Scand J Med Sci Sports 2011: 21: 129–136 & 2010 John Wiley & Sons A/S

doi: 10.1111/j.1600-0838.2009.01057.x

Psychological predictors of sport injuries among junior

soccer players
U. Johnson, A. Ivarsson
Urban Johnson, Centre for Sport and Health Research, Halmstad University, Halmstad, Sweden
Corresponding author: Urban Johnson, Center for Sport and Health Research, Halmstad University, Box 823, SE-301 18
Halmstad, Sweden. Tel: 146 351 672 61, Fax: 146 351 672 64, E-mail: [Link]@[Link]
Accepted for publication 16 October 2009

Previous researches have established models that specify Survey for Collegiate Athletes, Athletic Coping Skills
psychological factors that could predict sport injuries. One Inventory-28 and Swedish universities Scales of Personal-
example is Williams and Andersen’s stress–injury model ity. Injury record was collected by athletic trainers at the
stressing factors such as anxiety, negative life stress and few schools during a period of 8 months. The result suggested
coping resources. The purpose of the current study was to four significant predictors that together could explain
find psychological factors that could lead to an increased 23% of injury occurrence. The main factors are life
injury risk among junior soccer players, in addition to event stress, somatic trait anxiety, mistrust and ineffective
construct an empirical model of injury risk factors for coping. These findings partly support Williams and
soccer players. The participants were 108 male and female Andersen’s stress–injury model and are organized into an
soccer players (m 5 17, 6) studying at soccer high schools in empirical model. Recommendations are given to sport
southwest Sweden. Five questionnaires were used, State medicine teams and coaches concerning issues in sport injury
Trait Anxiety Inventory, Sport Anxiety Scale, Life Events prevention.

Participation in competitive sports sets high demands psychological predictors that could increase the in-
on athletes’ physical skills. As a result, injury fre- jury risk among soccer players.
quency is rather high (Pargman, 2007). For instance, There are at least a dozen models that try to
Hägglund (2007) found that between 65% and 95% establish a connection between psychological ante-
of Swedish elite soccer players (male) reported cedents and the occurrence of sport injuries. One of
at least one injury every year. International soccer the most well known is Williams and Andersen’s
players reported having had an injury frequency that (1998) ‘‘stress–injury model.’’ Among others, Rogers
was 9.4 injuries/1000 h of soccer practice (Waldén and Landers’s (2005) tested and found partial sup-
et al., 2005). Interest in the pathogenesis of sport port for relationships in the ‘‘stress–injury model.’’
injuries has led to the general conclusion that Williams and Andersen (1998) divided risk factors
two major factors influence injury vulnerability: into three main categories: personality, history of
external factors (e.g., type of sport and weather stressors and coping resources. Personality can affect
conditions) and internal factors (e.g., physiological what situations an athlete apprehends as stressful
and psychological factors) (see Williams & Andersen, (Petrie, 1993). An example is a positive relationship
2007). Kleinert (2007) suggests that a person’s between a sport injury occurrence and trait anxiety
psychological state could be related to the injury (Petrie, 1993), hardiness, locus of control, competi-
occurrence. For example, a dysfunctional psycho- tive trait anxiety and achievement motivation (Wil-
physiological process may affect an athlete’s risk of liams & Andersen, 1998). Moreover, Pedersen (2007)
being injured. Some examples are disturbance of stressed a relationship between perceived injury risk
attention, high or low arousal and poor muscle and aggregate aggression. Research clearly indicates
coordination. a positive connection between sport injuries and high
Well-designed sport injury etiology research focus- stress levels (Patterson et al., 1998; Maddison &
ing on psychological factors seems to have the Prapavessis, 2005). This also includes changes in
potential to discover psychological factors that might life and occurrence of sport injuries (Hardy & Riehl,
lead to increased or decreased risk of injury among 1988). More specifically, life event stress, daily ele-
athletes. The purpose of this research was to study ments of anxiety and past injury history seem to be

129
Johnson & Ivarsson
particularly related to injury outcome (Williams & Previous research clearly demonstrates both
Andersen, 1998). In a systematic review of 40 studies, positive and negative relationships between a large
it was found that approximately 85% of the studies number of different psychological factors and the
found some correlation between life event stress and occurrence of sport injuries. For example, studies
injury (Williams & Andersen, 2007). found a positive relationship between the dependent
Several studies have supported the link between variable injury occurrence, and personality factors
general coping resources and athletic injury. Wil- and life event stress, as well as negative relation-
liams et al. (1986) report a relationship between ships between coping resources and the occurrence
athletes low in coping resources and prediction of of sport injuries. While many cited studies apply
injury. Hanson et al. (1992) found that coping a cross-sectional design, often based on university-
resources were the best discriminator for both sever- age athletes at the individual sport level, it is
ity and number of injuries. Johnson et al. (2005) important for the advancement of the field to include
investigated the possibility of preventing sport inju- additional longitudinal studies on younger athletes
ries using psychological skills training among elite because of the potential of such studies to open
soccer players from 12 different teams in Sweden. A up for a life-span perspective of the field. Moreover,
group of 36 players, labelled as being in the risk zone it is essential to investigate additional psycholo-
of becoming injured according to a psychological gical factors and their relation to increased
screening, were divided into an experiment group injury risk, and to make recommendations to sport
and a control group. Based on a brief cognitive medicine teams and coaches for the prevention of
intervention (e.g., relaxation, stress management injuries.
and coping strategies), the result showed that players Accordingly, the purpose of the study is to find
in the experiment group reported significantly fewer significant psychological factors that could lead
injuries after the 5-month intervention than players to an increased injury risk among junior soccer
in the control group did. Much in the same way, players. In addition, it is necessary to investigate
Maddison and Prapavessis (2007) found it possible the relationship between specific types of personality
to prevent sport injuries by improving the athletes’ factors, coping skills and the relationship between
coping skills, especially through stress management the athletes’ history of stressors and frequency of
and self-confidence training. Perna et al. (2003) also sport injuries.
report that athletes practicing cognitive behavioral Four hypotheses are given:
stress management reported fewer injury days than
1. There are a few specific personality character-
athletes in the control group.
istics, such as trait anxiety and trait aggression,
Rogers and Landers (2005) carried out an exten-
which will be associated with an increased risk of
sive study investigating how different psychological
becoming injured.
factors influence the injury risk. Participants were
2. There is a positive relationship between negative
177 soccer players between 14 and 19 years old and
life event stress and injury occurrence among
the result showed that factors such as coping re-
soccer players.
sources (e.g., the ability to control arousal) had the
3. There is a relationship between a low number of
potential to reduce the occurrence of sport injuries.
coping skills and an increased risk of becoming
In their discussion, they argue that a potentially
injured.
stressful situation has a positive relationship leading
4. There are a few specific psychological factors,
to an increased level of state anxiety and/or periph-
such as negative life event stress and ineffective
eral narrowing. These factors also have a positive
coping strategies, which will be associated with
relationship with sport injuries. Moreover, findings
an increased risk of becoming injured.
(Rogers & Landers, 2005) also emphasize that nega-
tive life stress has positive relationships with both
sport injuries and peripheral narrowing. On the other
hand, a negative relationship was found between
coping resources and sport injuries. In other words, Materials and methods
athletes with a high number of coping resources are Participants
less exposed to injury risk in comparison with ath- The participants were 85 male and 23 female (n 5 108) soccer
letes with few coping resources. Kerr and Fowler players studying at three different soccer high schools located
(1988) also found that athletes with high levels of in southwest Sweden. Their ages were between 17 and 19 years
trait anxiety more frequently reported narrow con- old. Athletes below 18 years of age recieved parental approval
centration and attention problems. Finally, Ander- to take part in the study. The selection of the high schools was
made strategically and in cooperation with the Swedish Soccer
sen and Williams (1999) found a positive relation Association. All participation was voluntary. The research
between sport injuries and both peripheral narrowing design was authorized and approved by Halmstad University
and state anxiety. ethical committee for human studies.

130
 Injury prediction in Swedish Soccer
Measurements Sweden. SSP has been used in several studies (e.g., Magnusson
et al., 2007), however, not previously in a sport context. The
State Trait Anxiety Inventory (STAI) test consists of 91 items, classified into 13 categories. The
STAI (Spielberger et al., 1983) is used to measure current state categories, listed with a coefficients, are somatic trait anxiety
anxiety. The test consists of 40 items (20 state and 20 trait). (0.75), cognitive trait anxiety (0.82), mistrust (0.78), stress
The state items describe how the athletes feel just at the susceptibility, (0.74), submission (0.78), impulsiveness (0.73),
specific moment when the questionnaire is completed. On adventure-loving (0.84), interpersonal distance (0.77), social
the other hand, the trait items describe the athletes’ general conformity (0.59), bitterness (0.75), annoyance tendency
anxiety level. Questions were answered on a four-graded (0.78), verbal trait aggression (0.74) and physical trait aggres-
Likert scale, ranging from 1 (not at all) to 4 (very much so). sion (0.84) (Gustavsson et al., 2000). Questions were answered
The a coefficient was 0.93 for the state items and 0.90 for the on a four-graded Likert scale, ranging from 1 (not at all) to 4
trait items. Test–retest coefficients ranged from 0.73 to 0.86 for (very much so).
the trait items and 0.16 to 0.62 for the state items (Spielberger
et al., 1983).
Procedure
Sport Anxiety Scale (SAS) The sampling of the data occurred from October 1, 2007 to
May 1, 2008. Contact with coaches was established by phone
SAS (Smith et al., 1990) is used to measure an athlete’s anxiety and full information about the study was sent by letter. A test
level. The test consists of 20 items, classified in three cate- occasion was scheduled at each high school, just before soccer
gories. The categories are somatic anxiety (nine items) (0.88), practice. Thus, students were informed about the purposes of
worry (seven items) (0.82) and concentration disrupters (five the research, that they could stop their participation at any
items) (0.74). Questions were answered on a four-graded time and that all data would be confidential. During this
Likert scale, ranging from 1 (not at all) to 4 (very much so). period, the different questionnaires were completed. Injury
SAS has a test–retest reliability of 0.85 (Smith et al., 1990). records were collected continuously by athletic trainers at the
different schools during the entire test period. The coaches
were invited to register both the type of injury and the length
Athletic Coping Skills Inventory-28 (ACSI-28) of time the athlete was away from regular practice (i.e., days
ACSI-28 (Smith et al., 1995) is used to measure an athlete’s and/or weeks of missed practice). Injuries were defined as all
general coping skills. The test consists of 28 items, classified in types of injuries that occur in connection with sport participa-
seven categories. The categories are coping with adversity, tion; the severity of injuries was categorized according to the
peaking under pressure, goal setting, mental preparation, length of time the athlete was incapacitated: (a) minor (1–7
concentration, freedom of worry, confidence and achievement days), (b) moderate (1 week to 1 month) and (c) major injuries
against coachability. Questions were answered on a four- (41 month) (cf. Lysens et al., 1991). The injuries had all
graded Likert scale, ranging from 0 (not at all) to 3 (very occurred during training or a game.
much so). The five variables coping with adversity, peaking
under pressure, goal setting, mental preparation and concen-
tration are combined to form an effective (positive) coping Statistical analysis
skills category, and the two variables freedom of worry and One-way analysis of variance (ANOVA) was used for compar-
confidence and achievement against coachability are com- ing data between injured and non-injured groups of players.
bined to form an ineffective (negative) coping skills category. Hypotheses 1–4 were tested using linear regression analysis,
The authors chose to divide the coping factors into two backward method, with the dependent variable injury in order
subgroups because the five variables in the positive coping to find injury predictors. The models (one for each hypothesis)
category have a positive relationship with the occurrence while that were developed in the linear regression analysis were later
the two variables in the negative coping category have a tested using logistic regression analysis. Logistic regression
negative relationship with the injury factor (Johnson, 2007). allows multivariate analysis of a dichotomous dependent
The ACSI has a test–retest reliability of 0.87 and an internal variable into a probability statement, the so-called logic
consistency coefficient of 0.86 (Smith et al., 1995). transformation. Logistic regression was used to show how a
large group of players could be successfully predicted as
Life Events Survey For Collegiate Athletes (LESCA) injured or non-injured, according to the results from the linear
regression.
LESCA (Petrie, 1992) is used to measure an athlete’s life
history stressors. The test consists of 69 items. Athletes are
asked to indicate which events have occurred in the last 12 Results
months, and then, for each event, to rate the life event impact
that they have experienced on an eight-point Likert scale, with A total of 82 participants completed the five ques-
the anchoring 4 (extremely negative) to 14 (extremely
positive). Based on the results, the score will be divided into
tionnaires correctly. This indicates an internal drop-
three categories: negative life event stress, positive life event out rate of 24%. Of the 108 participating players, 42
stress and total life event stress. This procedure has been used (39%) athletes missed at least 1 day of sport practice
in several other studies (e.g., Rogers & Landers, 2005). due to an injury. In this group, 67 injuries were
LESCA has test–retest reliabilities ranging from 0.76 to 0.84 reported.
(Petrie, 1992).

Swedish universities Scales of Personality (SSP) Hypothesis 1

SSP (Gustavsson et al., 2000) is used to measure personality The result of the ANOVA analysis of somatic trait
factors and was developed by the Karolinska Institutet in anxiety between the injured and non-injured groups

131
Johnson & Ivarsson
of athletes [F(1, 99) 5 4, 79, P 5 0.031] showed that Table 1. Regression analysis of injury predictors
injured athletes have a higher level of somatic trait
Variable M SD b P
anxiety than the non-injured athletes. No other
significant relationship was found. Negative life event stress 18.68 13.74 0.22 0.047
A regression analysis with backward elimination Somatic trait anxiety 1.79 0.419 0.32 0.020
was conducted with the 13 personality variables (see Negative coping 9.62 2.93 0.24 0.019
Mistrust 1.84 0.488 0.30 0.008
SSP). This analysis showed that the two predictors Stress sensitivity 2.03 0.419 0.20 0.104
somatic trait anxiety and mistrust could explain 11%
of the total variance of injury occurrence Dependent variable: injury
[R2Adj 5 0, 11, F(2, 91) 5 6, 613, P 5 0.002]. Both fac-
tors were significant, somatic trait anxiety (b 5 0, 32,
P 5 0.003) and mistrust (b 5 0, 29, P 5 0.006). STAI, SAS, ACSI-28). This analysis showed that
A logistic regression analysis was performed using negative life event stress, somatic trait anxiety, nega-
somatic trait anxiety and mistrust as predictors. tive coping, mistrust and stress susceptibility could
Altogether, 101 cases were analyzed (w2 5 8, 57, explain 23% of the total variance [R2Adj 5 0, 23,
df 5 2, P 5 0.014). In this sample, 77.2% of the F(5, 76) 5 5, 73, P 5 o0.001]. Significant predictors
non-injured were successfully predicted while only (Po0.05) were negative life event stress (b 5 0.22,
36.4% for the injured group were accurate. In total, P 5 0.047), somatic trait anxiety (b 5 0.32,
59.4% of predictions were accurate. P 5 0.020), negative coping (b 5 0.24, P 5 0.019)
and mistrust (b 5 0.30, 0.008) (see Table 1).
Using the above-mentioned significant predictors
Hypothesis 2 and injury as a dependent variable, 86 cases were
The result of ANOVA analysis of negative life event analyzed (w2 5 12 182, df 5 4, P 5 0.016). Results
stress (N-LES) between the injured and non-injured showed that 78.3% of the non-injured was success-
groups of athletes [F(1, 97) 5 5, 525, P 5 0.021] fully predicted while 55% of the injury predictions
showed that injured athletes have a higher level of were accurate. In total, 67.4% of predictions were
N-LES than the non-injured athletes. No other accurate.
significant relationship was found.
A regression analysis with backward elimination
was conducted with the three life stress variables (see Discussion
LESCA). This analysis showed that the predictor
negative life event stress could explain 7% of the The purpose was to study psychological predictors
total variance on the dependent variable injury that could increase the injury risk among soccer
[R2Adj 5 0, 070, F(1, 97) 5 8.372, P 5 0.005]. The pre- players. The result showed that there are several
dictor negative life event stress was significant psychological factors that might predict the occur-
(b 5 0.21, P 5 0.005). rence of sport injuries. Significant factors are somatic
A logistic regression analysis was performed using trait anxiety, mistrust, negative life event stress and
the predictor negative life event stress and the de- negative coping. These factors will be discussed in the
pendent variable injury. In total, 98 cases were following hypotheses stated, and recommendations
analyzed (w2 5 5, 14, df 5 1, P 5 0.023). In this sam- will be provided for sports medicine teams and
ple, 81.8% were successfully predicted non-injured coaches for injury reduction.
while only 44.2% of the injury predictions were
accurate. In total, 65.3% of predictors were accurate.
Hypothesis 1
Williams and Andersen (1998) state in their stress–
Hypothesis 3 injury model that different personality variables,
The results of the one-way ANOVA showed no such as hardiness and locus of control, could increase
significant differences in positive or negative coping an athlete’s risk of becoming injured. The results of
between the injured and non-injured groups, as well the first hypothesis showed that injured athletes had
as the result from the linear regression analysis a significantly higher level of somatic trait anxiety
between the predictor’s negative coping, positive than non-injured athletes, much like the research by
coping and injury as the dependent variable. Petrie (1993) establishing a positive relationship
between competitive trait anxiety and sport injuries.
According to Kerr and Fowler (1988), athletes with a
Hypothesis 4 high level of trait anxiety more often reported nar-
A regression analysis with backward elimination was rowing concentration and attention than other ath-
conducted with 23 predictors (see SSP, LESCA, letes. However, no significant result was reported

132
 Injury prediction in Swedish Soccer
regarding aggressive behavior and injury occurrence. developmental stages. For 17- or 18-year-old en-
Rogers and Landers (2005) stressed that peripheral gaged in competitive sport, it is well established
narrowing is a predictor of sport injury. These that the most important role models in the psycho-
findings could explain why somatic trait anxiety is social development are the coach, peers and possibly
often found to be a predictor for sport injuries as parents. Thus, athletes at different age intervals
high anxiety levels seem to lead to peripheral narrow- activate different coping strategies dealing with age-
ing, thus higher injury vulnerability. The result from related stressors.
the regression analysis regarding personality factors
showed that a high level of somatic trait anxiety and
a low level of mistrust could explain 11% of the total Hypothesis 3
variance of injury occurrence. In other words, an The third hypothesis was that injured athletes should
athlete experiencing a high level of somatic trait have a lower number of coping resources than non-
anxiety and a low level of mistrust are more exposed injured athletes. Previous research noticeably reveals
to injuries. The finding that a low level of mistrust a positive relationship between a lack of coping
could be one predictor of sport injury had not been resources and the frequency of sport injuries (Han-
previously found. A potential explanation for the son et al. (1992). However, no differences where
findings is that athletes who are not sufficiently found in regard to coping resources between injured
apprehensive of threatening stimuli could be more and non-injured athletes. One reason could be that
likely to throw themselves into situations beyond the different questionnaires were not sensitive en-
their capacity to control. In such cases, the athletes ough. In particular ASCI-28 is designed to capture
could be exposed to a high risk of injury. Still coping strategies that athletes generally activate dur-
another, but related influencing factor is the coach. ing situations not associated with traumas such as
If an athlete who is low in mistrust places substantial being injured. Another possible explanation is that
decision-making responsibilities on the coach, for the participating players were slightly younger than
instance a soccer player coming into a critical one- athletes in most referred studies and thus have not
to-one situation seeking external approval for his/her yet developed mature coping strategies, making it
action, he or she might miss crucial cues, and thus be problematic to compare the results with other stu-
exposed to an elevated injury risk. dies. Taken together, few studies support the result of
this study
Hypothesis 2
Rogers and Landers (2005) suggest that negative life Hypothesis 4
event stress is a strongest predictor for injury, which The fourth hypothesis was about specifying what
the current study result supports. One reason for this psychological factors, taken together, could predict
is the fact that athletes with high stress levels prob- sport injuries. Both Williams and Andersen (1998)
ably have difficulty focusing on important or relevant and Rogers and Landers (2005) state several predic-
cues during sport participation, thus, ending up in tors of injury (e.g., negative life event stress, state
problematic situations such as being injured. How- anxiety and few coping strategies). In the current
ever, the present result showed that it could explain study, the result showed that five predictors (negative
7% of the injury occurrence. One possible reason life event stress, somatic trait anxiety, negative cop-
why negative life event stress could be a significant ing, mistrust and stress sensitivity) could explain
predictor even with the modest explanation as in this 23% of the total variance of injury occurrence.
study is that it influences other predictors. For Despite the fact that 23% is substantial, several
example, having few coping skills or having a factors have to be considered in exploring the result.
stress-sensitive personality could affect life event Williams and Andersen (2007) writes about external
stress negatively. When discussing the effect of life and internal factors that could affect injury occur-
event stressors, it is important to consider the fact rence. The current result relates to the internal and
that athletes, depending on their age, are exposed to psychological dimension, making it a noteworthy
different stressors. That is, stressors such as leaving high predictor. Combining results of controlled stu-
home, pressure from their parents to perform well in dies using both external and internal predictors
school and sport and building new social networks would most likely find even higher explanatory
could potentially be different from those of older values. However, no such studies are to be found.
athletes, for whom stressors such as building rela- Yet another challenging factor to consider is that,
tionships and developing future careers are more except for LESCA, none of the other questionnaires
prominent. Both Wylleman & Lavallee (2004) and used are specifically constructed for measuring stress-
Alferman & Stambulova (2007) emphasized that ful athletic situations, because no such instruments
athletes are going through different psychosocial seem to exist to date. It is the authors’ opinion that

133
Johnson & Ivarsson
more sensitive questionnaires would likely heighten each block (personality, history of stressors and
the total variance, making the results of this study coping) while the injury prediction model has only
even more attractive. one or two in each block. One other difference
The significant predictors could be divided into the between the models/findings is that the empirical
same main blocks as Williams and Andersen (1998) model of injury risk factors does not highlight the
suggest. In the current study, the personality pre- stress response. There are differences also between
dictors are somatic trait anxiety and mistrust, the the empirical model of injury risk factors and Rogers
history of stressors predictor is negative life event and Landers (2005) results regarding stress and
stress and the coping predictor is ineffective coping coping. The most distinct one might be that Rogers
skills. All significant predictors except mistrust have and Landers (2005) stress both peripheral narrowing
a positive relationship with sport injuries. The two and state anxiety as predictors. These two predictors
personality predictors that occurred in the current are not discussed in the empirical model of injury risk
study are sparsely discussed in the literature. One factors. Instead, the empirical model of injury risk
reason could be that the personality test (SSP) is factors highlights two personality factors. In an
rather new and used in only a few studies before, applied perspective, it is central for the sports med-
thus, psychological factors such as mistrust and icine team and coaches to consider the joint effect of
physical trait aggression have not yet been tested identified predictors and understand that high so-
methodically in a sport context. In line with Rogers matic trait anxiety as well as a low mistrust, anxiety
and Landers’s (2005) findings, one of the strongest about what the coach thinks and says and high life
predictors of sport injury is negative life event stress. event stress greatly influence injury vulnerability.
Also, the relationship between few coping resources Clearly, previous research suggests that brief cogni-
and an increased injury risk is well established tive intervention, especially through stress manage-
(Hanson et al., 1992). ment and self-confidence training, have the potential
The five significant predictors that are presented in to significantly lessen the amount of injuries for
the text above predict 67.4% of the cases success- athletes (Johnson et al., 2005; Maddison & Prapa-
fully. This indicates that approximately two out of vessis, 2007). Hence, the empirical model of injury
three athletes were sorted correctly into either the risk factor underline the importance of using such
injured or the non-injured group. Such a result factors in the applied work with injured athletes.
allows for the creation of an empirical model of Some suggestions for actions are given below.
injury risk factors (see Fig. 1). Development of an
empirical model has the advantage of summarizing
and integrating empirical findings, allowing for a Methodological issues
more systematic understanding of complex factors One limitation of the study is the fact that it is
influencing risk behavior. problematic to measure state levels using only two
The empirical model of injury risk factors supports test occasions during 8 months. State levels, by
significant parts of both Williams and Andersen’s definition, change over time, making it hazardous to
(1998) stress–injury model and Rogers and Landers’s receive a general rate. Another potential limitation is
(2005) findings. For example, both life stress and the sensitiveness of the different questionnaires. In
coping skills are discussed in the models. There are, particular, SSP has not been tested on a sport-related
however, some differences between the empirical population before. Thus, few reference points exist
model of injury risk factors and results from the comparing different studies. Dividing ACSI-28 vari-
two other studies. One distinct difference is that the ables into two categories is also a potential limitation
stress–injury model contains several predictors in of the study. However, this procedure has been used
successfully in other studies (e.g., Rogers & Landers,
2005). The rather low mean age of the soccer players
Personality (18 years) compared with most cited studies using
-SomaticTrait senior athletes could be regarded as a limitation of the
Anxiety (+) study as it is to be expected that senior athletes both
-Mistrust (–)
Ineffective Coping
experience and cope differently with age-related, stres-
Negative Life
Event Stress -Worry (+) sors such as relating to new partners and coaches, as
(+) -Coachability (+) well as education (Wylleman & Lavallee, 2004).

Increased Injury
Risk (+) = positive predictor
Perspectives
(–) = negative predictor
Following the findings, there are well-defined psy-
Fig. 1. Empirical model of injury risk factors. chological factors that affect the injury risk among

134
 Injury prediction in Swedish Soccer
soccer players, which have several implications for decreases. One promising area for future research is
sports medicine teams and coaches to consider. This to use the empirical model of injury risk factors as a
is especially true of negative life event stress. It is theoretical framework for studying combinations of
recommended that medicine teams and coaches be risk predictors for minor as well as major injuries in
observant if major negative life events occur for addition to gender, age and type of sport. Developing
players as these could have a devastating injury more situation-specific and sensitive instruments
impact on the athletes’ behavior in terms of disrupted would also be welcomed. Needless to say, a holistic
concentration and elevated anxiety levels. This is perspective on preventive issues has the potential to
especially important if the negative events have significantly lessen injury occurrence in sport.
occurred very recently. Thus, it is recommended
that the player be allowed to miss one or more Key words: coping strategies, Life Event Stress, per-
practices/games in order to restore psychological sonality, psychological predictors, sport injury.
and physical focus and balance. Another recommen-
dation, especially for coaches, is to help athletes
develop coping skills. That is, to decrease the players’
level of worry through increasing their self-efficacy Acknowledgements
through positive and reinforcing feedback, as well as
by creating an atmosphere of trust and openness to We thank the coaches and players for their cooperation. We
express their feelings. Still another recommendation would also thank the Swedish Soccer Associations Medical
Board for financial support. Moreover, we thank Natalia
is to help athletes learn muscle relaxation. Somatic Stambulova, at Halmstad University, and Anne-Marie Elbe,
anxiety often leads to increased muscle tension; if at Copenhagen University, for valuable feedback on earlier
athletes can control this phenomenon, injury risk drafts.

References
Alferman D, Stambulova N. Career Johnson U, Ekengren J, Andersen MB. issues. In: Pargman D, ed.
transitions and career termination. In: Injury prevention in Sweden. Helping Psychological bases of sport injuries.
Tenenbaum G, Eklund RC, eds. soccer players at risk. J Sport Exerc Morgantown: Fitness Information
Handbook of sport psychology. New Psychol 2005: 1: 32–38. Technology, 2007: xvii–xxvii.
York: John Wiley & Sons, 2007: 712– Kerr G, Fowler B. The relationship Patterson EL, Smith RE, Everett JJ.
733. between psychological factors and Psychosocial factors as predictors of
Andersen MB, Williams JM. Athletic sport injuries. Sports Med 1988: 6: 127– ballet injuries: interactive effects of life
injury, psychosocial factors, and 134. stress and social support. J Sport Behav
perceptual changes during stress. J Kleinert J. Mood states and perceived 1998: 21: 101–112.
Sport Sci 1999: 17: 735–741. physical states as short-term predictors Pedersen DM. Perceived aggression in
Gustavsson JP, Bergman H, Edman G, of sport injuries: two prospective sports and its relation to willingness to
Ekselius L, von Knorring L, Linder J. studies. Int J Sport Exerc Psychol 2007: participate and perceived risk of injury.
Swedish universities scales of 5: 340–351. Percept Mot Skills 2007: 104(1):
personality (SSP): construction, Lysens RJ, de Weerdt W, Nieuwboer A. 201–211.
internal consistency and normative Factors associated with injury Perna F, Antoni MH, Baum A, Gordon
data. Acta Psychiat Scand 2000: 102: proneness. Sports Med 1991: 12: P, Schneiderman N. Cognitive
217–225. 281–289. behavioral intervention effects on
Hanson SJ, McCullagh P, Tonymon P. Maddison R, Prapavessis H. A injury and illness among competitive
The relationship of personality psychological approach to the athletes: a randomized trial. Behav
characteristics, life stress, and coping prediction and prevention of athletic Medicine 2003: 25: 66–73.
resources to athletic injury. J Sport injury. J Sport Exerc Psychol 2005: 27: Petrie TA. Psychosocial antecedents of
Exerc Psychol 1992: 14: 262–272. 289–310. athletic injury: the effects of life stress
Hägglund M. Epidemiology and Maddison R, Prapavessis H. Preventing and social support on female collegiate
prevention of football injuries. Link ö sport injuries: a case for psychology gymnasts. Behav Med 1992: 18:
ping: Linköping University Medical intervention. In: Pargman D, ed. 127–138.
Dissertation No. 989, 2007. Psychological bases of sport injuries. Petrie TA. Coping skills, competitive trait
Hardy CJ, Riehl MA. An examination of Morgantown: Fitness Information anxiety and playing status: moderation
the life stress–injury relationship Technology, 2007: 25–38. effects on the life stress–injury
among noncontact sport participants. Magnusson Å, Göransson M, Heilig M. relationship. J Sport Exerc Psychol
Behav Med 1988: 14: 113–118. Hazardous alcohol users during 1993: 15: 261–274.
Johnson U. Psychological antecedents to pregnancy: psychiatric health and Rogers TM, Landers DM. Mediating
sport injury prevention. In: Pargman personality traits. Drug Alcohol effects of peripheral vision in the life
D, ed. Psychological bases of sport Depend 2007: 89: 275–281. event stress/athletic injury relationship.
injuries. Morgantown: Fitness Pargman D. Sport injury: an introduction J Sport Exerc Psychol 2005: 27: 271–
Information Technology, 2007: 39–52. and overview of related psychological 288.

135
Johnson & Ivarsson
Smith RE, Schutz RW, Smoll FL, Ptacek Waldén M, Hägglund M, Ekstrand J. Handbook of sport psychology, 3rd
JT. Development and validation of a UEFA champions league study: a edn. New York: John Wiley & Sons,
multidimensional measure of sport- prospective study of injuries in 2007: 379–403.
specific psychological skills: the athletic professional football during the 2001– Williams JM, Tonymon P, Wadsworth
coping skills inventory – 28. J Sport 2002 season. Br J Sports Med 2005: 39: WA. Relationship of stress to injury in
Exerc Psychol 1995: 17: 379–398. 542–546. intercollegiate volleyball. J Human
Smith RE, Smoll FL, Schutz RW. Williams JM, Andersen MB. Stress 1986: 12: 38–43.
Measurement and correlates of sport- Psychosocial antecedents of sport Wylleman P, Lavallee D. Develop-
specific cognitive and somatic trait injury: review and critique of the stress mental perspective on transitions
anxiety: the sport anxiety scale. and injury model. J App Sport Psychol faced by athletes. In: Weiss MR,
Anxiety Res 1990: 2: 263–280. 1998: 10: 5–25. ed. Developmental sport and
Spielberger CD, Gorsuch RL, Luschene R, Williams JM, Andersen MB. Psychosocial exercise psychology: A lifespan
Vagg PR, Jacobs GA. Manual for the antecedents of sport injury and inter- perspective. Morgantown: Fitness
state – trait anxiety inventory. Palo Alto, ventions for risk reduction. In: Information Technology, 2004:
CA: Consulting Psychologists, 1983. Tenenbaum G, Eklund RC, eds. 503–523.

136