The Journal of Early
Adolescence
http://jea.sagepub.com/
The Role of Individual- and Macro-Level Social Determinants on
Young Adolescents' Psychosomatic Complaints
Veronika Ottova, Michael Erhart, Wilma Vollebergh, Gyöngyi Kökönyei, Antony
Morgan, Inese Gobina, Helena Jericek, Franco Cavallo, Raili Välimaa, Margarida
Gaspar de Matos, Tania Gaspar, Christina W. Schnohr, Ulrike Ravens-Sieberer
and and the Positive Health Focus Group
The Journal of Early Adolescence 2012 32: 126 originally published online 14
November 2011
DOI: 10.1177/0272431611419510
The online version of this article can be found at:
http://jea.sagepub.com/content/32/1/126
Published by:
http://www.sagepublications.com
Additional services and information for The Journal of Early Adolescence can be found at:
Email Alerts: http://jea.sagepub.com/cgi/alerts
Subscriptions: http://jea.sagepub.com/subscriptions
Reprints: http://www.sagepub.com/journalsReprints.nav
Permissions: http://www.sagepub.com/journalsPermissions.nav
Citations: http://jea.sagepub.com/content/32/1/126.refs.html
>> Version of Record - Feb 22, 2012
OnlineFirst Version of Record - Nov 14, 2011
What is This?
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
419510
0Ottova et al.Journal of Early Adolescence
© The Author(s) 2012
JEA32110.1177/027243161141951
Reprints and permission:
sagepub.com/journalsPermissions.nav
The Role of
Individual- and
Macro-Level Social
Determinants on
Young Adolescents’
Psychosomatic
Complaints
Journal of Early Adolescence
32(1) 126­–158
© The Author(s) 2012
Reprints and permission:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0272431611419510
http://jea.sagepub.com
Veronika Ottova1, Michael Erhart1,
Wilma Vollebergh2, Gyöngyi Kökönyei3,4,
Antony Morgan5, Inese Gobina6,
Helena Jericek7, Franco Cavallo8, Raili Välimaa9,
Margarida Gaspar de Matos10,11, Tania Gaspar10,11,
Christina W. Schnohr12, Ulrike Ravens-Sieberer1,
and the Positive Health Focus Group
1
University Medical Center Hamburg-Eppendorf, Hamburg, Germany
University of Utrecht, Utrecht, Netherlands
3
Eotvos Lorand University, Budapest, Hungary
4
National Institute of Child Health, Hungary
5
National Institute of Health and Clinical Excellence, London, UK
6
Riga Stradins University, Riga, Latvia
7
Institute of Public Health, Ljubljana, Slovenia
8
University of Torino, Torino, Italy
9
University of Jyväskylä, Jyväskylä, Finland
10
FMH/Lisbon Technical University, Lisbon, Portugal
11
CMDT/IHMT/New University of Lisbon, Lisbon, Portugal
12
University of Copenhagen, Copenhagen, Denmark
2
Corresponding Author:
Ulrike Ravens-Sieberer, Department of Child and Adolescent Psychiatry, Psychotherapy,
and Psychosomatics, University Medical Center Hamburg-Eppendorf,
Martinistr. 52, Hamburg 20246, Germany
Email: Ravens-Sieberer@uke.uni-hamburg.de
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
127
Ottova et al.
Abstract
This study examines the social determinants of psychosomatic complaints
in young adolescents. Using data from the Health Behaviour in School-aged
Children (HBSC) study, psychosomatic complaints are studied in 98,773
adolescents (11- and 13-year-olds; 48% 11-year-olds, 52% 13-year-olds; 52%
females, 48% males) from 34 European countries. Individual-level determinants, including family-, peer- and school-related factors as well as countrylevel determinants (Human Development Index [HDI]) are considered. In
line with existing evidence, results revealed more psychosomatic complaints
in young adolescents experiencing stress inducing familial-, peer- and schoolrelated factors. Negative effects of poor friendships, negative class climate,
school pressure, and high media use were more pronounced for girls. After
controlling for these factors, a higher HDI was related to a lower risk for
psychosomatic complaints. Gender-specific intervention programs should
aim at improving the quality of relationships, especially among peers, to prevent psychosomatic complaints among young adolescents.
Keywords
well-being, family, peers, school context, stressors
Background
Psychosomatic complaints are important indicators of psychosocial health
(Piko, 2007) and well-being (Berntsson, Köhler, & Gustafsson, 2001). By
definition, they are “subjective physical complaints (headache, stomach
ache, back ache, and dizziness) as well as psychological complaints (feeling
low, irritability, nervousness, and difficulty in getting to sleep)” that are not
caused by an underlying physical disease (Natvig, Albrektsen, Anderssen, &
Qvarntstrøm, 1999, p. 362). Findings from studies with children and adolescents provide support for the relationship between psychosomatic complaints
and poor life satisfaction, depression (Piko, 2006), negative moods (Jellesma,
Rieffe, Terwogt, & Kneepkens, 2006), unhappiness (Natvig, Albrektsen, &
Qvarnstrom, 2003), and a lower health perception (Piko, 2007). The period
of adolescence is a vulnerable time of developmental changes which are
highly susceptible to psychosomatic complaints.
Early adolescence is an important developmental period as it marks the
transition between childhood and adolescence. It is characterized by increased
desire for “autonomy from adult control [. . .], peer orientation, self-focus,
self-consciousness, salience-of-identity issues; concern over sexual
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
128
Journal of Early Adolescence 32(1)
relationships; and capacity for abstract cognitive activity” (Eccles, Lord, &
Midgley, 1991, p. 534). Eccles and Midgley (1990) have suggested various
reasons for the negative changes which begin during early adolescence,
including “intrapsychic upheaval” which leads to motivational and behavioral problems, the simultaneous timing of school transition and major developmental changes, and the quality of school environments (p. 135). The last
20 years has seen an increase in the number of cross-cultural studies on child
and adolescent development across cultures (Gielen & Roopnarine, 2004).
Most frequently, these have been prevalence studies comparing age and gender differences; however, more recently, studies have taken account of
important social contexts, such as the family, peers, and school.
Psychosomatic complaints are thought to be stress related. Frequent or
sustained stress leads to emotional and physiological responses, which have
a clear effect on the development of frequent complaints (Brosschot, 2002).
Both individual-level and country-level factors, such as the nation’s wealth
(gross national product) or its distribution (Gini coefficient), play a role in
children’s well-being (Holstein et al., 2009). The present article extends this
knowledge by examining the relative importance of the social determinants
on health complaints in young adolescents. The robustness of the Health
Behaviour in School-aged Children (HBSC) study (large sample size, multinational data) provides an opportunity to investigate the impact of both
individual- and country-level factors on psychosomatic complaints across
Europe and North America.
Psychosomatic Complaints in Young Adolescents
Research has demonstrated an increased risk for psychosomatic complaints
in the presence of psychosocial stressors, such as parental conflicts
(Hurrelmann, Engel, Holler, & Nordlohne, 1988), bullying, lack of acceptance by peers, lack of support from parents and teachers (Gerber & Pühse,
2008; Natvig et al., 1999). Given the increase in psychosomatic complaints
(Berntsson & Köhler, 2001; Hagquist, 2009; Karvonen, Vikat, & Rimpelä,
2005), it is important to gain a better understanding of the distribution and
origin of this association. The most common symptoms reported by young
adolescents are back pain, neck pain, and headache (Bandell-Hoekstra et al.,
2001; Hakala, Rimpelä, Salminen, Virtanen, & Rimpelä, 2002; Haugland,
Wold, Stevenson, Aaroe, & Woynarowska, 2001; Jellesma et al., 2006).
Other complaints frequently reported are stomach complaints and loss of
appetite (Berntsson et al., 2001; Jellesma et al., 2006).
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
129
Ottova et al.
Evidence suggests age and gender to be important factors in reported complaints. Studies show that older adolescents have higher rates of complaints
(Cavallo et al., 2006; Ghandour, Overpeck, Huang, Kogan, & Scheidt, 2004;
Haugland et al., 2001; Karademas, Peppa, Fotiou, & Kokkevi, 2008; Tanaka,
Tamai, Terashima, Takenaka, & Tanaka, 2000) and that gender differences
increase with age. The highest rates of psychosomatic complaints are generally reported by 15-year-old girls (Haugland et al., 2001; West & Sweeting,
2003). However, while older girls report more complaints than their younger
peers, older boys tend to report fewer complaints than younger ones (Marschall,
1989). There is clear evidence that boys and girls differ in the types of psychosomatic complaints they commonly report (Natvig et al., 1999). The frequency
and intensity of complaints is higher in girls and they generally are more likely
to report having had more than two complaints at least once a week (Cavallo
et al., 2006).
Evidence on the relationship between stress and health complaints has been
confirmed by a number of studies; however, less frequent are the studies of
young people (Sundblad, Jansson, Saartok, Renström, & Engström, 2008).
Research also suggests that psychosocial complaints may be a stress reaction
to psychosocial tension (Holler-Nowitzki, 1994). Studies among adults have
pointed to an association between workplace psychosocial factors and stress
(Haukka et al., 2010). Likewise, increasing demands, academic pressure at
school, and fear of failure may result in the experience of stress among young
people (Eriksson & Sellström, 2010; Hurrelmann et al., 1988). Stress increases
with age, and girls are more affected than boys (Sundblad et al., 2008). Such
risk factors can of course be offset by the presence of protective factors.
Torsheim and Wold (2001) showed an inverse relationship between stress and
sense of coherence, suggesting that protective factors (or assets) have a role to
play in health and well-being during adolescence.
Clearly, psychosomatic complaints are determined by a number of factors
operating through social stress at a micro (individual) level, such as family
problems, peer pressure, unsuccessful coping strategies (Brill, Patel, &
MacDonald, 2001), as well as factors at the macro (societal or national) level
(e.g., Holstein et al., 2009; Torsheim, Currie, Boyce, & Samdal, 2006). The
complexity of these determinants and the rationale for their inclusion in the
present study are described below.
Individual-Level Factors Related
to Psychosomatic Complaints
Familial context. Studies confirm an association between adolescents’
health and well-being and family and school factors (Gaspar, Matos, Ribeiro,
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
130
Journal of Early Adolescence 32(1)
Leal, & Ferreira, 2009; Karademas et al., 2008), including good family climate and high social support (Wille, Bettge, Ravens-Sieberer, & the BELLA
Study Group, 2008), positive family, and school climate (Karademas et al.,
2008). In contrast, family conflicts and stress were found to be associated
with more health problems (Karvonen et al., 2005). Poor communication
with parents emerged as more strongly associated with psychological health
complaints in adolescents than poor peer communication (Moreno et al.,
2009). Family structure (in particular, single-parent families) and low educational level of the parents have an additional negative impact on children’s
health (Berntsson & Köhler, 2001; Brolin Laftman & Östberg, 2006; Östberg,
Alfven, & Hjern, 2006).
Peer relations. Research into peer relations has identified an association
between harassment and recurrent pain (Hjern, Alfven, & Östberg, 2008),
and studies have shown that higher rates of health complaints are experienced
by victims as well as by perpetrators of bullying (Gini, 2008; Richter, Bowles,
Melzer, & Hurrelmann, 2007). Girls who experience bullying report headaches, back aches, and morning fatigue more frequently than those with no
such negative experiences (Ghandour et al., 2004). Moreno et al. (2009) also
showed that good peer communication with same-sex friends is associated
with lower number of complaints.
School environment. School and extracurricular activities, such as homework, clubs, friendships, and other social activities, are some of the key factors for children’s psychosocial development (Samdal, Nutbeam, Wold, &
Kannas, 1998; Ravens-Sieberer, Freeman, Kokonyei, Thomas, & Erhart,
2009). Poor perception of the school environment can be a major stressor
during the school years in young adolescents (Hjern et al., 2008). Eccles &
Midgley (1990) showed a marked decrease in self-confidence in school
achievement and mastery of school work in the transition from childhood
into adolescence. Other studies have confirmed an association between
school-related stress, such as feeling pressured by school work, and psychosomatic complaints (Gerber & Pühse, 2008; Karademas et al., 2008; Natvig
et al., 1999). Older adolescents experience increased school pressure (Cavallo
et al., 2006). Poor academic achievement has been shown to be associated
with nearly a threefold greater chance of poor or fair self-perceived health
(Piko, 2007). Karvonen et al. (2005) found that both—the absence of as well
as academic demands—are associated with psychosomatic health complaints.
Other authors found that factors such as teacher support or resources at school
play an even greater role (Eriksson & Sellström, 2010). In fact, several studies have confirmed a protective effect of social support from peers (Natvig et al.,
1999; Santinello, Vieno, & De Vogli, 2009).
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
131
Ottova et al.
Physical activity and sedentary behaviors. Studies show that children who
engage in sedentary behaviors, such as playing TV games (Tanaka et al.,
2000), computer-related activities (Hakala, Rimpelä, Saarni, & Salminen,
2006), and screen-based activities in general (Torsheim et al., 2010), have a
higher rate of somatic complaints (Bakoula, Kapi, Veltsista, Kavadias, &
Kolaitis, 2006), and the risk for complaints increases with the number of
hours spent sedentary (Tanaka et al., 2000). One possible mechanism is that
sedentary behaviors sustained over a period of time cause muscle tension
which may then lead to pain (Straker et al., 2008; Torsheim et al., 2010). A
side effect of many hours spent on screen-based activities is that it potentially
reduces the number of hours for parent-child interaction and, in this way,
negatively affects child health (Tanaka et al., 2000). A distinct protective
effect of physical activity could not be confirmed (Gerber & Pühse, 2008;
Ghandour et al., 2004), but it may moderate the relationship between stress
and psychosomatic complaints (Haugland, Wold, & Torsheim, 2003). Sundblad
et al. (2008) found a correlation between physical activity and positive perceived health.
Socioeconomic indicators. Studies have pointed at associations between psychosomatic complaints and sociodemographic risk factors, such as low socioeconomic status (SES), a large family, or single motherhood (Egle, Hoffmann,
& Steffens, 1997), and low parental education (Karvonen et al., 2005; Piko &
Fitzpatrick, 2007; Sleskova et al., 2006). Unemployment of either mother or
father has also been identified as a risk factor for child health (Sleskova et al.,
2006), whereby economic stress, rather than social class, was a significant
determinant of psychosomatic complaints (Östberg et al., 2006). Still, Torsheim
and colleagues (2004) found a uniform pattern between health and social
inequalities across 22 countries, suggesting that previous inconsistent findings may in part be an artifact of methodological variance.
National Indicators
Research on children and adolescents which included national indicators has
found that both low household income and high material inequalities are
associated with poorer subjective health (Torsheim et al., 2006). Holstein
and colleagues (2009) found a relationship between material inequality
(measured by the Gini coefficient) and psychosomatic complaints, but not
for gross national product (GNP). In contrast, Olsen and Dahl (2007) did not
find a significant relationship between the Gini coefficient and health but
found one for gross domestic product (GDP). Another study comparing psy-
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
132
Journal of Early Adolescence 32(1)
chosomatic complaints in children in five Scandinavian countries revealed
largest country differences at the macro level, that is, economic factors, SES,
and economic resources of family (Berntsson et al., 2001). Interestingly,
despite variations in unemployment rates between the countries, there were
no notable country differences in the associations between psychosomatic
complaints, SES, family economy, and family type, indicating multidimensional risk factors for psychosomatic complaints and a complex relationship
between macro-level factors and individual health. A useful alternative to
traditional measures of development, such as GDP which only reflects one
dimension, is the Human Development Index (HDI) which covers the following dimensions: knowledge, healthy life, and standard of living (Sagar &
Najam, 1998). While a vast majority of research papers on child health have
used GDP (or GNP) and/or the Gini coefficient as indicators, none of them
has used the Human Development Index (HDI).
Research Aims
The purpose of this article was to extend previous research on psychosomatic
complaints from a stress perspective by investigating social stress factors at
an individual and a national (societal) level using a large international sample of young adolescents. The individual-level stress factors included familial factors (family structure, communication with parents), peer relations
(bullying, number of close friends), and school context (class climate, academic achievement). Physical activity and sedentary behaviors (media use)
were also included in the analyses to control for the effect of health behaviors. In addition, country-level effects were taken into account to further
understand the potential relationship between national factors and psychosomatic complaints. Based on previous research outlined above, we chose to
use HDI to analyze country-level effects as this multidimensional index
covers three relevant dimensions (healthy life, education, and economic
aspects) which are measured via life expectancy, adult literacy, and GDP. It
is a broad measure of national well-being (Moon, Welch, & Wong, 2005).
Currently, a majority of the studies on psychosomatic complaints and
stress come from Scandinavian countries (Eriksson & Sellström, 2010; Hjern
et al., 2008; Sundblad et al., 2008; Torsheim & Wold, 2001). Relatively little
is known about cross-national differences, in particular, on the combined
effects of individual- and country-level factors on psychosomatic complaints.
On this background, the following hypotheses were formulated and tested.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
133
Ottova et al.
Hypothesis 1: Stable family structure and easy communication with
parents is associated with lower rates of psychosomatic complaints
(i.e., better health).
Hypothesis 2: Good peer relationships (high number of close friends,
no or low occurrence of bullying) are associated with lower rates of
psychosomatic complaints (i.e., better health).
Hypothesis 3: A positive class climate and low pressure at school are
associated with lower rates of psychosomatic complaints (i.e., better
health).
Hypothesis 4: A high HDI is associated with a higher average level of
psychosomatic health in a country.
Method
Participants
The sample used in this study was obtained from the 2005/2006 HBSC
study. The HBSC study is an international survey conducted every 4 years in
a growing number of countries. Data were collected through a school-based
survey using classroom-administered self-completion questionnaires in each
participating country and region. Data collection and questionnaires were
based on a standardized research protocol. Participation in the survey was
voluntary, and assurance was provided in relation to confidentiality and anonymity. Each country was required to respect the ethical and legal requirements in their country for this type of survey. The target populations for the
study were 11-, 13- and 15-year-old children (Grades 5, 7, and 9). Participating
countries were required to include a minimum of 95% of the eligible target
population within their sample frame. The samples were drawn using the
cluster sampling procedure, whereby the primary sampling unit was school
class. The final international file contained 204,534 pupils across 41 countries or regions, consisting of 100,233 (49%) boys and 104,301 (51%) girls,
and 66,707 (33%) 11-year-olds, 69,954 (34%) 13-year-olds, and 67,873
(33%) 15-year-olds. The sampling response rates for all the countries in the
2005/2006 HBSC study ranged from 47% to 100% for school/class level and
from 34% to 97% at the pupil level. In most countries, school/class- and
pupil-level response rates were above 70% (Roberts et al., 2009). Full details
on the study and the participants can be found elsewhere (Currie et al., 2008a;
Roberts et al., 2007, 2009).
The present study included data from 98,773 11- and 13-year-olds from 34
countries. Due to lack of data on some covariates, Israel, Malta, Slovakia, and
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
134
Journal of Early Adolescence 32(1)
the United States were excluded from the analyses. In addition, we excluded
Greenland, Scotland, and Wales due to lack of information on HDI. We
focused on 11- and 13-year-olds to capture the transition from childhood to
adolescence, when youth are most likely to be susceptible to social stressors.
Hence, analyses for the present study were based on 98,773 adolescents.
Instruments
Psychosomatic complaints. In the HBSC study, the HBSC symptom checklist
(HBSC-SCL) was used to assess the occurrence of psychosomatic health complaints (Haugland et al., 2001). The HBSC-SCL is a reliable and valid instrument (Ravens-Sieberer et al., 2008) which measures eight symptoms (headache,
stomach ache, back ache, feeling low, irritability or bad temper, feeling nervous, difficulties in falling asleep, and feeling dizzy) during the past 6 months.
Answers were rated on a 5-point scale ranging from about every day to rarely
or never. Although previous research has suggested a two-factor solution, the
scale can also be thought to measure a unidimensional latent trait of psychosomatic complaints (Ravens-Sieberer et al., 2008). Multiple recurrent psychosomatic complaints were defined as at least two symptoms occurring more than
once weekly (Cronbach’s α = .80).
Physical activity and sedentary behaviors. Measures of physical activity and
sedentary behavior were included in the analysis. “Moderate to vigorous”
physical activity was assessed using the measure developed by Prochaska,
Sallis, & Long (2001). The valid and reliable measure asks the question: “Over
the past 7 days, on how many days were you physically active for a total of at
least 60 minutes per day?” with answer options ranging from 0 days to 7 days.
Information on sedentary behavior was assessed by asking how often respondents watched TV (including DVDs and videos), used the computer (for chatting online, Internet, emailing, homework, etc.), and played computer games
on weekdays and on weekends. The answer options were “not at all,” “about
half an hour a day,” “about 1 hour a day,” [. . .], “about 7 or more hours a day.”
Although results from a validation study on the HBSC items are not available,
a study by Schmitz et al. (2004) found high reliability and validity for similar
type of items with similar age groups. For this analysis, an index was calculated by averaging the responses for all three areas (TV watching, computer
use, and computer games) for weekdays and for weekend. This will be referred
to as the Media Use Index (Cronbach’s α = .59).
Familial context. The family context was assessed by two measures: communication with parents and family structure. The first asked “How easy is it
for you to talk to the following persons about things that really bother you?”
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
135
Ottova et al.
with option responses of “very easy,” “easy,” “difficult,” “very difficult,” or
“don’t have or see this person.” Separate questions were used for mother and
father. The item responses were dichotomized into “very easy/easy/don’t
have or see this person” and “difficult/very difficult.” Family structure was
assessed by asking adolescents to indicate who else usually lives with them
at home, also including whether they had siblings living with them and, if
yes, how many. In the analysis, we differentiated between families with “both
parents,” “single parent,” “step family,” or “other.” The number of female
and male siblings was combined for analysis. Both parental communication
and family structure are measures which were developed within HBSC.
Peer relations. Peer relations were assessed through two measures: number
of close friends and experience of bullying. The first was assessed by “At present, how many close male and female friends do you have?” with response
options of “none,” “one,” “two,” and “three or more.” Respondents answered
the question for each gender separately. Answers were later combined and the
average number of male and female friends calculated. Negative peer relations
were measured by asking “How often have you been bullied at school in the
past couple of months?” The answer options were “I have not been bullied at
school in the past couple of months,” “it has only happened once or twice,” “2
or 3 times a month,” “about once a week,” or “several times a week.” The
answer categories were dichotomized into “2 or 3 times a month/about once a
week/several times a week” versus “it has only happened once or twice/I have
not been bullied at school in the past couple of months.” The measure is based
on an adapted version developed by Olweus (1996).
School environment. School-related factors included class climate and
potential stressors at school (academic achievement and workload), all of
which have been developed within the HBSC study. Student relations were
measured by three items and a validation study has shown that these three
items function well as a subscale of a valid measurement model on support
(Osen, Torsheim, & Wold, 2000; Torsheim, Wold, & Samdal, 2000). The
items ask the participants to rate how much they agreed or disagreed with
each of three statements about students in their classes. The statements were
“students like being together,” “students are kind and helpful,” and “students
accept me.” Answers were rated on a 5-point scale ranging from strongly
agree to strongly disagree. The Class Climate Index used in the analyses was
calculated as the mean of the three items (Cronbach’s α = .70). Academic
achievement was assessed by the question “In your opinion, what does your
class teacher(s) think about your school performance compared to your classmates?” The response categories were “very good,” “good,” “average,” or
“below average.” Workload pressure was assessed by the question “How
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
136
Journal of Early Adolescence 32(1)
pressured do you feel by the schoolwork you have to do?” The answer options
were “not at all,” “a little,” “some,” or “a lot.” The School Pressure Index was
calculated in the same manner as the Class Climate Index.
Socioeconomic factors. The SES of the family was measured by the Family
Affluence Scale (FAS) which was based on four items: “Does your family
own a car, van, or truck?” (0 = no, 1 = yes, one, 2 = yes, two or more), “Do
you have your own bedroom for yourself?” (0 = no, 1 = yes), “During the past
12 months, how many times did you travel away on holiday [vacation] with
your family?” (0 = not at all, 1 = once, 2 = twice, 2 = more than twice), and
“How many computers does your family own?” (0 = none, 1 = one, 2 = two,
2 = more than two). The scores of the answers were summed (range 0-7), and
individuals were categorized into high (6-7), medium (4-5), and low (0-3) FAS
groups (Currie et al., 2008b). The FAS has been validated and can be used as
“an indicator of child material affluence” (Currie et al., 2008b, p. 1433).
Country-level indicators. The HDI was developed by the United Nations
Development Program (UNDP) and is a composite index formed by three
components: longevity, education, and income. Despite criticism of the construct (Neumayer, 2001; Sagar & Najam, 1998), it has proven to be a valid
index (Noorbakhsh, 1998). Information on the HDI was obtained from the
UNDP website at http://hdrstats.undp.org/en/indicators/.
Statistical Analyses
Linear multilevel regression analyses were performed using HLM 5.05 software. Multilevel analyses account for the fact that responders within a certain context (e.g., country) may be more similar to each other than to
individuals from a different context (i.e., intracontext correlation). As the
variability of effects across different contexts was estimated, the role of contextual factors for this variability could be studied.
To account for the hierarchical and clustered structure of the data, a hierarchical two-level model including separate variances at the individual level
and at the country level (second level) was specified with random variation
of the intercepts and slopes across countries. The reliability of the crosscountry variation was examined first, and an intraclass correlation coefficient
greater than .10 indicates that 10% of the observed variation in the coefficients between countries represents a “true” difference (Cheung & Keeves,
1990). Such coefficients (intercepts and slopes) were specified as “random”
and the coefficients and their distributions of variation across countries were
estimated. The variance partitioning coefficient denotes the proportion of
“reliable” variance in the outcome that is attributable to country differences.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
137
Ottova et al.
Consistent with Cohen’s recommendations, a proportion of .14 or greater
was classified as a large effect (Cohen, 1988).
Ravens-Sieberer et al. (2008) developed a unidimensional scoring algorithm based on seven of the eight items of the HBSC-SCL which enables
valid cross-cultural comparisons and interval-scaled assessment of subjective
health complaints. In this study, sociodemographic, socioeconomic, and
familial-, school- and peer-related covariates were regressed on the psychosomatic sum score (see Appendix A for details on the formulae). The random
variation of the intercept was regressed on HDI. The analyses were repeated
for multiple recurrent health complaints using multilevel logistic regression
analysis. Odds ratios (OR) were computed and are presented.
Where possible, we tested the assumptions for hierarchical linear modelling empirically (linear relationship, distribution of residuals, equal variances, and covariances; Ditton, 1998). The other assumptions were considered
from theoretical considerations of the model specifications. To the best of our
knowledge, these assumptions were not violated.
Results
Table 1 presents descriptive results of the total sample for the dependent
variable (psychosomatic complaints) as well as for some of the independent
variables (gender, family structure, communication with parents, FAS, being
bullied) including the HDI. The results for the remaining variables are presented in Appendix B. Girls comprised 52% of the total sample analyzed and
there were slightly fewer 11-year-olds (48%) than 13-year-olds (52%). The
mean age of the sample was 12.64 (SD = 1.05).
The intercept and slopes were estimated as random effects across the
countries. The intercept for the country-level indicator, age, and media use
were regressed on the HDI. The variance partitioning coefficient of the
dependent variable in the null model was .771.
A mean psychosomatic complaint value of 4.09 was observed in our sample. Statistically significant differences were seen in the average number of
psychosomatic complaints reported between different countries. The unconditional model with random intercept across country explained 4.33% of the
overall variance. Cross-country variation in the intercept accounted for
6.14% of the overall variance in psychosomatic complaints. It was estimated
that for 95% of the comparable countries, the mean number of psychosomatic
complaints is within the range of 3.74 to 4.44. The full final model explained
23.67% of the overall variance. The prevalence of young adolescents reporting multiple recurrent psychosomatic complaints (two or more complaints
more than once a week) varied between 17% (Austria) and 64% (Turkey).
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
138
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
Austria
Belgium
Bulgaria
Canada
Croatia
Czech
Republic
Denmark
Estonia
Finland
France
Germany
Greece
Hungary
Iceland
Ireland
Italy
Latvia
Lithuania
Luxembourg
Country
4.40
4.15
4.12
4.00
4.12
3.90
4.21
4.12
4.00
3.91
4.17
4.08
3.93
3.99
4.24
3.80
4.00
4.10
3.97
3,073
2,539
2,645
3,811
3,658
1,900
1,910
6,091
2,251
2,206
2,339
3,209
1,790
M
2,524
4,268
2,524
2,896
2,933
2,745
Na
0.67
0.76
0.69
0.76
0.72
0.76
0.79
0.81
0.67
0.77
0.76
0.83
0.83
0.69
0.76
0.82
0.79
0.70
0.75
SD
Psychosomatic
complaints sum
score
Table 1. Sample Characteristics
22.7
31.1
24.5
36.3
21.5
37.3
32.8
31.1
21.4
44.5
34.1
31.6
29.7
16.7
27.8
34.7
29.7
29.3
34.1
%
Multiple
recurrent
psychosomatic
complaints
52.6
52.9
52.4
53.1
50.4
53.7
52.6
51.3
52.8
51.9
53.2
49.7
52.3
51.2
50.4
51.9
54.8
51.8
51.0
%
Girls
18.1
17.3
14.8
14.2
14.9
9.7
14.6
14.5
12.3
8.8
22.6
17.6
13.6
14.2
13.7
12.0
18.3
7.6
14.6
%
Single
parent
14.6
10.0
11.0
21.6
12.5
10.6
9.1
11.1
13.2
15.5
15.0
15.3
16.6
9.9
19.7
9.5
16.4
10.8
18.0
%
Not
easy to
talk to
father
11.7
15.8
11.6
10.1
8.6
1.8
8.9
12.8
5.2
2.3
9.5
7.9
7.4
8.0
11.8
3.5
10.6
2.6
12.1
%
Not
easy to
talk to
mother
7.6
30.8
11.2
11.1
13.2
21.5
25.6
2.3
21.4
20.7
31.1
36.4
8.6
11.9
12.2
29.9
9.2
26.9
29.9
%
Low
FAS
8.9
26.6
9.2
14.9
14.7
24.8
8.7
5.7
8.7
9.4
23.5
29.5
15.5
16.9
13.7
15.1
17.1
8.8
6.0
%
0.955
0.883
0.959
0.961
0.947
0.942
0.879
0.969
0.965
0.951
0.866
0.870
0.960
0.955
0.953
0.840
0.966
0.871
0.903
HDI
(continued)
Being
bullied
139
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
4.29
4.23
4.07
3.98
3.59
3.85
4.25
4.09
4.08
2,993
4,674
2,105
2,475
2,501
2,418
3,016
7,997
98,773
0.68
0.77
0.72
0.72
0.83
0.77
0.73
0.75
0.77
0.65
0.69
0.79
0.63
0.86
0.85
SD
19.9
30.9
24.3
26.8
63.8
38.5
30.0
27.0
30.4
21.2
25.3
34.3
21.6
39.4
37.7
%
Multiple
recurrent
psychosomatic
complaints
51.9
52.9
51.5
51.9
49.9
55.1
51.9
53.7
52.2
51.7
50.8
52.6
52.7
52.9
53.1
%
Girls
9.4
10.9
13.3
11.3
10.7
16.7
4.9
18.0
14.6
12.8
14.6
10.9
10.1
38.5
21.3
%
Single
parent
6.6
12.1
9.8
14.8
13.7
8.9
9.2
14.9
12.8
6.7
14.2
7.5
16.5
5.1
14.0
%
Not
easy to
talk to
father
4.2
3.7
11.7
7.8
1.0
7.3
1.3
11.5
8.0
6.5
9.3
2.9
5.8
1.4
11.0
%
Not
easy to
talk to
mother
Note: FAS = Family Affluence Scale; HDI = Human Development Index.
a. Because of a large number of missing variables, the total N does not match the total population of 11- and 13-year-olds in each country.
4.39
4.14
4.05
4.41
3.94
4.05
2,283
1,860
2,849
1,904
2,179
3,870
Netherlands
Norway
Poland
Portugal
Romania
Russian
Federation
Slovenia
Spain
Sweden
Switzerland
Turkey
Ukraine
Macedonia
UK
Total
M
Na
Psychosomatic
complaints sum
score
Country
Table 1. (continued)
10.2
13.9
5.7
8.8
70.6
55.4
38.4
12.5
21.1
7.5
3.0
31.5
24.3
41.3
48.7
%
Low
FAS
10.2
4.9
4.3
12.9
29.6
22.4
10.2
10.6
13.6
9.8
9.5
10.7
15.9
20.3
19.1
%
Being
bullied
0.929
0.955
0.963
0.960
0.806
0.796
0.817
0.947
0.964
0.971
0.880
0.909
0.837
0.817
HDI
140
Journal of Early Adolescence 32(1)
Table 2 shows that a higher age, being a girl, not living with two original
parents (i.e., living with a single parent, step parent, or other parent), lack of
easy communication with one’s father or mother, low FAS, and high media
use were each associated with more psychosomatic complaints. Regarding
the school domain, a negative class climate, higher school pressure, and having been bullied two to three times or more in the past couple of months were
associated with more psychosomatic complaints.
Statistically significant interactions with gender were observed for the
Close Friends Index, the Class Climate Index, the School Pressure Index, and
the Media Use Index. Without consideration of interactions, girls have a
fixed effect coefficient of -.129, t(33) = 4.41, p < .001. When interaction
terms are entered into the model, the decreasing effects of poor close friends
relationships (-.012 [t(33) = -2.23, p = .033]), a negative class climate (-.031
[t(33) = -5.34, p < .001]), high school pressure (-.087 [t(33) = -11.25, p <
.001]), and a high media use (-.019 [t(33) = -4.99, p < .001]) on psychosomatic health were more pronounced for girls (indicated by negative interaction coefficients) which add to the decreasing effects of these predictors.
The results further show an association between cross-country variation and
HDI, wherein an increase of 0.1 HDI points was associated with an increase of
0.12 points on the Psychosomatic Complaints Index (indicating better health).
Moreover, we tested if the individual-level predictors contribute to individual psychosomatic health in the same way for countries with different
HDI. Within the multilevel regression model, it was tested if the cross-country
variability in individual-level predictors could be explained by HDI. Results
showed a significant covariation of individual-level coefficients with HDI
only for age and media use. The decreasing effect of higher age (-.017
[t(32) = -4.80, p < .01]) was less pronounced in countries with high HDI (age
on HDI: 0.131 [t(32) = 2.64, p < .05]). The decreasing effect of media use
(-.037 [t(32) = -16.41, p < .01]) on the other hand was more pronounced in
countries with high HDI (media use on HDI: -.163 [t(32) = -6.07, p < .001]).
Table 3 shows the results of the multilevel logistic regression of the effect
of the covariates shown in Table 2 on multiple recurrent psychosomatic complaints (at least two symptoms more than once a week).
Being a girl, not living with two original parents (i.e., living with a single
parent, step parent, or other parents), communication with father or mother
not being easy, low FAS, and high media use were all associated with an
increased risk (OR) of multiple recurrent health complaints. In the school
context, negative class climate, school pressure, and having been bullied 2 to
3 times or more in the past couple of months were also associated with an
increased risk of multiple recurrent health complaints. Frequent physical
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
141
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
4.090917
1.213696
−0.017491
0.131420
−0.129116
−0.058893
−0.102807
−0.142575
−0.169454
−0.159238
−0.028069
0.017354
−0.310972
0.004984
−0.084543
−0.205613
0.005001
−0.036813
−0.163414
−0.011629
−0.031254
−0.087051
−0.019656
0.032476
0.286952
0.003643
0.049817
0.029253
0.005937
0.008890
0.018837
0.0006609
0.011389
0.009145
0.005612
0.011714
0.006436
0.008174
0.011488
0.001655
0.002244
0.026913
0.005222
0.005857
0.007738
0.003936
Standard error of fixed
effect coefficient
.000
.000
.000
.013
.000
.000
.000
.000
.000
.000
.005
.004
.000
.444
.000
.000
.005
.000
.000
.033
.000
.000
.000
p value for fixed
effect coefficient
3.737
−0.053
−0.107
−0.100
−0.171
−0.299
−0.223
−0.266
−0.103
−0.030
−0.426
−0.056
−0.166
−0.329
−0.010
−0.057
−0.050
−0.073
−0.153
−0.056
<.001
<.001
>.500
>.500
.331
.143
.013
<.001
.003
.413
<.001
<.001
<.001
<.001
<.001
.028
>.500
.401
.103
.001
0.061
0.001
0.273
0.001
0.002
0.012
0.001
0.006
0.003
0.001
0.007
0.002
0.003
0.007
0.000
0.000
0.001
0.001
0.002
0.001
VPC
95% CI
lower-bound
country variation
p cross-country
variation
4.445
0.018
0.365
−0.018
−0.035
0.014
−0.115
−0.052
0.047
0.065
−0.196
0.066
−0.003
−0.083
0.020
−0.170
0.027
0.010
−0.021
0.017
95% CI
upper-bound
country variation
Note: VPC = variance partitioning coefficient; CI = confidence interval; HDI = Human Development Index; FAS = Family Affluence Scale. The reference group was
defined as being male, with a mean age of 12.64 years, living with two original parents, reporting easy communication with father/mother, with high FAS, not having been
bullied more than 1 to 2 times in the past couple of months, and with a mean value in the Close Friend Index, Class Climate Index, School Pressure Index, Media Use
Weekend Index, and Physical Activity.
Intercept
Intercept on HDI
Age
Age on HDI
Girl
Single parent
Stepparent
Other parent
Not easy to talk to father
Not easy to talk to mother
Low FAS
Medium FAS
Being bullied
Close Friend Index
Class Climate Index
School Pressure Index
Physical activity (days)
Media Use Index
Media Use Index on HDI
Girls × Close Friend Index
Girls × Class Climate Index
Girls × School Pressure Index
Girls × Media Use Index
Fixed effect
coefficient
Table 2. Multilevel Linear Regression of Familial, Peer-, and School-Related Social Determinants and Individual- and Country-Level
Socioeconomic Status on Psychosomatic Complaints Mean Item Score
142
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
0.288
0.015
0.980
0.691
1.140
1.220
1.365
1.471
1.625
1.426
1.248
1.013
2.078
1.010
1.235
1.793
0.981
1.125
1.980
1.022
1.053
1.116
1.018
Intercept
Intercept on HDI
Age
Age on HDI
Girl
Single parent
Stepparent
Other parent
Not easy to talk to father
Not easy to talk to mother
Low FAS
Medium FAS
Being bullied
Close Friend Index
Class climate Index
School Pressure Index
Physical activity (days)
Media Use Index
Media Use Index on HDI
Girls × Close Friend Index
Girls × Class Climate Index
Girls × School Pressure Index
Girls × Media Use Index
0.240
0.002
0.956
0.505
0.907
1.183
1.293
1.314
1.563
1.343
1.178
0.969
1.951
0.972
1.183
1.664
0.971
1.109
1.691
0.977
1.015
1.060
0.999
95% CI
lower-bound
0.346
0.101
1.004
0.944
1.432
1.258
1.440
1.645
1.691
1.513
1.321
1.059
2.213
1.049
1.289
1.932
0.991
1.140
2.318
1.070
1.092
1.175
1.037
95% CI
upper-bound
<.001
<.001
.095
.022
.253
<.001
<.001
<.001
<.001
<.001
<.001
.562
<.001
.609
<.001
<.001
<.001
<.001
<.001
.331
.008
<.001
.066
p value
1.090
2.752
1.327
1.672
2.076
1.903
1.663
1.568
1.198
2.784
1.163
1.504
2.633
1.018
1.186
1.182
1.223
1.361
1.089
0.472
1.121
1.114
1.042
1.389
1.222
0.993
0.857
1.551
0.877
1.014
1.221
0.945
1.066
0.884
0.906
0.915
0.952
0.715
95% CI
upper-bound
0.881
0.116
95% CI
lower-bound
Country variation
<.001
<.001
.253
>.500
>.500
.398
.015
.003
.072
.052
.001
.042
.002
0
.074
.197
.176
>.500
.217
.387
p value
Note: CI = confidence interval; HDI = Human Development Index; FAS = Family Affluence Scale. The reference group was defined as being male, with a mean age of 12.64
years, living with two original parents, reporting easy communication with father/mother, with high FAS, not having been bullied more than 1 to 2 times in the past couple
of months, and with a mean value in the Close Friend Index, Class Climate Index, School Pressure Index, Media Use Weekend Index, and physical activity.
OR
Fixed effects
Table 3. Multilevel Logistic Regression of Multiple Recurrent Psychosomatic Complaints on Familial, Peer-, and School-Related
Social Determinants and Individual- and Country-Level Socioeconomic Status
143
Ottova et al.
activity was associated with a lower risk for multiple recurrent health complaints (OR = 0.98, p < .001).
The risk for multiple recurrent health complaints associated with a poor
Close Friends Index (OR = 1.02, p = .331), negative class climate (OR = 1.05,
p < .01), high school pressure (OR = 1.12, p < .001), and high media use (OR =
1.02, p = .066) was more pronounced for girls as indicated by the additional
OR for the interaction terms.
A statistically significant cross-national variability was observed in the
percentage of young adolescents suffering from multiple recurrent health
complaints. This cross-national variability was associated with the HDI, and
a higher HDI was associated with a lower risk for multiple recurrent health
complaints.
Finally, we also tested if the inclusion of the control variables (physical
activity and Media Use Index) affected the estimation of the other coefficients, but we only found marginal differences.
Discussion
The purpose of this study was to investigate the relationship between both
individual-level (family, peers, school) and country-level factors with psychosomatic complaints in a large international sample of young adolescents. We
found that not living with both original parents, having difficulty talking with
either parent, being bullied 2 to 3 times or more in the past couple of months,
and experiencing high academic pressure at school was associated with poorer
psychosomatic health and a higher chance for multiple recurrent psychosomatic complaints. The next section will begin by addressing the hypotheses
formulated in the introduction and connecting these to the results found in this
study. Following a discussion of the main findings in light of past research in
this field, strengths and limitations of this study will be pointed out. The discussion concludes with implications for policy and practice.
As hypothesized, a stable family structure, good communication with
one’s parents, good peer relationships, absence of negative social interactions
(bullying), and a positive school environment (good class climate, low
schoolwork pressure) were all associated with better psychosomatic health.
At the country level, a positive relationship between HDI and young people’s psychosomatic health emerged as well as significant covariation
between the HDI and age, and between HDI and media use. Specifically,
decreasing effects of higher age were less pronounced in countries with
high HDI, while, for media use, the opposite was true. Our results
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
144
Journal of Early Adolescence 32(1)
confirmed the fourth hypothesis showing that countries with a high HDI
score had generally lower rates of psychosomatic complaints. By testing a
multidimensional country-level indicator—HDI—in a multinational data
set, our study makes an important contribution to research on young adolescents’ psychosomatic health.
From the stress perspective, psychosomatic complaints are heightened by
experiences that induce stress and are possibly relieved by protective factors.
In general, our results showed that most of the variance in psychosomatic
complaints was due to individual-level stressors found in all investigated
domains (family, peers, school) as well as the larger socioeconomic context.
These findings are consistent with other studies that also stressed the importance of family structure and child-parental relationships during adolescence
(Brolin Laftman & Östberg, 2006; Due, Lynch, Holstein, & Modvig, 2003).
Surprisingly, positive social relations did not have a significant effect in this
study, although previous research has shown that the quality of friendships
and supportive relationships are important for adolescent well-being (Rubin
et al., 2004). Negative peer relationships however did have an impact: adolescents who were frequently bullied were more likely to report multiple
recurrent psychosomatic complaints than those who were not bullied. This
negative effect of bullying on individual well-being is consistent with other
studies on this subject (Alfven, Őstberg, & Hjern, 2008; Ghandour et al.,
2004; Richter et al., 2007).
Consistent with previous studies, negative class climate and school pressure were both significantly associated with psychosomatic complaints
(Torsheim & Wold, 2001). Too much pressure from schoolwork can have a
negative effect, especially if there is too little free time to relax and participate in activities that benefit well-being (Gaspar, 2010). Consistent with previous research, being a girl was associated with a higher risk for psychosomatic
complaints (Cavallo et al., 2006). Different socialization strategies, as well as
higher attentiveness to their body (Verbruggev, 1982), may explain some of
the gender differences in complaint or symptom reporting. In fact, in many
cultures, the expression of physical and emotional states is more acceptable
for girls than for boys (Gijbers van Wijk, van Vilet, Kolk, & Everaerd, 1991).
In this study, age did not have significant effects, but this could be due to the
narrow age span under investigation (11- and 13-year-olds only).
After controlling for individual-level factors, an effect of national characteristics (HDI) still remained. This demonstrates that individual psychosomatic complaints are also predicted by the social state of one’s country.
People living in less privileged countries report more psychosomatic
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
145
Ottova et al.
complaints even after controlling for other stressors. Previous research on
this subject has shown the effect of certain country-level indicators, such as
the Gini and GDP (Holstein et al., 2009; Olsen & Dahl, 2007; Torsheim et
al., 2006). In this study, we looked at the effect of the HDI to see whether it
would explain more of the variance in the cross-national differences in psychosomatic complaints. Indeed, an effect of HDI was identified, and our
results further show that the decreasing effect of higher age on psychosomatic complaints was lower in countries with a high HDI, whereas high
media use had a greater decreasing effect on psychosomatic complaints in
countries with a high HDI. Thus, as adolescents get older, they tend to
experience more psychosomatic complaints, but this effect is most pronounced in countries with low HDI. We may only speculate why this is the
case. Perhaps, the effect of living in a low-HDI country is felt most by older
adolescents as they tend to be more aware of the social and socioeconomic
context in which they live. With regard to media use, the effect we found
may be due to another third variable which is related to HDI and psychosomatic complaints.
The prevalence of psychosomatic complaints varied greatly between
countries, whereby it was lowest in Austria (17%) and highest in Turkey
(64%). The reason for these differences is not clear but it can be assumed that
cultural issues play a part—an aspect that could be clarified by further qualitative research. Although psychosomatic complaints occur in all cultures, not
all cultures may be accepting of psychological symptoms and mental illness
(Araya, Rojas, Fritsch, Acuna, & Lewis, 2001). Cultural factors may influence the way a somatic complaint is described, expressed, and categorized.
The key difference is the nature of the social adversities and the types of illness models that might predispose individuals to or precipitate these complaints (Patel & Sumathipala, 2006). The HBSC-SCL is sensitive to cross-cultural
comparisons, and previous analyses have identified cross-cultural variations in psychosomatic complaints among 35 countries (Ravens-Sieberer
et al., 2008).
Strengths and Limitations
This study has several strengths that deserve attention. First, we used data
from a large international survey (HBSC) that is carried out every 4 years
and provides valuable information on health status and health behavior in a
large sample of children. Second, obtaining data from children themselves
through self report is advantageous when exploring subjective constructs
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
146
Journal of Early Adolescence 32(1)
such as well-being. Third, country-level indicators such as HDI have been
rarely used in cross-sectional analyses and can be useful in further explaining
cross-country patterns. We consider the use of the HDI a strength of this
study as it introduces an innovative aspect in research on child health. Even
though it has been criticized (McGillivray, 1991), it adds a different perspective in the interpretation of country differences as it measures human development along three dimensions (knowledge, healthy life, standard of living;
Sagar & Najam, 1998). In this way, it broadens the discussion and contributes to research.
At the same time, there are also some limitations that need to be stressed.
The use of single-item scales for capturing various health behaviors (e.g.,
physical activity, sedentary behavior), as well as for the other contexts (family, school, peers), posed some challenges. Generally, multiple item measures are preferred as they are more reliable and this is important in statistical
analyses when errors of variance need to be minimized (Loo, 2002). However,
most of the single-item measures have long-standing use in HBSC and many
have been subjected to validation studies (Prochaska et al., 2001). The crosssectional nature of HBSC prohibits any conclusions about causal relationships. Last, we did not control for chronic diseases or for emotional disturbances
(such as anxiety, depressive mood). Although most of the complaints during
adolescence have no organic cause (Taylor, Szatmari, Boyle, & Offord,
1996), lack of information on chronic diseases and emotional disturbances,
both of which have somatic and emotional correlates, may be considered a
weakness. Chronic disease may be a stressor in itself (Compas & Boyer,
2001) and thus could have an indirect effect on complaints.
Implications of the Findings and Future Directions
Given the fact that psychosomatic complaints are higher in girls and are
associated with the HDI of a country, one implication of this study is that
gender-specific health promotion programs should be adapted to the cultural
context within each country. The availability and use of media should be
taken into greater consideration in programs targeting health behavior modifications as media use emerged as an important factor associated with health
complaints—particularly in girls—and moreover had a more negative effect
in countries with high HDI. Thus, programs in countries with higher HDI
should take media use into greater account.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
147
Ottova et al.
Appendix A
Level 1 Model
Psychosomatic complaint sij = β0 j + β1 j age + β2 j girl
+ β3 j singleparent + β 4 j stepparent + β5 j otherparent + β6 j talkfather
+ β7 j talkmother + β8 j lowFAS + β9 j medFAS + β10 j bullied
+ β11 j closefriend + β12 j classclimate + β13 j schoolpressure
+ β14 j physicalactivity + β15 j mediause + β16 j Girl × Closefriend
+ β17 j Girl × Classclimate + β18 j Girl × Schoolpressure
+ β19 j Girl × Mediause + eij
Level 2 Model
β0 j = γ 00 + γ 01 HDI + γ 0 j
β1 j = γ10 + γ11 HDI + γ1 j
β15 j = γ150 + γ151 HDI + γ15 j
For the model with continuous outcome, the link function is identical. For the
model with dichotomous outcome, the link function is as follows:
 pij 
ηij = log 

1 - pij 
where pij is the probability of recurrent psychosomatic health complaints. For
the model with dichotomous outcome, ηij replaces Psychosomaticcomplaintsij in
the Level 1 formula.
Appendix B
Authors’ Notes
The members of the Positive Health Focus Group are as follows: Ulrike RavensSieberer (DE; coordinator), Torbjørn Torsheim (NO), Bogdana Alexandrova (BG),
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
148
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
N
2,524
4,268
2,524
2,896
2,933
2,745
3,073
2,539
2,645
3,811
3,658
1,900
1,910
6,091
2,251
2,206
2,339
3,209
1,790
Country
Austria
Belgium
Bulgaria
Canada
Croatia
Czech Republic
Denmark
Estonia
Finland
France
Germany
Greece
Hungary
Iceland
Ireland
Italy
Latvia
Lithuania
Luxembourg
a
Sample Characteristics
1.2
1.3
1.8
3.1
0.9
2.0
3.1
1.6
1.0
1.3
1.0
1.0
1.6
1.9
1.3
1.7
2.6
2.1
1.6
%
Stepparent
43.7
41.5
52.3
37.9
48.9
46.8
39.8
43.6
44.2
37.8
39.0
48.3
47.5
25.4
57.4
45.4
46.9
47.1
31.8
%
Other
parent
25.1
35.6
25
31.1
24.9
34.7
30.1
22.9
26.1
40.4
28.6
31.4
20.6
21.6
25.9
35.4
28.6
30.8
34.4
%
FAS
med
3.18
3.50
3.42
3.54
3.65
3.60
3.45
3.28
3.08
3.45
3.16
2.94
3.60
3.44
3.57
3.40
3.36
3.47
3.30
M
0.74
0.71
0.76
0.70
0.65
0.66
0.73
0.79
0.79
0.74
0.74
0.79
0.63
0.73
0.69
0.80
0.80
0.77
0.78
SD
Close Friend
Index
1.91
1.93
2.54
2.30
2.03
2.36
1.87
2.14
2.16
2.23
1.82
2.20
2.18
1.97
2.01
1.94
2.46
2.33
1.86
M
0.69
0.78
0.90
0.76
0.70
0.78
0.73
0.85
0.74
0.80
0.67
0.77
0.78
0.75
0.71
0.76
0.89
0.89
0.73
SD
Class
Climate
Index
4.52
4.07
4.22
4.76
4.65
4.18
4.62
4.11
4.96
3.90
4.19
4.08
4.19
4.50
5.21
3.91
4.37
4.06
3.88
M
1.90
2.09
2.36
1.91
1.93
2.12
1.95
2.02
1.89
1.94
1.93
2.05
2.04
1.98
1.81
2.04
1.99
2.08
1.94
SD
Physical
activity
4.22
4.19
4.70
4.20
3.82
3.73
3.88
4.55
3.77
3.91
3.98
3.89
4.39
3.84
3.19
3.24
4.15
3.70
4.03
M
1.77
1.67
1.89
1.65
1.54
1.50
1.43
1.60
1.37
1.55
1.62
1.48
1.74
1.39
1.27
1.32
1.59
1.44
1.62
SD
Media Use
Index
0.57
0.64
0.67
0.66
0.61
0.63
0.65
0.59
0.59
0.65
0.55
0.62
0.61
0.69
0.68
0.64
0.59
0.61
0.64
SD
(continued)
2.03
2.14
2.04
2.16
1.82
2.31
2.10
2.26
2.24
2.20
2.24
1.92
2.20
2.19
2.00
2.37
2.28
2.35
2.14
M
School
Pressure
Index
Fiona Brooks (GB), Antony Morgan (GB), Cath Fenton (GB), Kädi Lepp (EE), Raili Välimäa (FI), Céline Vignes (FR), Mariane
149
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
2,283
1,860
2,849
1,904
2,179
3,870
2,993
4,674
2,105
2,475
2,501
2,418
3,016
7,997
98,773
N
0.3
1.8
1.5
1.4
2.4
1.4
1.3
1.2
0.2
0.5
2.5
1.9
1.4
1.4
1.6
%
Stepparent
38.9
28.0
43.6
44.1
43.6
39.7
43.8
46.1
38.2
41.5
24.2
37.7
44.6
37.0
40.9
%
Other
parent
16.4
27.3
19.4
32.7
19.5
26.3
15.8
27.1
20.4
31.9
37.2
21.5
19.0
30.0
27.3
%
FAS
med
3.38
3.53
3.12
3.61
3.43
3.30
3.33
3.60
3.44
3.27
3.37
3.47
3.42
3.60
3.42
M
0.74
0.69
0.92
0.66
0.77
0.81
0.83
0.67
0.72
0.78
0.81
0.71
0.79
0.65
0.76
SD
Close Friend
Index
1.87
1.78
2.23
1.80
2.03
2.36
1.97
1.88
1.87
1.87
2.23
2.32
1.80
2.03
2.07
M
0.65
0.68
0.76
0.66
0.69
0.85
0.69
0.72
0.67
0.64
0.78
0.81
0.67
0.79
0.78
SD
Class
Climate
Index
4.54
4.33
4.18
3.66
3.73
3.42
4.15
4.11
4.37
3.81
3.96
4.32
4.37
4.54
4.25
M
1.94
1.91
2.01
2.01
2.17
2.13
1.98
2.16
1.92
1.95
2.26
2.07
2.05
1.86
2.04
SD
Physical
activity
4.28
3.83
4.57
4.45
4.42
4.02
3.90
3.65
3.79
3.43
3.62
3.86
4.39
4.07
3.98
M
1.57
1.37
1.75
1.75
1.86
1.58
1.47
1.44
1.42
1.44
1.71
1.46
1.81
1.60
1.60
SD
Media Use
Index
Note: FAS = Family Affluence Scale
a. Because of a large number of missing variables, the total N does not match the total population of 11- and 13-year-olds in each country.
Netherlands
Norway
Poland
Portugal
Romania
Russian Federation
Slovenia
Spain
Sweden
Switzerland
Turkey
Ukraine
Macedonia
United Kingdom
Total
Country
a
Appendix B (continued)
2.04
2.10
2.26
2.34
3.43
3.31
3.33
3.61
3.44
3.27
3.37
3.47
3.42
3.60
3.42
M
0.55
0.62
0.61
0.64
0.58
0.63
0.59
0.67
0.59
0.61
0.71
0.61
0.61
0.65
0.64
SD
School
Pressure
Index
150
Journal of Early Adolescence 32(1)
Sentenac (FR), Veronika Ottova (DE), Christina Schnohr (GL), Gyöngyi Kökönyei
(HU), Kjartan Unak (IS), Franco Cavallo (IT), Inese Gobina (LV), Wilma
Vollebergh (NL), Saskia van Dorsselaer (NL), Jørn Hetland (NO), Joanna Mazur
(PL), Tania Gaspar (PT), Viorel Mih (RO), Aurora Szentagotai (RO), Eva Kallay
(RO), Andrea Geckova (SK), Zuzana Katreniakova (SK), Helena Jericek (SI), Eva
Stergar (SI), Vesna Pucelj (SI), Pilar Ramos (ES), Mia Danielson (SE), Lilly
Eriksson (SE), Mujgan Alikasifoglu (TR), Ethem Erginoz (TR).
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research,
authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research,
authorship, and/or publication of this article: Regarding the research, the authors are
grateful for the financial support offered towards the data collection by the national
government ministries, research foundations and other funding bodies in the participating countries.
References
Alfven, G., Őstberg, V., & Hjern, A. (2008). Stressors, perceived stress and recurrent
pain in Swedish schoolchildren. Journal of Psychosomatic Research, 65, 381-387.
Araya, R., Rojas, G., Fritsch, R., Acuna, J., & Lewis, G. (2001). Common mental disorders in Santiago, Chile: Prevalence and socio-demographic correlates. British
Journal of Psychiatry, 178, 228-233.
Bakoula, C., Kapi, A., Veltsista, A., Kavadias, G., & Kolaitis, G. (2006). Prevalence
of recurrent complaints of pain among Greek schoolchildren and associated factors: A population-based study. Acta Paediatrica, 95, 947-951.
Bandell-Hoekstra, I. E., Abu-Saad, H. H., Passchier, J., Frederiks, C. M., Feron, F. J.,
& Knipschild, P. (2001). Prevalence and characteristics of headache in Dutch
schoolchildren. European Journal of Pain, 5, 145-153.
Berntsson, L. T., & Köhler, L. (2001). Long-term illness and psychosomatic complaints in children aged 2-17 years in the five Nordic countries. European Journal
of Public Health, 11(1), 35-42.
Berntsson, L. T., Köhler, L., & Gustafsson, J.-E. (2001). Psychosomatic complaints
in schoolchildren: A Nordic comparison. Scandinavian Journal of Public Health,
29, 44-54.
Brill, S. R., Patel, D. R., & MacDonald, E. (2001). Psychosomatic disorders in pediatrics. Indian Journal of Pediatrics, 68, 597-603.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
151
Ottova et al.
Brolin Laftman, S. B., & Östberg, V. (2006). The pros and cons of social relations:
An analysis of adolescents’ health complaints. Social Science & Medicine, 63,
611-623.
Brosschot, J. F. (2002). Cognitive-emotional sensitization and somatic health complaints. Scandinavian Journal of Psychology, 43, 113-121.
Cavallo, F., Zambon, A., Borraccino, A., Ravens-Sieberer, U., Torsheim, T., Lemma, P.,
& The HBSC Positive Health Group. (2006). Girls growing through adolescence
have a higher risk of poor health. Quality of Life Research, 15, 1577-1585.
Cheung, K. C., & Keeves, J. P. (1990). Hierarchical linear modelling. International
Journal of Educational Research, 14, 289-297.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ:
Lawrence Erlbaum.
Compas, E. B., & Boyer, M. C. (2001). Coping and attention: Implications for child
health and pediatric conditions. Developmental and Behavioral Pediatrics, 22,
323-333.
Currie, C., Nic Gabhainn, S., Godeau, E., Roberts, C., Smith, R., Currie, D., . . .
Barnekow, V. (Eds.). (2008a). Inequalities in young people’s health: HBSC international report from the 2005/2006 Survey. Copenhagen, Denmark: WHO Regional
Office for Europe.
Currie, C., Molcho, M., Boyce, W., Holstein, B. E., Torsheim, T., & Richter, M.
(2008b). Researching health inequalities in adolescents: The development of the
Health Behaviour in School-aged Children (HBSC) Family Affluence Scale. Social
Science & Medicine, 66, 1429-1436.
Ditton, H. (1998). Mehrebenenanalyse. Grundlagen und Anwendungen des hierarchischen linearen Modells [Multilevel analysis: Fundamentals and applications
of hierarchical linear model]. Weinheim, Germany: Juventa.
Due, P., Lynch, B., Holstein, B., & Modvig, J. (2003). Socioeconomic health inequalities among a nationally representative sample of Danish adolescents: The role
of different types of social relations. Journal of Epidemiology and Community
Health, 57, 692-698.
Eccles, J. S., & Midgley, C. (1990). Changes in academic motivation and self-perception
during early adolescence. In R. Montemayor, G. R. Adams, & T. P. Gullotta
(Eds.), From childhood to adolescence: A transitional period? (pp. 134-155).
Newbury Park, CA: SAGE.
Eccles, J. S., Lord, S., & Midgley, C. (1991). What are we doing to early adolescents?
The impact of educational contexts on early adolescents. American Journal of
Education, 99, 521-542.
Egle, U. T., Hoffmann, S. O., & Steffens, M. (1997). Psychosoziale Risiko—und
Schutzfaktoren in Kindheit und Jugend als Prädisposition für psychische Störungen
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
152
Journal of Early Adolescence 32(1)
im Erwachsenenalter. Gegenwärtiger Stand der Forschung. Nervenarzt, 68,
683-695.
Eriksson, U., & Sellström, E. (2010). School demands and subjective health complaints among Swedish schoolchildren: A multilevel study. Scandinavian Journal
of Public Health, 38, 344-350.
Gaspar, T. (2010). Health-related quality of life in children and adolescents: Personal and social factors that promote quality of life. Saarbrücken, Germany: LAP
LAMBERT Academic Publishing.
Gaspar, T., Matos, M., Ribeiro, J., Leal, I., & Ferreira, A. (2009). Health-related quality of life in children and adolescents and associated factors. Journal of Cognitive
and Behavioral Psychotherapies, 9(1), 33-48.
Gerber, M., & Pühse, U. (2008). “Don’t crack under pressure!”—Do leisure time
physical activity and self-esteem moderate the relationship between school-based
stress and psychosomatic complaints? Journal of Psychosomatic Research, 65,
363-369.
Ghandour, R. M., Overpeck, M. D., Huang, Z. J., Kogan, M. D., & Scheidt, P. C.
(2004). Headache, stomachache, backache, and morning fatigue among adolescent
girls in the United States. Archives of Pediatrics Adolescent Medicine, 158, 707-803.
Gielen, U. P., & Roopnarine, J. (2004). Childhood and adolescence: Cross-cultural
perspectives and applications. Westport, CT: Praeger.
Gijbers van Wijk, C. M., van Vilet, K. P., Kolk, A. M., & Everaerd, W. T. (1991).
Symptom sensitivity and sex differences in physical morbidity: A review of health
survey in the United States and the Netherlands. Women Health, 17, 91-124.
Gini, G. (2008). Associations between bullying behaviour, psychosomatic complaints,
emotional and behavioural problems. Journal of Paediatrics and Child Health, 44,
492-497.
Hagquist, C. (2009). Psychosomatic health problems among adolescents in Sweden—
Are the time trends gender related? European Journal of Public Health, 19, 331-336.
Hakala, P., Rimpelä, R., Salminen, J. J., Virtanen, S. M., & Rimpelä, M. (2002).
Back, neck, and shoulder pain in Finnish adolescents: National cross sectional
surveys. British Medical Journal, 325, 743-745.
Hakala, P. T., Rimpelä, A. H., Saarni, L. A., & Salminen, J. J. (2006). Frequent computer-related activities increase the risk of neck-shoulder and low back pain in
adolescents. European Journal of Public Health, 16, 536-541.
Haugland, S., Wold, B., Stevenson, J., Aaroe, L. E., & Woynarowska, B. (2001).
Subjective health complaints in adolescence. A cross-national comparison of
prevalence and dimensionality. European Journal of Public Health, 11(1), 4-10.
Haugland, S., Wold, B., & Torsheim, T. (2003). Relieving the pressure? The role of
physical activity in the relationship between school-related stress and adolescent
health complaints. Research Quarterly for Exercise & Sport, 74(2), 127-135.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
153
Ottova et al.
Haukka, E., Leino-Arjas, P., Ojajärvi, A., Takala, E. P., Viikari-Juntura, E., &
Riihimäki, H. (2010). Mental stress and psychosocial factors at work in relation to multiple-site musculoskeletal pain: A longitudinal study of kitchen
workers. Advance online publication. European Journal of Pain. doi:10.1016/j.
ejpain.2010.09.005
Hjern, A., Alfven, G., & Östberg, V. (2008). School stressors, psychological complaints and psychosomatic pain. Acta Paediatrica, 97, 112-117.
Holler-Nowitzki, B. (1994). Psychosomatische Beschwerden im Jugendalter. Schulische Belastungen, Zukunftsangst und Streßreaktionen [Psychosomatic symptoms
in adolescence: Curricular pressures, future anxiety and stress reactions]. Weinheim,
Germany: Juventa.
Holstein, B. E., Currie, C., Boyce, W., Damsgaard, M. T., Gobina, I., Kökönyei, G., . . .
HBSC Social Inequalities Focus Group. (2009). Socio-economic inequality in
multiple health complaints among adolescents: International comparative study
in 37 countries. International Journal of Public Health, 54, 260-270.
Hurrelmann, K., Engel, U., Holler, B., & Nordlohne, E. (1988). Failure in school,
family conflicts, and psychosomatic disorders in adolescence. Journal of Adolescence, 11, 237-249.
Jellesma, F. C., Rieffe, C., Terwogt, M. M., & Kneepkens, C. M. F. (2006). Somatic
complaints and health care use in children: Mood, emotion awareness and sense
of coherence. Social Science & Medicine, 63, 2640-2648.
Karademas, E. C., Peppa, N., Fotiou, A., & Kokkevi, A. (2008). Family, school and
health in children and adolescents. Journal of Health Psychology, 13, 1012.
Karvonen, S., Vikat, A., & Rimpelä, M. (2005). The role of school context in the
increase in young people’s health complaints in Finland. Journal of Adolescence,
28, 1-16.
Loo, R. (2002). A caveat on using single-item versus multiple-item scales. Journal of
Managerial Psychology, 17(1), 68-75.
Marschall, P. (1989). Self-report and stability of physical symptoms by adolescents.
Adolescence, 24, 209-216.
McGillivray, M. (1991). The Human Development Index: Yet another redundant
composite development indicator? World Development, 19, 1461-1468.
Moon, M. J., Welch, E. W., & Wong, W. (2005). What drives global e-governance?
An exploratory study at a macro level. Proceedings of the 38th Hawaii International Conference on System Sciences, Waikoloa, Big Island, Hawaii.
Moreno, C., Sánchez-Queija, I., Muñoz-Tinoco, V., Gaspar de Matos, M., Dallago, L.,
Ter Bogt, T., . . . the HBSC Peer Culture Focus Group. (2009). Cross-national
associations between parent and peer communication and psychological complaints. International Journal of Public Health, 54, 235-242.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
154
Journal of Early Adolescence 32(1)
Natvig, G. K., Albrektsen, G., Anderssen, N., & Qvarntstrøm, U. (1999). Schoolrelated stress and psychosomatic symptoms among school adolescents. Journal of
School Health, 69, 362-368.
Natvig, G. K., Albrektsen, G., & Qvarnstrom, U. (2003). Associations between psychosocial factors and happiness among school adolescents. International Journal
of Nursing Practice, 9, 166-175.
Neumayer, E. (2001). The Human Development Index and sustainability—A constructive proposal. Ecological Economics, 39, 101-114.
Noorbakhsh, F. (1998). The Human Development Index: Some technical issues and
alternative indices. Journal of International Development, 10, 589-605.
Olsen, K. M., & Dahl, S.-Ǻ. (2007). Health differences between European countries.
Social Science & Medicine, 64, 1665-1678.
Olweus, D. (1996). Bullying at school: Knowledge base and an effective intervention
program. Annals of the New York Academy of Sciences, 794, 265-276.
Osen, E. M., Torsheim, T., & Wold, B. (2000). Skolerelatert stotte som moderator av
kjonnsforskjeller i subjective helseplager hos norsk skoleungdom [School-related
support moderating gender differences in subjective health complaints in Norwegian
students]. Tidsskrift for Norsk Psykologforening, 37, 900-907.
Östberg, V., Alfven, G., & Hjern, A. (2006). Living conditions and psychosomatic
complaints in Swedish schoolchildren. Acta Paediatrica, 95, 929-934.
Patel, V., & Sumathipala, A. (2006). Psychological approaches to somatisation in
developing countries. Advances in Psychiatric Treatment, 12, 54-62.
Piko, B. F. (2006). Satisfaction with life, psychosocial health and materialism among
Hungarian youth. Journal of Health Psychology, 11, 827.
Piko, B. F. (2007). Self-perceived health among adolescents: The role of gender and
psychosocial factors. European Journal of Pediatrics, 166, 701-708.
Piko, B. F., & Fitzpatrick, K. M. (2007). Socioeconomic status, psychosocial health
and health behaviours among Hungarian adolescents. European Journal of Public
Health, 17, 353-360.
Prochaska, J. J., Sallis, J. F., & Long, B. (2001). A physical activity screening measure for use with adolescents in primary care. Archives of Pediatrics & Adolescent
Medicine, 155, 554-559.
Ravens-Sieberer, U., Erhart, M., Torsheim, T., Hetland, J., Freeman, J., Danielson, M.,
. . . HBSC Positive Health Group. (2008). An international scoring system for
self-reported health complaints in adolescents. European Journal of Public
Health, 18, 294-299.
Ravens-Sieberer, U., Freeman, J., Kokonyei, G., Thomas, C. A., & Erhart, M. (2009).
School as determinant for health outcomes—A structural equation model analysis. Health Education, 109, 342-356.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
155
Ottova et al.
Richter, M., Bowles, D., Melzer, W., & Hurrelmann, K. (2007). Bullying, psychosoziale Gesundheit und Risikoverhalten im Jugendalter [Bullying, psychosocial
health and risk behaviour in adolescence]. Gesundheitswesen, 69, 475-482.
Roberts, C., Currie, C., Samdal, O., Currie, D., Smith, R., & Maes, L. (2007). Measuring the health and health behaviours of adolescents through cross-national survey
research: Recent developments in the Health Behaviour in School-aged Children
(HBSC) study. Journal of Public Health, 15, 179-186.
Roberts, C., Freeman, J., Samdal, O., Schnohr, C. W., de Looze, M. E., Nic Gabhainn, S.,
. . . International HBSC Study Group. (2009). The Health Behaviour in School-aged
Children (HBSC) study: Methodological developments and current tensions. International Journal of Public Health, 54, 140-150.
Rubin, K. H., Dwyer, K. M., Kim, A. H., Burgess, K. B., Booth-Laforce, C., &
Rose-Krasnor, L. (2004). Attachment, friendship, and psychosocial functioning
during early adolescence. Journal of Early Adolescence, 24, 326-356.
Sagar, A. D., & Najam, A. (1998). The Human Development Index: A critical review.
Ecological Economics, 25, 249-264.
Samdal, O., Nutbeam, D., Wold, B., & Kannas, L. (1998). Achieving health and educational goals through schools—A study of the importance of the school climate
and the students’ satisfaction with school. Health Education Research, 13, 383-397.
Santinello, M., Vieno, A., & De Vogli, R. (2009). Primary headache in Italian early
adolescents: The role of perceived teacher unfairness. Headache, 49, 366-374.
Schmitz, K. H., Harnack, L., Fluton, J. E., Jacobs, D. R., Jr., Gao, S., Lytle, L. A.,
& Van Coevering, P. (2004). Reliability and validity of a brief questionnaire to
assess television viewing and computer use by middle school children. Journal of
School Health, 74, 370-377.
Sleskova, M., Salonna, F., Madarasova Geckova, A., Nagyova, I., Stewart, R. E., van
Dijk, J. P., & Groothoff, J. W. (2006). Does parental unemployment affect adolescents’ health? Journal of Adolescence Health, 38, 527-535.
Straker, L. M., Coleman, J., Skoss, R., Maslen, B. A., Burgess-Limerick, R., & Pollock, C. M.
(2008). A comparison of posture and muscle activity during tablet computer, desktop
computer and paper use by young children. Ergonomics, 51, 540-555.
Sundblad, G. B., Jansson, A., Saartok, T., Renström, P., & Engström, L.-M. (2008).
Self-rated pain and perceived health in relation to stress and physical activity
among school-students: A 3-year follow-up. Pain, 136, 239-249.
Tanaka, H., Tamai, H., Terashima, S., Takenaka, Y., & Tanaka, T. (2000). Psychosocial factors affecting psychosomatic symptoms in Japanese schoolchildren. Pediatrics International, 42, 354-358.
Taylor, D. C., Szatmari, P., Boyle, M. H., & Offord, D. R. (1996). Somatization and
the vocabulary of everyday bodily experiences and concerns: A community study
of adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 491-499.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
156
Journal of Early Adolescence 32(1)
Torsheim, T., Currie, C., Boyce, W., Kalnins, I., Overpeck, M., & Haugland, S.
(2004). Material deprivation and self-rated health: A multi-level study of adolescents from 22 European and North American countries. Social Science & Medicine, 59, 1-12.
Torsheim, T., Currie, C., Boyce, W., & Samdal, O. (2006). Country material distribution and adolescents’ perceived health: Multilevel study of adolescents in 27
countries. Journal of Epidemiology & Community Health, 60, 156-161.
Torsheim, T., Eriksson, L., Schnohr, C. W., Hansen, F., Bjarnason, T., & Välimaa, R.
(2010). Screen-based activities and physical complaints among adolescents from
the Nordic countries. BMC Public Health, 10, 324.
Torsheim, T., & Wold, B. (2001). School-related stress, support, and subjective health
complaints among early adolescents: A multilevel approach. Journal of Adolescence, 24, 701-713.
Torsheim, T., Wold, B., & Samdal, O. (2000). The Teacher and Classmate Support
Scale: Factor structure, test-retest reliability and validity in samples of 13 and 15
year old adolescents. School Psychology International, 21, 195-212.
Verbruggev, L. M. (1982). Sex differentials in health. Public Health Report, 97,
417-437.
West, P., & Sweeting, H. (2003). Fifteen, female and stressed: Changing patterns of
psychological distress over time. Journal of Child Psychology and Psychiatry,
44, 399-411.
Wille, N., Bettge, S., Ravens-Sieberer, U., & the BELLA Study Group. (2008). Risk
and protective factors for children’s and adolescents’ mental health: Results of the
BELLA study. European Journal of Child and Adolescent Psychiatry, 17(Suppl. 1),
133-147.
Bios
Veronika Ottova is a research assistant and doctoral student at the University
Medical Center Hamburg-Eppendorf in Hamburg, Germany. Her research interests
include psychosocial health, well-being, and health-related quality of life in children
and adolescents.
Michael Erhart has a background in psychology and a PhD in public health. His
research areas include health-related quality of life, chronic diseases, and international studies on child and adolescent health. His area of expertise is statistical modelling, psychometry, and epidemiology.
Wilma Vollebergh is a professor of social and behavioral sciences at the University
of Utrecht, Utrecht, Netherlands and head of the research program Youth in Changing
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
157
Ottova et al.
Cultural Contexts, an interdisciplinary research program that focuses on trajectories of
mental health, risk behaviors, and cultural lifestyles of adolescents and young adults.
Gyöngyi Kökönyei received her PhD in health psychology at Eotvos Lorand University
in Budapest, Hungary. She is an assistant professor in the personality and health psychology department at Eotvos Lorand University. She is also a researcher at the National
Institute of Child Health in Budapest. Her research interests include somatization, emotion regulation, chronic pain, and adolescents’ well-being and deviant behavior.
Antony Morgan is an epidemiologist and fellow of the UK Faculty of Public Health.
He currently works as an associate director at the National Institute of Health and
Clinical Excellence and is the chair of the HBSC Policy Development Group.
Inese Gobina is a public health specialist and a lecturer at the Department of Public
Health and Epidemiology at Riga Stradins University, Riga, Latvia. Her research
interests include subjective health issues and medicine use among adolescents.
Helena Jericek received her PhD in social pedagogy at Faculty of Education at
University in Ljubljana. She is currently a researcher at the Centre for Health
Promotion at the National Institute of Public Health, Ljubljana, Slovenia, where she
is involved into development of different promotional and preventional programs.
Her research interests include mental health promotion, health and health behavior of
children and young people.
Franco Cavallo graduated in medicine in Torino, where he also received his specialty degree in child psychiatry and in hygiene and preventive medicine. He is currently a full professor in clinical epidemiology with the Faculty of Medicine at the
University of Torino, Turin, Italy. His research interests are prevention and health
promotion in childhood. He is currently active in the domain of adolescent behavior,
growth, and development.
Raili Välimaa is a senior lecturer in the health sciences at the University of
Jyväskylä, Jyväskylä, Finland and an adjunct professor in health promotion at the
University of Eastern Finland. She holds a PhD in health education. Her research
interests include self-rated health and health behavior, health as a concept, health
promotion practices in families, and health promotion in schools, including research
on school health education and health education teacher training.
Margarida Gaspar de Matos is a health and clinical psychologist with a PhD in
special education and rehabilitation (1993) and a professor in international health
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012
158
Journal of Early Adolescence 32(1)
(2004). She is the principal investigator for HBSC in Portugal and coordinates international EU research there (7th Framework). She has an academic position at the
FMH/Technical University of Lisbon and a research position at the CMDT/Institute of
Hygiene and Tropical Medicine/New University of Lisbon. Her area of expertise is
youth mental health and healthy lifestyles. She works in close connection with the
Portuguese government in the area of health and education and also supervises the
Psychological Mental Health Center/Technical University of Lisbon.
Tania Gaspar is a health and clinical psychologist with a master’s degree in public
health and a PhD in psychology. She is the Portuguese executive coordinator of the
European TEMPEST project (7th Framework) and a member of HBSC and
KIDSCREEN in Portugal. She is the director of the Psychology and Education
Sciences Institute/Lisbon Lusíada University, where she coordinates and supervises
the Psychological Mental Health and Health Promotion Center. She also has a
research position at the CMDT/Institute of Hygiene and Tropical Medicine/New
University of Lisbon and at the FMH/Technical University of Lisbon. Her areas of
expertise are youth mental health, healthy lifestyles, health-related quality of life, and
socioeconomic and migration status.
Christina W. Schnohr is consultant at Ministry of Social Affairs (Greenland) and
associated researcher at Department of Public Health, University of Copenhagen,
Copenhagen, Denmark. PhD project concerned socioeconomic inequalities in adolescent health smoking behaviors and how policy may affect this behavior. General
research interests are methodological and measurement issues, in particular, use of
scales and scale validation.
Ulrike Ravens-Sieberer is a professor of public health, health psychology, and
health care services research for children and adolescents at the University Medical
Center Hamburg-Eppendorf. She is the deputy director of the Department of
Psychosomatics of Children and Adolescents and head of the research group “Child
Public Health.” She coordinates and conducts national and European health surveys
on mental health, health-related quality of life, and health behavior in children and
adolescents. Her focuses lie in prevention and health promotion, social determinants
of children’s mental health, and the development and evaluation of supportive programs for mental health promotion.
Downloaded from jea.sagepub.com at NUI, Galway on June 12, 2012