González-Gross and Pietrzik^{57, 66}. The concentration of total homocysteine in serum and plasma is elevated in both folate and cobalamin deficiencies. High homocysteine levels are associated with an increased risk of several chronic diseases in adulthood, such as cardiovascular diseases and dementia, and are higher in adolescent males than in females^66, Reference Borke and Ueland⁶⁷. The risk of high homocysteine levels in adolescence must be established, although available data seem to indicate that they correlate with poorer cardiovascular fitnessReference Ruiz, Sola, González-Gross, Ortega, Vicente-Rodríguez and García-Fuentes⁶⁸ and increased risk for future chronic diseases.

Dietary intake of nutrients does not always correlate with serum values. Serum ferritin and serum transferrin receptor concentrations, growth and food habits were studied in healthy Swedish boys and girls at the age of 17 years and compared with those in the same adolescents at age 15. The results indicate insufficient iron stores in the 17-year-old girls in relation to erythropoiesis and iron needs, but more favourable iron stores in the boys. The absence of a significant decrease in mean serum ferritin despite rapid growth suggests that the earlier iron fortification of flour only marginally contributed to the iron status of Swedish adolescents of this age groupReference Samuelson, Lonnerdal, Kempe, Elverby and Bratteby⁶⁹. It has been proposed that risk of poor iron status could be reduced by consuming (particularly lean red) meat or enhancers of non-haem iron absorption (e.g. fruit or fruit juice) in vegetariansReference Thane, Bates and Prentice⁶⁵. Although research suggests that adolescents, particularly girls, may avoid dairy products due to concerns that these foods are ‘fattening’, avoidance of dairy foods due to a possible association with relative body weight is not supported by the available resultsReference Phillips, Bandini, Cyr, Colclough-Douglas, Naumova and Must⁷⁰. In Finnish peripubertal girls, vitamin D supplementation daily with 20 μg is needed to prevent hypovitaminosis D in winter. Sunlight exposure in summer is more effective than approximately 20 mg of vitamin D₂ supplementation daily in winter to raise serum 25-hydroxyvitamin D levels. Both the daily supplementation with 20 mg of vitamin D₂ and summertime sunlight exposure had more effect on those who had severe hypovitaminosis D than in those who had a normal vitamin D statusReference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala and Viikari^61, Reference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala, Leino and Viikari⁶². Pubertal girls with hypovitaminosis D seem to be at risk of not reaching maximum peak bone mass, particularly at the lumbar spine. Dietary enrichment or supplementation with vitamin D should be considered to ensure an adequate vitamin D statusReference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala and Viikari^61, Reference Lehtonen-Veromaa, Mottonen, Nuotio, Irjala, Leino and Viikari⁶². A positive association between dietary calcium/phosphorus ratio and markers of collagen formation (N-terminal propeptide of type 1 collagen) has been observed. This relationship can be attributable to a higher calcium intake per se, a critical balance between calcium and phosphorus intake, or high dairy product consumption. A higher incidence of vitamin D insufficiency in older adolescents may reflect a more sedentary lifestyle or increased utilisation of 25-hydroxyvitamin DReference Ginty, Cavadini, Michaud, Burckhardt, Baumgartner and Mishra⁷¹.

The voluntary addition of micronutrients to the appropriate foods may help address the risks associated with low micronutrient intakes. Folic acid fortification of food has shown to be responsible for about 25% of folate intake in the German adolescent populationReference González-Gross, Prinz-Langenohl and Pietrzik⁵⁹. Folic acid fortification could be the explanation for the differences observed between folate intake and blood values. Folate intake from fortified food or from supplements is not taken into account in most of the studies, which is a variable that can lead to confusion. Nutrition surveys should include fortified commercial foods in their composition tables, in order to make folate intake more accurate. However, concerns need to be addressed regarding the potential for unacceptably high intakes, particularly for those people consuming very large amounts of foodReference Flynn, Moreiras, Stehle, Fletcher, Muller and Rolland⁷², as it is in adolescents.

It is accepted that food and nutrient intake data are not enough to establish nutritional status and must be complemented with biochemical data. For micronutrient status, it is necessary to establish reference values obtained in large population samples analysed under the same methodological conditions. This will permit a correct interpretation of results in further nutritional studies. As there have not been any nutritional studies on adolescents performed at European level (as proposed in HELENA), these reference values and cut-off points for most parameters, and specifically for vitamins, are missing.

Immunological characteristics related to nutritional status

The study of nutritional immunology is a relatively new discipline; however, it is widely accepted that normal functioning of the immune system is crucial for health and that diet is one of the major exogenous factors modulating individual immunocompetence. Malnutrition causes significant alteration in immune response but even subclinical deficits may be associated with an impaired immune response, which makes immune parameters sensitive biomarkers of nutritional statusReference Marcos, Nova and Montero^73, Reference Nova, Samartin, Gomez, Morande and Marcos⁷⁴.

A growing body of evidence implicates inflammatory markers in pathophysiological mechanisms of chronic diseases such as atherosclerotic diseases, diabetes, obesity, cancer, allergies, arthritis and rheumatic diseases. These associations have mainly been studied in adults or diseased subjects, but moderately increased levels of novel risk markers, such as the ones that will be examined in the HELENA Study, have also been found during past years in young populations with no apparent ongoing chronic disease.

Confirming several studies, including a larger American study in childrenReference Ford⁷⁵, the AVENA Study showed that overweight is associated with higher production of inflammatory proteins, characterising this otherwise apparently healthy young overweight population by a state of chronic low-grade inflammationReference Wärnberg, Moreno and Mesana^55, Reference Warnberg, Nova, Moreno, Romeo, Mesana and Ruiz⁷⁶. There is also increasing evidence to show that chronic subclinical inflammation is associated with metabolic dysfunction, which links causally to insulin resistance and the metabolic syndrome in adolescentsReference De Ferranti, Gauvreau, Ludwig, Newburger and Rifai⁷⁷. Dietary factors such as fatty acids, antioxidants or fibre can potentially modulate the association between adiposity and subclinical inflammationReference Aeberli, Molinari, Spinas, Lehmann, l’Allemand and Zimmermann^78, Reference King, Egan and Geesey⁷⁹, as can physical activity and fitnessReference Wärnberg⁸⁰, but more studies of the determinants of subclinical inflammation in this age group are needed to clarify how a healthier lifestyle affects these novel risk markers.

The interpretation of immunological markers depends critically on knowledge of the behaviour of normal adolescent immune function, as well as the understanding of determinants of variability. While there are substantial data on immune measures and functioning among the adult and newborn population and more recently in the elderly, there has been little research on immunonutrition in young populations. The HELENA Study will provide nutrition scientists with reference values of immunological parameters for future research on the nutritional status of European adolescents. As all laboratory assessments will be centralised, the HELENA Study will provide unique comparable immunological data related to the nutritional status in European adolescents.

Nucleotide polymorphisms and phenotype heterogeneity

The molecular pathogenesis of paediatric obesity remains unknown for the vast majority of overweight children and adolescents. More than 100 candidate genes or regions are potentially implicatedReference Snyder, Walts, Perusse, Chagnon, Weisnagel, Rankinen and Bouchard⁸¹. Single formal genetic studies suggest a higher heritability of body weight in adolescence and genes that influence body weight in adulthood might not be the same as those that are relevant in childhood and adolescence. Multiple measurements of body mass index from childhood to adulthood in white siblings showed considerably greater heritability than a cross-sectional measurement. Significant and suggestive linkage with long-term burden and trend of body mass index was observed on chromosomes 1, 5, 7, 12, 13 and 18. Candidate genes involved in the pathophysiology of obesity are identified in all regions on six chromosomes except for 7q11.1Reference Chen, Li, Cook, Rosner, Srinivasan and Boerwinkle⁸².

Some individuals appear to be relatively sensitive to dietary or lifestyle intervention, whereas others are quite insensitive. There is strong evidence that variability to changes in environmental factors is partly determined by genetic factors. This can be relevant for obesity at least at three different levels: they could be involved in determining the susceptibility to gain fat in response to environmental risk factors (high-fat diet, low physical activity, etc.); they may influence the response of the phenotype to interventions; and they can be involved in the susceptibility of obese individuals to develop co-morbidities associated with obesity.

There are more than 430 chromosomic regions with gene variants involved in body weight regulation and obesity development. Polymorphisms in genes related to energy expenditure (such as uncoupling proteins), adipogenesis and insulin resistance (such as hormone-sensitive lipase, peroxisome proliferator-activated receptor-γ, β-adrenergic receptors 2 and 3 and tumour necrosis factor-α) and food intake (such as ghrelin) appear to be associated with obesity phenotypes.

The genotype–environment interactions could also be involved in the susceptibility of obese individuals to develop co-morbidites associated with obesity (diabetes, hyperlipidaemia, hypertension and coronary heart disease). Definition of these interaction effects for phenotypes related to obesity is therefore important because it will eventually allow the identification of individuals at risk for the development of complications and the identification of those likely to be resistant to dietary interventions. The study of these genetic markers in adolescents and their relationship to several phenotypic characteristics of the population will permit a better understanding of the pathogenic mechanisms that are involved in non-communicable diseases, specifically cardiovascular diseases.

There are also several genetic factors which determine blood lipid profile, specifically several apoproteins. These apoproteins, which are synthesised by polymorphic genes, present several isoforms which are relatively frequent among the population and can influence blood lipid profile by interacting with specific exogenous factors like dietary habits.

The HELENA Study will provide data on the relationship between genetic markers and phenotypic characteristics among male and female European adolescents. Two major lines of investigation will be developed. The first will test whether single-nucleotide polymorphisms, located in genes related to energy expenditure, adipogenesis, insulin resistance and food intake, interact with indicators of physical activity to explain part of body mass composition variability in adolescents. The second will look for possible interactions between known single-nucleotide polymorphisms in genes involved in plasma lipid and glucose homeostasis, and blood pressure regulation and body composition, in the regulation of major cardiovascular risk factors and components of the metabolic syndrome in adolescents.

Food preferences and development of new healthy foods

There are many concerns surrounding adolescent food choice behaviour, including low intake of fruits and vegetables, and high intake of foods that are high in fat, sugar and saltReference Alexy, Sichert-Hellert and Kersting^15, Reference Croll, Neumark-Sztainer and Story^83, Reference Shepherd and Dennison⁸⁴. Reasons for these preferences and consumption patterns can range from innate food preferences and familiarityReference Popper and Kroll^85, Reference Birch and Fisher⁸⁶ to social and environmental influencesReference Story, Neumark-Sztainer and French⁸⁷.

Although many adolescents demonstrate awareness and knowledge of nutrition and healthy eating, it does appear they find it difficult putting this theory into practiceReference Shepherd and Dennison^84, Reference Birch and Fisher^86, Reference Brown, McIlveen and Strugnell^88, Reference Warwick, McIlveen and Strugnell⁸⁹.

A lack of understanding about how to communicate dietary messages effectively is hindering the innovation of products that can contribute to consumer health, well-being and enhanced industrial competitivenessReference Carbone, Campbell and Honess-Morreale⁹⁰. Small- and medium-sized enterprises play an important role in producing the great diversity of foods in Europe and the retail sector increasingly contributes to strengthening the links between production, processing and the consumer. Therefore, there is a need to better understand consumer requirements and preferences, and to provide a healthy, safe and high-quality food supply – in this case, specifically for adolescents.

There are a multitude of factors that can influence adolescent food choices and preferences. In general terms these may include availability, convenience, cost, influence of peers, parents, hunger and health concernsReference Story, Neumark-Sztainer and French^87–Reference Warwick, McIlveen and Strugnell⁸⁹. However, one of the most important determinants of food choice is the adolescents’ food preferences. Taste, in particular, plays an important role in food choice. Generally speaking, adolescents won’t eat what they don’t likeReference Shepherd and Dennison^84, Reference Birch and Fisher^86–Reference Brown, McIlveen and Strugnell⁸⁸.

The HELENA Study will seek to further understand the impact of these and other factors on adolescent food choices and preferences, and utilise this information to guide the development of healthy new food products. In particular, given that adolescents are highly driven by taste, close attention needs to be given to the sensory aspects of the newly developed food products.

In addition to understanding the opinions and views of the adolescents themselves, a multidisciplinary research effort bringing together a wide range of expertise is essential to address the whole problem surrounding nutrition and lifestyle of adolescents in Europe. The HELENA consortium will also develop new healthy foods, attractive for male and female European adolescents. One of these foods will be a low-glycaemic, high-fibre biscuit.

The glycaemic index has proved to be a useful nutritional concept, providing new insights into the relationship between foods and chronic diseaseReference Jimenez-Cruz, Gutierrez-Gonzalez and Bacardi-Gascon⁹¹. Observational studies suggest that diets with a high glycaemic load are independently associated with increased risk of type 2 diabetes and cardiovascular diseasesReference Brand-Miller⁹². Some evidence suggests that a low-glycaemic-index diet may also protect against obesity, colon cancer and breast cancer. In adolescents, the effect of an ad libitum, reduced-glycaemic-load diet has been compared with that of an energy-restricted, reduced-fat diet; at 12 months, body mass index and fat mass decreased more in the low-glycaemic-index diet compared with the reduced-fat diet. In post hoc analyses, the glycaemic load was a significant predictor of treatment response among both groups, whereas dietary fat was notReference Ebbeling, Leidig, Sinclair, Hangen and Ludwig⁹³.

Low-glycaemic-index diets influence body weight and resting energy expenditure independently of caloric intake. In a short cross-over study, Agus et al.Reference Agus, Swain, Larson, Eckert and Ludwig⁹⁴ compared a high-glycaemic-index, energy-restricted diet with an isocaloric, low-glycaemic-index diet in moderately overweight young men and showed that resting energy expenditure declined by 10.5% on the former diet compared with 4.6% on the latter diet.

Interventions on eating habits and physical activity

Most studies dealing with nutritional status and physical activity among European adolescents conclude that nutritional interventions and interventions to enhance physical activity are strongly neededReference Contento, Balch and Bronner^95, Reference Lytle, Jacobs, Perry and Klepp⁹⁶.

School-based health and nutrition education intervention studies in Europe and the USA have had mixed results in effecting physiological changesReference De Bourdeaudhuij, Sallis and Vandelanotte^97, Reference Manios, Moschandreas, Hatzis and Kafatos⁹⁸. The majority of trials have been conducted in the USA, and there are doubts whether these can be extrapolated to the great diversity of the European areaReference Anderson, Cox, McKellar, Reynolds, Lean and Mela⁹⁹.

New and modern tools for health promotion in adolescents need to be developed focusing on this specific population and considering gender differences. Adolescence is a unique period in life. Health promotion should not force models of behaviour onto individuals or groups. Adolescents need a food culture based on foods to eat, rather than foods to avoid, and an understanding of suitable weight-control measuresReference Nowak¹⁰⁰. Computer-tailored nutrition and physical activity education is an innovative, promising and cost-effective tool to motivate people to make healthy dietary and physical activity changes. It provides respondents with individualised feedback about dietary behaviour and physical activity. The available evidence indicates that computer-tailored education is more effective in motivating people to make changes than general nutritional and physical activity educationReference Brug, Oenema and Campbell¹⁰¹. Until now most computer-tailored nutrition programmes have focused on one or a few aspects of nutrition behaviour such as fat intake or fruit and vegetable intake. Most programmes are aimed at adults. The present project will develop and evaluate a web-based intervention focusing on the usual eating and physical activity habits of adolescents. Adolescents will be stimulated to use the web-based intervention by teachers in schools; the first use of the intervention will take place during class time. For further use the intervention will be available on the web. The diet optimisation approach, not used until now in computer-tailored interventions, will make it possible to write an individualised advice to optimise the usual eating habits of adolescents. To date, there is no experience in the assessment of the efficacy of such a tool in Europe.