Content uploaded by Rogerio Teixeira Silva
Author content
All content in this area was uploaded by Rogerio Teixeira Silva
Content may be subject to copyright.
Tennis elbow:
survey among
839 tennis
players with
and without
injury
Abstract
Background: Lateral epicondylitis, which is more
commonly known as tennis elbow, is prevalent
in the world of sports; however this injury still
causes confusion among doctors and other
healthcare professionals. Numerous studies have
evaluated treatment options and prognosis, but
few have taken into consideration the extent to
which tennis players, themselves, understand this
ailment.
Objective: To determine to what level Brazil-
ian tennis players understand the aetiology and
courses of treatment for tennis elbow.
Subjects: 839 tennis players who had been playing
tennis for a mean age of 11 years.
Methods: A simple, multi-choice questionnaire
was made available on an internet site dedicated
to tennis enthusiasts.
Results: Among the players enrolled in this study,
41.8 % reported being affected by this injury. De-
spite the fact that nearly half of all tennis players
had been affected, most players were unaware
of the causes of tennis elbow. Even among those
who had been affected by tennis elbow, only 39.3
% believed that the cause of the lesion was the
backhand stroke, and almost all (94.9 %) had un-
dergone some form of treatment before seeking
advice from a physician.
Conclusions: We believe that this study demon-
strates the need to properly educate players,
coaches, and physicians so that they can better
understand the clinical management of tennis
elbow. Without such an understanding, tennis
players tend to treat themselves and only seek
medical advice during the chronic phase of the
injury, when the prognosis is worse.
Original research
Rogerio Teixeira Silva and Marcelo Bannwart Santos
Med Sci Tennis 2008;13(1):36-41
Photography: Henk Koster
36
Keywords: tennis elbow, tennis, injury, epidemiology
Introduction
e term tennis elbow is commonly used to describe an injury,
lateral epicondylitis, which occurs on the lateral face of the elbow.
Despite the fact that tennis elbow is highly prevalent among rec-
reational tennis players - nearly half will suff er this injury - a large
percentage of tennis elbow studies have focused on individuals
who do not participate in sports.1-6 ese studies however, fail
to address the diff erences in tennis elbow aetiology and injury
characteristics between athletes and non-athletes. erefore, it is
important to understand this injury with regard to its aetiology
among tennis players.
While the clinical diagnosis of tennis elbow is simple, there ap-
pears to be a lack of consensus on the clinical characteristics and
aetiology of the injury as well as the eff ectiveness of various treat-
ments.3,7 ere is currently no uniform treatment for this injury
among tennis players and several recent studies have reported
a lack of evidence for justifying the various treatments that are
currently prescribed.8-10 It is well established that tennis elbow
incidence correlates with improper stroke technique, particularly
improperly executed backhand strokes performed with one
hand.5,11,12 Although these fi ndings are well understood within
the scientifi c community, many tennis players are unaware of
these characteristics of tennis elbow, which leads a large number
of players to treat this injury without seeking medical assistance,
most often with rest or using homemade therapies.
Since recreational tennis players are largely unaware of the aetiol-
ogy and courses of treatment for tennis elbow, we attempted to
determine to what level Brazilian tennis players understood this
injury. e principal objective of this study was to determine the
opinions of recreational Brazilian tennis players regarding their
understanding of tennis elbow, taking into account that this term
is common to recreational tennis athletes. e results of this
survey will contribute to the establishment of policies to properly
educate coaches, players, and physicians on the management and
prevention of this injury.
Methods
We developed a simple, multiple-choice questionnaire in order
to analyze Brazilian tennis players’ knowledge of the aetiology
and available courses of treatment for tennis elbow. is survey
was available on an internet website specializing in tennis news.
Participation by the responding tennis players was voluntary and
those who completed the questionnaire consented to the use of
the collected data for the purposes of scientifi c research including
statistical analysis and publication. e research ethics committee
of our sector gave its approval for the work to be performed, and
the methods used in this study followed the norms for research
involving living beings that are in force in our country.
In the questionnaire general data, including the respondent’s age,
sex, years playing recreational tennis, and whether they had ever
suff ered from tennis elbow diagnosed by a doctor which required
treatment were collected. Subjective questions, with a single cor-
rect answer, were then asked about these injuries and the tennis
player’s method(s) of treatment. e questions and possible
responses for the questionnaire are shown in Table 1.
Table 1. Questions and possible responses for the questionnaire
relating to the aetiology of tennis elbow.
1. In your opinion, what is the meaning of tennis elbow?
a. Any kind of pain that the tennis player feels in the elbow
region, because of tennis practice
b. It is a pain on the external (lateral) side of the elbow
c. It is a pain on the internal (medial) side of the elbow
d. It is a kind of arthrosis of the elbow in tennis players
2. In your opinion, what type of movement is the main cause of
the injury?
a. Backhand
b. Forehand
c. Serve
d. All of them
e. e injury is unrelated to inadequate technique in a specifi c
stroke >>
37
Drawing Frans Bosch
38
3. In your opinion, what is the most important cause that leads
to tennis elbow?
a. Inadequate stroke technique
b. A stronger racquet for playing tennis
c. High string tension
d. None of these
4. Do you think that all elbow pains should lead tennis players to
seek immediate medical assessments?
a . Ye s
b. No
5. Have you ever used treatments without medical guidance
when you had elbow pains resulting from tennis?
a . Ye s
b. No
e correct responses for questions 1, 2 and 3 were: B, A and
A, respectively. e other responses to these questions were
considered to be incorrect for the purposes of statistical power
analysis. e responses A and B were compared for the purposes
of statistical analysis in questions four and five.
Statistical analysis
e qualitative variables were represented by absolute and
relative frequencies (%) while the quantitative variables were
represented by the means, standard deviations (sd), minimums
and maximums. e mean age and length of time playing tennis
for each group was compared using the Student’s t test for inde-
pendent samples. e presence of association between qualitative
variables was calculated using a chi-squared test. e significance
level of 0.05 (α = 5%) was adopted, and described values lower
than this were considered to be significant.
e statistical power of the analysis performed in this study was
calculated for each question using a confidence interval of 95%,
a beta error of 20% (power of 0.80), and alpha error of 0.05
(5%).13 To calculate the statistical power in questions 1, 2, and
3, we compared correct (single answer) and incorrect responses
(all three alternative answers together). e subjective questions
relating to treatment (questions four and five) were analyzed
according to the responses given and compared in a two-tailed
model with the hypothesis that the responses given by the group
reporting injuries would be different from the responses given by
the uninjured group.
Results
Demographic data
A total of 839 tennis players responded to the questionnaire and
sent it for our evaluation. Of the
respondents, 762 (90.8%) were
male and 77 (9.2%) were female.
e tennis players’ ages ranged
from 10 to 78 years, with a mean
age of 37.2 years and a standard
deviation of 12.4 years. e dis-
tribution of these players by age
groups is presented in Table 2.
e tennis players included in this study had played tennis
for a mean of 145.5 months, with a minimum of three and a
maximum of 792 months. Out of the total number of players,
351 (41.8%) had been afflicted with a tennis elbow injury while
the remaining 488 (58.2%) reported that they had not had tennis
elbow in the past.
Questions relating to the aetiology of this tennis injury
When asked about the characteristics of tennis elbow, the major-
ity of responding tennis players (446 or 53.2%) believed correctly
that the injury was a pain on the external side of the elbow. e
remaining respondents were incorrect in believing that tennis
elbow was described as any elbow pain (156/839 or 18.6%) ,
pain on the internal face of the elbow (148/839 or 17.6%), or a
type of joint arthrosis (89/ 839 or 10.6%).
Respondent were less clear when asked about which movement
caused the injury. Two hundred and eighty nine (34.1 %) of the
responding tennis players correctly indicated that the backhand
stroke was responsible; however, 269 (32.1 %), a statistically
equivalent number, indicated that all strokes were responsible. Of
the remaining one-third of respondents, ninety-six tennis play-
ers (11.4%) thought the injury was caused by the forehand, 93
(11.1%) thought it was the serve, and 95 (11.3%) believed it was
unrelated to any specific stroke.
When asked what was the most important cause of tennis elbow
among players, a majority (553 or 65.9%) responded that it was
due to playing incorrectly. e remaining players responded in-
correctly; 17.6% (148/839) believed that high string tension was
the main cause of the injury, 7.3% (61/839) believed that racquet
weight was the problem, and 9.2% (77/839) believed that none
of these reasons adequately expressed what caused the injury.
Questions relating to care and treatment for the symptoms
e majority of the responding tennis players believed that they
should seek medical advice from a physician when they had
elbow pain resulting from tennis. Out of the total respondents,
70.4% (591/839) answered yes to question 4 of the question-
naire while the remaining 29.6% (248/839) believed that medical
care should not be sought for the first episode. When asked
about treatments, 67.3% (565/839) reported that they had used
homemade treatments or undergone treatment without seeking
medical advice when they had an episode of elbow pain.
Age (years) n (%)
10 to 19 96 (11.4)
20 to 29 128 (15.3)
30 to 39 202 (24.1)
40 to 49 279 (33.3)
50 to 59 116 (13.8)
60 to 69 16 (1.9)
70 to 79 2 (0.2)
Table 2. Distribution of the ten-
nis players by age groups
39
Statistical analysis among players with and without tennis
elbow
We statistically analyzed the data collected from the survey by
comparing the responses give by players previously afflicted with
tennis elbow (351 players or 41.8%) with those given by the
players who did not report having had an injury (488 players
or 58.2%). e players who reported having had tennis elbow
were on average older and had been playing tennis for a longer
time. e comparison between the demographic data for the two
groups is shown in Table 3.
Table 3. Characteristics of the tennis players with and without
episodes of tennis elbow
e two groups of respondents (those with and without episodes
of tennis elbow) differed with regard to the percentage of re-
sponses that were considered correct for each question. However,
the differences were only statistically significant (p<0.001) for
questions 1, 3 and 5 (Table 4).
Table 4. Data obtained from the 839 tennis players
Variables Tennis elbow injury
Years No (n = 488) Yes (n = 351)
Mean (sd) 32.7 (12.1) 43.6 (9.8)
Minimum – Maximum 9 – 73 15 – 78
Student’s t test p < 0.001 *
Sex
Female 51 (10.5%) 26 (7.4%)
Male 437 (89.5%) 325 (92.6%)
Chi-squared test p = 0.132
Length of time playing tennis
Mean (sd) 126.3 (113.0) 172.3 (125.5)
Minimum – Maximum 3 – 600 5 – 792
Student’s t test p < 0.001 *
Tennis elbow
Question 1 NO (n = 488) YES (n = 351)
A 101 (20.7%) 55 (15.7%)
B 229 (46.9%) 217 (61.8%)
C 97 (19.9%) 51 (14.5%)
D 65 (12.5%) 28 (8.0%)
p < 0.001 * / SP = 99%
Question 2 NO (n = 488) YES (n = 351)
A 148 (30.3%) 138 (39.3%)
B 55 (11.3%) 41 (11.7%)
C 56 (11.5%) 37 (10.5%)
D 167 (34.2%) 102 (29.1%)
E 62 (12.7%) 33 (9.4%)
p = 0.069
Question 3 NO (n = 488) YES (n = 351)
A 300 (61.5%) 253 (72.1%)
B 50 (10.2%) 11 (3.1%)
C 87 (17.8%) 61 (17.4%)
D 51 (10.5%) 26 (7.4%)
p < 0.001 * / SP = 89.6%
Question 4 NO (n = 488) YES (n = 351)
1 342 (70.1%) 249 (70.9%)
2 146 (29.9%) 102 (29.1%)
p = 0.788
Question 5 NO (n = 488) YES (n = 351)
1 232 (47.5%) 333 (94.9%)
2 56 (11.5%) 17 (4.8%)
3 200 (41.0%) 1 (0.3%)
p < 0.001 * / SP = 100%
p: probability (alpha error); SP: statistical power
DISCUSSION
Tennis elbow is a common injury although the majority of cases
do not occur among tennis players. Despite the prevalence of this
injury, conflicting opinions about the best treatment options exist
within the scientific community. Numerous studies have been
published that discuss the problems relating to such lesions, how-
ever many of these studies do not include tennis players in their
analysis.2,3,10 It is well documented that playing tennis greatly
increases the incidence of tennis elbow injuries. In fact, Mens et
al.14 determined that tennis players are 2.8 times more likely to
develop a tennis elbow injury as compared to individuals who do
not play tennis. According to the literature, approximately half of
all tennis players will develop elbow pains at some point of their
sporting lives.4-6 ese figures are consistent with our findings, >>
in which 41.8% of the players in our survey had already suffered
tennis elbow requiring them to seek treatment. Collectively, these
data demonstrate the importance of gaining a better understand-
ing of this injury especially among recreational tennis players.
Many biomechanical studies have been performed to study this
injury among tennis players. Kelley et al.12 compared the electro-
myographic activity of arm muscles in players with and without
lateral epicondylitis during a single-handed backhand tennis
stroke. ey concluded that the injured tennis players had signifi-
cantly greater activity in the wrist extensors and pronator teres
muscles during ball impact and early follow-through. ey also
reported that a leading elbow was a type of motion that could
lead to such an injury. Furthermore, Roetert et al.5 described the
importance of high muscle activity in the extensor carpi radialis
brevis (ECRB) at ball contact during the one-handed backhand
stroke. Collectively, these studies, together with findings reported
by Riek et al.15, demonstrate that the main causes of tennis
elbow among recreational tennis players are the movements
associated with a one-handed backstroke and improper stroke
technique. erefore, we utilized the conclusions of these studies
to design our questionnaire.
In our study, the majority of tennis players believed that the term
tennis elbow refers to a pain on the external side of the elbow.
However, there was a highly significant difference (p < 0.001), with
high statistical power (99%), between the responses of players who
reportedly had never suffered this injury and those who had been
injured. While the majority (61.8%) of previously injured players
correctly defined tennis elbow, only 46.9% of uninjured players
were able to do so. ese results are especially disconcerting con-
sidering that even after suffering this injury, more than one third of
patients (39.2%) did not fully understand their injury. In our view,
this deserves attention because it suggests that, even when tennis
players with such injuries are treated, they do not receive adequate
guidance from doctors and physical therapists.
Unfortunately, few healthcare professionals have studied biome-
chanical concepts specifically applied to tennis. In our opinion,
this is necessary to adequately treat the injuries of tennis players
and advise them on how to prevent re-injury. Although such
advice is important, we agree with Nirschl and Ashman,16 who
made it clear that tennis players themselves do not fully under-
stand all aspects of this type of injury. It is unknown, for example,
how to adequately determine the ideal racquet for each tennis
player, and more studies in this field are necessary.
In this study, slightly more of the previously injured players
(39.3%) than uninjured players (30.5%) - a statistically insignifi-
cant difference (p=0.069) - responded that they believed that the
backhand stroke was the main tennis movement responsible for
tennis elbow. Consistent with the conclusions derived from our
analysis of the first question in our survey, this finding further
demonstrates that players do not have an adequate understanding
of the causes of this injury. erefore, they are unable to prevent
these injuries from occurring or re-occurring. .
When tennis players were asked to identify the most important
for cause of tennis elbow, the majority of players, regardless of
whether they had been previously afflicted with tennis elbow, cor-
rectly identified inadequate stroke technique as the major cause.
However, significantly more players (p < 0.001) who had been
previously injured correctly identified the cause. Again, these
results illustrate the need to better educate tennis coaches and
physicians in order to manage the rehabilitation of such patients.
We propose that only through better education and training for
tennis coaches will be possible to improve tennis players’ under-
standing of this injury.
When we analyzed the data pertaining to the methods of treat-
ment used by the survey respondents, we were surprised to learn
the vast majority (94.9%) of the 351 tennis players who reported
having had tennis elbow stated that they had undertaken some
treatment on their own before seeking medical advice from a
physician. Among the tennis players who reported that they had
not had this injury, a nearly equivalent number sought treatment
advice from a physician (52.5%) as underwent some treatment
without medical guidance (47.5%). e difference between the
two groups (with and without the injury) was highly significant
(p < 0.001), with a statistical power of 100%. As commonly
observed in our daily practice, different patients have different
levels of pain tolerance and this partially dictates when they will
seek a physician’s advice. During a patient’s first episode of elbow
pain, most tennis players tend to treat their symptoms by using
an analgesic cold gel after matches, by changing their racquet,
or by limiting their play for a few weeks. Affected individuals
tend to seek assistance only after the problem persists for some
amount of time, often weeks or months. is may mean that
they lose the opportunity for treatment during the acute phase
of the injury, which has been shown to correlate with a better
prognosis. erefore, we believe that patients, especially tennis
players, should be made aware of the advantages of receiving early
treatment from a physician for lateral epicondylitis.
Knudson and Blackwell11 used the term “leading elbow” to define
the position of the elbow that may favor this injury. As predicted,
when we question our patients suffering from tennis elbow, we
observe that this technical error was frequently reported. Wei et
al.17 recently proposed that control of the follow-through phase of
the movements is a critical factor for reducing shock transmission
and recommended that clinicians or trainers instruct beginner
tennis players to quickly release the tightness of their grip after
ball-to-racquet impact in order to reduce shock impact transmis-
sion to the wrist and elbow. Unfortunately, only anecdotal evidence
exists to support the claim that altering one’s technique is sufficient
to prevent injury or re-injury, therefore additiona studies will be
required to examine this possibility.10, 17, 19
In the current study, no distinction was made between whether
the tennis player performed backhand strokes with one or two
hands. is may be of particular importance since Roetert et
al.5 proposed that players who utilize the two-handed backhand
stroke have a lower incidence of tennis elbow and may be the
most effective backhand stroke for preventing lateral tennis elbow.
is was not examined in our study since an individual’s sports
technique often changes over time. When some tennis players
have this injury, or feel elbow pain, they change their playing
technique; for example changing from a one to a two handed
backhand stroke. Since our evaluation did not involve a clinical
examination of the players, we believe that this issue would be
subject to error during an evaluation by questionnaire alone.
e methods employed to collect data in this study were flawed
only in that the survey was made available on an internet site
dedicated to tennis enthusiasts. erefore, individuals without
access to the internet or who did not visit this particular website,
due to lack of interest, were excluded from the study. Despite this,
40
a large sample of tennis players were surveyed (839), which made
the sample a statistically relevant representation of the popula-
tion. Due, in part to the large number of survey respondents, we
were unable to examine each tennis players in person. However,
with a greater number of tennis players responding, we were able
to generate a larger volume of data and the survey was able to
be more wide-ranging and representative. Since the term tennis
elbow is well known among tennis players, it would not have been
diffi cult for these players to report whether they had suff ered this
injury and had sought medical treatment. Krosshaug et al.20 have
indicated that when an injury is well known in biomechanical
terms and the injury mechanism is well established in the literature,
a data survey by questionnaire alone is valid.
Conclusions
Unfortunately, no offi cial data on recreational tennis players
is available in Brazil, but it is believed that around one million
people are regular recreational players. e fact that nearly half of
these athletes, an estimated 500,000 individuals, will develop ten-
nis elbow at some point demonstrates the importance of having
more adequate player, coach, and physician educational programs
in order to prevent this injury and enable athletes to have a better
prognosis when this injury occurs. According to the data gath-
ered in our survey, we determined that Brazilian tennis players
are largely uniformed as to the causes and available treatments
for tennis elbow, therefore current educational programs need
to be enhanced. We believe that it may be possible to facilitate
education programs through clubs, gyms and other organizations
that give sports instruction to our tennis players. We believe that
this study illustrates the inadequacies of the current programs
in educating Brazilian tennis players to the risks of tennis elbow
and hope that the results of this survey will contribute to the
establishment of policies to properly educate coaches, players, and
physicians on the management and prevention of this injury.
Acknowledgements
We would like to acknowledge José Nilton Dalcim and Bruna
Dalcim for their assistance in the online formatting of the ques-
tionnaire as well as PCE Company for their help with the English
translation and revision of the manuscript. We would especially
like to thank the tennis players who participated in this study.
References
1. Fedorczyk JM. Tennis elbow: blending basic science with clinical practice. J Hand
er 2006;19(2):146-53.
2. Balk ML, Hagberg WC, Buterbaugh GA, Imbriglia JE. Outcome of surgery for
lateral epicondylitis (tennis elbow): eff ect of worker’s compensation. Am J Orthop
2005;34(3):122-6.
3. Haahr JP, Andersen JH. Physical and psychosocial risk factors for lateral
epicondylitis: a population based case-referent study. Occup Environ Med
2003;60(5):322-9.
4. Kamien M. A rational management of tennis elbow. Sports Med.
1990;9(3):173-91.
5. Roetert EP, Brody H, Dillman CJ, Groppel JL, Schultheis JM. e biomechanics
of tennis elbow. An integrated approach. Clin Sports Med 1995;14(1):47-57.
6. Perkins RH, Davis D. Musculoskeletal injuries in tennis.. Phys Med Rehabil Clin
N Am 2006;17(3):609-31.
7. Korthals-de Bos IBC, Smidt N, van Tulder MW, et al. Cost Eff ectiveness of
Interventions for Lateral Epicondylitis Results from a Randomised Controlled
Trial in Primary Care. Pharmacoeconomics 2004;22(3):185-95.
8. Hong QN, Durand MJ, Loisel P. Treatment of lateral epicondylitis: where is the
evidence? Joint Bone Spine 2004;71(5):369-73.
Rogerio Teixeira Silva, MD, PhD, is an orthopedic surgeon with a special interest
in shoulder and knee surgery. He is the Chairman of the South American Com-
mittee of the STMS, Vice-President of the Orthopedic Sports Medicine Commit-
tee of the Brazilian Orthopedic Society, and also acts as the Chief Medical Offi cer
for the Brazilian Tennis Federation. He is responsible for the medical services of
the Brazil Davis Cup and Fed Cup Teams. After completing his PhD he created
the NEO – Orthopedic Sports Medicine Research Center, a private institute to
promote research projects in the fi eld of orthopedic sports medicine. e group is
sponsored by Merck Sharp Dohme, Ache and Asics, and is working on projects for
sports injuries prevention, advances in treatment of cartilage and tendon injuries,
and pre-emptive analgesia for surgical procedures around the knee and shoulder.
Address for correspondence: Rogerio Teixeira Silva, MD, PhD, Rua Carmelo
Damato 40, São Paulo, SP, Brazil. Email: rgtsilva@uol.com.br
Marcelo Bannwart Santos, PT, i s Physical erapist and Coordinator of the
Sports Medicine Rehabilitation Section, at CETE - Federal University of Sao
Paulo / and NEO - Orthopedic Sports Medicine Center. He is Director of the
Brazilian Society of Sports Physical erapy (SONAFE).
About the author
41
9. Boyer MI, Hastings H 2nd. Lateral tennis elbow: “ Is there any science out there?”.
J Shoulder Elbow Surg 1999;8(5):481-91.
10. Alvarez-Nemegyei J, Canoso JJ. Evidence-based soft tissue rheumatology - epi-
condylitis and hand stenosing tendinopathy. J Clin Reumathol 2004;10:33-40.
11. Knudson D, Blackwell J. Upper extremity angular kinematics of the one-handed
backhand drive in tennis players with and without tennis elbow. Int J Sports
Med 1997;18(2):79-82.
12. Kelley JD, Lombardo SJ, Pink M, Perry J, Giangarra CE. Electromyographic
and cinematographic analysis of elbow function in tennis players with lateral
epicondylitis. Am J Sports Med 1994;22(3):359-63.
13. Eston RG, Rowlands AV. Stages in the development of a research project: put-
ting the ideas together. Br J Sports Med 2000;34(1):54-69.
14. Mens JM, Stoeckart R, Snijders CJ, Verhaar JA, Stam HJ. Tennis elbow, natural
course and relationship with physical activities: an inquiry among physicians. J
Sports Med Phys Fitness. 1999;39(3):244-8.
15. Riek S, Chapman AE, Milner T. A simulation of muscle force and internal
kinematics of extensor carpi radialis brevis during backhand tennis stroke:
implications for injury. Clin Biomech 1999;14: 477–483.
16. Nirschl RP, Ashman ES. Elbow tendinopathy: tennis elbow. Clin Sports Med
2003;22:813-36.
17. Wei SH, Chiang JY, Shiang TY, Chang HY. Comparison of shock transmission
and forearm electromyography between experienced and recreational tennis
players during backhand strokes. Clin J Sport Med 2006;16(2):129-35.
18. Assendelft W, Green S, Buchbinder R, Struijs P, Smidt N. Extracts from concise
clinical evidence: tennis elbow. Br Med J 2003;327:329.
19. Bisset L, Paungmali A, Vicenzino B, Beller E. A systematic review and meta-
analysis of clinical trials on physical interventions for lateral epicondylalgia. Br J
Sports Med 2005;39:411-22.
20. Krosshaug T, Andersen TE, Olsen O-EO, Myklebust G, Bahr R. Research
approaches top describe the mechanism of injuries in sport: limitations and
possibilities. Br J Sports Med 2005;39:330-9.
