Discussion
The present study investigated the validity and reliability of the ASQ for assessing stress in European adolescents. The ASQ, originally developed for Australian adolescents, is documented to assess subjective stressor load and consists of 10 different stress dimensions covering a broad domain of adolescent stress experiences [
6]. The reliability analysis in the present study indicated a moderate internal reliability of the ASQ scales, as five of 10 Cronbach α-values were > 0.8. The Cronbach α-values were lower than those in the validation study of the original ASQ carried out by Byrne
et al. [
6]. Test-retest ICCs of the ASQ scales in the present study were, with the exception of stress from romantic relationships, < 0.8, indicative of poor stability over time. They were lower than those reported by Byrne
et al. [
6], probably because of the greater within-measurement period (two weeks in the present study rather than one week in the study of Byrne
et al. [
6]) and the lack of power owing to a small sample size (37 in the present study compared with 105 in the study of Byrne
et al. [
6]).
The present study demonstrated that adolescent girls experienced higher stress levels than boys, confirming the gender differences in the ASQ as presented by Byrne
et al. [
6], and confirming previously reported gender differences in adults [
26] and adolescents [
27]. In addition, the present study demonstrated that pubertal stage was also a predictor for the ASQ, with the adolescents that are already in a post-pubertal stage of development having higher scores than those still in a pre-pubertal stage or in full pubertal development. This is in line with the hypothesis of Dahl and Gunnar that pubertal development is the driving force behind increasing stress sensitivity during adolescence [
28] and with findings from Sumter
et al. that the biological stress sensitivity increased with pubertal status [
29]. The differential functioning of the ASQ according to the adolescents' gender and pubertal stage give strength to the validity of the ASQ. Age was no systematic predictor of the ASQ scores, suggesting that the ASQ functions similarly across the whole age range (12.5-17.5 years). This is not in agreement with observations of Byrne
et al., where bivariate correlations with age were weak positive for five out of ten scales (significant correlations varied from 0.12 to 0.35) [
6]. This might be indicative of differential scales functioning across cultures (see below). Surprisingly, no systematic association was evident between SES and the ASQ scores, conflicting with previous evidence that resource- and prestige-based indices of family SES were associated with adolescents' risk for uncontrollable and controllable negative life events, respectively [
30]. These controversial results are probably because the ASQ assesses daily stressful situations or events, which have a lower impact on life and more frequent occurrence than the negative life events assessed in the study of Brady & Matthews [
30].
The results of the second-order CFA suggest that the theoretical model of the ASQ fits moderately well with the observed data in the present population. The absolute goodness of fit index χ
2 was not indicative of a good fit. However, this fit index is widely recognized to be problematic as it is highly sensitive to sample size. The comparative fit indices, Bentler's CFI and RMSEA, were indicative of a marginally good or acceptable fit. The factor loadings of the majority of the observed items (51/56) were indicative of high correlations with the extracted factors. The factor loadings of the component scales on the summary score were higher (> 0.6), suggesting very good correlations between the scales and the summary score. The moderate construct validity could be explained by the fact that the nature of adolescent stress differs across culture boundaries. European adolescents may experience stress in a different way from Australian adolescents owing to differences in cultural background, education, living conditions, norms and standards. It could be hypothesized that the greater these differences, the lower the applicability of the original scale construct will be. This finding is in agreement with the concern outlined by Byrne
et al. that
'the capacity of scales of adolescent stress to cross boundaries of culture is not yet well understood' [
6].
Linear regression analyses demonstrated that in boys, four out of 10 ASQ scale scores were found to be a significant positive predictor for their BWSF cortisol values, while in girls this positive association was not observed for any ASQ scale. These results indicate a rather poor criterion validity of the ASQ when compared with BWSF cortisol, especially in girls. A possible explanation for the weak or lack of distinct positive association between the ASQ and BWSF cortisol could be that the ASQ assesses only one aspect of the stress concept, namely the personal perception of stress from certain situations and events in adolescent life (so-called cognitive appraisal in the 'Stress and Coping Theory' developed by Lazarus and Folkman [
31]), while the adolescents' coping mechanisms (defined as the second process in the 'Stress and Coping Theory' [
31]) are not determined by the ASQ. Salivary cortisol is an indicator of Hypothalamic-Pituitary-Adrenal (HPA) axis activity and can be considered as a reflection of the adaptational outcome, which is the result of the appraisal of and coping with stress [
13]. No controls for coping mechanisms could be used, and this could explain why increasing ASQ scores are not associated with increased salivary cortisol levels. This limitation of the ASQ restricts its utility to the assessment of cognitive appraisal of stress and not chronic stress, as a result of which its associations with stress-related health outcomes could be weakened or non-existent. Therefore, the information obtained with the ASQ should be complemented with information concerning previous lifetime stressful events (e.g. death of family members, sexual, psychological or physical abuse) as these are likely to influence the test results. Moreover, this additional information would complement the picture of the adolescents' chronic exposure to stress and provide more opportunities for investigating the potential role of chronic stress in the etiology of certain disorders. The lack of association between self-reported subjective measures of stress and cortisol measurements in biological samples has previously been denominated the 'lack of psychoendocrine covariance' [
32]. Hellhammer
et al. indicated that this phenomenon is not surprising given the complex interplay of neurobiological events that link perceived stress to HPA axis activity and the difficulties in assessing perceived stress [
33].
In addition, in children with post-traumatic stress disorder (PTSD), long-term cortisol levels were no longer elevated as a reaction to trauma, but normalized [
34]. The adolescents in the present study were not screened for previous development of PTSD or other severe traumas in their lifetime, and this could have reduced the expected positive association between perceived stress and cortisol levels. Therefore, future research is recommended to include all lifetime traumatic/stressful events in addition to focusing on a certain time period.
Strengths and limitations
The multi-national character of HELENA-CSS and the stress sub-study, together with the strict standardization of the fieldwork across all European cities are great advantages of this project and make this project unique in its kind. The integration of the stress module in the HELENA-CSS had the advantage of investigating in-time associations between the experience of psychosocial stress and other health-related parameters (socio-demographic, clinical, food-related and physical activity parameters) conducted in HELENA. Unfortunately, the study also faces some limitations, which are specified below.
For feasibility reasons, the study population within the HELENA-CSS was determined on the basis of random cluster sampling. Inherent in cluster sampling, the standard errors for estimates are greater than for simple random samples, the so-called 'design effect' [
35]. In view of these possible higher standard errors, it could be the case that some differences in ASQ scores between groups were not revealed.
Test-retest reliability analysis and criterion validity analysis were performed on a convenient sample of adolescents and are subject to limitations inherent to convenience sampling (e.g. non-generalization, and a potentially large and unmeasured bias). These data were collected for validation purposes only and the consequences of this convenience sampling are assumed to be minimal. In addition, the samples for these analyses were small and the lack of power could have attenuated the test results and contributed to the rather poor test-retest reliability and criterion validity.
Other limitations included the time limits for the fieldwork of HELENA (only a well-defined period could be spent in the schools) and the high load of the measurements. Owing to these logistic limitations, BWSF cortisol was measured as this could be carried out at home rather than during school hours. Alternative procedures and biomarkers, recently suggested to be promising biomarkers in the context of stress assessment, are the cortisol awakening response (defined by Clow
et al. as
'the period of cortisol secretory activity in the first 45-60 min immediately post-awakening') [
12,
14,
36], hair cortisol [
37] and salivary α-amylase [
38]. These alternative procedures and biomarkers for chronic stress may have led to better criterion validity. However, this large body of evidence was not available at the beginning of the present study-. In addition, several methodological challenges associated with the measurement of these alternative biomarkers have been documented [
14,
39], which would have been difficult to address in the school setting of the HELENA project. Previously, several methodological difficulties related to salivary cortisol assessment have been addressed in the literature, which are hard to control in epidemiological surveys [
25,
40]. In the present study, researchers experienced some difficulties in collecting and analysing salivary cortisol. For instance, it was impossible to standardize the time of arousal and therefore sampling time, a factor that is documented to influence cortisol values [
41]. This was partly controlled by using samples taken on awakening between 6 am and 8 am. The adolescents' compliance was also an issue of concern, as it might have been weakened by the high burden of measurements in HELENA. Sometimes individuals forgot to take a sample and there was no guarantee of correct timing (immediate after awakening) when samples were provided. The issue of protocol compliance for saliva sampling has previously been outlined for children and adolescents [
12] and in adults [
40]. Samples were taken on different weekdays, another factor that influences cortisol values [
41]. To what extent these methodological shortcomings have influenced cortisol values and further analyses in this study is difficult to evaluate. Therefore, it is recommended that salivary sampling protocols are strictly standardized in future research (a particular challenge in epidemiological surveys).
Acknowledgements
The writing group takes sole responsibility for the content of this article. The authors want to thank the participating adolescents and their parents for their voluntary participation to the HELENA study. The authors also want to acknowledge D. Byrne for his permission to use the Adolescent Stress Questionnaire and the service of BioMedES for copyediting the manuscript. The HELENA Study takes place with the financial support of the European Community Sixth RTD Framework Programme (Contract FOOD-CT-2005-007034). The content of this article reflects only the authors' views and the European Community is not liable for any use that may be made of the information contained therein. Tineke De Vriendt is financially supported by the Research Foundation - Flanders (Grant number: 1.1.746.09.N.01). The authors want to acknowledge the HELENA Study Group for succesfully finalizing this project. A complete list of HELENA Study Group members is provided here:
HELENA Study Group
Co-ordinator: Luis A. Moreno.
Core Group members: Luis A. Moreno, Fréderic Gottrand, Stefaan De Henauw, Marcela González-Gross, Chantal Gilbert.
Steering Committee: Anthony Kafatos (President), Luis A. Moreno, Christian Libersa, Stefaan De Henauw, Sara Castelló, Fréderic Gottrand, Mathilde Kersting, Michael Sjöstrom, Dénes Molnár, Marcela González-Gross, Jean Dallongeville, Chantal Gilbert, Gunnar Hall, Lea Maes, Luca Scalfi.
Project Manager: Pilar Meléndez.
1. Universidad de Zaragoza (Spain): Luis A. Moreno, Jesús Fleta, José A. Casajús, Gerardo Rodríguez, Concepción Tomás, María I. Mesana, Germán Vicente-Rodríguez, Adoración Villarroya, Carlos M. Gil, Ignacio Ara, Juan Revenga, Carmen Lachen, Juan Fernández Alvira, Gloria Bueno, Aurora Lázaro, Olga Bueno, Juan F. León, Jesús Mª Garagorri, Manuel Bueno, Juan Pablo Rey López, Iris Iglesia, Paula Velasco, Silvia Bel, Luis A. Gracia Marco, Theodora Mouratidou.
2. Consejo Superior de Investigaciones Científicas (Spain): Ascensión Marcos, Julia Wärnberg, Esther Nova, Sonia Gómez, Ligia Esperanza Díaz, Javier Romeo, Ana Veses, Belén Zapatera, Tamara Pozo, David Martínez.
3. Université de Lille 2 (France): Laurent Beghin, Christian Libersa, Frédéric Gottrand, Catalina Iliescu, Juliana Von Berlepsch.
4. Research Institute of Child Nutrition Dortmund, Rheinische Friedrich-Wilhelms-Universität Bonn (Germany): Mathilde Kersting, Wolfgang Sichert-Hellert, Ellen Koeppen.
5. Pécsi Tudományegyetem (University of Pécs) (Hungary): Dénes Molnar, Eva Erhardt, Katalin Csernus, Katalin Török, Szilvia Bokor, Mrs. Angster, Enikö Nagy, Orsolya Kovács, Judit Répasi.
6. University of Crete School of Medicine (Greece): Anthony Kafatos, Caroline Codrington, María Plada, Angeliki Papadaki, Katerina Sarri, Anna Viskadourou, Christos Hatzis, Michael Kiriakakis, George Tsibinos, Constantine Vardavas, Manolis Sbokos, Eva Protoyeraki, Maria Fasoulaki.
7. Institut für Ernährungs- und Lebensmittelwissenschaften - Ernährungphysiologie. Rheinische Friedrich Wilhelms Universität (Germany): Peter Stehle, Klaus Pietrzik, Marcela González-Gross, Christina Breidenassel, Andre Spinneker, Jasmin Al-Tahan, Miriam Segoviano, Anke Berchtold, Christine Bierschbach, Erika Blatzheim, Adelheid Schuch, Petra Pickert.
8. University of Granada (Spain): Manuel J. Castillo, Ángel Gutiérrez, Francisco B Ortega, Jonatan R Ruiz, Enrique G Artero, Vanesa España, David Jiménez-Pavón, Palma Chillón, Cristóbal Sánchez-Muñoz, Magdalena Cuenca
9. Istituto Nazionalen di Ricerca per gli Alimenti e la Nutrizione (Italy): Davide Arcella, Elena Azzini, Emma Barrison, Noemi Bevilacqua, Pasquale Buonocore, Giovina Catasta, Laura Censi, Donatella Ciarapica, Paola D'Acapito, Marika Ferrari, Myriam Galfo, Cinzia Le Donne, Catherine Leclercq, Giuseppe Maiani, Beatrice Mauro, Lorenza Mistura, Antonella Pasquali, Raffaela Piccinelli, Angela Polito, Raffaella Spada, Stefania Sette, Maria Zaccaria.
10. University of Napoli "Federico II" Dept of Food Science (Italy): Luca Scalfi, Paola Vitaglione, Concetta Montagnese.
11. Ghent University (Belgium): Ilse De Bourdeaudhuij, Stefaan De Henauw, Tineke De Vriendt, Lea Maes, Christophe Matthys, Carine Vereecken, Mieke de Maeyer, Charlene Ottevaere, Inge Huybrechts.
12. Medical University of Vienna (Austria): Kurt Widhalm, Katharina Phillipp, Sabine Dietrich, Birgit Kubelka Marion Boriss-Riedl.
13. Harokopio University (Greece): Yannis Manios, Eva Grammatikaki, Zoi Bouloubasi, Tina Louisa Cook, Sofia Eleutheriou, Orsalia Consta, George Moschonis, Ioanna Katsaroli, George Kraniou, Stalo Papoutsou, Despoina Keke, Ioanna Petraki, Elena Bellou, Sofia Tanagra, Kostalenia Kallianoti, Dionysia Argyropoulou, Katerina Kondaki, Stamatoula Tsikrika, Christos Karaiskos.
14. Institut Pasteur de Lille (France): Jean Dallongeville, Aline Meirhaeghe.
15. Karolinska Institutet (Sweden): Michael Sjöstrom, Jonatan R Ruiz, Francisco B. Ortega, María Hagströmer, Anita Hurtig Wennlöf, Lena Hallström, Emma Patterson, Lydia Kwak, Julia Wärnberg, Nico Rizzo.
16. Asociación de Investigación de la Industria Agroalimentaria (Spain): Jackie Sánchez-Molero, Sara Castelló, Elena Picó, Maite Navarro, Blanca Viadel, José Enrique Carreres, Gema Merino, Rosa Sanjuán, María Lorente, María José Sánchez.
17. Campden BRI (United Kingdom): Chantal Gilbert, Sarah Thomas, Elaine Allchurch, Peter Burgess.
18. SIK - Institutet foer Livsmedel och Bioteknik (Sweden): Gunnar Hall, Annika Astrom, Anna Sverkén, Agneta Broberg.
19. Meurice Recherche & Development asbl (Belgium): Annick Masson, Claire Lehoux, Pascal Brabant, Philippe Pate, Laurence Fontaine.
20. Campden & Chorleywood Food Development Institute (Hungary): Andras Sebok, Tunde Kuti, Adrienn Hegyi.
21. Productos Aditivos SA (Spain): Cristina Maldonado, Ana Llorente.
22. Cárnicas Serrano SL (Spain): Emilio García.
23. Cederroth International AB (Sweden): Holger von Fircks, Marianne Lilja Hallberg, Maria Messerer
24. Lantmännen Food R&D (Sweden): Mats Larsson, Helena Fredriksson, Viola Adamsson, Ingmar Börjesson.
25. European Food Information Council (Belgium): Laura Fernández, Laura Smillie, Josephine Wills.
26. Universidad Politécnica de Madrid (Spain): Marcela González-Gross, Jara Valtueña, David Jiménez-Pavón, Ulrike Albers, Raquel Pedrero, Agustín Meléndez, Pedro J. Benito, Juan José Gómez Lorente, David Cañada, Alejandro Urzanqui, Rosa María Torres, Paloma Navarro.