Too little sleep and an unhealthy diet could increase the risk of sustaining a new injury in adolescent elite athletes

Abstract

Little is known about health variables and if these variables could increase the risk of injuries among adolescent elite athletes. The primary aim was to present overall data on self-perceived stress, nutrition intake, self-esteem, and sleep, as well as gender and age differences, on two occasions among adolescent elite athletes. A secondary aim was to study these health variables as potential risk factors on injury incidence. A questionnaire was e-mailed to 340 adolescent elite athletes on two occasions during a single school year: autumn semester and spring semester. The results show that during autumn semester, the recommended intake of fruits, vegetables, and fish was not met for 20%, 39%, and 43% of the adolescent elite athletes, respectively. The recommended amount of sleep during weekdays was not obtained by 19%. Multiple logistic regression showed that athletes sleeping more than 8 h of sleep during weekdays reduced the odds of injury with 61% (OR, 0.39; 95% CI, 0.16-0.99) and athletes reaching the recommended nutrition intake reduced the odds with 64% (OR, 0.36; 95% CI, 0.14-0.91). Our findings suggest that nutrition intake and sleep volume are of importance in understanding injury incidence.

Full text

Too little sleep and an unhealthy diet could increase the risk of
sustaining a new injury in adolescent elite athletes
P. von Rosen
1
, A. Frohm
1,2
, A. Kottorp
1,3
, C. Frid
en
1,4
, A. Heijne
1
1
Department of Neurobiology, Care Sciences, and Society (NVS), Division of Physiotherapy, Karolinska Institutet, Huddinge,
Sweden,
2
Swedish Sports Confederation Centre, Bos€
on Sports Clinic, Liding€
o, Sweden,
3
Department of Occupational Therapy,
University of Illinois at Chicago, Chicago, IL, USA,
4
St Erik Academic Primary Healthcare Centre, Stockholm, Sweden
Corresponding author: Philip von Rosen, MSc, Division of Physiotherapy, NVS, Karolinska Institutet, Alfred Nobels All
e 23,
SE-141 83 Huddinge, Sweden. Tel: +46-8-524 888 37, Fax: +46-8-524 888 13, E-mail: philip42195@yahoo.com
Accepted for publication 29 June 2016
Little is known about health variables and if these
variables could increase the risk of injuries among
adolescent elite athletes. The primary aim was to present
overall data on self-perceived stress, nutrition intake, self-
esteem, and sleep, as well as gender and age differences, on
two occasions among adolescent elite athletes. A secondary
aim was to study these health variables as potential risk
factors on injury incidence. A questionnaire was e-mailed
to 340 adolescent elite athletes on two occasions during a
single school year: autumn semester and spring semester.
The results show that during autumn semester, the
recommended intake of fruits, vegetables, and fish was not
met for 20%, 39%, and 43% of the adolescent elite
athletes, respectively. The recommended amount of sleep
during weekdays was not obtained by 19%. Multiple
logistic regression showed that athletes sleeping more than
8 h of sleep during weekdays reduced the odds of injury
with 61% (OR, 0.39; 95% CI, 0.16–0.99) and athletes
reaching the recommended nutrition intake reduced the
odds with 64% (OR, 0.36; 95% CI, 0.14–0.91). Our
findings suggest that nutrition intake and sleep volume are
of importance in understanding injury incidence.
Recent research has focused on identifying injuries,
illness, and associated risk factors in elite sports
based on surveillance reports (Dijkstra et al., 2014).
However, factors that may influence the overall
health of elite athletes, such as stress behavior, self-
esteem, nutrition, and sleep have been less in focus,
especially among adolescent elite athletes. Exploring
such health variables could deepen our knowledge
regarding overall health, injury occurrence, and ath-
letic performance among adolescent elite athletes
(The National Sleep Foundation [NSF], 2010; Potgi-
eter, 2013).
During adolescence, psychosocial stress can
increase, owing to conflicts with parents, school,
vocational demands, and relationship problems
(Byrne et al., 2007). Unlike their non-athletes peers,
adolescent athletes may face psychosocial stress from
sports coaches, parents, competitors, and even them-
selves concerning athletic performance expectations
(Ommundsen et al., 2006). Epidemiology reports
have shown that adolescent elite athletes are at high
risk for injuries (Le Gall et al., 2008; Jacobsson
et al., 2013; Von Rosen et al., 2015). High stress
levels and daily hassle, that is, problems at school/
work, have been found to be a predictor for injury
occurrence in athletes (Fawkner et al., 1999; Kolt &
Kirkby 1996; Ivarsson et al., 2014). Data on stress
and to what extent stress affect the risk of sustaining
injuries are lacking for adolescent elite athletes.
Self-esteem, defined as a personal judgment of one’s
worthiness, is of importance in a health and athletic
performance perspective. For instance, if an athlete’s
self-esteem is competence based, that is, based on
competition results, the general well-being could be
affected, with consequences of higher levels of anxiety
and lower levels of self-confidence (Koivula et al.,
2002; Blom et al., 2011). Low self-esteem could also
increase the risk of eating disorders (Bratland-Sanda
& Sundgot-Borgen, 2013). In contrary, a high self-
esteem serves as a protective factor. Low self-esteem
has been associated with depression, antisocial behav-
ior, poorer mental and physical health (Trzesniewski
et al., 2006), and possibly affecting athletic perfor-
mance. Self-esteem increases as adolescents grow
older (Erol & Orth, 2011). There is also a gender dif-
ference, with women experience lower values of self-
esteem compared to men (Blom et al., 2011).
Reports have shown that a high proportion of
adolescent elite athletes have a low energy intake and
are at risk for eating disorders (Sundgot-Borgen &
Torstveit, 2004; Bratland-Sanda & Sundgot-Borgen,
2013; Knapp et al., 2014). Although nutritional
1
Scand J Med Sci Sports 2016: :–
doi: 10.1111/sms.12735
ª2016 John Wiley & Sons A/S.
Published by John Wiley & Sons Ltd

knowledge seems to be high among athletes (Heaney
et al., 2011), it is not clear to what degree adolescent
athletes follow nutritional guidelines, likely affecting
their overall health, recovery levels, and athletic per-
formance (Meyer et al., 2007). These factors are also
influenced by sleep habits. As adolescents grow older
the amount of sleep decreases, even though their
sleep need is still high (NSF, 2000). Lack of sleep
increases the risk of developing depression, anxiety,
and of suicide attempts among adolescents (Owens,
2014). A high proportion of adolescent non-athletes
experience sleep deprivation and irregular sleep pat-
terns (Milewski et al., 2014). However, few reports
have studied sleep deprivation in adolescent athletes
and to what extent it affects the risk of sustaining
injuries.
In summary, data on self-perceived stress, nutri-
tion, self-esteem, and sleep are limited or not
reported at all for adolescent elite athletes. In addi-
tion, gender and age differences have not been
explored for several of these variables. Examining
these variables provides a wider perspective on the
life situation of adolescent elite athletes and may
therefore be relevant from an injury and athletic per-
formance perspective (Gila et al., 2005; NSF, 2010;
Potgieter, 2013). The primary aim was to present
overall data on self-perceived stress, nutrition intake,
self-esteem, and sleep, as well as gender and age
differences, on two occasions among adolescent
elite athletes. A secondary aim was to study these
health variables as potential risk factors on injury
incidence.
Materials and methods
This study is part of the KASIP study (Karolinska Athlete
Screening Injury Prevention), aiming to understand injury
occurrence in Swedish adolescent elite athletes, and is
approved by the Regional Ethical Committee in Sweden (No:
2011/749-31/3).
Contact was taken with the National Federation of
Basketball, Skiing, Orienteering, Handball, Volleyball, Ten-
nis, and Athletics to inform about this project. The Vol-
leyball, Tennis, and the Basketball Federation rejected
participation. A total of 21 National Sports High Schools
were invited to participate and six schools declined partici-
pation. The available sample was therefore 439 athletes
from 15 schools. Each school was visited by one author,
and orally informed of the purpose of the study as well
as the voluntary nature of their participation. Written con-
sent was obtained from the athletes. Ninety-nine athletes
declined to participate, which made the final cohort
consists of 340 adolescent elite athletes (men =178,
women =162) from seven different sports: athletics, cross-
country skiing, orienteering, handball, downhill skiing, ski
orienteering, and freestyle skiing (Fig. 1). To attend the
National Sports High Schools, the adolescent athletes
must exhibit high national performance achievement and
practice at the highest national level for their age group.
Therefore, these athletes are considered to be elite
athletes.
A web-based questionnaire was e-mailed to the athletes
on two occasions, autumn semester (October–November
2013) and spring semester (February–March 2014), using
the Questback online survey (Questback V. 9.9, Questback
AS, Oslo, Norway). If no response was registered, a
reminder e-mail was sent 4 days later. The questionnaire
contained background questions (age, sex, anthropomet-
rics, sports participation, training variables, alcohol intake,
etc.), as well as validated and reliable sub-questionnaires
about sleep (Kecklund &

Akerstedt, 1992), self-perceived
stress (Cohen et al., 1983), nutrition (Sepp et al., 2004),
and competence-based self-esteem (Johnson & Blom,
2007). These variables were chosen to target injury inci-
dence in a holistic approach, including multiple health
variables, previously sparsely studied in relation to injury
risk. Complete data were collected for 313 (92%) athletes
in the autumn semester and for 260 (76%) athletes in the
spring semester. A total of 105 (31%) athletes responded
only once—that is, in either the autumn or the spring.
These athletes consisted of a higher proportion of male
athletes (65%) compared to the main cohort. They did
not differ significantly (P>0.05) with regard to age, BMI,
or sports participation from the main cohort.
21 schools were invited
Avaliable sample: 439
athletes from 15 schools
340 athletes
enrolled (men =
178, women = 162)
Freestyle skiing (n =
10, men = 8, women =
2)
Ski orienteering (n = 8,
men = 4, women = 4)
Orienteering (n = 57,
men = 29, women =
28)
Athletics (n = 136,
men = 71, women =
65)
Cross-country skiing
(n = 72, men = 39,
women = 33)
Handball n = 45 (men
= 22, women = 23)
Downhill skiing (n =
12, men = 5, women
= 7)
99 athletes declined
participation
Six schools declined
participation
Fig. 1. Flowchart of participant enrollment.
2
von Rosen et al.

Outcomes
Stress
The Perceived Stress Scale (PSS), first described by Cohen
et al. (1983), contains 14 items, for which scores are obtained
using a four-grade Likert-type scale from 0 (never)to4(very
often). The scale addresses general feelings and thoughts about
unpredictable and uncontrollable life events and to what
extent these situations are appraised as stressful. The score
ranges between 0 and 56. High scores indicate high levels of
self-perceived stress. The PSS scale has been found to have
good internal reliability as well as satisfactory construct, con-
current, criterion, and predictive validity (Nordin & Nordin,
2013; Eklund et al., 2014).
Nutrition
The Swedish Nutrition Food Agency index (SNFA index) was
developed as an indicator of a diet’s nutritional quality and
has been used in epidemiological studies of the Swedish popu-
lation (Sepp et al., 2004). The questionnaire comprises 14
items describing a diet’s contents in terms of butter, cooking
oils, bread, and one’s daily intake of fruits, vegetables, fish,
French fries, sausage, sweets, cookies, and soda. The response
to each item summarizes to an index of 0–12; high scores rep-
resent a healthy diet. The SNFA index has been shown to
have acceptable reliability and criterion validity (Sepp et al.,
2004). The proportion of athletes who did not meet the
national recommended intake, that is, a fruit and vegetable
intake of less than once a day and a fish intake of less than
twice a week (24), was calculated.
Self-esteem
The Competence-Based Self-Esteem scale (CBSE scale)
describes contingent self-esteem dependent on competence. It
contains 12 items, each ranked on a Likert-type scale from 1
(strongly disagree)to5(completely disagree). The CBSE has
shown high reliability and concurrent validity (Johnson &
Blom, 2007). The result is determined by calculating the aver-
age score of all questions. High scores indicate competence-
dependent self-esteem, as opposed to non-contingent
self-esteem, which is preferable (Johnson, 1998).
Sleep
Athletes were asked the average amount of sleep during week-
days and weekends. The proportion of athletes who did not
meet the recommendations of more than 8 h of sleep per night
was calculated (NSF, 2000).
Injury
Injury was defined any physical complaint resulting in reduced
training volume, experience of pain, difficulties participating
in normal training or competition, or reduced performance in
sports (Clarsen et al., 2013), and was self-reported by the
athletes.
Statistical analyses
All data were self-reported in line with recommendations from
consensus statement concerning pre-diagnostic data (Timpka
et al., 2014). Descriptive statistics for outcomes are presented
as mean and standard deviation (SD) for continuous variables
and as median with 25th–75th percentiles (p25–p75) for non-
normally distributed data. Based on previous research (Cohen
et al., 1983; Johnson & Blom, 2007) and the characteristics of
data, parametric statistics were used to calculate the PSS and
CBSE scores, whereas non-parametric statistics were used for
SNFA index.
Proportions, mean, and median values for main outcomes
were analyzed according to sex and age using the chi-square test,
the t-test, and the Mann–Whitney Utest, respectively. Differences
in main outcomes between the youngest athletes (age 16) and the
oldest athletes (age 18–19) were calculated. The analyses were car-
ried out separately for the autumn (AU) and spring semester
(SP). Effect size between semesters, sex, and age groups was com-
puted using Cohen’s dcalculation (Cohen, 1988). Throughout
calculations, the significance level was set to P≤0.05.
A binary logistic regression analysis was performed to
study the influence of health variables (measured during AU)
as potential risk factors on injury incidence (measured during
SP). All injured athletes (32.9%) during AU were excluded in
the analysis, meaning that only athletes at risk for a new
injury were included. Possible risk factors were self-perceived
stress, self-esteem, SNFA index, the proportion athletes reach-
ing the recommended intake of fruits, vegetables, and fish,
and the proportion athletes reaching the recommended
amount of sleep during weekdays. The proportion of athletes
meeting the recommended amount of sleep during weekends
was not included in the analysis because all except four ath-
letes reached the recommendation of sleep during weekends.
The SNFA index was treated as a categorical variable, with
four levels, based on sample sizes. All variables were first
assessed by univariate logistic regression analysis and there-
after by multiple logistic regression analysis, using a forward
stepwise procedure. Possible risk factors with P≤0.10 were
included in the multiple logistic regression analysis. Risk fac-
tors with P≤0.05 were chosen in the final model. The final
model was controlled for the influence of sex and age. Odds
ratios (OR) are presented with a 95% confidence interval (CI).
All analyses were performed using the SPSS software for Win-
dows, version 22.0 (SPSS, Evanston, IL, USA).
Results
Demographics
The mean age of the athletes was 17.1 years (SD 0.9;
female 17.1, SD 0.9; male 17.1, SD 1.0), with an
average BMI of 21.8 (SD 2.3) (Table 1). Alcohol
intake was equally distributed for male (46.1%) and
female (45.1%) athletes, but found in a higher pro-
portion of the older athletes (82.6% for age 19) than
younger ones (12.6% for age 16). Few athletes
(0.9%) used tobacco. No gender or age differences
were observed for the use of nutrition supplements,
drugs, or in training data. The most commonly used
nutrition supplements were protein powder, high-
carbohydrate sport drinks, and supplements includ-
ing iron, vitamin D, zinc, and magnesium.
Self-perceived stress
Female athletes reported significantly higher self-per-
ceived stress than male athletes did in both semesters
3
Health variables in adolescent elite athletes

(AU: P<0.001; SP: P<0.001) (Table 2). The PSS
score was also significantly higher for athletes age
18–19 than for 16-year-old athletes (AU: P<0.05;
SP: P<0.001).
Nutrition
No significant gender (AU: P=0.35; SP: P=0.16)
or age (AU: P=0.23; SP: P=0.20) differences were
observed for the SNFA index (Table 3).
The recommended national guidelines regarding
intake of fruits, vegetables, and fish were not met
for 20%, 39%, and 43% of the athletes, respec-
tively, in AU. A significantly (AU: P<0.05; SP:
P<0.01) higher proportion of male athletes than
female athletes did not meet the recommended
intake of vegetables. The percentage of male ath-
letes not meeting the recommended intake of
fruits was significantly (P<0.01) higher than that
among female athletes during SP. This was not
evident for AU (P=0.14). A lower proportion of
the athletes age 18–19 did not meet the recom-
mended intake of vegetables during AU, in com-
parison to athletes age 16 (P<0.01). However,
the difference was not significant (P=0.76)
during SP.
Self-esteem
A significantly (AU: P<0.05; SP: P<0.05) higher
CBSE score was observed for female athletes than
for male athletes (Table 4). No significant (AU:
P=0.17; SP: P=0.18) age difference was observed
for the CBSE score.
Sleep
Among all the athletes, the recommended amount of
sleep (more than 8 h of sleep) during weekdays and
weekends was not obtained by 18.5% and 1.0%,
respectively, for AU. No significant differences
between gender or age for sleep during weekdays and
weekends were observed (Table 5).
Risk factors for injury incidence
During AU 32.9% (n=103) of the athletes were
injured and during SP 28.4% (n=74) were
Table 1. Background data for adolescent elite athletes for BMI, use of nutrition supplements, tobacco, alcohol, drugs during the last 6 months, and
training variables. Presented for all athletes, gender, and age
All athletes
(n=340)
Male athletes
(n=178)
Female athletes
(n=162)
Age 16
(n=110)
Age 17
(n=113)
Ages 18–19
(n=117)
BMI, mean (SD) 21.8 (2.3) 22.2 (2.4) 21.3 (2.2) 21.3 (2.4) 21.5 (2.0) 22.5 (2.5)
Male/female (n) 178/162 178/––/162 59/51 54/59 65/52
Use of nutrition supplements (%) 24.4 23.6 25.3 18.2 31.0 23.9
Use of tobacco (snuff*, cigarettes) (%) 0.9 1.7 0 0 0.9 1.7
Use of alcohol (%) 45.6 46.1 45.1 12.7 38.1 83.8
Use of drugs
†
(%) 34.4 31.5 37.7 32.7 37.2 33.3
Training sessions base training
‡,§
5(4–6) 5 (4–6) 4 (4–5) 4 (4–5) 5 (4–6) 5 (4–6)
Training sessions competition season
‡,§
4(4–5) 4 (4–5) 4 (3–5) 4 (3–4) 4 (4–5) 4 (4–5)
Rest days base training
§,¶
2(2–3) 2 (2–3) 2 (2–3) 3 (2–3) 2 (2–3) 2 (2–3)
Rest days competition season
§,¶
3(2–3) 3 (2–3) 3 (2–3) 3 (2–3) 3 (2–3) 3 (2–3)
*Swedish smokeless tobacco.
†
Examples of drugs: NSAIDs, penicillin, paracetamol.
‡
Number of training sessions per week.
§
Median with 25th–75th percentiles in parenthesis.
¶
Number of days not training per week.
Table 2. Self-perceived stress (PSS) score with mean (SD) for adolescent elite athletes. Results presented for all athletes, male and female, and for
athletes ages 16 and 18–19 for the autumn and the spring semester. Statistical significance tested for gender and age differences in each semester.
High scores indicate high levels of self-perceived stress
Autumn Spring Autumn Spring Autumn Spring
All
athletes
All
athletes
Male Female Male Female Age 16 Age 18–19 Age 16 Age 18–19
PSS 22.2 (7.6) 22.9 (7.4) 20.0 (7.4)*** 24.7 (7.0) 20.5 (6.9)*** 24.9 (7.2) 21.0 (7.4)* 23.2 (8.0) 22.1 (6.4)*** 24.4 (7.9)
0.04
†
0.31
†
0.30
†
0.14
†
0.15
†
†
Cohen’s effect size, calculated on between-group differences.
*P<0.05; ***P<0.001.
4
von Rosen et al.

injured. The uninjured athletes (n=162) at the
AU were followed to the second measurement
(during SP), with a median number of 88 days
for the uninjured athletes between the two occa-
sions. Univariate logistic regression showed that
athletes sleeping more than 8 h per day during
weekdays (OR, 0.41; 95% CI, 0.17–0.96) and
reaching the recommended nutrition intake (OR,
0.38; 95% CI, 0.16–0.90) significantly (P<0.05)
reduced the odds of a new injury during SP.
These factors were still identified as risk factors
for sustaining a new injury in the multivariate
logistic regression analysis. Athletes reaching the
sleep recommendation during weekdays reduced
the odds of injury with 61% (OR, 0.39; 95% CI,
0.16–0.99) and athletes reaching the recommended
nutrition intake decreased the odds with 64%
(OR, 0.36; 95% CI, 0.14–0.91), adjusted for sex
and age category (Table 6).
Discussion
The principal findings in the present study were that
the recommended intake of fruits, vegetables, and
fish was not met among 20%–43% of the adolescent
elite athletes. The recommended amount of sleep
during weekdays was not obtained by 19% during
the autumn semester. In addition, athletes sleeping
more than 8 h during weekdays and reached the rec-
ommended nutrition intake during the autumn seme-
ster reduced the odds of sustaining a new injury
during the spring semester.
Few reports have analyzed multiple health vari-
ables in one model related to injury occurrence.
Consistent with Milewski et al. (2014), we found
too little sleep to be a risk factor of sustaining a new
injury. Athletes sleeping more than 8 h in average
during weekdays reduced the odds of a new injury,
whereas Milewski et al. (2014) showed a significant
Table 3. SNFA index for adolescent elite athletes with median (25th–75th percentiles) and percentages of athletes not meeting the recommended
intake of fruits, vegetables, and fish. Statistical significance tested for gender and age differences in each semester. High SNFA scores represent a
healthy diet
Autumn Spring Autumn Spring Autumn Spring
All athletes All athletes Male Female Male Female Age 16 Age 18–19 Age 16 Age 18–19
SNFA index 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6) 5 (4–6)
Fruits
†
(%) 20.1 20.4 23.3 16.7 28.5* 13.1 21.4 18.5 23.4 22.4
Vegetables
†
(%) 39.3 42.7 45.4* 32.7 53.7** 32.8 53.4** 35.2 45.7 43.4
Fish
†
(%) 42.5 41.5 45.4 39.3 44.7 38.7 48.5 42.6 48.9 36.8
†
Athletes not meeting the national recommended intake.
*P<0.05, **P<0.01.
Table 4. Competence-Based Self-Esteem scale (CBSE) with mean (SD) for adolescent elite athletes, presented for all athletes, males and females,
and athletes ages 16 and 18–19 for the autumn and the spring semester. Statistical significance tested for gender and age differences in each
semester. High scores indicate a competence-dependent self-esteem
Autumn Spring Autumn Spring Autumn Spring
All athletes All athletes Male Female Male Female Age 16 Age 18–19 Age 16 Age 18–19
CBSE 2.7 (0.7) 2.6 (0.7) 2.6 (0.7)* 2.8 (0.7) 2.5 (0.7)* 2.7 (0.8) 2.6 (0.7) 2.7 (0.7) 2.6 (0.7) 2.7 (0.7)
0.01
†
0.13
†
0.12
†
0.09
†
0.10
†
†
Cohen’s effect size, calculated on between-group differences.
*P<0.05.
Table 5. Athletes not sleeping the recommended amount during weekdays and weekends
Autumn Spring Autumn Spring Autumn Spring
All athletes All athletes Male Female Male Female Age 16 Age 18–19 Age 16 Age 18–19
Sleep weekdays* (%) 18.5 20.4 17.3 19.6 19.5 21.2 13.9 16.5 18.1 18.4
Sleep weekends* (%) 1.0 1.5 1.8 0 1.6 1.5 1.0 0 0 1.3
*Athletes not sleeping recommended amount of sleep.
5
Health variables in adolescent elite athletes

correlation between decreased injury risk and
increased amount of sleep (6–9 h of sleep). Irregular
sleep pattern, causing temporary impaired recovery,
may also be associated with injury risk. The NSF
(2000) recommends ≥9 h of sleep to be optimal for
high school students which should apply to adoles-
cent elite athletes too.
Several reports have shown that athletes’ diets are
inadequate compared to national recommendations
and to sports nutrition recommendations (Zalcman
et al., 2008; Gibson et al., 2011). Our results confirm
that a great proportion of adolescent elite athletes do
not meet even the national diet recommendations,
especially in terms of fish and vegetable intake. In
addition, a lower proportion of our athletes (70%)
compared to the general population (80%) reached a
SNFA score of 5 or higher in terms of nutrition
intake (Livsmedelsverket [LSV], 2013). Most impor-
tant, we found that a healthy diet, meeting the rec-
ommendation of fruits, vegetables, and fish, reduce
the odds of injury. A healthy diet, involving such
adequate intake of protein, essential vitamins, and
minerals, is likely to enhance recovery between train-
ing sessions and competitions and thereby reduce the
risk of injury and illness [American Dietetic Associa-
tion (ADA), Dietitians of Canada, American College
of Sports Medicine et al., 2009; Heaney et al., 2011].
The SNFA index differs from the recommended
intake of fruits, vegetables, and fish in terms of tak-
ing into account the intake of empty calories and the
fat quality. Possibly due to limited variance in this
sample, the SNFA index was not identified as a sig-
nificant risk factor. At the studied schools, there are
generally no employed dieticians and the responsibil-
ity to maintain a healthy diet is therefore put on the
athletes themselves or the coaches. Creating medical
teams including nutrition experts, accessible to the
schools, may be valuable in improving the diet in
athletes and thereby reduce the risk of injury and
other unhealthy variables.
Normative data for students (mean age 19) pre-
sented by Cohen et al. (1983) showed a mean
self-perceived stress score of 23.2, and female stu-
dents found themselves more stressed (23.6) than
male students (22.4). Our cohort showed a similar
distribution of scores for the entire group, however a
greater significant difference of mean scores between
females (24.7) and males (20.0) with a medium effect
size (0.31) was found. There is a lack of reference
data in terms of what is a high degree of stress. High
stress levels have been related to depressive symp-
toms and higher scores of trait anxiety in adolescent
elite athletes (Gerber et al., 2011). Stress levels have
also been found to be a predictor of sustaining inju-
ries among adolescent as well as adult elite athletes
(Kolt & Kirkby, 1996; Galambos et al., 2005).
Whether a high level of self-perceived stress over a
long period interferes with an athlete’s recovery is
unknown. High stress levels were not found as a risk
factor in this cohort possibly due to stress is not con-
stant over time and especially not in an adolescent
population.
A competence-dependent self-esteem in an athlete
refers to a self-esteem based on performance and
competition results. Such self-esteem can be consid-
ered fragile since sport performance is likely to vary
over a season due to different reasons. In contrary, a
non-contingent self-esteem does not fluctuate over
time and is not dependent on performance results or
earned by achievements. Having a self-esteem based
on competence could increase the risk of unhealthy
behaviors, such as eating disorders (Bratland-Sanda
& Sundgot-Borgen, 2013), depression, and poorer
mental and physical health (Trzesniewski et al.,
2006), stressing the importance of identifying athletes
at risk. A limitation with the CBSE scale is the lack
of cut-off scores. Consistent with Blom et al. (2011),
we found a higher competence based self-esteem in
female compared to male athletes.
The cohort in the present study consists of athletes
with similar characteristics, which strengthen the
generalizability of our findings to similar cohorts. All
participants were elite athletes at sports high schools
likely to experience similar demands regarding
Table 6. Odds ratios (OR) from univariate and multiple logistic regression analyses. The calculation includes the uninjured athletes (n=162) at the
autumn semester
Univariate logistic regression* Pvalue OR (95% CI) Multiple logistic regression* Pvalue OR (95% CI)
Sleep weekdays
†
0.04 0.41 (0.17–0.96) Sleep weekdays
†
0.05 0.39 (0.16–0.99)
Nutrition recommendation
‡
0.03 0.38 (0.16–0.90) Nutrition recommendation
‡
0.03 0.36 (0.14–0.91)
SNFA index (score 0–4) 0.72 Reference
SNFA index (score 5) 0.31 2.10 (0.51–8.69)
SNFA index (score 6) 0.37 1.88 (0.47–7.47)
SNFA index (score 7–12) 0.62 1.44 (0.34–5.99)
CBSE 0.53 0.83 (0.46–1.49)
PSS 0.55 1.02 (0.96–1.07)
*Adjusted for sex and age category.
†
Proportion athletes reaching the recommendation of sleep during weekdays.
‡
Proportion athletes reaching the recommendation of nutrition intake of fish, fruits and vegetables.
6
von Rosen et al.

schoolwork and sport performance. Including ath-
letes from multiple schools and sports also con-
tributed to the generalizability of the results. The
athletes who responded during either the autumn or
the spring semester showed characteristics similar to
the main cohort (aside from the higher proportion of
male athletes). To identify risk factors, only athletes
at risk for a new injury were included in the logistic
regression analysis. Still, the risk factors identified
may represent a healthy behavior of athletes, mean-
ing that other factors associated with these risk fac-
tors could potentially explain or partly explain the
reduced risk of injury found in this study. Associated
factors to a healthy behavior in elite sports might for
instance be season planning, progressive training,
enough rest between training sessions, etc. (Luke
et al., 2011). However, identifying risk factors is of
importance, even though the mechanism of how
these factors is contributing to an increased risk of
injury is not yet fully understood (van Mechelen
et al., 1992).
A possible limitation of the present study is the
self-reporting method. The accuracy of self-reporting
data depends on the participants and both over- and
underestimation need to be considered. However,
the used questionnaires were constructed for self-
assessment. In addition, collecting self-report data
follows the recommendations from a consensus
statement concerning reports of pre-diagnostic data
(Timpka et al., 2014), which approach has been
applied in recent injury epidemiology (Clarsen et al.,
2013; Jacobsson et al., 2013; Nilstad et al., 2014;
Von Rosen et al., 2015). Regarding the risk factor
analysis, we did not control for injury events occur-
ring between the two measures in this study. Because
of the multiple sports considered and the size of the
cohort, the results can usefully be generalized to
other adolescent elite athletes in especially individual
sports. Finally, we believe our results throw new
light on the need to include different health variables
in future studies of injury incidence. We therefore
suggest that future studies should focus on exploring
different aspects of nutrition intake and sleep data,
for example, acute sleep deprivation, in a context of
injury occurrence and athletic performance.
Perspectives
In summary, this is one of the first studies analyzing
multiple health variables in one model related to
injury incidence in adolescent elite athletes. The
results show that sleeping more than 8 h during
weekdays and having a healthy diet during the
autumn semester reduce the odds of sustaining a new
injury during the spring semester. In addition, a con-
siderable number of adolescent elite athletes do not
have a healthy diet, and this proportion seems to be
higher compared to the general population. Our
findings suggest that nutrition intake and sleep vol-
ume are of importance in understanding injury inci-
dence. This strengthens the need to educate coaches
and athletes regarding nutrition intake and sleep or
even to use an interdisciplinary approach in order to
reduce the risk of injury in adolescent elite athletes.
Further understanding of the relationship and the
predictive values of these health variables to injury
risk may be valuable in improving elite sports partic-
ipation for adolescent elite athletes.
Key words: Children, elite sports, prevention, self-
confidence, surveillance, youth.
Acknowledgements
We would like to express our gratitude to all adolescent elite
athletes participating in the study. We also gratefully
acknowledge Swedish Sports Confederation for the support
during this project. Special thanks to Wim Grooten of
Department of Neurobiology, Care Sciences, and Society
(NVS) Division of Physiotherapy, Karolinska Institutet for
contributing with valuable input in the data analysis.
References
American Dietetic Association (ADA),
Dietitians of Canada, American
College of Sports Medicine, Rodriguez
NR, Di Marco NM, Langley S.
American College of Sports Medicine
position stand. Nutrition and athletic
performance. Med Sci Sports Exerc
2009: 41: 709–731.
Blom V, Johnson M, Patching GR.
Physiological and behavioral
reactivity when one’s self-worth is
staked on competence. Indiv Diff
Res 2011: 9: 138–152.
Bratland-Sanda S, Sundgot-Borgen J.
Eating disorders in athletes: overview
of prevalence, risk factors and
recommendations for prevention and
treatment. Eur J Sport Sci 2013: 13:
499–508.
Byrne DG, Davenport SC, Mazanov J.
Profiles of adolescent stress: the
development of the adolescent stress
questionnaire (ASQ). J Adolesc 2007:
30: 393–416.
Clarsen B, Myklebust G, Bahr R.
Development and validation of a
new method for the registration of
overuse injuries in sports injury
epidemiology: the Oslo Sports
Trauma Research Centre (OSTRC)
overuse injury questionnaire. Br J
Sports Med 2013: 47: 495–502.
Cohen J. Statistical Power Analysis for
the Behavioral Sciences. Hillsdale,
NJ: Lawrence Erlbaum, 1988.
Cohen S, Kamarck T, Mermelstein R.
A global measure of perceived stress.
J Health Soc Behav 1983: 24: 385–
396.
Dijkstra HP, Pollock N, Chakraverty
R, Alonso JM. Managing the health
of the elite athlete: a new integrated
performance health management and
coaching model. Br J Sports Med
2014: 48: 523–531.
Eklund M, B€
ackstr€
om M, Tuvesson H.
Psychometric properties and factor
structure of the Swedish version of
7
Health variables in adolescent elite athletes

the Perceived Stress Scale. Psychiatry
2014: 68: 494–499.
Erol RY, Orth U. Self-esteem
development from age 14 to
30 years: a longitudinal study. J Pers
Soc Psychol 2011: 101: 607–619.
Fawkner HJ, McMurrary NE,
Summers JJ. Athletic injury and
minor life events: a prospective
study. J Sci Med Sport 1999: 2: 117–
124.
Galambos SA, Terry PC, Moyle GM,
Locke SA. Psychological predictors
of injury among elite athletes. Br J
Sports Med 2005: 39: 351–354.
Gerber M, Holsboer-Trachsler E,
P€
uhse U, Brand S. Elite sport is not
an additional source of distress for
adolescents with high stress levels.
Percept Mot Skills 2011: 112: 581–
599.
Gibson JC, Stuart-Hill L, Martin S,
Gaul C. Nutrition status of junior
elite Canadian female soccer athletes.
Int J Sport Nutr Exerc Metab 2011:
21: 507–514.
Gila A, Castro JM, G
omez J, Toro J.
Social and body self-esteem in
adolescents with eating disorders. Int
J Psychol Psycholog Ther 2005: 1:
63–71.
Heaney S, O’Connor H, Michael S,
Gifford J, Naughton G. Nutrition
knowledge in athletes: a systematic
review. Int J Sport Nutr Exerc
Metab 2011: 21: 248–261.
Ivarsson A, Johnson U, Lindwall M,
Gustafsson H, Altemyr M.
Psychosocial stress as a predictor of
injury in elite junior soccer: a latent
growth curve analysis. J Sci Med
Sport 2014: 17: 366–370.
Jacobsson J, Timpka T, Kowalski J,
Nilsson S, Ekberg J, Dahlstr€
om O,
Renstr€
om PA. Injury patterns in
Swedish elite athletics: annual
incidence, injury types and risk
factors. Br J Sports Med 2013: 47
(15): 941–952.
Johnson M. Self-esteem stability: the
importance of basic self-esteem and
competence strivings for the stability
of global self-esteem. Eur J Pers
1998: 12: 103–116.
Johnson M, Blom V. Development and
validation of two measures of
contingent self-esteem. Indiv Diff
Res 2007: 5: 300–328.
Kecklund G,

Akerstedt T. The
psychometric properties of the
Karolinska sleep questionnaire. J
Sleep Res 1992: 1: 113.
Knapp J, Aerni G, Anderson J. Eating
disorders in female athletes: use of
screening tools. Curr Sports Med
Rep 2014: 13: 214–218.
Koivula N, Hassm
en P, Fallby J. Self-
esteem and perfectionism in elite
athletes: effects on competitive
anxiety and self-confidence. Pers
Indiv Diff 2002: 32: 865–875.
Kolt G, Kirkby R. Injury in Australian
female competitive gymnasts: a
psychological perspective. Aust J
Physiother 1996: 42: 121–126.
Le Gall F, Carling C, Reilly T. Injuries
in young elite female soccer players:
an 8-season prospective study. Am J
Sports Med 2008: 36: 276–284.
Livsmedelsverket [LSV] (2013) Synen
p

a bra matvanor och kostr

ad- en
utv€
ardering av Livsmedelsverkets
r

ad: rapport 22. Available at http://
www.livsmedelsverket.se/globalassets/
rapporter/2013/2013_livsmedelsverket
_22_utvardering_kostrad.pdf?
_t_id=1B2M2Y8AsgTpgAmY7PhC
fg%3D%3D&_t_q=2005%3A22
&_t_tags=language%3Asv%2Csiteid
%3A67f9c486-281d-4765-ba72-ba39
14739e3b&_t_ip=66.249.78.212&_t_
hit.id=Livs_Common_Model_
MediaTypes_DocumentFile/_6ee
7bbbd-2177-4799-ac6b-5de0cbaf2a
59&_t_hit.pos=19 (Accessed
October 25, 2014).
Luke A, Lazaro RM, Bergeron MF,
Keyser L, Benjamin H, Brenner J,
d’Hemecourt P, Grady M, Philpott
J, Smith A. Sports-related injuries in
youth athletes: is overscheduling a
risk factor? Clin J Sport Med 2011:
21: 307–314.
Meyer F, O’Connor H, Shirreffs SM.
Nutrition for the young athlete. J
Sports Sci 2007: 25: S73–S82.
Milewski MD, Skaggs DL, Bishop GA,
Pace JL, Ibrahim DA, Wren TA,
Barzdukas A. Chronic lack of sleep
is associated with increased sports
injuries in adolescent athletes. J
Pediatr Orthop 2014: 34: 129–133.
Nilstad A, Bahr R, Andersen T. Text
messaging as a new method for
injury registration in sports: a
methodological study in elite female
football. Scand J Med Sci Sports
2014: 24: 243–249.
Nordin M, Nordin S. Psychometric
evaluation and normative data of the
Swedish version of the 10-item
perceived stress scale. Scand J
Psychol 2013: 54: 502–507.
Ommundsen Y, Roberts GC, Lemyre
PN, Miller BW. Parental and coach
support or pressure on psychosocial
outcomes of pediatric athletes in soccer.
Clin J Sport Med 2006: 16: 522–526.
Owens J. Insufficient sleep in
adolescents and young adults: an
update on causes and consequences.
Pediatrics 2014: 134: e921–e932.
Potgieter S. Sport nutrition: a review of
the latest guidelines for exercise and
sport nutrition from the American
College of Sport Nutrition, the
International Olympic Committee
and the International Society for
Sports Nutrition. S Afr J Clin Nutr
2013: 26: 6–16.
Sepp H, Ekelund U, Becker W (2004).
Enk€
atfr

agor om kost och fysisk
aktivitet –Underlag till urval av
fr

agor i befolkningsinriktade enk€
ater.
Livsmedelsverkets rapport 21.
Available at: http://www.slv.se/
upload/dokument/rapporter/
kostundersokningar/Rapp%2021%
20hela.pdf (Accessed October 31,
2014).
Sundgot-Borgen J, Torstveit MK.
Prevalence of eating disorders in elite
athletes is higher than in the general
population. Clin J Sport Med 2004:
14: 25–32.
The National Sleep Foundation [NSF]
(2000). Adolescent sleep needs and
patterns. Available at: http://
sleepfoundation.org/sites/default/
files/sleep_and_teens_report1.pdf
(Accessed October 20, 2014).
Timpka T, Alonso JM, Jacobsson J,
Junge A, Branco P, Clarsen B,
Kowalski J, Mountjoy M, Nilsson
S, Pluim B, Renstr€
om P, Rønsen
O, Steffen K, Edouard P. Injury
and illness definitions and data
collection procedures for use in
epidemiological studies in Athletics
(track and field): consensus
statement. Br J Sports Med 2014:
48: 483–490.
Trzesniewski KH, Donnellan MB,
Moffitt TE, Robins RW, Poulton R,
Caspi A. Low self-esteem during
adolescence predicts poor health,
criminal behavior, and limited
economic prospects during
adulthood. Dev Psychol 2006: 42:
381–390.
Von Mechelen W, Hlobil H, Kemper
HC. Incidence, severity, aetiology and
prevention of sports injuries. A review
of sports injuries. A review of
concepts. Sports Med 1992: 14: 82–99.
Von Rosen P, Heijne A, Frohm A.
Injuries and associated risk factors
among adolescent elite orienteerers: a
26-week prospective registration
study. J Athl Train 2015: 51: 321–
328.
Zalcman I, Guarita HV, Juzwiak CR,
Crispim CA, Antunes HK, Edwards
B, Tufik S, de Mello MT.
Nutritional status of adventure
racers. Nutrition 2008: 23:
404–411.
8
von Rosen et al.