Load management in athletes: What is the evidence and where does it fit in an injury prevention paradigm?

Conference Paper (PDF Available) · October 2015with 2,923 Reads 
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
·
Conference: Australian Physiotherapy Conference, At Gold Coast, Australia
Cite this publication
Advertisement
1 23
Sports Medicine
ISSN 0112-1642
Sports Med
DOI 10.1007/s40279-015-0459-8
The Relationship Between Training
Load and Injury, Illness and Soreness: A
Systematic and Literature Review
Michael K.Drew & Caroline F.Finch
1 23
Your article is protected by copyright and
all rights are held exclusively by Springer
International Publishing Switzerland. This e-
offprint is for personal use only and shall not
be self-archived in electronic repositories. If
you wish to self-archive your article, please
use the accepted manuscript version for
posting on your own website. You may
further deposit the accepted manuscript
version in any repository, provided it is only
made publicly available 12 months after
official publication or later and provided
acknowledgement is given to the original
source of publication and a link is inserted
to the published article on Springer's
website. The link must be accompanied by
the following text: "The final publication is
available at link.springer.com”.
SYSTEMATIC REVIEW
The Relationship Between Training Load and Injury, Illness
and Soreness: A Systematic and Literature Review
Michael K. Drew
1,2,3
Caroline F. Finch
2
ÓSpringer International Publishing Switzerland 2016
Abstract
Background Clinically it is understood that rapid
increases in training loads expose an athlete to an increased
risk of injury; however, there are no systematic reviews to
qualify this statement.
Objective The aim of this systematic review was to
determine training and competition loads, and the rela-
tionship between injury, illness and soreness.
Methods The MEDLINE, SPORTDiscus, CINAHL and
EMBASE databases were searched using a predefined
search strategy. Studies were included if they analysed the
relationship between training or competition loads and
injury or illness, and were published prior to October 2015.
Participants were athletes of any age or level of competi-
tion. The quality of the studies included in the review was
evaluated using the Newcastle–Ottawa Scale (NOS). The
level of evidence was defined as strong, ‘consistent find-
ings among multiple high-quality randomised controlled
trials (RCTs)’; moderate, ‘consistent findings among mul-
tiple low-quality RCTs and/or non-randomised controlled
trials (CCTs) and/or one high-quality RCT’; limited, ‘one
low-quality RCT and/or CCTs, conflicting evidence’;
conflicting, ‘inconsistent findings among multiple trials
(RCTs and/or CCTs)’; or no evidence, ‘no RCTs or CCTs’.
Results A total of 799 studies were identified; 23 studies
met the inclusion criteria, and a further 12 studies that were
not identified in the search but met the inclusion criteria
were subsequently added to the review. The largest number
of studies evaluated the relationship between injuries and
training load in rugby league players (n=9) followed by
cricket (n=5), football (n=3), Australian Football
(n=3), rugby union (n=2),volleyball (n=2), baseball
(n=2), water polo (n=1), rowing (n=1), basketball
(n=1), swimming (n=1), middle-distance runners
(n=1) and various sports combined (n=1). Moderate
evidence for a significant relationship was observed
between training loads and injury incidence in the majority
of studies (n=27, 93 %). In addition, moderate evidence
exists for a significant relationship between training loads
and illness incidence (n=6, 75 %). Training loads were
reported to have a protective effect against injury (n=9,
31 %) and illness (n=1, 13 %). The median (range) NOS
score for injury and illness was 8 (5–9) and 6 (5–9),
respectively.
Limitations A limitation of this systematic review was
the a priori search strategy. Twelve further studies were
included that were not identified in the search strategy, thus
potentially introducing bias. The quality assessment was
completed by only one author.
Conclusions The results of this systematic review high-
light that there is emerging moderate evidence for the
relationship between the training load applied to an athlete
and the occurrence of injury and illness.
Implications The training load applied to an athlete
appears to be related to their risk of injury and/or illness.
Sports science and medicine professionals working with
athletes should monitor this load and avoid acute spikes in
loads. It is recommended that internal load as the product
of the rate of perceived exertion (10-point modified Borg)
&Michael K. Drew
Michael.drew@ausport.gov.au
1
Department of Physical Therapies, Australian Institute of
Sport, Canberra, Australia
2
Australian Centre for Research into Injury in Sport and its
Prevention (ACRISP), Federation University Australia,
Ballarat, VIC, Australia
3
c/o AIS Physical Therapies, Australian Institute of Sport,
Leverrier Cr, Bruce, ACT 2614, Australia
123
Sports Med
DOI 10.1007/s40279-015-0459-8
Author's personal copy
and duration be used when determining injury risk in team-
based sports. External loads measured as throw counts
should also be monitored and collected across a season to
determine injury risk in throwing populations. Global
positioning system-derived distances should be utilised in
team sports, and injury monitoring should occur for at least
4 weeks after spikes in loads.
Key Points
There is emerging moderate evidence for a
relationship between training load and risk of injury
and illness.
Both absolute and relative training loads are
recommended when examining the relationship
between training load and risk of injury and/or
illness.
The product of rate of perceived exertion and length
of training and/or competition is commonly used to
calculate training load. It is recommended that this
measure be included in future prospective injury
studies and adjusted for in the analyses.
Athletes’ training data can be used to quantify the
risk of injury and/or illness. This data should be
reviewed and utilised to ensure a safe and successful
environment is achieved.
1 Introduction
Quantification and monitoring of training load and athletes’
responses to it is imperative to maximise the likelihood of
optimal athletic performance at a specific time and place.
The response to a load stimulus applied to an athlete can
either be positive (increased physical capacity) or negative
(injury, illness, and overtraining or underperformance). It is
generally reported that the outcomes follow a dose-re-
sponse relationship with negative outcomes following
either a parabolic [1] or exponential curve [2,3] and
parabolic relationships being apparent for performance [4].
Remaining injury and illness free is a fundamental com-
ponent of ideal preparation for sporting performances. This
was illustrated in a recent study in European football
(soccer) [5] which showed that both lower season injury
burden and higher match availability rates correlated with
higher team success. Emerging evidence supports the
clinical hypothesis that the amount of training and com-
petition undertaken is related to the incidence of injuries
and illnesses in competitive athletes [2,6,7]. However,
there have been no previously published systematic
reviews and/or meta-analyses evaluating the association
between these two variables.
1.1 Quantification of Training Volumes
It is imperative to have a clear definition of what consti-
tutes a ‘load’ on an athlete prior to evaluation of the
relationship between that load and subsequent events such
as injury or illness. Training loads can broadly be defined
as ‘internal’ or ‘external’ workloads based on what data is
available. Importantly, both can lead to different risk pro-
files [3]. Internal workloads can be described as a measure
of perception of effort by the athlete themselves (e.g. rate
of perceived exertion [RPE] or heart rate [HR] response to
the stimulus), and external workloads are more typically
the quantification of workloads external to the athlete by
someone else (e.g. balls bowled in cricket or running dis-
tance covered).
1.2 Internal and External Workloads
External workloads describe the quantification of work
external to the athlete. Examples of these workloads
include distance covered in locomotion sports such as
middle-distance or endurance running, the number of balls
bowled in cricket, the number of pitches delivered in
baseball, or the distance swum in swimming events.
Internal workloads quantify the physical loadings experi-
enced by an athlete, as directly perceived by them (RPE),
and are therefore a subjective index of effort when this
measure is utilised. Internal loads are the response to the
external loads placed upon an athlete. Internal loads may
also consist of HR, HR to RPE ratio, training impulse
(TRIMP, a unit of physical effort that is calculated using
maximal, resting or average HR during the session multi-
plied by duration), lactate concentrations, lactate to RPE
ratio, HR recovery (the rate of HR decline following
exercise), HR variability or biochemical, hormonal or
immunological assessments [8]. Appropriate measures of
internal workload should have an exposure (such as dis-
tance or time) component of the activity and also the ath-
lete’s response to that activity. The most widely utilised
measure is the resultant of the multiplication of a 10-point
RPE with the duration of the training session or competi-
tion [9,10].
1.3 Absolute and Relative Workloads
Training loads can be primarily analysed in two ways.
Absolute training loads are the sum of all training sessions,
or a particular domain of training, over a given period such
as a day or week. Relative workloads describe the change
M. K. Drew, C. F. Finch
123
Author's personal copy
in training either expressed as a percentage increase over a
period (such as week to week) or as a ratio of recent and
historical loads (such as a week to month ratio). Both
absolute and relative workloads have been examined for
their relationship to injuries [3,7] and illnesses [11,12].
Absolute workloads can be expressed as the sum of either
the internal or external loads usually contained to a set
period (Monday to Sunday) [7] or a rolling accumulation of
loads (e.g. sum of workload in last 7 days). Absolute
workloads have been shown to be related to injury occur-
rence in elite Australian Football [13], rugby union [14]
and baseball [15,16].
Relative workloads were introduced by Banister and
colleagues [17,18] to account for the workload that an
athlete had achieved in the acute (‘fatigue’) period of
training in comparison to the workload achieved in the
chronic period (‘fitness’), thus allowing athletes of differ-
ing training levels to be compared with respect to perfor-
mance and physiological outcomes. Originally described as
the product of sessional HR and duration [17], it has since
been adopted conceptually to be used with the load variable
described by Foster et al. [10] whereby the training load is
determined as the product of a 10-point RPE by duration.
This concept is commonly referred to as the ‘training stress
balance’ (TSB) whereby the average daily workload of the
last 7-day period (acute load) is compared with the average
daily workload in the last 28-day period (chronic load) and
expressed as a percentage [3]. This has more recently be
described as the ‘‘‘acute:chronic workload ratio’ to reflect
what is being calculated [19]. The TSB has only recently
been used as a measure to quantify the risk of injury, with
internal and external load measures differing in their ability
to quantify risk. Hulin et al. [3] considered this approach in
the context of cricket. They explained that counting the
number of balls bowled (external load) does not encompass
all aspects of training that produce a total workload, such
as batting, fielding and strength training. The results of
observations are reinforced by subsequent studies [20].
Because this only represents a partial quantification of
loads, it is likely to lead to a partial quantification of risk of
injury [13]. This is an important concept that highlights
that the training load achieved in the last 7 days must be
proportional to the training undertaken over the last month,
otherwise an increase in risk of injury is observed.
1.4 Definition of an Injury and Illness
Several recent studies have defined sporting injuries and
illnesses for athletics [21], football [22,23], rugby union
[24], tennis [25] and both the Winter and Summer Olympics
[26]. Injury definitions can broadly be divided into medical
attention (any time an athlete accesses medical attention) and
time-loss definitions (often matches lost but also training
days lost). There has been a recent increase in the uptake of
prevalence-based definitions for overuse injuries where the
prevalence of symptoms irrespective of medical attention or
time loss is quantified and measured [2729]. This definition
relates to athlete-reported symptomatology, such as ‘sore-
ness’ to a body region, limiting their ability to participate,
and therefore articles relating to ‘soreness’ may represent an
‘athlete’s self-reported injury’. The conceptual foundation of
sports injury and illness has also recently been revised on the
basis aligning the definition of injury to the notion of
impairment used by the World Health Organisation [30]. In
doing so, three domains have been suggested: clinical
examination reports, athlete self-reports and sports perfor-
mance [30]. This is now known as the Injury Definitions
Concept Framework (IDCF) [21].
1.5 Objectives
The objective of this study was to perform a systematic
review to determine the relationship between injury and
training loads, relative to the definition of injury used. In
doing so, this review gives recommendations on their
appropriateness and improvements for future studies evalu-
ating the link between training load and injury and/or illness.
2 Methods
2.1 Literature Search
The Preferred Reporting Items for Systematic Reviews and
Meta-Analyses (PRISMA) guidelines were followed [31].
Studies for review were identified through a systematic
search of four databases (MEDLINE, SPORTDiscus,
CINAHL and EMBASE), which were searched using com-
binations of the following keywords: injury, illness, strain$,
sprain$, muscl$ or muscle, bone, tendon or ligament, tend$,
overuse, overtrain$, overreach$, athlete or athlet$, sport$,
cohort study, comparative study, prospective study, epi-
demiology, epidem$, relative risk, odds ratio, training,
train$, load, volume. A full search strategy is available from
the authors. The search was limited to the English language
and studies published prior to October 2015. Despite the
adopted systematic search strategy, it was observed that
some relevant articles had not been identified. To ensure this
review is comprehensive, we have included these additional
articles and have labelled them as ‘not identified from search
strategy’ for transparency.
2.2 Selection Criteria
The titles and abstracts were initially scanned, and all
duplicates and articles clearly outside the scope of this
Training Load and Injury
123
Author's personal copy
study were excluded. The definition of load was taken from
a recent review [32] and was defined as ‘‘the cumulative
amount of stress placed on an individual from multiple
training sessions and games over a period of time, external
workloads performed or the internal response to that
workload’’. Studies were included if they were published or
were ‘in press’ prior to the search date. Abstracts were
assessed and articles were excluded if (1) the participants
were not monitored longitudinally; (2) they reported case
studies, case series or surveys; (3) the articles did not report
any measures of load or injury; (4) the relationship between
load and injury, illness or soreness was not analysed; (5)
they were review articles; or (6) they were related to ani-
mal studies. The full-text versions of the studies were then
retrieved and analysed under the same criteria. The refer-
ences of the selected articles were reviewed for identifi-
cation of potentially relevant studies not captured in the
original search strategy.
2.3 Quality Assessment
The quality of the studies included in the review was
evaluated by a single author (MD) using the Newcastle–
Ottawa Scale (NOS) for Cohort Studies [33], for assessing
the quality of non-randomised studies in meta-analyses. As
all included studies were non-randomised studies, this
quality assessment was chosen post hoc as the most
appropriate quality assessment tool. The NOS is a highly
cited assessment tool [34] and is reported as one of two
useful tools by the Cochrane Handbook for Systematic
Reviews of Interventions, and allows customisation to the
review question of interest [35]. All studies were scored in
three domains: selection of the study groups; comparability
of cohorts on the basis of the design and analysis; and
outcome, including adequate follow-up for outcome to
occur. A maximum score of 4, 2, and 3 for each respective
area was allocated for a total possible score of 9. Selection
of the study groups was assessed in terms of its represen-
tativeness of the exposed and non-exposed cohorts, ascer-
tainment of exposure, and demonstration that the outcome
of interest (injury/illness/soreness) was not present at the
start of the study. The comparability of the cohorts was
assessed based on the design of the study, particularly
around the factors that the study controlled for. The out-
come was assessed on the basis of the manner in which it
was assessed, the length of follow-up and the adequacy of
follow-up of the cohort.
2.4 Levels of Evidence
The a priori level of evidence for monitoring training and
competition loads was evaluated using the van Tulder et al.
method [36]. The level of evidence was defined as strong,
‘consistent findings among multiple high-quality ran-
domised controlled trials (RCTs)’; moderate, ‘consistent
findings among multiple low-quality RCTs and/or non-
randomised controlled trials (CCTs) and/or one high-
quality RCT’; limited, ‘one low-quality RCT and/or CCTs,
conflicting evidence’; conflicting, ‘inconsistent findings
among multiple trials (RCTs and /or CCTs)’; no evidence,
‘no RCTs or CCTs’. The Oxford Centre of Evidence-based
Medicine—Levels of Evidence [37] was utilised to deter-
mine the hierarchical level of evidence according to the
type of research question with the highest level of evidence
(‘1a’) pertaining to a systematic review (with homogene-
ity) of RCTs, and lowest level of evidence (‘5’) being
expert opinion without critical appraisal, or based on
physiology, bench research or ‘first principles’.
2.5 Summary Measures
Where possible, relative risks and odds ratios for sustaining
an injury or illness were extracted and reported, with no
limits placed on the summary measures. The latent period,
defined as the period between the training load (dose) and
onset of injury or illness (negative response), was also
extracted where possible. To illustrate the concept of the
latent period it is best considered in the context of an
athlete/patient whereby they have been exposed to an
infectious agent that leads to a respiratory illness. The
latent period in this example is the period between the
exposure and the time when the patient gains a diagnosis or
symptoms.
3 Results
A total of 787 studies were retrieved through the four
database searches. After screening the titles, abstracts and
full-texts, 23 studies met the inclusion criteria in the review
(Fig. 1), with a further 12 articles included that were not
identified in the search strategy. The summary of included
articles relating to injuries is presented in Table 1, and
included studies of illness are presented in Table 2. The
largest number of included studies evaluated the relation-
ship between training load and injury. The sports evaluated
were rugby league players (n=9), cricket (n=5), football
(soccer, n=3), Australian Football (n=3), volleyball
(n=2), rugby union (n=2), baseball (n=2), water polo
(n=1), rowing (n=1), basketball (n=1), swimming
(n=1), and various sports (n=1). Eight studies evaluated
the relationship between training load and illness, with
studies of athletes in Australian Football (n=1), middle
distance runners (n=1), football (soccer, n=1), speed
skating (n=1), swimming (n=1), basketball (n=1),
rugby league (n=1) and various (n=1) included. For
M. K. Drew, C. F. Finch
123
Author's personal copy
injury-related studies, the relationship was evaluated using
internal loads (n=12), external loads (n=15) or both
(n=2). For illness-related studies, the relationship was
evaluated using internal loads (n=5), external loads
(n=1) or both (n=2). Of the studies reporting internal
load measures, 13 (86 %) reported using the load quan-
tification method described by Foster et al. [14].
The results of the NOS for the quality of the included
studies are presented in Table 3. Injury-based studies were
of slightly better quality than studies of illness (median
[range] NOS, 8 [59] vs. 6 [59], respectively). Only three
studies demonstrated that the outcome of interest (injury or
illness) was not present at the start of the study. Three
studies reached the maximum score [12,38,39].
No RCTs evaluated the use of load monitoring as an
injury and/or illness prevention programme. Moderate
evidence for a significant relationship was observed
between training loads and injury incidence in the majority
of studies (n=25). Training loads were observed to be
protective against injury in ten studies. Moderate evidence
was found for the relationship between training loads and
soreness in throwing sports. In addition, moderate evidence
was observed for the relationship between training loads
and illness, with one study [12] showing a protective effect.
No studies were of an RCT design and therefore the level
of evidence was unable to be determined as strong.
4 Discussion
The results of this systematic review highlight that there is
emerging moderate evidence for the relationship between
the training load applied to an athlete and the occurrence of
injury. There was conflicting evidence to link training loads
to illness, and the majority of studies utilised the quan-
tification of training load proposed by Foster et al. [10].
Based on the repeated identification of relationship
between both absolute and relative training loads in this
review, it is recommended that this type of load quantifi-
cation be utilised as an indication of injury risk when
Fig. 1 Study selection process for inclusion in the systematic review
Training Load and Injury
123
Author's personal copy
Table 1 Summary of the included results indicating the relationship between training and competition load with injury and soreness in athletes
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Rugby league
Gabbett [40]
(2004)
Prospective
cohort, 2b
79 sub-elite
rugby league
players, age
not reported
Medical attention RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil NA, relative A significant relationship (p\0.05) between
changes in the incidence of training injuries
and changes in training intensity (r=0.83),
training duration (r=0.79) and training load
(r=0.86)
Significant correlation between the incidence of
match-play injuries with changes in match
intensity (r=0.74) and match load
(r=0.86)
Gabbett [41]
(2004)
Non-RCT, 2b 220 sub-elite
rugby league
players, age
21.3 ±5.5
Medical attention RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil NA, absolute The 2002 and 2003 pre-season training loads
were significantly lower (p=0.001) than the
2001 pre-season training loads
Significantly higher injury incidence
(p=0.001) in the 2001 pre-season period
(156.7/1000 training hours, 95 % CI
136.3–177.1) than the 2002 (94.4/1000
training hours, 95 % CI 76.7–112.0) and 2003
(78.4/1000 training hours, 95 % CI 64.2–92.7)
pre-season periods
Gabbett and
Domrow [42]
(2005)
a
Prospective
cohort, 2b
153 sub-elite
rugby league
players, age
not explicitly
reported
Competition time loss: any
pain, disability, or
something that occurred
as a result of a
competition match that
caused the player to miss
a subsequent match
Nil Training weeks NA, absolute Completing\18 weeks of training increased the
risk of injury during a match (OR 8.69,
95 % CI 1.33–56.76)
Gabbett and
Domrow [43]
(2007)
Prospective
cohort, 2b
183 semi-
professional
rugby league
players, age
21.4 ±5.1
Any pain or disability
suffered that resulted in
the athlete being unable to
continue training or
competition
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil NA, absolute Incidence of injury was higher (p=0.001) in
the pre-season training phase (137.7 [95 % CI
120.6–154.7] per 1000 training hours) than the
early-competition (76.0 [95 % CI 64.7–87.3]
per 1000 training hours) and late-competition
(62.6 [95 % CI 52.5–72.7] per 1000 training
hours) training phases
Significant relationship (p=0.01) between
training load and incidence of injury within
the training load range of 155 and 590 AU
(0.35 increase in injury incidence with each
arbitrary unit increase in training load)
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Gabbett [45]
(2010)
Non-RCT, 2b 91 Professional
rugby league
players, age
23.7 ±3.8
Non-contact, soft-tissue
injury that resulted in the
athlete being unable to
continue training or
competition. Contact
injuries excluded
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil \1 day, absolute 50–80 % likely to sustain a pre-season injury
within the training load range of 3000–5000
arbitrary units
Players who exceeded the training load
threshold were 70 times more likely to sustain
a non-contact, soft-tissue injury, whereas
players who did not exceed the training load
threshold had a 90 % reduced risk
Killen et al. [44]
(2010)
Prospective
cohort, 2b
36 professional
rugby league
players,
mean age not
reported
(range
17–32)
Any pain or disability that
occurred during
participation in a rugby
league training activity
that was sustained by a
player, irrespective of the
need for training time loss
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Psychological
data of a
player’s
perception
relating to
sleep, food,
energy, mood
and stress
Scale in each
category
(1–10)
Nil Nil, absolute The higher training loads in the first half of the
pre-season corresponded to a higher injury
rate (p[0.05). No relationship between
injury and psychological monitoring data
Gabbett and
Jenkins [46]
(2011)
Prospective
cohort, 2b
79 sub-elite
rugby league
players, age
22.3 ±3.8
Any pain or disability
suffered during a training
session, confirmed by
physiotherapist (medical
attention)
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil Nil, absolute Total training load was significantly related
(p\0.05) to overall injury (r=0.82), non-
contact field injury (r=0.82), and contact
field injury (r=0.80) rates
Significant relationships between field training
load and overall field injury (r=0.68), non-
contact field injury (r=0.65), and contact
field injury (r=0.63)
Significant relationship between strength and
power training loads to incidence of strength
and power injuries (r=0.63)
No significant relationship between field
training loads and incidence of strength and
power injuries
Strength and power training loads were
significantly associated with the incidence of
contact (r=0.75) and non-contact (r=0.82)
field training injuries (p\0.01)
Training Load and Injury
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Gabbett and Ullah
[62] (2012)
a
Prospective
cohort, 2b
34 Professional
rugby league
players,
23.6 ±3.8
Any non-contact, lower
body soft-tissue injury
suffered by a player
during a training session
These were classified as a
transient (no training
missed), time loss (any
injury resulting in missed
training), or a missed
match (any injury
resulting in a subsequent
missed match) injury.
Injury was verified by the
presence of one or more
of the following
characteristics: pain,
tenderness,swelling, and
restricted range of motion
Nil GPS: total distance (m) was
stratified by intensity/
speed, as was distance
covered while
accelerating at different
intensities. Number of
repeated high-intensity
effort bouts
Nil, absolute Frailty model showed that the risk of a no-time-
loss injury was 2.7 (95 % CI 1.2–6.5) times
higher when very-high-intensity running
exceeded 9 m per session compared
with \9 m per session. Higher low-intensity
distances were protective: a 60 % lower risk
of time loss injury was observed (RR 0.4,
95 % CI 0.2–0.9) when very-low-intensity
running exceeded 542 m per session
compared with \542 m per session, and
when the distance covered in moderate-
acceleration activity was [217 m per session
compared with B217 m per session. The RR
of injury was lower (RR 0.5, 95 % CI 0.2–0.9)
when the distance covered in low-intensity
running was [2342 m per session and
maximum acceleration distance was [143 m
persession
Hulin et al. [19]
(2015)
a
Prospective
cohort, 2b
53 Professional
rugby league
players,
23.4 ±3.5
Any time-loss injury that
resulted in a player being
unable to complete full
training or missing match
time
Nil GPS: total distance (m) for
field training sessions and
matches
0–1 weeks, both Absolute workloads alone were not sufficient in
calculating risk. High chronic workloads
([16,095 m) combined with high (2-week
average acute:chronic workload ratio [[1.54])
showed higher risk (28.6 %, 90 % CI
10.5–46.7 likelihood of injury). A very high
acute:chronic workload was associated with a
very high likelihood of injury (97 %). High
chronic workload combined with a moderate
workload ratio (1.02–1.18) showed the
smallest risks of injury
Cricket
Dennis et al. [2]
(2003)
Prospective
cohort, 2b
90 male state-
level fast
bowlers, age
27 (range
18–38)
A condition that affects
availability for team
selection, limits
performance during a
major match or requires
surgery. Injuries also had
to be ‘gradual bowling’
mechanism. Acute onset
or collision-type injuries
were not analysed
Nil Number of balls bowled 1 week, absolute Fast bowlers with an average number of days\2
or C5 between bowling sessions were
significantly at risk (RR 2.4, 95 % CI 1.6–3.5;
RR 1.8, 95 % CI 1.1–2.9, respectively) when
compared with an average of 3–3.99 days.
Compared with those bowlers with an average
of 123–188 deliveries per week, bowlers with
an average of fewer than 123 deliveries per
week (RR 1.4, 95 % CI 1.0–2.0) or more than
188 deliveries per week (RR 1.4, 95 % CI
0.9–1.6) may be at an increased risk of injury
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Dennis et al. [49]
(2005)
Prospective
cohort, 2b
44 Elite junior
cricket fast
bowlers, age
14.7 ±1.4
A condition that affected
availability for team
selection, limited
performance during a
match, or required
surgery. Minor injuries
that only affected
participation in training
sessions were not
examined in this study
Nil Number of match and
training balls bowled
Nil, absolute Fast bowlers with \3.5 days rest between
bowling were at increased risk (RR 3.1,
95 % CI 1.1–8.9). No association was found
between the average number of deliveries
bowled per week and injury
Orchard et al. [47]
(2009)
Retrospective
cohort, 2b
129 National
cricket
players, age
not reported
Competition loss;
unavailable for selection
or a complaint that
resulted in the athlete
being unable to bat, bowl,
or keep wicket during a
match
Nil Number of overs bowled 4 weeks, absolute [50 overs per match resulted in increased risk
for up to 28 days (14 days, OR 1.77 [95 % CI
0.95–3.32]; 21 days, OR 1.77 [95 % CI
1.05–2.98]; 28 days, OR 1.62 [95 % CI
1.02–2.57])
Hulin et al. [3]
(2014)
Retrospective
cohort, 2b
28 State-level
cricket fast
bowlers, age
26.0 ±5.0
Any non-contact injury
resulting in either loss of
match-time or [1
training session in a
1-week period. ‘Soreness’
was excluded from
analysis
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Total number of balls
bowled/week
1 week, relative Internal workload TSB[200 % =RR of injury
of 4.5 (CI 3.43–5.90, p=0.009) compared
with TSB of 50 % and 99 %. External
workload TSB [200 % had RR of injury of
3.3 (CI 1.50–7.25, p=0.033) compared with
TSB of 50 and 99 %
Orchard et al. [48]
(2015)
a
Prospective
cohort, 2b
235
Professional
fast bowlers
The definition of a cricket
injury (to a bowler) is one
that either (i) prevents a
player from being fully
available for selection in a
major match (which is a
List A or first-class
match); or (ii) during a
major match, causes a
player to be unable to bat
or bowl when required by
either the rules or the
team’s captain
Nil Match overs bowled.
Training and club overs
not available
3–4 weeks, absolute Tendon injuries in the next 3 weeks: acute
match overs C50 (OR 3.69, 95 % CI
1.82–8.24); career overs C1200 (OR 2.38,
95 % CI 1.65–3.42); overs in previous season
C00 (OR 2.01, 95 % CI 1.38–2.94); overs in
previous 3 months C150 (OR 0.29, 95 % CI
0.17–0.50); career overs C3000 (OR 0.24,
95 % CI 0.11–0.52)
Bone stress injuries in next 4 weeks: overs in
previous 3 months C150 (OR 2.10, 95 % CI
1.48–2.99); career overs C1200 (OR 0.31,
95 % CI 0.21–0.45)
Muscle injuries in next 3 weeks: overs in
previous season C400 (OR 0.71, 95 % CI
0.53–0.95)
Joint injuries in the next 4 weeks: overs in
previous season C450 (OR 1.96, 95 % CI
1.14–3.37); career A list overs C3000
(OR 1.84, 95 % CI 1.02–3.31)
Training Load and Injury
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Football (soccer)
Brink et al. [50]
(2010)
a
Prospective
cohort, 2b
53 Junior
national-
level football
players, age
16.5 ±1.2
Injury: any physical
complaint sustained by a
player that results from a
soccer match or soccer
training, irrespective of
the need for medical
attention or time loss
from soccer activities
RPE 9time
(mins)
RPE (6–20)
Time =total
training session
minutes
Nil 1 week, absolute Traumatic injuries: injured athlete had higher
duration (mins; OR 1.14, 95 % CI 1.06–1.23),
training load (OR 1.01, 95 % CI 1.00–1.02),
monotony (OR 2.59, 95 % CI 1.22 5.50) and
strain (OR 1.01, 95 % CI 1.00–1.01)
compared with the uninjured athletes in the
preceding week
Lovell et al. [51]
(2006)
a
Prospective
cohort, 2b
19 Elite junior
football
players, age
16.1 (range
15–17)
Groin injury: pubic groin
pain, had local tenderness
over the pubic symphysis,
had painful resisted
bilateral hip adduction,
and (on history and
examination) had no other
cause for his groin pain
(such as acute adductor
injury or inguinal groin
pain suggesting incipient
inguinal hernia)
Nil Number of sessions per
week prior to scholarship
Unknown, absolute Decreased risk with increased number of
training sessions completed prior to the
scholarship period. Unclear statistics utilised
and reported
Ehrmann et al.
[52] (2015)
a
Prospective
cohort, 2b
19 Professional
football
players, age
25.7 ±5.1
Any physical complaint
sustained by a player that
resulted from a soccer
match or soccer training,
irrespective of the need
for medical attention or
time loss from soccer
activities
Nil GPS: total distance, high-
intensity running
(14.3–19.7 km.h
-1
),
very-high-intensity
running ([19.7 km.h
-1
),
metres per minute and
new total body load (sum
of tri-axial
accelerometry) Note:
competition loads are
estimates
1–4 weeks, relative Higher metres/min in the 1-week and 4-week
periods prior to injury (?9.6 % and ?7.4 %,
respectively; p=0.008; ES 0.52). New total
body load significantly lower in the 1 week
(p=0.06; ES 0.54) and 4 weeks (p=0.01;
ES 0.58) preceding injury
Australian football
Piggott [11]
(2009)
Prospective
cohort, 2b
16 Elite
Australian
Football
players, age
23.8 ±5.1
Injury: any pain or
disability suffered by a
player during a training
session or game that
restricted full
participation in the
general training
programme
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Minutes [80 %
of maximum
heart rate
Total distance run (km) and
total distance run
[12 km/h
1 week, relative 40 % of injuries (n=6) were associated with a
preceding spike in training load ([10 %
increase compared with the preceding week)
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Rogalski et al. [7]
(2013)
Prospective
cohort, 2b
46 Elite
Australian
Football
players, age
22.2 ±2.9
Time loss of training
session or match
confirmed by
physiotherapist (medical
attention)
RPE 9time
(mins)
RPE (0–10)[10]
Time =total
training session
minutes
Nil 1–2 weeks, both When comparing weekly load, athletes with
[1750 AU vs. \1250 AU were more likely
to be injured (OR 2.44, 95 % CI 1.28–4.66);
absolute weekly change in training load
[1250 AU was a risk factor (OR 2.58,
95 % CI 1.43–4.66)
Colby et al. [13]
(2014)
a
Prospective
cohort, 2b
46 Elite
Australian
Football
players, age
25.1 ±3.4
Medical attention and/or
time-loss injuries. Only
intrinsic injuries included
in analysis
Nil GPS: total distance, sprint
distance ([75 % of
athlete’s maximum
speed), V1 distance,
velocity load, relative
velocity change load,
force load
1–4 weeks, both Pre-season: 3-weekly distance between 73,721
and 86,662 m was associated with higher risk
of injury when compared with \73,721 m or
[86,662 m (OR 5.49, 95 % CI 1.57–19.16)
Moderate 3-weekly sprint distance and velocity
load were protective (OR 0.239, 95 % CI
0.06–0.92 and OR 0.23, 95 % CI 0.05–0.97,
respectively)
In-season: higher 3-weekly force load
[5397 AU vs. \4561 AU and 4-weekly RVC
load [102 AU was associated with increased
risk (OR 2.52, 95 % CI 1.09–5.87 and
OR 2.24, 95 % CI 1.06–4.77, respectively),
whereas moderate 2-weekly total distance and
high 2-weekly V1 distance were protective
(OR 0.43, 95 % CI 0.20–0.89 and OR 0.28,
95 % CI 0.11–0.69, respectively)
Rowing
Wilson et al. [53]
(2010)
Prospective
cohort, 2b
20 Olympic
rowers, age
26.25 ±4.18
Modification of Waller
et al. [68]
Missed competition, missed
two training sessions or
medical attention
Nil Training and competition
hours
Nil, absolute When examining the specific training type, there
were significant associations between monthly
ergometer time and injury (r=0.75,
p=0.01), time spent training with heavy
weights and injury (r=0.66, p=0.02) and
time spent on core stability and injury
(r=0.68, p=0.01)
Training Load and Injury
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Baseball
Lyman et al. [16]
(2001)
a
Prospective
cohort, 2b
398 Pitcher
seasons
(across two
seasons in
298 junior
athletes), age
10.8 (range
8.1–12.4)
Shoulder or elbow pain
classified as mild (pain
without time loss), minor
(competition loss in the
game where onset
occurred), moderate
(practice or competition
loss following onset,
medical attention or
stopping pitching
for [2 weeks), serious
(cessation of pitching for
the season)
Nil Competition pitches Nil, absolute; cumulative
pitches before onset
evaluated. No timeframe
mentioned
Moderate cumulative season pitches (300–599)
before the game was protective (OR 0.54,
95 % CI 0.32–0.92), low (\300; referent
OR 1.00) and high ([600; OR 2.07, 95 % CI
0.68–6.26) was an increased risk. Increasing
game pitchers, per ten pitches, increased the
risk of both elbow and shoulder pain
(OR 1.06, 95 % CI 1.00–1.12 and OR 1.15,
95 % CI 1.08–1.23, respectively)
Lyman et al. [15]
(2002)
Prospective
cohort, 2b
476 Junior
baseball
pitchers, age
12 (range
9–14)
The outcomes of interest
were post-game
complaints of elbow or
shoulder pain. Injuries
requiring medical
treatment are rare for this
age group and this
criterion was therefore
not used as a measure of
outcome in the study
Nil Game pitches thrown, no
training logs were
recorded
Cumulative across one
season, absolute
Increases in the number of pitches per game
were associated with increased joint pain in
shoulders (1–24 vs. [100, OR 1.77,
p\0.01). Cumulative season pitches were
also associated with both shoulder and elbow
pain (1–200 vs. 800?, elbow OR 2.61;
shoulder OR 3.29, p\0.01)
Water polo
Wheeler et al.
[54] (2013)
Prospective
cohort, 2b
7 National
water polo
players, age
23 (range
18–29)
Shoulder soreness for their
dominant arm (athlete
reported definition)
Nil Number of shots taken NA, absolute Volume of shots accounted for 74 % variance in
shoulder pain
Basketball
Anderson et al.
[55] (2003)
Prospective
cohort, 2b
12 Female
NCAA
Division III
basketball
athletes,
mean age
unreported
(range
18–22)
An injury was defined as a
circumstance in which the
athlete received an
evaluation from the
team’s student athletic
trainer and ATC, and
required limiting their
practice for at least 1 day
The product of
the session
RPE and
session
duration was
defined as the
‘session
load’,which
was averaged
over each week
of training
Nil 2 weeks, absolute A moderately positive correlation was found
between weekly injuries and total weekly
training load (p=0.01; r=0.68) and
between strain and monotony (p=0.01;
r=0.67). Injuries tended to increase
following breaks in training
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Rugby union
Brooks et al. [56]
(2008)
Prospective
cohort, 2b
502
Professional
rugby union
players
Any injury that prevented a
player from taking a full
part in all training
activities typically
planned for that day and
match-play for more than
24 h from midnight at the
end of the day the injury
was sustained
Nil Accumulative training
h/week (quintiles; \5h;
5–6.2 h; 6.3–7.3 h;
7.4–9.1 h; [9.1 h/week)
1 week, absolute Increasing training volumes ([9.1 h/week) did
not correlate with an increased incidence of
injury (injury/1000 training hours). Higher
training volumes ([9.1 h/week) increased the
severity of injuries (days missed) sustained in
competition
Cross et al. [14]
(2015)
a
Prospective
cohort, 2b
173
Professional
rugby union
players
Competition time loss or
24-h time loss
RPE 9time
(mins)
RPE (0–10) [10]
Time =total
training session
minutes
Nil Absolute: Cumulative 2-, 3-
and 4-weekly loads
calculated by the sum of
the previous weeks’
training loads; week-to-
week change in loads
(absolute change in a
players current load from
that of the previous
week); weekly training
monotony) weekly
training strain; relative:
TSB
Reductions in injury risk were observed with
moderate monthly loads (5932–8651 AU,
OR 0.55, 95 % CI 0.22–1.38), with high loads
([8651 AU/month) associated with a higher
risk (OR 1.39, 95 % CI 0.98–1.98) compared
with low training (\3684 AU/month).
Absolute week-to-week changes [1069 AU
were associated with increased risk (OR 1.58,
95 % CI 0.98–2.54)
Swimming
Hellard et al. [12]
(2015)
Prospective
cohort, 2b
28 National-
(n=8) and
international-
level
(n=20)
swimmers,
14 male, 14
female, age
range 16–30
Medical attention by sports
physician
Swimming
intensity
defined blood
lactate
concentration.
Meters per
week at each
intensity
Nil Nil, absolute For every 10 % higher loads of training and
general conditioning, increased risk of injury
was observed (OR 1.49, 95 % CI 1.14–1.96
and OR 1.63, 95 % CI 1.20–2.21,
respectively). No difference was observed
between intensive and moderate training loads
for risk of injury; however, during tapering
athletes were of less risk (OR 0.46, 95 % CI
0.22–0.96)
Volleyball
Visnes and Bahr
[39] (2013)
Prospective
cohort, 2b
141 School
volleyball
players, 69
male, 72
female, age
range 16–18
Standardised clinical
assessment of jumper’s
knee: a history or pain in
the quadriceps or patella
tendons at their patella
insertions in connection
with training or
competition, and
tenderness to palpation
corresponding to the
painful area
Nil Number of sets played in
matches per week. Hours
of volleyball, beach
volleyball, strength, jump
training and other
training
For every extra hour trained in the week, the
odds of developing jumper’s knee increased
(OR 1.72, 95 % CI 1.18–2.53)
Match exposure was a predictor for developing
jumper’s knee (OR 3.88, 95 % CI 1.80–8.40)
for every extra set played per week
Training Load and Injury
123
Author's personal copy
Table 1 continued
References (year) Study design,
hierarchical
level of
evidence
Number of
participants,
sport(s), level
of competition
and age
(mean ±SD)
Definition of an injury Internal load
measure
External load measure Latent period,
absolute or relative
loads
Findings
Bahr and Bahr
[38] (2014)
Prospective
cohort, 2b
44 School
volleyball
players, 26
male
(jumper’s
knee,
n=12), 18
female
(jumper’s
knee, n=1),
age range
15–16
Standardised clinical
assessment of jumper’s
knee: a history of pain in
the quadriceps or patella
tendons at their patella
insertions in connection
with training or
competition, and
tenderness to palpation
corresponding to the
painful area
Nil Jump frequency in training
and competition: number
of jumps performed by
each player divided by
the hours of training and
competition of each
player
Nil, absolute No difference in jump frequency between
asymptomatic and symptomatic athletes
(p=0.28). Limited conclusions as this study
only monitored athletes for 1 week and five
matches
Various
Malisoux et al.
[57] (2013)
Prospective
cohort, 2b
154 (102
compliant
athletes)
national-
level
adolescent
athletes from
various
sports, age
14.1 (range
12–19)
Any physical complaint
resulting from a match or
training session that
forced an athlete to
interrupt or modify the
usual training plan for at
least one session (time
loss)
4-point
RPE 9time
(mins)
Nil 1 week, relative Sport category was the only significant risk
factor (p\0.001). Racket and individual
sports being associated with a lower injury
risk compared with team sports (HR 0.37 and
0.34, p=0.001 and p\0.001, respectively).
Intensity was found to be significantly higher
during the week prior to injury when
compared with the 4 preceding weeks
2b ‘Individual cohort study’, determined by the Oxford Centre of Evidence-based Medicine [37], RPE rate of perceived exertion, SD standard deviation, HR hazard ratio, OR odds ratio, AU arbitrary units of load, RR
relative risk, ES effect size, GPS global positioning system, V1 distance, ‘total distance covered (m) above the individual player’s aerobic (blood lactate approximately 2 mmol L
-1
) threshold speed, as determined from a
pre-season incremental speed (1 % gradient) treadmill running test to exhaustion’, velocity load ‘a measurement of running power and momentum; a more continuous and higher velocity equates to a higher velocity load’,
RVC relative velocity change load, ‘a calculated function (algorithm) of accelerations, decelerations, and changes of direction, which are summed together to produce an overall ‘acceleration load’ value, force load ‘a
cumulative measurement that sums the forces produced from both foot strikes and collisions’, NA Not applicable, NCAA National Collegiate Athletic Association, ATC athletic trainer, CI confidence interval, TSB training
stress balance, RCT randomised controlled trial, NA not applicable
a
Not identified in the search strategy
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 2 Summary of the results of the included studies indicating the relationship between training and competition load with illnesses in athletes
References (year) Study
design,
hierarchical
level of
evidence
Number of participants, sport(s), level
of competition and age (mean ±SD)
Definition of an illness Internal load measure External load measure Latent
period,
absolute or
relative
loads
Findings
Australian football
Piggott [11]
(2009)
Prospective
cohort, 2b
16 Elite Australian Football players,
age 23.8 ±5.1
Illness: occurrence where the
player missed a training
session or game due to a
medical condition that was
diagnosed by the club’s
doctor
RPE 9time (mins)
RPE (0–10) [10]
Time =total training
session minutes
Minutes [80 % of
maximum HR
Total distance run (km)
and total distance run
[12 km/h
1 week,
relative
42 % of illnesses (n=3) were
associated with a preceding spike
in training load ([10 % increase
compared with the preceding week)
Middle-distance runners
Fricker et al. [59]
(2005)
Prospective
cohort, 2b
20 national- to international-level
middle-distance and distance runners,
age 24.2 ±3.1
Athlete self-reported illness
symptoms for [2
consecutive days
confirmed by a sports
physician’s clinical
examination
Distance 9intensity
Distance (km)
Intensity scale (1–5;
1=light,
5=maximal)
Total distance (km) 1 week,
absolute
No substantial correlation observed
between weekly mileage (km),
intensity of training or training
load, and the number of illnesses
reported
Basketball
Anderson et al.
[55] (2003)
Prospective
cohort, 2b
12 Female NCAA Division III
basketball athletes, mean age
unreported (range 18–22)
An illness was defined as a
circumstance in which the
athlete or medical doctor
felt the athlete was limited
or unable to perform the
drills as directed by the
coach
The product of the
session RPE and
session duration was
defined as the
‘session load’, which
was averaged over
each week of training
Nil 2 weeks,
absolute
No correlation was found between
total weekly training loads and
illness rates (r=0.099)
Football (Soccer)
Brink et al. [50]
(2010)
a
Prospective
cohort, 2b
53 Junior national-level football
players, age 16.5 ±1.2
Injury: any physical
complaint sustained by a
player that results from a
soccer match or soccer
training, irrespective of the
need for medical attention
or time loss from soccer
activities
RPE 9time (mins)
RPE (6–20)
Time =total training
session minutes
Nil 1 week,
absolute
Ill athlete had higher duration (mins;
OR 1.12, 95 % CI 1.00–1.26) but
not training load, monotony and
strain compared with the healthy
athletes in the preceding week
Speed skating
Foster [9] (1998)
a
Prospective
cohort, 2b
25 Competitive speed skaters; national
level, n=8; Olympic level, n=3,
unknown, n=13, age 26.3 ±3.2
Not defined RPE [70] 9time
(mins)
Time =total training
session minutes
Nil \10 days,
absolute
84 % of illnesses could be explained
by a preceding spike in training
load above the individual training
threshold, with 55 % of excursions
above the threshold not associated
with illness. 77 % of illness was
associated with a monotony spike,
and 52 % of monotony spikes were
not associated with illness. 89 % of
illness was associated with a strain
spike, and 59 % of strain spikes
were not associated with illness
Training Load and Injury
123
Author's personal copy
Table 2 continued
References (year) Study
design,
hierarchical
level of
evidence
Number of participants, sport(s), level
of competition and age (mean ±SD)
Definition of an illness Internal load measure External load measure Latent
period,
absolute or
relative
loads
Findings
Rugby league
Thornton et al.
[69] (2015)
a
Prospective
cohort, 2b
32 Professional (n=18) or semi-
professional (n=14) rugby league
players, age 26 ±4.8
Self-reported symptoms of
URTI
RPE 9time (mins)
RPE (0–10) [2]
Time =total training
session minutes
Nil Nil,
absolute
The predictive models developed
(decision trees, random forest and
boosting models) showed increases
in internal training load (strain
values of [2282 AU, weekly
training load [2786 AU and
monotony [0.78 AU) to best
predict when athletes are at
increased risk of self-reported
illness
Swimming
Hellard et al. [12]
(2015)
Prospective
cohort, 2b
28 National- (n=8) and international-
level (n=20) swimmers, 14 male,
14 female, age range 16–30
Medical attention by sports
physician
Swimming intensity
defined blood lactate
concentration.
Meters/week at each
intensity
Nil Nil,
absolute
The risk of URTPI was slightly but
significantly increased with
resistance (OR 1.08, 95 % CI
1.01–1.16) and high-load
(OR 1.10, 95 % CI 1.01–1.19)
training. Intensive training showed
a higher risk of URTPI (referent)
compared with moderate (OR 0.74,
95 % CI 0.49–0.70), tapering
(OR 0.30, 95 % CI 0.13–0.70) and
competition (OR 0.50, 95 % CI
0.23–1.06)
Various
Gleeson et al.
[63] (2013)
Prospective
cohort, 2b
90 (75 completed the study) healthy
university students of various
sporting involvement (predominantly
endurance-based activities such as
running, cycling, swimming,
triathlon, team games and racquet
sports), 50 male, 40 female, age
22.5 ±4.0
Self-reported symptoms of
URTI
Nil Training hours per
week
Nil,
absolute
High ([11 h/week) and moderate
(7–10 h/week) training had a
higher proportion of URTI than
low (3–6 h/week; p\0.001 and
p=0.01, respectively). 81 % of
participants stated training was
negatively affected when suffering
a URTI
2b ‘Individual cohort study’, determined by the Oxford Centre of Evidence-based Medicine [31], RPE rate of perceived exertion, monotony average daily training load/standard deviation of the daily load for each week,
strain monotony 9weekly load, URTI upper respiratory tract infection, URTPI upper respiratory tract and pulmonary infections, NCAA National Collegiate Athletic Association, HR heart rate, OR odds ratio, CI
confidence interval, AU arbitrary units of load
a
Not identified in the search strategy
M. K. Drew, C. F. Finch
123
Author's personal copy
Table 3 Quality of the reviewed studies according to the Newcastle–Ottawa Assessment Scale for cohort studies
References NOS score
b
Selection Comparability Outcome Total score
Studies reporting injury and soreness
Rugby league
Gabbett [40] (2004) 3 1 3 7
Gabbett [41] (2004) 3 2 3 8
Gabbett and Domrow [42] (2005)
b
32 38
Gabbett and Domrow [43] (2007) 3 2 3 8
Gabbett [68] (2010) 3 2 3 8
Killen et al. [44] (2010) 3 2 3 8
Gabbett and Jenkins [46] (2011) 3 2 3 8
Gabbett and Ullah [62] (2012)
b
41 38
Hulin et al. [19] (2015) 3 2 3 8
Cricket
Dennis et al. [2] (2003) 3 2 3 8
Dennis et al. [49] (2005) 3 2 2 7
Orchard et al. [47] (2009) 3 2 2 7
Hulin et al. [3] (2014) 3 2 3 8
Orchard et al. [48] (2015)
b
32 38
Football
Lovell et al. [51] (2006)
b
30 36
Brink et al. [50]
b
32 38
Ehrmann et al. (2015) [52] (2010)
b
30 36
Australian football
Piggott [11] (2009) 3 0 2 5
Rogalski et al. [7] (2013) 3 2 3 8
Colby et al. [13] (2014)
b
32 38
Rowing
Wilson et al. [53] (2010) 3 1 2 6
Rugby union
Brooks et al. [56] (2008) 3 2 3 8
Cross et al. [14] (2015)
b
32 38
Baseball
Lyman et al. [16] (2001)
b
22 15
Lyman et al. [15] (2002) 3 1 2 6
Volleyball
Visnes and Bahr [39] (2013) 4 2 3 9
Bahr and Bahr [38] (2014) 4 2 3 9
Swimming
Hellard et al. [12] (2015) 4 2 3 9
Water polo
Wheeler et al. [54] (2013) 3 2 2 7
Basketball
Anderson et al. [55] (2003) 3 1 3 7
Various
Malisoux et al. [57] (2013) 2 2 2 6
Median (range) 3 (2–4) 2 (0–2) 3 (1–3) 8 (5–9)
Training Load and Injury
123
Author's personal copy
designing internal load measures in a practical setting. This
study addresses the need for level 1 (systematic review)
hierarchical evidence [37] exploring the relationships
between the training loads of athletes (dose) and the neg-
ative outcomes (injury and/or illness).
4.1 Relationship Between Training Load and Injury
The relationship of injury incidence and training loads was
evaluated in 31 of the included studies [2,3,7,1116,19,
3857,62]. Across the studies, increased workload (com-
petition and training loads) was related to subsequent
injuries and injury incident rates in 28 studies, with three
studies [38,49,56] finding no significant relationship. In
cricket, there was a relationship in the relative/absolute
workload between balls bowled (external load) and injury
risk. The TSB of 200 % was shown to be associated with a
relative risk of 3.3 (95 % confidence interval [CI]
1.50–7.25) for injury in the following week [3], and this risk
was sustained for up to 3–4 weeks [48]. In the same study
population, internal loads (RPE 9duration) were shown to
be twice as predictive as external loads (balls bowled) [3].
This study highlighted a key factor, namely that external
loads only partially quantify training, and can thus only
provide a partial quantification of the risk of injury. Future
studies should therefore include an internal load measure
across all modalities of training as a minimum. Replication
of the risk profiles identified in Hulin et al. [3] is also
warranted to validate the previous statement.
Important practical implications arising from this review
include inappropriate prescription of training loads (train-
ing error) on a given day could influence injury risk of the
athlete for up to 1 month [47]. Furthermore, the amount of
training possible should be derived from recent historical
workloads, as illustrated by Hulin et al. [19], who showed
that the chronic load of an athlete combined with the rel-
ative load has a higher predictive capacity than absolute
loads alone, therefore introducing the concept of the
‘workload-injury paradoxwhereby higher chronic work-
load is protective against injury when acute workload is
similar to chronic workload. This highlights the need for
proper (or careful) planning training sessions and review-
ing the plan within the context of the current training loads.
This point was previously shown by Gabbett [45], where an
injury prediction model of planned and actual training
loads was developed for rugby league players. He found
that players who exceeded the threshold of the model
(higher actual training loads than planned) were 70 times
more likely to sustain a non-contact, soft-tissue injury. In
Australian Football, lower levels of accumulated load led
to higher injury rates in novice players compared with
more experienced players, with the former experiencing
injury events much sooner than the latter [58]. This is
contradictory to previous work [7] and is likely to be
explained in the statistical methodology whereby Forting-
ton et al. [58] utilised survival analysis rather than relative
risk calculations. On the basis of these collective results,
daily re-evaluation of the training plan, taking into account
Table 3 continued
References NOS score
b
Selection Comparability Outcome Total score
Studies reporting illness
Rugby league
Thornton et al. [69] (2015) 2 1 2 5
Australian football
Piggott [11] (2009) 3 0 2 5
Football (soccer)
Brink et al. [50] (2010)
b
32 38
Speed skating
Foster [9] (1998)
b
21 25
Swimming
Hellard et al. [12] (2015) 4 2 3 9
Middle-distance runners
Fricker et al. [59] (2005) 3 1 2 6
Basketball
Anderson et al. [55] (2003) 3 1 3 7
Various
Gleeson et al. [63] (2013) 3 1 2 6
Median (range) 3 (2–4) 1 (0–2) 2 (2–3) 6 (5–9)
M. K. Drew, C. F. Finch
123
Author's personal copy
recent workloads, is recommended for elite athletes and
must be proportional to their age.
4.2 Relationship Between Training Load and Illness
The relationship between training loads and illness was
found to be moderate. A positive relationship was found in
six studies (75 %) [9,11,12,50,63,69], with only two
studies finding no relationship [55,59]. No relationship
was observed between internal and external load measures
in an elite running population [59] and a possible causal
link found in collegiate female basketballers [55], with
42 % of illnesses associated with a preceding spike in
training load ([10 % increase compared with the preced-
ing week) in Australian Football [11]. One study [12]
received a maximum NOS score and reported a dose-re-
sponse relationship. In this study, swimmers who trained in
moderate training loads compared with intensive training
were protected, with a reduction in risk of 26 %.
4.3 Relationship Between Training Load
and Soreness
Shooting volume accounted for 74 % of the variance in
shoulder pain, indicating a dose-response relationship
between the external load and negative outcome (shoulder
soreness) in elite female water polo players [54]. Similarly,
in youth baseball, the number of pitches in competition was
related to shoulder and elbow pain when pitching [15,16,
60]. This, in combination with the evidence pertaining to
cricket fast bowlers [2,3,48] and other similar papers that
were not included in this review [60,61], highlights the
strong need to monitor the number of throws/pitches/balls
bowled in these types of sports.
4.4 Reported Internal and External Load Measures
In relation to injuries, internal load measures were utilised
in the majority of included studies, with the product of
RPE and time being the most frequently reported method
(n=25). External loads were primarily utilised in studies
based on sporting actions with throwing as the major
component, such as cricket fast bowling, baseball pitch-
ing, and water polo shots at goal. Distance run was found
to have no relationship with illness, whereas weekly
accumulated time of training was associated with injuries
in rowers and Australian Football players but not rugby
union. This is in contrast to recent studies that have
highlighted a significant relationship [52,62], and is best
explained by the methods of data collection, i.e. instru-
mentation improves the relationship, indicating self-re-
ports of training distances by the athlete may be
unreliable.
When considering the evidence for the relationship
between illness and training load, it is important to note the
different method of quantification of internal training loads
used across the studies. Fricker et al. [59] defined internal
load as the product of distance and a 5-point RPE scale
[59], and found no relationship between this internal load
measure and illness incidence. In contrast, two studies [11,
55] utilised a 10-point RPE scale by time, with an asso-
ciation observed in one study [11]. In another study, which
had the highest quality of those reviewed, Hellard et al.
[12] found a significant relationship between training load
and illness, and utilised an internal loading measure as the
product of intensity (determined by lactate concentration
training zones) by duration. Two studies evaluated external
loads and illness incidence. In middle-distance runners
[59], no relationship was found between total distance and
injury, with spikes in training loads preceding illnesses in
Australian Football [11]; however, caution should be
exercised when considering these results as this study had
low participant numbers and incidence rate. Higher training
duration was significantly associated with illness in foot-
ball (soccer) [50] and in Gleeson et al. [63] who evaluated
upper respiratory tract infections (URTIs) in a variety of
university athletes.
4.5 Latent Period Between Training Load
and the Onset of Injury and/or Illness
Future studies should analyse the relationship between the
dose (training load) and injury/illness response, acknowl-
edging a lag between the two events up to a period of
4 weeks [19,48]. To illustrate this point, consider injured
players who are removed from training (time-loss injury)
or have training reduced as a component of their man-
agement. If the injury occurs in the same week as the data
being analysed then an artificially low training load will be
recorded if the data are analysed over a specified Monday–
Sunday time period rather than as a rolling 7-day load. This
situation may likely to be a type II error. In simple terms,
the injury led to reporting of lower training loads, rather
than the injury causing low training loads. Future studies
should report and analyse this latent period with respect to
the type of tissue injury or illness, i.e. bone, muscle and
tendon versus medical complaints such as URTIs.
4.6 Training Load as a Protectant Against Injury,
Illness or Soreness
This review has focused on the negative consequences of
inappropriate loading of an athlete. However, it must be
stated that training load can also be protective against
injury and illness. In thirteen studies, a protective effect
was reported [2,3,1214,16,19,4244,48,51,62]. In
Training Load and Injury
123
Author's personal copy
cricket, ‘under-loading’ or insufficient level of fast-bowl-
ing overs has been shown to increase the risk of injury.
This can be expressed as either absolute external loads,
such as overs bowled [2], or a relative load expressed as
TSB (acute:chronic workload ratio) [3]. Contemporary
publications highlighted that moderate chronic loads are
protective, whereas high and low loads are not [14,48].
The combination of absolute and relative workloads is
more predictive than absolute workloads in isolation and
should be considered in future studies [19]. This is
important because the avoidance of training within mod-
erate load zone ranges could expose the athlete to an
increased risk of injury.
4.7 Level and Quality of Evidence
The NOS was chosen post hoc as all included studies were of
non-randomised designs and therefore this was the most
appropriate scale to assess the quality of the included studies.
The quality of studies was excellent, with three (9 %)
achieving nine out of a possible nine points on the NOS. Very
few (n=3) studies reported information that injuries, ill-
nesses or soreness were present at the start of the study.
Future studies should report the presence or absence of injury
or illnesses at the start of the surveillance period to achieve
maximum NOS score. The comparability section of the NOS
was scored the poorest, with studies generally failing to
control for one or more factors. This is best evidenced by
studies only analysing data pertaining to training loads rather
than reporting or controlling for other factors which may
influence the risk of injury/illness. Future studies should
attempt to control for risk factors of the injuries and illnesses
in their sport to improve this subsection of the NOS. For
example, studies evaluating hamstring injury in Australian
Football should control for previous injury [64], eccentric
weakness [65] and high-speed distances [13].
Although 35 studies of good quality were identified in
this review, the level of evidence was determined as
moderate pertaining to the relationship between injuries
and training loads. It was unable to be declared as strong
because of the lack of RCTs available to fulfil the van
Tulder et al. [36] criteria. This is an area of future research;
however, we acknowledge the practical and ethical limi-
tations of such study designs at the elite and sub-elite level
of sport, such as withholding management of loads to
athletes who are reliant on this to perform in their sport.
While the van Tulder et al. [36] criteria led to the
conclusion of moderate evidence, there is significant scope
for increased numbers of studies evaluating this relation-
ship as only 33 studies were identified in this review. These
studies covered only eight sports, highlighting the need for
future research across a wider variety of sports as well as
comparisons of risk profiles between sports using the same
methodology. All included studies were rated as being
hierarchical level of evidence [37] ‘2b’. This relates to
individual cohort studies.
4.8 Recommendations for Future Studies
4.8.1 Definition of an Injury and/or Illness
The definition of an injury ranged from reports of soreness
in a joint, medical attention injuries by qualified medical
practitioners, and medical attention injuries by unqualified
medical practitioners, to time loss injuries of various
severity scales. Future studies should report the definition
of an injury and illness in the context of the ICDF [21] and
use qualified medical personnel to diagnose and report their
epidemiological characteristics. It is therefore recom-
mended that studies report their definition in relationship to
sports performance, clinical examination and athlete self-
report, or a combination of all three [21].
4.8.2 Use of Internal and External Loads as Risk Factors,
Effect Modifiers and Confounding Variables
There is moderate evidence for the relationship of both
internal and external loads to sustaining an injury or illness. It
is recommended that external loads of ‘throwing units’ (balls
bowled, pitches or shots at goal) be measured for future
studies of injury risk in throwing athletes and that surveil-
lance periods following spikes in training and competition
loads must be longer than 4 weeks in duration to ensure all
risk periods are captured. For rugby league and other team
sports, it is recommended that future injury risk studies also
account for the training and competition loads using the
product of the 10-point RPE scale and duration of the session
[10]. However, there are limitations to this method and care
should be taken. For example, this internal load measure-
ment does not account for the modality/type of training, with
equal loading given to short, intense sessions as to low-in-
tensity, long events. To illustrate this point, a session with an
RPE of 9, lasting 30 min, will produce an internal load of 270
units, whereas a session with an RPE of 5, lasting 60 min,
will produce 300 units. Similarly, when comparing sports we
must acknowledge that endurance-based sports will gener-
ally have longer durations at lower intensity, with other
sports such as football focusing on higher intensity training
with lower duration.
4.8.3 Modelling Data on the Acute and Chronic Load
Difference
Several studies included in the review [3,7,13,14,19,47,
62] examined the risk of injury as an index of recent and
historical workloads utilising the TSB (acute:chronic
M. K. Drew, C. F. Finch
123
Author's personal copy
workload ratio) as a metric for risk. This is a central con-
cept and has implications for future studies. Relative loads
measured using the TSB could hold important information
that has been overlooked; for example, in this method,
injury risk is determined by the magnitude of the ratio
between the acute load (7-day average daily loads) com-
pared with the chronic (28-day average daily loads) rather
than absolute loads per se, or preferably in combination
with the chronic load [19]. This explains the clinical
observation that highly trained athletes sustain injuries on
lower-than-normal workloads following a break from
training [55,66]. Future studies should examine the rela-
tionship with respect to both absolute and relative training
loads and their interdependence.
4.8.4 Monitor Latent Periods Between Load and Adverse
Outcomes
A latent period should also be built into the studies as the
risk of spikes in loads is actualised in the weeks following
these periods [3,19,47,48]. This review identified studies
which show that a latent period can exist for up to 4 weeks
post spikes in training load [47,48]. Future studies must
include a potential lag between the dose (e.g. spike in
training loads) and negative outcome (injury or illness). It
is recommended that these be reported relative to the body
tissue, and the timelines of the pathological processes [48]
has been shown to be related to inconsistent loading [48]
and bone injury up to 4 weeks following acute spikes in
training and competition loads [47,48].
4.9 Strengths and Limitations
Ten studies were identified through searching references of
the identified studies, leading to a low risk of studies not
being included. Furthermore, studies were assessed for
their quality, with the included studies synthesised to report
the level of evidence for the relationship between training
loads and injuries and illnesses. The van Tulder et al.
method [36] is an accepted method of measuring strength
of evidence and is highly cited. However, without RCTs
the strength of evidence is unable to be determined as
strong. This is a real-world limitation of research in sport
and is not confined to training load research per se. The
inclusion of the NOS strengthens the review results. The
included studies are generally of a high quality. A limita-
tion of this systematic review was that the NOS was
completed by a single author. The NOS limitations inclu-
ded the number of items and limited scope of assessment,
which may reduce sensitivity to the appropriate design,
blinding, statistical analysis, handling of missing data and
internal bias when compared with Downs and Black [67],
the other Cochrane Handbook [35] recommended
assessment tool. It was not possible to receive level 1a
evidence as review articles were excluded.
5 Conclusions
The link between training and competition loads and both
injury risk and illness is an emerging field of research interest,
with 35 studies published prior to October 2015. There is
moderate evidence indicating a dose-response relationship
between the amount of training and competition loading that
an athlete undertakes and the incidence of injury, illness and
soreness. Training loads can either be positive (increased or
decreased load reduces the chance of injury/illness) or nega-
tive (increased risk of injury/illness with increases or
decreases in training load), or a combination of these
depending on the method of quantification. It appears this is
also dependent on whether this training load is viewed as
absolute or relative to an athlete’s recent training history. In
team-based sports it is recommended that internal loads are
routinely measured and reviewed, with throw counts recom-
mended for all throwing sports. Where an athlete has a spike in
training, the consequences may not be realised until
3–4 weeks following this loading; therefore, studies must
include surveillance periods for at least 3–4 weeks following
these periods. Clinically, the coach, athlete and support staff
must be aware of the risks imposed following spikes in
training loads, with an increased requirement to manage the
training loads and environment when a spike is identified to
ensure athlete safety is maximised.
Compliance with Ethical Standards
Funding Funding Caroline Finch was supported by a National
Health and Medical Research Council (of Australia) Principal
Research Fellowship (ID: 1058737). No other sources of funding
were used to assist in the preparation of this article. The Australian
Centre for Research into Injury in Sport and its Prevention (ACRISP)
is one of the International Research Centres for Prevention of Injury
and Protection of Athlete Health supported by the International
Olympic Committee (IOC).
Conflict of interest Michael Drew and Caroline Finch declare that
they have no conflicts of interest relevant to the content of this sys-
tematic review.
References
1. Gamble P. Reducing injury in elite sport-is simply restricting
workloads really the answer? NZ J Sports Med. 2013;40(1):34–6.
2. Dennis R, Farhart R, Goumas C, et al. Bowling workload and the
risk of injury in elite cricket fast bowlers. J Sci Med Sport.
2003;6(3):359–67.
3. Hulin BT, Gabbett TJ, Blanch P, et al. Spikes in acute workload
are associated with increased injury risk in elite cricket fast
bowlers. Br J Sports Med. 2014;38:708–12.
Training Load and Injury
123
Author's personal copy
4. Busso T. Variable dose-response relationship between exercise
training and performance. Med Sci Sports Exerc. 2003;35(7):
1188–95.
5. Ha
¨gglund M, Walde
´n M, Magnusson H, et al. Injuries affect team
performance negatively in professional football: an 11-year fol-
low-up of the UEFA Champions League injury study. Br J Sports
Med. 2013;47(12):738–42.
6. Ekstrand J, Walde
´nM,Ha
¨gglund M. A congested football cal-
endar and the wellbeing of players: correlation between match
exposure of European footballers before the World Cup 2002 and
their injuries and performances during that World Cup. Br J
Sports Med. 2004;38(4):493–7.
7. Rogalski B, Dawson B, Heasman J, et al. Training and game
loads and injury risk in elite Australian footballers. J Sci Med
Sport. 2013;16(6):499–503.
8. Halson SL. Monitoring training load to understand fatigue in
athletes. Sports Med. 2014;44(2):139–47.
9. Foster C. Monitoring training in athletes with reference to over-
training syndrome. Med Sci Sports Exerc. 1998;30(7):1164–8.
10. Foster C, Florhaug JA, Franklin J, et al. A new approach to
monitoring exercise training. J Strength Cond Res. 2001;15(1):
109–15.
11. Piggott B. The relationship between training load and incidence
of injury and illness over a pre-season at an Australian Football
League Club. J Aust Strength Cond. 2009;17(3):4–17.
12. Hellard P, Avalos M, Guimaraes F, et al. Training-related risk of
common illnesses in elite swimmers over a 4-yr period. Med Sci
Sports Exerc. 2015;47(4):698–707.
13. Colby MJ, Dawson B, Heasman J, et al. Accelerometer and GPS-
derived running loads and injury risk in elite Australian foot-
ballers. J Strength Cond Res. 2014;28(8):2244–52.
14. Cross M, Williams S, Trewartha G, et al. The influence of in-
season training loads on injury risk in professional rugby union.
Int J Sports Physiol Perform. 2015. (Epub 26 Aug 2015).
15. Lyman S, Fleisig GS, Andrews JR, et al. Effect of pitch type,
pitch count, and pitching mechanics on risk of elbow and
shoulder pain in youth baseball pitchers. Am J Sports Med.
2002;30(4):463–8.
16. Lyman S, Fleisig GS, Waterbor JW, et al. Longitudinal study of
elbow and shoulder pain in youth baseball pitchers. Med Sci
Sports Exerc. 2001;33(11):1803–10.
17. Banister EW. Modeling elite athletic performance. In: Green HJ,
editor. Physiological testing of the high-performance athlete.
Champaign: Human Kinetics Books; 1991. p. 403–24.
18. Banister EW, Calvert TW. Planning for future performance:
implications for long term training. Can J Appl Sport Sci.
1980;5(3):170–6.
19. Hulin BT, Gabbett TJ, Lawson DW, et al. The acute:chronic
workload ratio predicts injury: high chronic workload may
decrease injury risk in elite rugby league players. Br J Sports
Med. 2015. (Epub 28 Oct 2015).
20. Veugelers KR, Young WB, Fahrner B, et al. Different methods of
training load quantification and their relationship to injury and
illness in elite Australian football. J Sci Med Sport.
2016;19(1):24–8.
21. Timpka T, Jacobsson J, Ekberg J, et al. Meta-narrative analysis of
sports injury reporting practices based on the Injury Definitions
Concept Framework (IDCF): a review of consensus statements
and epidemiological studies in athletics (track and field). J Sci
Med Sport. 2015;18(6):643–50.
22. Ha
¨gglund M, Walde
´n M, Bahr R, et al. Methods for epidemio-
logical study of injuries to professional football players: devel-
oping the UEFA model. Br J Sports Med. 2005;39(6):340–6.
23. Fuller CW, Ekstrand J, Junge A, et al. Consensus statement on
injury definitions and data collection procedures in studies of
football (soccer) injuries. Scand J Med Sci Sports.
2006;16(2):83–92.
24. Fuller C, Laborde F, Leather R, et al. International rugby board
rugby world cup 2007 injury surveillance study. Br J Sports Med.
2008;42(6):452–9.
25. Pluim BM, Fuller CW, Batt ME, et al. Consensus statement on
epidemiological studies of medical conditions in tennis. Br J
Sports Med. 2009;43(12):893–7.
26. Junge A, Engebretsen L, Alonso JM, et al. Injury surveillance in
multi-sport events: the International Olympic Committee
approach. Br J Sports Med. 2008;42(6):413–21.
27. Timpka T, Alonso J-M, Jacobsson J, et al. Injury and illness
definitions and data collection procedures for use in epidemio-
logical studies in athletics (track and field): consensus statement.
Br J Sports Med. 2014;48(7):483–90.
28. Clarsen B, Krosshaug T, Bahr R. Overuse injuries in professional
road cyclists. Am J Sports Med. 2010;38(12):2494–501.
29. 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(8):495–502.
30. Timpka T, Jacobsson J, Bickenbach J, et al. What is a sports
injury? Sports Med. 2014;44(4):423–38.
31. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items
for systematic reviews and meta-analyses: the PRISMA state-
ment. Ann Intern Med. 2009;151(4):264–9.
32. Gabbett TJ, Whyte DG, Hartwig TB, et al. The relationship between
workloads, physical performance, injury and illness in adolescent
male football players. Sports Med. 2014;44(7):989–1003.
33. Wells G, Shea B, O’Connell D, et al. The Newcastle-Ottawa
Scale (NOS) for assessing the quality of nonrandomised studies
in meta-analyses. 2000. Available at: http://www.ohri.ca/
programs/clinical_epidemiology/oxford.asp. Accessed 17 Oct
2015.
34. Quigley JM, Thompson JC, Halfpenny N, et al. Critical appraisal
of non-randomized controlled studies: a review of recommended
and commonly used tools. Value Health. 2014;17(3):A203.
35. Higgins JP, Green S. Cochrane handbook for systematic reviews
of interventions. New York: Wiley Online Library; 2008.
36. Van Tulder M, Furlan A, Bombardier C, et al. Updated method
guidelines for systematic reviews in the Cochrane Collaboration
Back Review Group. Spine. 2003;28(12):1290–9.
37. Phillips B, Ball C, Sackett D, et al. Oxford centre for evidence-
based medicine—levels of evidence. CEBM levels of evidence
(March 2009). Available at: http://www.cebm.net/oxford-centre-
evidence-based-medicine-levels-evidence-march-2009/. Acces-
sed 17 Oct 2015
38. Bahr MA, Bahr R. Jump frequency may contribute to risk of jum-
per’s knee: a study of interindividual and sex differences in a total of
11,943 jumps video recorded during training and matches in young
elite volleyball players. Br J Sports Med. 2014;48(17):1322–6.
39. Visnes H, Bahr R. Training volume and body composition as risk
factors for developing jumper’s knee among young elite volley-
ball players. Scand J Med Sci Sports. 2013;23(5):607–13.
40. Gabbett TJ. Influence of training and match intensity on injuries
in rugby league. J Sports Sci. 2004;22(5):409–17.
41. Gabbett TJ. Reductions in pre-season training loads reduce
training injury rates in rugby league players. Br J Sports Med.
2004;38(6):743–9.
42. Gabbett TJ, Domrow N. Risk factors for injury in subelite rugby
league players. Am J Sports Med. 2005;33(3):428–34.
43. Gabbett TJ, Domrow N. Relationships between training load,
injury, and fitness in sub-elite collision sport athletes. J Sports
Sci. 2007;25(13):1507–19.
M. K. Drew, C. F. Finch
123
Author's personal copy
44. Killen NM, Gabbett TJ, Jenkins DG. Training loads and inci-
dence of injury during the preseason in professional rugby league
players. J Strength Cond Res. 2010;24(8):2079–84.
45. Gabbett TJ. The development and application of an injury pre-
diction model for noncontact, soft-tissue injuries in elite collision
sport athletes. J Strength Cond Res. 2010;24(10):2593–603.
46. Gabbett TJ, Jenkins DG. Relationship between training load and
injury in professional rugby league players. J Sci Med Sport.
2011;14(3):204–9.
47. Orchard JW, James T, Portus M, et al. Fast bowlers in cricket
demonstrate up to 3-to 4-week delay between high workloads and
increased risk of injury. Am J Sports Med. 2009;37(6):1186–92.
48. Orchard JW, Blanch P, Paoloni J, et al. Cricket fast bowling
workload patterns as risk factors for tendon, muscle, bone and
joint injuries. Br J Sports Med. 2015;49:1064–8.
49. Dennis R, Finch CF, Farhart P. Is bowling workload a risk factor
for injury to Australian junior cricket fast bowlers? Br J Sports
Med. 2005;39(11):843–6.
50. Brink MS, Visscher C, Arends S, et al. Monitoring stress and
recovery: new insights for the prevention of injuries and illnesses
in elite youth soccer players. Br J Sports Med. 2010;44(11):
809–15.
51. Lovell G, Galloway HR, Hopkins W, et al. Osteitis pubis and
assessment of bone marrow edema at the pubic symphysis with
MRI in an elite junior male soccer squad. Clin J Sport Med.
2006;16(2):117–22.
52. Ehrmann FE, Duncan CS, Sindhusake D, et al. GPS and injury
prevention in professional soccer. J Strength Cond Res. 2015.
(Epub 11 Jul 2015).
53. Wilson F, Gissane C, Gormley J, et al. A 12-month prospective
cohort study of injury in international rowers. Br J Sports Med.
2010;44(3):207–14.
54. Wheeler K, Kefford T, Mosler A, et al. The volume of goal
shooting during training can predict shoulder soreness in elite
female water polo players. J Sci Med Sport. 2013;16(3):255–8.
55. Anderson L, Triplett-Mcbride T, Foster C, et al. Impact of
training patterns on incidence of illness and injury during a
women’s collegiate basketball season. J Strength Cond Res.
2003;17(4):734–8.
56. Brooks JHM, Fuller CW, Kemp SPT, et al. An assessment of
training volume in professional rugby union and its impact on the
incidence, severity, and nature of match and training injuries.
J Sports Sci. 2008;26(8):863–73.
57. Malisoux L, Frisch A, Urhausen A, et al. Monitoring of sport
participation and injury risk in young athletes. J Sci Med Sport.
2013;16(6):504–8.
58. Fortington LV, Berry J, Buttifant D, et al. Shorter time to first
injury in first year professional football players: a cross-club
comparison in the Australian Football League. J Sci Med Sport.
2016;19(1):18–23.
59. Fricker PA, Pyne DB, Saunders PU, et al. Influence of training
loads on patterns of illness in elite distance runners. Clin J Sport
Med. 2005;15(4):244–50.
60. Fleisig GS, Andrews JR, Cutter GR, et al. Risk of serious injury
for young baseball pitchers: a 10-year prospective study. Am J
Sports Med. 2011;39(2):253–7.
61. Olsen SJ, Fleisig GS, Dun S, et al. Risk factors for shoulder and
elbow injuries in adolescent baseball pitchers. Am J Sports Med.
2006;34(6):905–12.
62. Gabbett TJ, Ullah S. Relationship between running loads and
soft-tissue injury in elite team sport athletes. J Strength Cond Res.
2012;26(4):953–60.
63. Gleeson M, Bishop N, Oliveira M, et al. Influence of training load
on upper respiratory tract infection incidence and antigen-stim-
ulated cytokine production. Scand J Med Sci Sports. 2013;
23(4):451–7.
64. Opar DA, Williams MD, Timmins RG, et al. The effect of pre-
vious hamstring strain injuries on the change in eccentric ham-
string strength during preseason training in elite Australian
footballers. Am J Sports Med. 2014;43(2):377–84.
65. Bourne MN, Opar DA, Williams MD, et al. Eccentric knee flexor
strength and risk of hamstring injuries in rugby union: a
prospective study. Am J Sports Med. 2015;43(11):2663–70.
66. Purdam C, Drew MK, Blanch PD, et al. Prescription of training
load in relation to loading and unloading phases of training. In:
Departments of Physical Therapies, Sports Medicine and Physi-
ology, Purdam C, Drew MK, editors. Bruce: Australian Sports
Commission; 2015.
67. Downs SH, Black N. The feasibility of creating a checklist for the
assessment of the methodological quality both of randomised and
non-randomised studies of health care interventions. J Epidemiol
Community Health. 1998;52(6):377–84.
68. Waller AE, Feehan M, Marshall SW, et al. The New Zealand
rugby injury and performance project: I. Design and methodology
of a prospective follow-up study. Br. J. Sports Med.
1994;28(4):223–228
69. Thornton HR, Delaney JA, Duthie GM, et al. Predicting self-
reported illness for professional team-sport athletes. Int J Sports
Physiol Perform. 2015. [Epub ahead of print].
Training Load and Injury
123
Author's personal copy
ResearchGate has not been able to resolve any citations for this publication.
  • Article
    Full-text available
    Systematic reviews should build on a protocol that describes the rationale, hypothesis, and planned methods of the review; few reviews report whether a protocol exists. Detailed, well-described protocols can facilitate the understanding and appraisal of the review methods, as well as the detection of modifications to methods and selective reporting in completed reviews. We describe the development of a reporting guideline, the Preferred Reporting Items for Systematic reviews and Meta-Analyses for Protocols 2015 (PRISMA-P 2015). PRISMA-P consists of a 17-item checklist intended to facilitate the preparation and reporting of a robust protocol for the systematic review. Funders and those commissioning reviews might consider mandating the use of the checklist to facilitate the submission of relevant protocol information in funding applications. Similarly, peer reviewers and editors can use the guidance to gauge the completeness and transparency of a systematic review protocol submitted for publication in a journal or other medium.
  • Article
    Full-text available
    Aim: Investigate whether acute workload (1 week total distance) and chronic workload (4-week average acute workload) predict injury in elite rugby league players. Methods: Data were collected from 53 elite players over two rugby league seasons. The ‘acute:chronic workload ratio’ was calculated by dividing acute workload by chronic workload. A value of greater than 1 represented an acute workload greater than chronicworkload. All workload data were classified into discrete ranges by z-scores. Results Compared with all other ratios, a very-high acute:chronic workload ratio (≥2.11) demonstrated the greatest risk of injury in the current week (16.7% injury risk) and subsequent week (11.8% injury risk). High chronic workload (>16 095 m) combined with a very high 2-week average acute:chronic workload ratio (≥1.54) was associated with the greatest risk of injury (28.6% injury risk). High chronic workload combined with a moderate workload ratio (1.02–1.18) had a smaller risk of injury than low chronic workload combined with several workload ratios (relative risk range from 0.3 to 0.7×/÷1.4 to 4.4; likelihood range=88–94%, likely). Considering acute and chronic workloads in isolation (ie, not as ratios) did not consistently predict injury risk. Conclusions: Higher workloads can have either positive or negative influences on injury risk in elite rugby league players. Specifically, compared with players who have a low chronic workload, players with a high chronic workload are more resistant to injury with moderate-low through moderate-high (0.85–1.35)acute:chronic workload ratios and less resistant to injury when subjected to ‘spikes’ in acute workload, that is, very-high acute:chronic workload ratios ∼1.5.
  • Article
    Full-text available
    Purpose: The aim of this study was to identify contributing factors to the incidence of illness for professional team-sport athletes, utilizing training load (TL), self-reported illness and wellbeing data. Methods: Thirty-two professional rugby league players (26.0 ± 4.8 yr; 99.1 ± 9.6 kg; 1.84 ± 0.06 m) were recruited from the same club. Players participated in prescribed training and responded to a series of questionnaires to determine the presence of self-reported illness and markers of wellbeing. Internal-TL was determined using the session rating of perceived exertion (sRPE) method. These data were collected over 29 weeks, across the preparatory and competition macrocycles. Results: The predictive models developed recognized increases in internal-TL (strain values of >2282 AU, weekly-TL >2786 AU and monotony >0.78 AU) to best predict when athletes are at increased risk of self-reported illness. In addition, a reduction in overall wellbeing (<7.25 AU) in the presence of increased internal-TL as previously stated, was highlighted as a contributor to self-reported illness occurrence. Conclusions: These results indicate that self-report data can be successfully utilized to provide a novel understanding of the interactions between competition-associated stressors experienced by professional team-sport athletes and their susceptibility to illness. This may assist coaching staff to more effectively monitor players during the season and to potentially implement preventative measures to reduce the likelihood of illnesses occurring.
  • Article
    Full-text available
    To explore the association between in-season training load measures and injury risk in professional Rugby Union players. Methods This was a one-season prospective cohort study of 173 Professional Rugby Union players from four English Premiership teams. Training load (duration x session-RPE) and time-loss injuries were recorded for all players for all pitch and gym based sessions. Generalised estimating equations were used to model the association between in-season training load measures and injury risk in the subsequent week. Injury risk increased linearly with one-week loads and week-to-week changes in loads, with a 2 standard deviation (SD) increase in these variables (1245 AU and 1069 AU, respectively) associated with odds ratios of 1.68 (95% CI 1.05-2.68) and 1.58 (95% CI: 0.98-2.54). When compared with the reference group (<3684 AU), a significant non-linear effect was evident for four-week cumulative loads, with a likely beneficial reduction in injury risk associated with intermediate loads of 5932 to 8651 AU (OR: 0.55, 95% CI: 0.22-1.38) (this range equates to around four weeks of average in-season training load), and a likely harmful effect evident for higher loads of >8651 AU (OR: 1.39, 95% CI: 0.98-1.98). Players had an increased risk of injury if they had high one-week cumulative loads (1245 AU), or large week-to-week changes in load (1069 AU). In addition, a 'U-shaped' relationship was observed for four-week cumulative loads, with an apparent increase in risk associated with higher loads (>8651 AU). These measures should therefore be monitored to inform injury risk reduction strategies.
  • Article
    This study investigated the relationship between GPS variables measured in training and gameplay and injury occurrences in professional soccer. Nineteen professional soccer players competing in the Australian Hyundai A-League were monitored for one entire season using 5Hz Global Positioning System (GPS) units (SPI-Pro GPSports, Canberra, Australia) in training sessions and pre-season games. The measurements obtained were Total Distance, High Intensity Running Distance, Very High Intensity Running Distance, New Body Load and Metres per Minute. Non-contact soft tissue injuries were documented throughout the season. Players' seasons were averaged over one and four week blocks according to when injuries occurred. These blocks were compared to each other and to players' seasonal averages. Players performed significantly higher Metres per Minute in the weeks preceding an injury compared to their seasonal averages (+9.6 % and +7.4 % for one and four week blocks respectively) (p<0.01), indicating an increase in training and gameplay intensity leading up to injuries. Furthermore, injury blocks showed significantly lower average New Body Load compared to seasonal averages (-15.4 % and -9.0 % for one and four week blocks respectively) (p<0.01 and p=0.01). Periods of relative under-preparedness could potentially leave players unable to cope with intense bouts of high intensity efforts during competitive matches. Although limited by FIFA regulations, the results of this study isolated two variables predicting soft tissue injuries for coaches and sports scientist to consider when planning and monitoring training.
  • Article
    Hamstring strain injuries (HSIs) represent the most common cause of lost playing time in rugby union. Eccentric knee flexor weakness and between-limb imbalance in eccentric knee flexor strength are associated with a heightened risk of HSIs in other sports; however, these variables have not been explored in rugby union. To determine if lower levels of eccentric knee flexor strength or greater between-limb imbalance in this parameter during the Nordic hamstring exercise are risk factors for HSIs in rugby union. Cohort study; Level of evidence, 2. This prospective study was conducted over the 2014 Super Rugby and Queensland Rugby Union seasons. In total, 178 rugby union players (mean age, 22.6 ± 3.8 years; mean height, 185.0 ± 6.8 cm; mean weight, 96.5 ± 13.1 kg) had their eccentric knee flexor strength assessed using a custom-made device during the preseason. Reports of previous hamstring, quadriceps, groin, calf, and anterior cruciate ligament injuries were also obtained. The main outcome measure was the prospective occurrence of HSIs. Twenty players suffered at least 1 HSI during the study period. Players with a history of HSIs had a 4.1-fold (95% CI, 1.9-8.9; P = .001) greater risk of subsequent HSIs than players without such a history. Between-limb imbalance in eccentric knee flexor strength of ≥15% and ≥20% increased the risk of HSIs by 2.4-fold (95% CI, 1.1-5.5; P = .033) and 3.4-fold (95% CI, 1.5-7.6; P = .003), respectively. Lower eccentric knee flexor strength and other prior injuries were not associated with an increased risk of future HSIs. Multivariate logistic regression revealed that the risk of reinjuries was augmented in players with strength imbalances. Previous HSIs and between-limb imbalance in eccentric knee flexor strength were associated with an increased risk of future HSIs in rugby union. These results support the rationale for reducing imbalance, particularly in players who have suffered a prior HSI, to mitigate the risk of future injuries. © 2015 The Author(s).
  • Article
    To assess workload-related risk factors for injuries to particular tissue types in cricket fast bowlers. 235 fast bowlers who bowled in 14600 player innings over a period of 15 years were followed in a prospective cohort risk factor study to compare overs bowled in each match (including preceding workload patterns) and injury risk in the 3-4 weeks subsequent to the match. Injuries were categorised according to the affected tissue type as either: bone stress, tendon injuries, muscle strain or joint injuries. Workload risk factors were examined using binomial logistic regression multivariate analysis, with a forward stepwise procedure requiring a significance of <0.05. High acute match workload and high previous season workload were risk factors for tendon injuries, but high medium term (3-month workload) was protective. For bone stress injuries, high medium term workload and low career workload were risk factors. For joint injuries, high previous season and career workload were risk factors. There was little relationship between muscle injury and workload although high previous season workload was slightly protective. The level of injury risk for some tissue types varies in response to preceding fast bowling workload, with tendon injuries most affected by workload patterns. Workload planning may need to be individualised, depending on individual susceptibility to various injury types. This study supports the theory that tendons are at lowest risk with consistent workloads and susceptible to injury with sudden upgrades in workload. Gradual upgrades are recommended, particularly at the start of a bowler's career to reduce the risk of bone stress injury. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://group.bmj.com/group/rights-licensing/permissions.
  • Article
    Consistency in routines for reporting injury has been a focus of development efforts in sports epidemiology for a long time. To gain an improved understanding of current reporting practices, we applied the Injury Definitions Concept Framework (IDCF) in a review of injury reporting in a subset of the field. Meta-narrative review, An analysis of injury definitions reported in consensus statements for different sports and studies of injury epidemiology in athletics (track and field) published in PubMed between 1980 and 2013 was performed. Separate narratives for each of the three reporting contexts in the IDCF were constructed from the data. Six consensus statements and 14 studies reporting on athletics injury epidemiology fulfilled the selection criteria. The narratives on sports performance, clinical examination, and athlete self-report contexts were evenly represented in the eligible studies. The sports performance and athlete self-report narratives covered both professional and community athletes as well as training and competition settings. In the clinical examination narrative, data collection by health service professionals was linked to studies of professional athletes at international championships. From an application of the IDCF in a review of injury reporting in sports epidemiology we observed a parallel usage of reporting contexts in this field of research. The co-existence of reporting methodologies does not necessarily reflect a problematic situation, but only provided that firm precautions are taken when comparing studies performed in the different contexts. Copyright © 2014 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
  • Article
    Objectives: To compare different methods of training load (TL) quantification and their relationship to injury and illness in elite Australian footballers. Design: Prospective cohort study. Methods: Forty-five elite Australian footballers (mean±standard deviation: age=23.4±3.8 years) from one elite club participated in this 15 week pre-season study. TL was quantified every session for each individual using four different methods involving rating of perceived exertion (RPE). Two of these methods enabled the quantification of TL for all exercise modalities whilst two were applicable only to outdoor field activities. One- and two-weekly cumulative TL was investigated against injury and illness data using a logistic regression model where the low TL group was considered as the reference group. Results: A general trend existed across all TL methods which suggested lower odds of injury and illness in high TL groups. The one-week RPE (all) and one-week RPExDuration (all) methods detected reduced odds of injury in high TL groups compared to low TL groups (p<0.05, OR=0.199-0.202). Similarly, the one-week RPE (field) method identified lower illness odds in the high TL groups (p<0.05, OR=0.083-0.182). Conclusions: Higher TL appeared to provide a protective effect against both injury and illness. The inclusion of duration in the quantification of TL via RPE did not improve the ability of RPE to predict change in odds of injury or illness.